Total parameters: 1,488
| idea_id | cell_type_code | parameter | description |
|---|---|---|---|
| 094tbl12fffo | T | MEAN_CYTOPLASMIC_BASOPHILIA | MEAN_CYTOPLASMIC_BASOPHILIA is a numeric parameter that quantifies the average intensity of hematoxylin staining in the cytoplasm of tumor cells within a patch. It reflects the degree of cytoplasmic basophilia, where lower values indicate higher basophilia. This parameter is normalized through averaging and is suitable for comparing different patient cases. |
| 094tbl12fffo | T | SD_CYTOPLASMIC_BASOPHILIA | SD_CYTOPLASMIC_BASOPHILIA is a numeric parameter that measures the variability in hematoxylin staining intensity among tumor cells in a patch, computed as the standard deviation. This statistic illustrates the degree of inter-cell heterogeneity in cytoplasmic basophilia and is normalized, making it appropriate for cross-patient comparisons. |
| 09oganc9lq1b | TF | GRADIENT_FIBROBLAST_RESPONSE | GRADIENT_FIBROBLAST_RESPONSE quantifies the response of fibroblast density to changes in the tumor cell density gradient. It is computed as the slope of a linear regression where the independent variable is the magnitude of the tumor density gradient and the dependent variable is the fibroblast density. This normalized metric enables comparisons across different patient cases by measuring the rate at which fibroblast density changes relative to variations in tumor density gradients. |
| 09oganc9lq1b | TF | GRADIENT_INTERCEPT | GRADIENT_INTERCEPT represents the baseline fibroblast density when there is no tumor cell density gradient. Derived as the y-intercept from the linear regression analysis, it provides a normalized value that reflects the inherent fibroblast density independent of tumor cell gradient effects. |
| 09oganc9lq1b | TF | GRADIENT_R_SQUARED | GRADIENT_R_SQUARED is the coefficient of determination from the linear regression, indicating how well the model explains the variability in fibroblast density based on the tumor cell density gradient. As a normalized metric ranging between 0 and 1, it facilitates fair comparisons across different patches and patient cases. |
| 09oganc9lq1b | TF | MEAN_FIBROBLAST_DENSITY | MEAN_FIBROBLAST_DENSITY is the average density of fibroblasts calculated across all grid bins within a patch. This value is normalized by the bin area, ensuring that the measure is comparable between patches with different sizes or from different patient cases. |
| 09oganc9lq1b | TF | MEAN_TUMOR_GRADIENT | MEAN_TUMOR_GRADIENT represents the average magnitude of the tumor cell density gradient across all grid bins in the patch. As a value derived from normalized density differences, it enables the evaluation of tumor cell distribution variations in a way that is comparable across different regions and patients. |
| 09oganc9lq1b | TF | MAX_TUMOR_GRADIENT | MAX_TUMOR_GRADIENT captures the highest magnitude of the tumor cell density gradient observed in a patch. This parameter provides an insight into the most extreme local variations in tumor cell density, and its normalization makes it suitable for cross-patient analyses. |
| 09t5d9fkl0s6 | TPEF | MEAN_TUMOR_TO_OTHER_RATIO | MEAN_TUMOR_TO_OTHER_RATIO represents the average ratio of tumor cells to the combined number of plasma cells, eosinophils, and fibroblasts across all valid clusters (hotspots) within a patch. This measure is normalized since it is expressed as a ratio, allowing for comparisons across different patient cases. |
| 09t5d9fkl0s6 | TPEF | SD_TUMOR_TO_OTHER_RATIO | SD_TUMOR_TO_OTHER_RATIO is the standard deviation of the tumor-to-other cell ratio among valid clusters in a patch. It provides insight into the variability of the heterotypic ratios and is a normalized metric derived from the relative proportions. |
| 09t5d9fkl0s6 | TPEF | MIN_TUMOR_TO_OTHER_RATIO | MIN_TUMOR_TO_OTHER_RATIO is the smallest observed ratio of tumor cells to the combined count of plasma cells, eosinophils, and fibroblasts among all valid clusters within a patch. As a normalized ratio, it enables comparison between patches from different patient cases. |
| 09t5d9fkl0s6 | TPEF | MAX_TUMOR_TO_OTHER_RATIO | MAX_TUMOR_TO_OTHER_RATIO is the highest observed ratio of tumor cells to the combined count of plasma cells, eosinophils, and fibroblasts among valid clusters in a patch. Being a ratio, it is normalized and suitable for comparative analysis across different patient cases. |
| 0deksrl5dju0 | TNP | TNP_RATIO | TNP_RATIO is a normalized metric that quantifies the tricellular infiltration at the invasive border zone. It is calculated by multiplying the counts of tumor cells, neutrophils, and plasma cells located within the defined invasive border region, and then normalizing this product by dividing by the cube of the total number of these cells in the same area. This normalization allows for the comparison of infiltration patterns across different patient cases and patch sizes, compensating for variations in overall cellular density. |
| 0fuzp53opov4 | P | CYTOPLASM_BASOPHILIA_VARIANCE | CYTOPLASM_BASOPHILIA_VARIANCE quantifies the variability in the cytoplasmic basophilia intensity of plasma cells. This metric is derived from the variance of the pixel intensity values obtained from the hematoxylin channel after isolating the cytoplasmic region via image masks. The variance reflects the heterogeneity in staining, potentially indicating variations in cellular function or tumor behavior, and is normalized to allow comparisons across different patient patches. |
| 0fuzp53opov4 | P | CYTOPLASM_BASOPHILIA_MEAN | CYTOPLASM_BASOPHILIA_MEAN measures the average intensity of the cytoplasmic basophilia in plasma cells. It is calculated by averaging the pixel intensities within the cytoplasmic region (isolated using mask inversion) in H&E stained images after color deconvolution. This average provides a normalized, quantitative assessment of staining intensity for comparative analysis across different patient cases. |
| 0fuzp53opov4 | P | CYTOPLASM_BASOPHILIA_STD | CYTOPLASM_BASOPHILIA_STD represents the standard deviation of the cytoplasmic basophilia intensities in plasma cells. It provides a normalized measure of dispersion around the mean intensity, offering insight into the consistency or variability of staining across cells within a patch. This parameter can help in assessing the underlying cellular heterogeneity and functional status in a tumor sample. |
| 0gyycvnhqc3z | NTFM | VARIANCE_NT_RATIO | VARIANCE_NT_RATIO represents the variance of the neutrophil-to-tumor cell cytoplasmic red channel intensity ratios within a patch. This metric is calculated by first measuring the red channel (a proxy for cytoplasmic eosin staining) intensities for each neutrophil and tumor cell in regions that are rich in both fibroblasts and macrophages. For each valid neutrophil-tumor cell pair (ensuring no division errors by using only tumor cell intensities greater than zero), the ratio of neutrophil intensity to tumor cell intensity is computed. The variance of these ratios is then calculated to capture the heterogeneity in cytoplasmic staining between neutrophils and tumor cells. Since it is derived from intensity ratios and subsequently normalized by its computation across patches, it can be compared across different patient cases, making it a valid parameter for multi-case statistical analysis. |
| 0qvjnks544pk | LMPT | TUMOR_TO_IMMUNE_RATIO | The TUMOR_TO_IMMUNE_RATIO parameter represents a normalized measure that compares the count of tumor cells (defined in this context as epithelial cells within the tumor region) to the count of immune cells (specifically lymphocytes, macrophages, and plasma cells) within a given image patch. This ratio, calculated for each patch, allows for meaningful comparisons across different patient cases by mitigating the effects of varying patch sizes or absolute cell numbers, as it reflects the relative abundance of tumor cells to immune cells, thereby supporting assessments of disease aggressiveness based on local cellular composition. |
| 0slldfjtvc0p | TLPMNEF | CO_CLUSTERING_INDEX | CO_CLUSTERING_INDEX quantifies the proportion of tumor-containing clusters that also include stromal cells within a patch. It is determined by dividing the number of clusters that contain both epithelial tumor cells and stromal cells by the total number of tumor-containing clusters, resulting in a normalized value between 0 and 1. This normalization ensures the parameter is comparable across different patient cases by adjusting for variability in raw cluster counts. |
| 0yj6dlmloe3x | N | VACUOLATION_FREQUENCY | VACUOLATION_FREQUENCY is the ratio of vacuolated neutrophils to the total number of neutrophils in a patch. This normalized metric is obtained by dividing the count of neutrophils with at least one detected vacuole by the overall neutrophil count, making it suitable for comparisons between different patient cases since it represents a relative frequency rather than an absolute count. |
| 0yj6dlmloe3x | N | MEAN_VACUOLE_AREA_UM2 | MEAN_VACUOLE_AREA_UM2 represents the average area of detected vacuoles in square micrometers within a patch. It is calculated by taking the mean area of all valid vacuoles (those meeting specific size and circularity criteria) in the patch, after converting from pixel area to μm². This metric provides a normalized and numeric measure of the typical vacuole size in the analyzed regions. |
| 0yj6dlmloe3x | N | MEAN_VACUOLES_PER_CELL | MEAN_VACUOLES_PER_CELL estimates the average number of vacuoles per vacuolated neutrophil in a patch. It is computed by averaging the number of valid vacuoles detected in each neutrophil that presents vacuolation, offering a normalized measure that facilitates comparison across patches by reflecting the intensity of vacuolation independent of raw cell counts. |
| 12eyvvxtixgy | TLPMNE | COMPOSITE_SCORE | COMPOSITE_SCORE is a normalized metric that combines the z-scored infiltration and morphological irregularity values computed for each patch. It reflects the adjusted measure of both cell density and cell shape irregularity, allowing for direct comparison between different patient cases by accounting for scaling differences across patches. |
| 17olz8zlao69 | TLPMNEF | TUMOR_CLEARING_FRACTION | TUMOR_CLEARING_FRACTION represents the fraction of tumor (epithelial) cells within a patch that exhibit the intranuclear clearing phenotype. This measure is obtained by analyzing the nucleus of each tumor cell, computing the intensity gradient between its central and peripheral regions, and then normalizing by the total number of tumor cells in the patch. |
| 17olz8zlao69 | TLPMNEF | LYMPHO_CLEARING_FRACTION | LYMPHO_CLEARING_FRACTION quantifies the proportion of lymphocytes in a given patch that display the intranuclear clearing phenotype. Cells are evaluated based on their grayscale intensity distribution within the nucleus, with a normalized fraction computed from the number of lymphocytes meeting the clearing criteria relative to the total lymphocyte count. |
| 17olz8zlao69 | TLPMNEF | PLASMA_CLEARING_FRACTION | PLASMA_CLEARING_FRACTION calculates the ratio of plasma cells showing intranuclear clearing within a patch. This parameter is derived by comparing the peripheral and central nuclear intensities of plasma cells and determining the fraction of cells that exceed the predefined intensity gradient threshold. |
| 17olz8zlao69 | TLPMNEF | MACRO_CLEARING_FRACTION | MACRO_CLEARING_FRACTION is the normalized measure of macrophages in a patch that exhibit the intranuclear clearing phenotype. The assessment involves extracting the nucleus of each macrophage, analyzing its intensity gradient, and computing the fraction of macrophages classified as clearing-positive relative to the total macrophage count. |
| 17olz8zlao69 | TLPMNEF | EOSINO_CLEARING_FRACTION | EOSINO_CLEARING_FRACTION represents the proportion of eosinophils displaying intranuclear clearing in a patch. The parameter is computed by determining the intensity difference within the nucleus of each eosinophil, checking if it surpasses the set threshold, and then normalizing by the total number of eosinophils. |
| 17olz8zlao69 | TLPMNEF | FIBRO_CLEARING_FRACTION | FIBRO_CLEARING_FRACTION reflects the fraction of connective tissue cells (fibroblasts) demonstrating intranuclear clearing in a patch. It is calculated by assessing the nuclear intensity gradient for each fibroblast and normalizing the count of clearing-positive cells by the total fibroblast count in the patch. |
| 18emi7f3skx6 | LPE | MEAN_COCLUSTER_DENSITY | MEAN_COCLUSTER_DENSITY measures the average density of immune cells in valid co-clusters. Each co-cluster is defined as a group of cells, identified using clustering on spatial coordinates, that includes at least two different immune cell types. The density is calculated by dividing the number of cells in a cluster by the area of that cluster (computed using the convex hull of the cluster’s centroids), yielding a normalized metric (cells per μm²) suitable for comparing across different patches and patients. |
| 18emi7f3skx6 | LPE | MAX_COCLUSTER_DENSITY | MAX_COCLUSTER_DENSITY captures the highest cell density among the valid co-clusters within a patch. Similar to the mean density, it is determined by calculating the density of each valid cluster and then selecting the maximum value observed. This normalized metric highlights the most intense area of immune cell aggregation in the tissue. |
| 18emi7f3skx6 | LPE | PERCENT_CELLS_IN_CLUSTERS | PERCENT_CELLS_IN_CLUSTERS represents the ratio of immune cells that are part of valid co-clusters to the total number of stromal immune cells within a patch, expressed as a percentage. This provides a normalized view of the proportion of immune cells involved in co-clustering, facilitating reliable comparisons between different tumor cases. |
| 1i39mtnj4oa6 | TMP | MACRO_SPECKLE_CONTRAST_MEAN | MACRO_SPECKLE_CONTRAST_MEAN measures the average level of contrast derived from the Gray-Level Co-occurrence Matrix computed on normalized and filtered grayscale images of macrophage cell nuclei within deep invasive regions. This parameter quantitatively reflects the texture variation in nuclear speckle patterns, facilitating comparisons across different patient cases. |
| 1i39mtnj4oa6 | TMP | MACRO_SPECKLE_CONTRAST_VAR | MACRO_SPECKLE_CONTRAST_VAR captures the variability (variance) of the contrast values in macrophage nuclear speckles within deep invasive areas. It indicates how dispersed the texture measurements are among these cells, thereby providing insights into heterogeneity that is comparable across samples. |
| 1i39mtnj4oa6 | TMP | PLASMA_SPECKLE_CONTRAST_MEAN | PLASMA_SPECKLE_CONTRAST_MEAN quantifies the mean contrast of nuclear speckle patterns in plasma cells found in deep invasive patches. By using standardized image processing methods, this metric offers a normalized measure that facilitates reliable cross-case comparisons. |
| 1i39mtnj4oa6 | TMP | PLASMA_SPECKLE_CONTRAST_VAR | PLASMA_SPECKLE_CONTRAST_VAR reflects the dispersion in contrast values of plasma cell nuclear speckles. It assesses the variability in texture features, acting as a normalized indicator of the heterogeneity in nuclear speckle patterns among plasma cells. |
| 1iv4oahx25zs | TLPMNEF | TUMOR_ELONG_DISP | TUMOR_ELONG_DISP: This parameter measures the circular standard deviation of tumor cell elongation angles in radians. It quantifies the variability in the orientation (angle) of the elongated shape of tumor cells within tissue patches, facilitating comparisons between different patient cases. |
| 1iv4oahx25zs | TLPMNEF | LYMPHO_ELONG_DISP | LYMPHO_ELONG_DISP: This parameter quantifies the circular standard deviation of lymphocyte elongation angles in radians. It captures how consistently or variably lymphocyte cells are oriented in tissue patches, which is important for assessing cell shape heterogeneity. |
| 1iv4oahx25zs | TLPMNEF | PLASMA_ELONG_DISP | PLASMA_ELONG_DISP: This parameter represents the circular standard deviation of plasma cell elongation angles in radians. It reflects the dispersion in the orientation of plasma cells, serving as an indicator of cell shape variability across different patches. |
| 1iv4oahx25zs | TLPMNEF | MACRO_ELONG_DISP | MACRO_ELONG_DISP: This parameter captures the circular standard deviation of macrophage elongation angles in radians. It measures the spread of the major axis orientations of macrophage cells, providing insights into their structural heterogeneity in a normalized manner. |
| 1iv4oahx25zs | TLPMNEF | NEUTRO_ELONG_DISP | NEUTRO_ELONG_DISP: This parameter measures the circular standard deviation of neutrophil elongation angles in radians. It reflects the variation in the orientation of neutrophil cells, with values normalized for cross-case comparisons. |
| 1iv4oahx25zs | TLPMNEF | EOSINO_ELONG_DISP | EOSINO_ELONG_DISP: This parameter calculates the circular standard deviation of eosinophil elongation angles in radians. It assesses the variability in cell elongation orientations, indicating the degree of dispersion in cell shape among eosinophils in tissue patches. |
| 1iv4oahx25zs | TLPMNEF | FIBRO_ELONG_DISP | FIBRO_ELONG_DISP: This parameter assesses the circular standard deviation of fibroblast elongation angles in radians. It quantifies the variation in the orientation of the elongated axis of fibroblast cells, providing a normalized measure of cell shape variability. |
| 1my2c4o8jszj | TLPMNEF | GLOBAL_SPECKLE_INDEX | GLOBAL_SPECKLE_INDEX represents the overall average nuclear speckling variation across all cells in a given patch. It is computed by averaging the cell-level variance measurements obtained from segmented dark-staining speckles, making it a normalized metric suitable for cross-patient or cross-case comparison. |
| 1my2c4o8jszj | TLPMNEF | MEAN_SPECKLE_INDEX | MEAN_SPECKLE_INDEX is the average of the nuclear speckling indices calculated at the single-cell level within a patch. It characterizes the central tendency of the nuclear speckle variance measurements and is normalized to allow comparisons between different tumor regions. |
| 1my2c4o8jszj | TLPMNEF | SD_SPECKLE_INDEX | SD_SPECKLE_INDEX denotes the standard deviation of cell-level nuclear speckling indices within the patch, quantifying the variability of nuclear speckle intensities among cells. This numeric measure is normalized and reflects heterogeneity in the subnuclear features across cells. |
| 1my2c4o8jszj | TLPMNEF | MAX_SPECKLE_INDEX | MAX_SPECKLE_INDEX corresponds to the maximum cell-level nuclear speckling index observed in a patch, indicating the highest degree of speckling variance among the cells. It is a normalized, numeric metric that highlights the most pronounced alteration in nuclear speckle intensity within the region. |
| 1n5felm8naxy | F | FIBROBLAST_SPINDLE_INDEX_MEAN | FIBROBLAST_SPINDLE_INDEX_MEAN represents the average spindle elongation ratio calculated for fibroblasts in a specific patch. This ratio is determined by measuring the major axis and minor axis of each fibroblast nucleus (derived from the minimum rotated rectangle), then averaging these ratios across all fibroblasts. It provides a normalized, dimensionless measure of nuclear elongation indicative of fibroblast reactivity within the tumor stroma. |
| 1n5felm8naxy | F | FIBROBLAST_SPINDLE_INDEX_STD | FIBROBLAST_SPINDLE_INDEX_STD represents the standard deviation of the spindle elongation ratios computed for the fibroblasts within a patch. This measure quantifies the variability or dispersion of the elongation ratios around the mean value across fibroblasts, offering insight into the consistency of fibroblast morphology in the tumor microenvironment. |
| 1n5felm8naxy | F | FIBROBLAST_SPINDLE_INDEX_MIN | FIBROBLAST_SPINDLE_INDEX_MIN represents the minimum spindle elongation ratio observed among the fibroblasts in the patch. As a normalized dimensionless metric, it indicates the lower bound of nuclear elongation, highlighting the least elongated fibroblast nuclei within the analyzed region. |
| 1n5felm8naxy | F | FIBROBLAST_SPINDLE_INDEX_MAX | FIBROBLAST_SPINDLE_INDEX_MAX represents the maximum spindle elongation ratio observed among the fibroblasts in the patch. This parameter, being a dimensionless ratio, identifies the upper bound of nuclear elongation, reflecting the highest degree of fibroblast spindle shape observed in the tissue patch. |
| 1obt4qu2bans | TLPMNF | AVG_CLUSTERING_COEFF | AVG_CLUSTERING_COEFF measures the average local connectivity among cells in a patch by calculating the average clustering coefficient over the network graph. This coefficient, bounded between 0 and 1, indicates the likelihood that cells in contact with a given cell are also in contact with each other. Its normalization makes it comparable across different patient cases. |
| 1obt4qu2bans | TLPMNF | NETWORK_COMPLEXITY_SCORE | NETWORK_COMPLEXITY_SCORE is a composite metric that integrates the average clustering coefficient and the normalized average degree of cell interactions in the patch. The normalized average degree is obtained by scaling the average number of cell connections by the maximum possible connections. This combined score, ranging from 0 to 1, quantitatively reflects the complexity of cell contact networks, serving as an indicator for the microenvironment's role in tumor progression. |
| 1ox2jkoph1p4 | T | POLARITY_DEVIATION_SCORE | POLARITY_DEVIATION_SCORE: This metric quantifies the average angular deviation, expressed in degrees, between the primary orientation of tumor cells and a computed reference epithelial polarity axis within a tissue patch. The evaluation involves determining each cell's main axis using statistical methods on its geometric outline, computing the reference polarity axis from the spatial distribution of tumor cell centroids, and then calculating the absolute angular difference between these orientations. The angle difference is normalized into a calibrated range (0 to 90 degrees) to reflect the inherent bidirectional symmetry, making the measure comparable across different patient cases. |
| 1p45i3q5init | TLPF | MEAN_TUMOR_HEMATOXYLIN_INTENSITY | MEAN_TUMOR_HEMATOXYLIN_INTENSITY quantifies the average hematoxylin staining intensity of tumor cell nuclei within pre-selected patches that are identified as fibroblast-rich and have dense immune cell infiltration. This measurement is computed by isolating tumor cells within a standard 1000x1000 micrometer region, applying a cell-specific mask to the grayscale image to extract pixel values, and then averaging these pixel intensities. As a normalized parameter, it enables direct comparisons across different patient cases by reflecting variations in staining intensity related to tumor aggressiveness. |
| 1qmkdkato8bq | TLPMNEF | HYPOXIA_SCORE_TUMOR | HYPOXIA_SCORE_TUMOR is a normalized and numeric parameter representing the mean hypoxia-like morphology deviation score for tumor cells within a patch. The score is based on the combined binary evaluation of nuclear enlargement and faint eosin staining within the tumor compartment, with values ranging between 0 and 1. This allows for consistent comparison across different patient cases. |
| 1qmkdkato8bq | TLPMNEF | HYPOXIA_SCORE_STROMA | HYPOXIA_SCORE_STROMA is a normalized and numeric parameter representing the mean hypoxia-like morphology deviation score for stromal cells within a patch. Similar to the tumor score, it is calculated by averaging the binary scores derived from the detection of nuclear enlargement and faint eosin staining among stromal cells, resulting in values between 0 and 1. This normalization enables comparison across different patient images. |
| 1rsunedae44j | LNTF | MAX_IMMUNE_STRUCTURAL_RATIO | This parameter measures the maximum local density ratio of immune cells to structural cells within a patch. Specifically, it calculates the highest ratio, observed in any 50μm radius neighborhood, of the number of immune cells (lymphocytes and neutrophils) to the number of tumor or fibroblast cells (epithelial and connective tissue cells). The use of a ratio makes the parameter normalized, allowing for direct comparisons across different patient cases and patches, as it indicates relative cell densities rather than absolute counts. |
| 1smqxig1fjic | NE | NE_RATIO | NE_RATIO is a normalized parameter that represents the ratio of neutrophils to eosinophils within stromal regions of patches that exhibit high tumor cell density. This parameter is computed by dividing the count of neutrophils by the count of eosinophils in each qualifying patch. Because it is a ratio, it provides a normalized measure suitable for comparison across different patient cases, avoiding issues related to absolute cell count variations. |
| 210pdq3rjftr | L | MEAN_CHROMATIN_CLUMPING | MEAN_CHROMATIN_CLUMPING represents the average chromatin clumping score calculated from lymphocyte nuclei within a patch. The score is derived by analyzing the grayscale intensity histogram of nuclear pixels, detecting local peaks (which indicate areas of dense chromatin), and then normalizing the number of these peaks by the nuclear area, scaled by a constant. This normalization allows for comparable measurements across different tissue patches and patient cases. |
| 210pdq3rjftr | L | MEDIAN_CHROMATIN_CLUMPING | MEDIAN_CHROMATIN_CLUMPING indicates the median value of the chromatin clumping scores across all lymphocytes in a patch. It serves as a robust central tendency measure, essential for comparisons between patches where the distribution of scores might be asymmetric due to variations in chromatin density. |
| 210pdq3rjftr | L | SD_CHROMATIN_CLUMPING | SD_CHROMATIN_CLUMPING denotes the standard deviation of chromatin clumping scores among lymphocytes in a patch. This parameter quantifies the variability in nuclear chromatin patterns, providing insight into the heterogeneity of the cellular response within the tumor microenvironment. |
| 210pdq3rjftr | L | MIN_CHROMATIN_CLUMPING | MIN_CHROMATIN_CLUMPING identifies the minimum chromatin clumping score found in lymphocytes within a patch. This parameter helps in identifying the lower bound of chromatin density alterations, potentially reflecting areas with less pronounced nuclear changes. |
| 210pdq3rjftr | L | MAX_CHROMATIN_CLUMPING | MAX_CHROMATIN_CLUMPING reflects the maximum chromatin clumping score observed in lymphocytes of a patch. It highlights the highest degree of nuclear chromatin aggregation detected, which could correlate with areas of heightened cellular activity or abnormal chromatin organization. |
| 29px7lf5a3wx | TLPMNEF | INNER_LAYER_SLACK | INNER_LAYER_SLACK quantifies the average slack ratio within the innermost concentric layer of a tumor patch. The metric is derived by calculating the gap area to occupied area ratio for each cell type within the layer (from the tumor centroid to one-third of the maximum distance) and then taking the mean across these cell types. This normalized value facilitates comparison between different tumor regions. |
| 29px7lf5a3wx | TLPMNEF | MIDDLE_LAYER_SLACK | MIDDLE_LAYER_SLACK represents the average slack ratio for the middle concentric layer of a tumor patch. This layer spans from one-third to two-thirds of the maximum distance from the tumor centroid. The value is computed as the average of the gap area to occupied area ratios across various cell types in this region, providing a normalized measure of tissue slack that aids in clinical comparisons. |
| 29px7lf5a3wx | TLPMNEF | OUTER_LAYER_SLACK | OUTER_LAYER_SLACK measures the average slack ratio in the outermost layer of the tumor patch, defined from two-thirds of the maximum radius to the outer boundary. It is calculated by averaging the gap-to-occupied area ratios for each cell type in that layer, resulting in a normalized parameter that reflects the tissue organization at the periphery of the tumor region. |
| 29px7lf5a3wx | TLPMNEF | MEAN_SLACK_RATIO | MEAN_SLACK_RATIO is the overall slack ratio computed as the arithmetic mean of the slack ratios from all three concentric layers (inner, middle, and outer) within a tumor patch. This aggregate metric provides a single, normalized value summarizing the tissue slack across the entire tumor region, making it easier to compare different patient cases. |
| 2iewr5vwzbnt | MN | CONTACT_PROBABILITY | CONTACT_PROBABILITY represents the likelihood of direct contact between macrophages and neutrophils within a given patch of tumor tissue. It is calculated by dividing the number of contact events—where the distance between a macrophage and a neutrophil is within a defined threshold equivalent to 10 micrometers—by the total number of possible unique pairs of these cells. This normalized metric enables comparisons across different patient cases and patches, as it accounts for variations in cell density and provides a ratio rather than a raw count. |
| 2pbeobp6dvn1 | TLPMNEF | NUCLEAR_AREA_VARIANCE | NUCLEAR_AREA_VARIANCE measures the variability in nuclear areas across all cells in a patch. It quantifies how much individual nuclear area measurements differ from the mean, providing insight into cellular heterogeneity within the tumor microenvironment. |
| 2pbeobp6dvn1 | TLPMNEF | MEAN_NUCLEAR_AREA | MEAN_NUCLEAR_AREA calculates the average nuclear area of all cells within a patch, serving as an indicator of the typical nuclear size in that localized region of tumor tissue. |
| 2pbeobp6dvn1 | TLPMNEF | SD_NUCLEAR_AREA | SD_NUCLEAR_AREA represents the standard deviation of the nuclear areas in a patch, offering a measure of dispersion around the mean, which helps to assess the spread of nuclear size values. |
| 2pbeobp6dvn1 | TLPMNEF | MIN_NUCLEAR_AREA | MIN_NUCLEAR_AREA identifies the smallest nuclear area observed among the cells in a patch, which can be useful in detecting outliers or specific subpopulations with unusually small nuclei. |
| 2pbeobp6dvn1 | TLPMNEF | MAX_NUCLEAR_AREA | MAX_NUCLEAR_AREA captures the largest nuclear area found among cells in a patch, providing information about potential abnormalities or subpopulations with larger nuclei and contributing to the overall assessment of cellular heterogeneity. |
| 2txrpjgl75eu | LM | ENTROPY_MEAN | ENTROPY_MEAN is the average Shannon entropy calculated for each patch based on the local spatial mixing of lymphocytes and macrophages. This metric quantifies the degree of heterogeneity in cell-cell interactions and is normalized, making it suitable for comparing different patient cases. |
| 2txrpjgl75eu | LM | ENTROPY_STD | ENTROPY_STD represents the standard deviation of the calculated Shannon entropy values across cells within a patch. It captures the variability in spatial entropy among cells, reflecting differences in local cell arrangement, and, being derived from a normalized measure, it supports inter-patient comparisons. |
| 2txrpjgl75eu | LM | ENTROPY_MAX | ENTROPY_MAX is the maximum Shannon entropy value observed in a patch. It indicates the region where the most extensive mixing of the two cell types occurs. As a normalized metric, it allows for consistent comparisons of cell mixing dynamics across different patient samples. |
| 324jgf82vbjx | LME | LYMPHOCYTE_PCT | LYMPHOCYTE_PCT measures the percentage of lymphocytes relative to the total number of inflammatory cells (which include lymphocytes, macrophages, and eosinophils) in a tumor patch. This normalized metric allows for inter-patient comparisons of immune infiltration patterns in fibro-inflammatory bridging zones. |
| 324jgf82vbjx | LME | MACROPHAGE_PCT | MACROPHAGE_PCT represents the normalized percentage of macrophages among the total inflammatory cell population within each tumor patch. This metric is designed for consistent comparisons across different patient cases by standardizing the measure relative to the sum of key inflammatory cell counts. |
| 324jgf82vbjx | LME | EOSINOPHIL_PCT | EOSINOPHIL_PCT calculates the normalized percentage of eosinophils relative to the total inflammatory cells in a tumor patch. It provides a standardized measure for evaluating the relative presence of eosinophils in areas of fibro-inflammatory bridging, enabling meaningful comparisons between patients. |
| 32v2q5w03ghp | TFME | TUMOR_AREA_MEAN | TUMOR_AREA_MEAN represents the average area of tumor cells within a patch, quantified in square micrometers. It is computed by converting the area derived from each cell's polygon representation and then taking the average, offering insight into the typical size of tumor cells in that spatial segment. |
| 32v2q5w03ghp | TFME | TUMOR_AREA_STD | TUMOR_AREA_STD measures the variability in tumor cell sizes within a patch by calculating the standard deviation of their areas in square micrometers. This metric provides an indication of the heterogeneity in tumor cell morphology across the patch. |
| 32v2q5w03ghp | TFME | FIBRO_AREA_MEAN | FIBRO_AREA_MEAN denotes the average area of fibroblast cells within a patch, measured in square micrometers. It is calculated from the fibroblasts’ polygon-derived areas, reflecting the typical cell size in the stromal compartment. |
| 32v2q5w03ghp | TFME | FIBRO_AREA_STD | FIBRO_AREA_STD captures the standard deviation of fibroblast cell areas in a patch. It quantifies the dispersion in cell sizes among fibroblasts, thereby indicating the degree of morphological variability within the patch. |
| 32v2q5w03ghp | TFME | MACRO_AREA_MEAN | MACRO_AREA_MEAN is the average area of macrophages within a patch, expressed in square micrometers. This parameter is derived from the areas determined via the cell polygons, serving to summarize the characteristic size of macrophages in that tissue region. |
| 32v2q5w03ghp | TFME | MACRO_AREA_STD | MACRO_AREA_STD assesses the variability in macrophage cell areas within the patch by computing the standard deviation of their sizes. It helps to understand the heterogeneity in macrophage morphology across different regions of the tumor. |
| 32v2q5w03ghp | TFME | EOSINO_AREA_MEAN | EOSINO_AREA_MEAN represents the average area of eosinophil cells within a patch, measured in square micrometers. This average is derived from each cell’s area, providing a baseline for the typical size of eosinophils in the analyzed region. |
| 32v2q5w03ghp | TFME | EOSINO_AREA_STD | EOSINO_AREA_STD indicates the standard deviation of eosinophil cell areas in a patch, describing the spread or variability in these cell sizes. This metric reflects how varied the eosinophils are in terms of their morphological features. |
| 32v2q5w03ghp | TFME | DIST_TO_CORE | DIST_TO_CORE calculates the Euclidean distance, in micrometers, from the center of a patch to the tumor core. The tumor core is determined as the weighted centroid of all tumor cells over the whole-slide image. This normalized metric is used to assess spatial gradients relative to the tumor center, which may be indicative of temporal subclone infiltration. |
| 3agbviytdfon | TLMFP | AVG_GRADIENT_SLOPE | AVG_GRADIENT_SLOPE quantifies the average slope derived from the nuclear intensity gradient computed across all cells in a patch. It is determined by fitting a linear regression to the normalized H&E intensity values from the nuclear center to the boundary, providing a normalized metric that facilitates comparison across different patient cases. |
| 3agbviytdfon | TLMFP | STD_GRADIENT_SLOPE | STD_GRADIENT_SLOPE represents the variability in the nuclear intensity gradient slopes within a patch. This parameter captures the dispersion of the gradient measurements among individual cells, thereby reflecting heterogeneity in chromatin organization in the tumor microenvironment. |
| 3agbviytdfon | TLMFP | TUMOR_GRADIENT_SLOPE | TUMOR_GRADIENT_SLOPE calculates the average nuclear intensity gradient slope specifically for tumor (epithelial) cells. This measurement isolates the chromatin reorganization aspects of tumor cells, offering insights into tumor progression and biological aggressiveness. |
| 3agbviytdfon | TLMFP | LYMPHO_GRADIENT_SLOPE | LYMPHO_GRADIENT_SLOPE computes the average nuclear intensity gradient slope for lymphocytes. By focusing on these immune cells, it provides a measure of the nuclear intensity distribution that may correlate with immune response within the tumor region. |
| 3agbviytdfon | TLMFP | MACRO_GRADIENT_SLOPE | MACRO_GRADIENT_SLOPE provides the average slope of the nuclear intensity gradient for macrophages. This parameter reflects the degree of chromatin reorganization in these cells and helps elucidate their role in the tumor microenvironment. |
| 3agbviytdfon | TLMFP | FIBRO_GRADIENT_SLOPE | FIBRO_GRADIENT_SLOPE determines the average nuclear intensity gradient slope for fibroblasts. It quantifies changes in the nuclear intensity profile of connective tissue cells, which can be indicative of alterations in the tumor stroma. |
| 3agbviytdfon | TLMFP | PLASMA_GRADIENT_SLOPE | PLASMA_GRADIENT_SLOPE measures the average nuclear intensity gradient slope for plasma cells. This normalized parameter captures the subtle variations in nuclear intensity, potentially highlighting the functional state of these cells within the patch. |
| 3k6bidenj9ok | TLPMNEF | ORIENTATION_ENTROPY | The ORIENTATION_ENTROPY parameter measures the Shannon entropy of the distribution of cell orientation angles within a tissue patch. This measurement is obtained by determining the major orientation of cell nuclei using principal component analysis on the cells' polygonal representations, converting the resultant angle to a normalized range of [0, 180) degrees, and then building a normalized histogram based on a set number of bins. The entropy calculated from this histogram reflects the degree of randomness in cell directions, where higher values indicate a more disordered arrangement and lower values suggest more uniform alignment. This parameter is normalized and numeric, making it a valid measure for comparing different patient cases. |
| 3lsqtllp99st | E | MEAN_LOBE_DISTANCE_UM | MEAN_LOBE_DISTANCE_UM represents the average distance between the centroids of nuclear lobes in eosinophilic granulocytes within each tumor patch. This metric is computed by measuring the Euclidean distance between identified lobes for each eosinophil and then averaging these distances across the patch, providing a normalized comparison across different patient cases. |
| 3lsqtllp99st | E | MEDIAN_LOBE_DISTANCE_UM | MEDIAN_LOBE_DISTANCE_UM is the median value of the distances between nuclear lobes measured across eosinophils in the patch. It offers a robust central tendency metric that minimizes the influence of any outliers, ensuring consistency in inter-patient comparisons. |
| 3lsqtllp99st | E | STD_LOBE_DISTANCE_UM | STD_LOBE_DISTANCE_UM quantifies the variability or dispersion in the distances between nuclear lobes among eosinophils in the patch. This standard deviation helps to understand the heterogeneity in nuclear morphology within tumor regions, and is normalized to facilitate comparisons between cases. |
| 3lsqtllp99st | E | MIN_LOBE_DISTANCE_UM | MIN_LOBE_DISTANCE_UM indicates the smallest measured distance between the nuclear lobes of eosinophilic granulocytes in a patch, reflecting the closest proximity of lobes detected. As a normalized metric, it allows the comparison of minimal stretching or merging of lobes across different patient samples. |
| 3lsqtllp99st | E | MAX_LOBE_DISTANCE_UM | MAX_LOBE_DISTANCE_UM denotes the largest separation distance between nuclear lobes observed in the patch. This metric captures the extreme value of nuclear lobe separation and is used in a normalized manner to compare cellular morphology across varied patient cases. |
| 3udb5gk7vihj | TE | MEAN_COALIGNMENT_SCORE | MEAN_COALIGNMENT_SCORE is a normalized metric that quantifies the average co-alignment between tumor cells and eosinophils within a localized patch of tissue. It is derived by first computing the minimal angular difference between the orientation of each tumor cell and each eosinophil, then converting that difference into a co-alignment score ranging from 0 (indicating perpendicular orientation) to 1 (indicating perfect alignment). The final value is obtained by taking the mean of these scores across all valid cell pairs in the patch, ensuring comparability across different patient cases and patches. |
| 3yi5cd0asf6y | T | MEAN_HYPERCHROMASIA | MEAN_HYPERCHROMASIA represents the average grayscale intensity of tumor cell nuclei within a patch. Lower values indicate higher hyperchromasia, which is linked with more aggressive tumor phenotypes. This metric is normalized at the patch level, allowing for comparisons across different patient cases. |
| 3yi5cd0asf6y | T | STD_HYPERCHROMASIA | STD_HYPERCHROMASIA is the standard deviation of the grayscale intensities measured in tumor cell nuclei within a patch. It quantifies the variability or heterogeneity in chromatin density among the cells, and being a normalized value, it is useful for comparing nuclear intensity dispersion across patients. |
| 3yi5cd0asf6y | T | MIN_HYPERCHROMASIA | MIN_HYPERCHROMASIA indicates the minimum intensity value recorded among tumor cell nuclei in a patch. It reflects the most hyperchromatic (darkest stained) nucleus in that region and is normalized by being computed per patch, facilitating comparison between cases. |
| 3yi5cd0asf6y | T | MAX_HYPERCHROMASIA | MAX_HYPERCHROMASIA represents the maximum intensity value among tumor cell nuclei in a patch, highlighting the cell with the least degree of hyperchromasia (lightest stained). As a normalized, patch-level metric, it enables direct comparisons across different patient samples. |
| 4138zwcy1x8o | EF | MAJOR_AXIS_CORRELATION | MAJOR_AXIS_CORRELATION measures the Pearson correlation coefficient between the major axis lengths of paired fibroblast and eosinophil cells. It quantifies the degree of linear correlation between the elongated dimensions of these cells within tumor infiltration zones. As a normalized metric ranging from -1 to 1, it facilitates comparisons across different patient cases. |
| 4138zwcy1x8o | EF | MINOR_AXIS_CORRELATION | MINOR_AXIS_CORRELATION measures the Pearson correlation coefficient between the minor axis lengths of paired fibroblast and eosinophil cells. This parameter evaluates the consistency in the shorter dimensions of the cells' elliptical shapes in infiltration zones. Being normalized between -1 and 1, it is suitable for comparative analysis across various patient samples. |
| 42kyma4bof9o | N | MEAN_NUCLEAR_INTENSITY | MEAN_NUCLEAR_INTENSITY represents the average hematoxylin intensity measured from the nuclear regions of neutrophils within a patch. It captures the central tendency of the nuclear chromatin staining, allowing for meaningful comparison across different patient cases because the value is normalized by averaging over the selected nuclear pixels. |
| 42kyma4bof9o | N | SD_NUCLEAR_INTENSITY | SD_NUCLEAR_INTENSITY reflects the standard deviation of neutrophil nuclear intensities in a patch. It quantifies the variability in staining intensity among neutrophils, thereby providing insight into the heterogeneity of nuclear chromatin intensity measurements. |
| 42kyma4bof9o | N | MEDIAN_NUCLEAR_INTENSITY | MEDIAN_NUCLEAR_INTENSITY is the median value of the nuclear hematoxylin intensity for neutrophils within a patch. This robust statistic offers a center measure that is less affected by outliers, ensuring reliable comparisons among different cases. |
| 42kyma4bof9o | N | Q25_NUCLEAR_INTENSITY | Q25_NUCLEAR_INTENSITY captures the 25th percentile of the neutrophil nuclear intensity distribution within a patch. It indicates the lower bound of the staining intensity distribution and helps in understanding the distribution's skew. |
| 42kyma4bof9o | N | Q75_NUCLEAR_INTENSITY | Q75_NUCLEAR_INTENSITY represents the 75th percentile of the neutrophil nuclear intensity distribution in a patch, providing a measure of the upper bound of the intensity values. This metric assists in assessing the spread of intensity values in a normalized manner. |
| 43g3vrkosbmf | M | FUSION_PROPORTION | FUSION_PROPORTION is a normalized metric that represents the ratio of macrophages showing lysosomal fusion within a tissue patch. It is calculated by dividing the number of macrophages with detected fused lysosomal structures by the total number of macrophages in that patch. This ensures that the value can be used for comparison across different tumor regions and patient cases, making it a robust and numeric parameter for evaluating lysosomal fusion efficiency in macrophages. |
| 45mw24q36eza | TPLMF | PLASMA_EOSIN_CV | PLASMA_EOSIN_CV represents the coefficient of variation of eosin intensity measured in the cytoplasmic region of plasma cells within a 1x1 mm patch. This metric is computed as the ratio of the standard deviation to the mean of the eosin intensities extracted from the plasma cell cytoplasm. It is a normalized measure allowing comparison across different patient cases since it captures the relative variability in eosin staining, independent of the absolute values. |
| 4721jee2x6ca | TLME | CONCURRENCY_HOTSPOT_RATIO | CONCURRENCY_HOTSPOT_RATIO is a normalized metric that indicates whether a patch is identified as a concurrency hotspot. It is calculated as a binary indicator, set to 1.0 when all four cell types (tumor cells, lymphocytes, macrophages, and eosinophils) are present in numbers greater than or equal to a predefined threshold, and 0.0 otherwise. This parameter captures the spatially relevant concurrent infiltration and is normalized as it represents a relative measure, making it suitable for comparing different patient cases. |
| 4e99o588p8u9 | PE | AVG_NN_DIST_UM | AVG_NN_DIST_UM measures the average nearest neighbor distance between plasma cells and eosinophils in micrometers. This metric is computed by determining the minimum distance from each plasma cell to the nearest eosinophil and vice versa, then averaging these distances to capture the typical spatial proximity between these two cell types within an advanced infiltration zone, making it normalized across different patient cases. |
| 4e99o588p8u9 | PE | CONFIG_ENTROPY | CONFIG_ENTROPY represents the configurational (Shannon) entropy of the spatial distribution of plasma cells and eosinophils. It reflects the level of order or disorder in the spatial arrangement by dividing the patch into a fixed grid, computing the probability distribution of cells across the grid, and calculating entropy from these probabilities. This normalized metric enables the comparison of spatial organization patterns across different tumor regions. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_MEAN | NUCLEOLI_VISIBILITY_MEAN measures the average contrast score of nucleoli across all cells in a patch. The score is derived by comparing the mean intensity of candidate nucleoli regions against the surrounding nuclear intensity, making the value normalized and suitable for comparing different patient cases. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_SD | NUCLEOLI_VISIBILITY_SD represents the standard deviation of the nucleoli contrast scores across all cells in a patch. It quantifies the variability in nucleoli contrast measurements, computed as differences between nucleoli and surrounding nuclear intensities, and is normalized to enable cross-case comparisons. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_MAX | NUCLEOLI_VISIBILITY_MAX indicates the maximum contrast score observed among the nucleoli in a patch. This parameter highlights the cell with the highest difference in staining intensity between its nucleoli and the remainder of the nucleus, ensuring consistency across patches. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_MIN | NUCLEOLI_VISIBILITY_MIN corresponds to the minimum nucleoli contrast score in a patch, reflecting the least pronounced nucleoli contrast among cells. It is determined by the difference in mean intensities and remains normalized, enabling valid comparisons across samples. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_TUMOR | NUCLEOLI_VISIBILITY_TUMOR represents the average nucleoli contrast score specifically for tumor cells (epithelial cells). This measure is obtained by aggregating nucleoli contrast values from cells classified as tumor cells, making it a normalized metric useful for inter-patient analysis. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_LYMPHO | NUCLEOLI_VISIBILITY_LYMPHO computes the mean nucleoli contrast score for lymphocytes. It aggregates contrast scores from lymphocyte cells and is normalized by comparing the intensity differences, which allows for valid comparisons between different patient patches. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_PLASMA | NUCLEOLI_VISIBILITY_PLASMA measures the average nucleoli contrast score among plasma cells. The contrast is defined as the difference between the brighter nucleoli regions and their surrounding nuclear background, normalized for consistent cross-case analysis. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_MACRO | NUCLEOLI_VISIBILITY_MACRO calculates the mean nucleoli contrast score for macrophages. It reflects the intensity difference between nucleoli and surrounding regions within these cells, ensuring the value is a normalized metric for comparison. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_NEUTRO | NUCLEOLI_VISIBILITY_NEUTRO indicates the mean nucleoli contrast score for neutrophils. This parameter is derived from intensity differences between the segmented nucleoli regions and the remaining nucleus, normalized to provide consistent comparisons. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_EOSINO | NUCLEOLI_VISIBILITY_EOSINO represents the average nucleoli contrast score for eosinophils. It measures the difference in staining intensity between potential nucleoli and their surrounding nuclear regions, yielding a normalized metric suitable for cross-patient analysis. |
| 4ke9y5huroel | TLPMNEF | NUCLEOLI_VISIBILITY_FIBRO | NUCLEOLI_VISIBILITY_FIBRO computes the mean nucleoli contrast score for fibroblasts (connective cells). By assessing the intensity contrast between candidate nucleoli regions and the rest of the nucleus, this normalized parameter enables reliable comparisons among different tissue samples. |
| 4m5wd7doqbgz | TFPE | CV_INFILTRATION_THICKNESS | CV_INFILTRATION_THICKNESS quantifies the relative variability in the thickness of fibroblast and plasma cell infiltration zones around tumor clusters that have at least one eosinophil in their vicinity. This parameter is calculated as the coefficient of variation (standard deviation divided by the mean) of radial measurements taken along multiple directions from the tumor cluster’s boundary, thereby providing a normalized, dimensionless measure that facilitates comparison across different patient cases. |
| 4m5wd7doqbgz | TFPE | MEAN_INFILTRATION_THICKNESS | MEAN_INFILTRATION_THICKNESS represents the average distance of the infiltrative front of fibroblasts and plasma cells from the tumor boundary in regions where eosinophils are present. It is derived from the mean of multiple radial measurements (converted to micrometers) of the infiltration thickness, offering a numeric quantification of the typical extent of infiltration that can be compared across patient samples. |
| 4qc36pxcas88 | NE | TOTAL_COINFILTRATION_RATIO | TOTAL_COINFILTRATION_RATIO provides a normalized metric by calculating the ratio of the total number of neutrophils to the total number of eosinophils in a patch, allowing comparisons across different patient cases by reflecting the overall co-infiltration status. |
| 4qc36pxcas88 | NE | TUMOR_COINFILTRATION_RATIO | TUMOR_COINFILTRATION_RATIO is a normalized parameter that measures the balance between neutrophils and eosinophils specifically within the tumor region. It is computed by dividing the tumor-specific neutrophil count by the tumor-specific eosinophil count, yielding a comparative value for the tumor microenvironment. |
| 4qc36pxcas88 | NE | STROMA_COINFILTRATION_RATIO | STROMA_COINFILTRATION_RATIO offers a normalized measurement of the relative abundance of neutrophils versus eosinophils in the stroma region, calculated as the ratio of stromal neutrophils to stromal eosinophils, which facilitates the assessment of immune cell infiltration in the supporting tissue. |
| 4r56uk96xdt3 | TPME | CV_NUCLEAR_PERIMETER_DIFF | CV_NUCLEAR_PERIMETER_DIFF is the coefficient of variation derived from the unique pairwise absolute differences between the mean nuclear perimeters of the considered cell types. It is a unitless metric that normalizes variation, enabling comparisons across different patient cases and patches. |
| 4r56uk96xdt3 | TPME | MEAN_NUCLEAR_PERIMETER_DIFF | MEAN_NUCLEAR_PERIMETER_DIFF represents the average of the absolute differences in nuclear perimeters between the various cell types present in a patch, expressed in micrometers. This metric is calculated from intra-patch averages and permits standardized comparison of nuclear size differences across patient samples. |
| 4r56uk96xdt3 | TPME | SD_NUCLEAR_PERIMETER_DIFF | SD_NUCLEAR_PERIMETER_DIFF indicates the standard deviation of the pairwise absolute differences in nuclear perimeters between cell types within a patch, measured in micrometers. This measure of dispersion provides insight into the consistency of nuclear differences and is computed in a normalized manner to allow comparisons among different patient cases. |
| 4s70lwybxyzb | M | PHAGOCYTOSIS_INDEX | PHAGOCYTOSIS_INDEX measures the ratio of the total number of red blood cells detected within macrophages to the total number of macrophages in a patch. This normalized metric reflects the prevalence of red blood cell phagocytosis in individual patches, and allows for comparisons between different patient cases by adjusting for differences in macrophage density. |
| 4s70lwybxyzb | M | MEAN_RBC_PER_MACROPHAGE | MEAN_RBC_PER_MACROPHAGE indicates the average number of red blood cells detected per macrophage in a patch. This normalization, achieved by averaging the phagocytic counts across all macrophages, provides a clear metric for assessing the typical phagocytic activity at the cellular level within each tumor patch. |
| 4swhpkih8ntn | MF | MACROPHAGE_SKEWNESS | MACROPHAGE_SKEWNESS measures the average skewness of the curvature distribution of the nuclear boundary for macrophages found in the invasive tumor margin. This parameter is computed by first extracting the polygonal nuclear boundary, calculating the discrete curvature at sampled points along the boundary, and then using a statistical function to calculate the skewness of these curvature values for the selected macrophage cells. |
| 4swhpkih8ntn | MF | FIBROBLAST_SKEWNESS | FIBROBLAST_SKEWNESS quantifies the average skewness of nuclear boundary curvature for fibroblasts located in the invasive margin. The measure is derived by discretizing each fibroblast's nuclear boundary into points, computing the curvature at these points, and then summarizing the asymmetry of their curvature distribution via a skewness calculation. |
| 4swhpkih8ntn | MF | COMBINED_SKEWNESS | COMBINED_SKEWNESS represents a cell-count weighted average of the skewness measurements from both macrophages and fibroblasts in the invasive margin. It integrates the individual skewness values by weighting them according to the number of cells contributing to each mean, thereby providing an overall metric of nuclear boundary asymmetry across both cell types. |
| 4vco7v980tcs | M | SOLIDITY_MEAN | SOLIDITY_MEAN represents the average solidity of macrophage cells in a patch, where solidity is defined as the ratio of the cell's polygon area to its convex hull area. This metric evaluates the compactness and degree of concavity of the cell shape in a normalized manner, enabling comparisons across different patient cases. |
| 4vco7v980tcs | M | SOLIDITY_SD | SOLIDITY_SD measures the variability in the solidity values of macrophage cells within a patch. It provides insight into the dispersion of cell shape compactness values, which is useful for assessing heterogeneity in cell morphology in a normalized format. |
| 4vco7v980tcs | M | CONVEXITY_MEAN | CONVEXITY_MEAN is the average convexity of macrophage cells in a patch, defined as the ratio of the convex hull perimeter to the cell's actual perimeter. This parameter provides a normalized measure of how closely the cell shape approximates a convex form, facilitating comparisons across different samples. |
| 4vco7v980tcs | M | CONVEXITY_SD | CONVEXITY_SD quantifies the standard deviation of the convexity values for macrophage cells in a patch. It reflects the degree of variation in cell shape convexity within the patch, serving as a normalized indicator of heterogeneity in cell morphology. |
| 4x9275j83oa8 | TP | TUMOR_PLASMA_INTENSITY_DIFF | TUMOR_PLASMA_INTENSITY_DIFF represents the difference between the average H&E staining intensity of tumor cells and that of plasma cells within the epithelial compartment of a tumor patch. This metric is calculated by first determining the mean intensity for tumor cells and plasma cells separately, and then subtracting the plasma cells' mean intensity from that of the tumor cells. The resulting value is a normalized measure that can be compared across different patient cases to assess differences in cellular staining properties that may indicate distinct biological or functional states. |
| 4x9275j83oa8 | TP | MEAN_TUMOR_INTENSITY | MEAN_TUMOR_INTENSITY is a parameter that quantifies the average H&E staining intensity of tumor cells specifically within the epithelial compartment of a tumor patch. By calculating the mean intensity across all relevant tumor cells, this parameter provides a normalized measure of the staining quality and characteristics for tumor cells, which can be reliably compared between different patient samples. |
| 4x9275j83oa8 | TP | MEAN_PLASMA_INTENSITY | MEAN_PLASMA_INTENSITY measures the average H&E staining intensity for plasma cells in the epithelial compartment of a tumor patch. This parameter is computed by averaging the intensity values of plasma cells only, resulting in a normalized metric that facilitates comparison of plasma cell staining characteristics across various patient cases. |
| 4xcvlt7mybkd | T | MEAN_NUCLEOLI_COUNT | MEAN_NUCLEOLI_COUNT represents the average number of nucleoli per tumor cell within each patch. This parameter is derived by first detecting and counting the nucleoli in each tumor cell using an intensity-based segmentation method on cell-level grayscale images, followed by aggregating these counts by computing the mean across the tumor cells in a patch. The use of the average per cell normalizes the measurement, making it comparable across different patient cases. |
| 4xcvlt7mybkd | T | VAR_NUCLEOLI_COUNT | VAR_NUCLEOLI_COUNT represents the variance in nucleoli counts across tumor cells within the same patch. After obtaining the nucleoli counts for each tumor cell, the variability is quantified by calculating the variance. This metric captures the heterogeneity in nucleoli distribution among tumor cells and is normalized per patch, facilitating meaningful comparisons across different patient cases. |
| 4xe2sxdwbfom | LM | MEAN_HOTSPOT_DENSITY | MEAN_HOTSPOT_DENSITY quantifies the average density of immune cells, specifically lymphocytes and macrophages, within the stromal hotspots of a 1x1 mm patch. This value is normalized to represent the number of cells per 1000 square micrometers and is computed by averaging the densities of all identified clusters in the patch. |
| 4xe2sxdwbfom | LM | MAX_HOTSPOT_DENSITY | MAX_HOTSPOT_DENSITY represents the highest density found among the immune cell hotspots in a patch. It reflects the peak concentration of immune cells per 1000 square micrometers in the region where cells are most strongly clustered. |
| 4xe2sxdwbfom | LM | MIN_HOTSPOT_DENSITY | MIN_HOTSPOT_DENSITY indicates the lowest density of immune cells among the identified hotspots within a patch. This value, expressed as cells per 1000 square micrometers, captures the minimal concentration of immune cells in any localized cluster. |
| 4xe2sxdwbfom | LM | STD_HOTSPOT_DENSITY | STD_HOTSPOT_DENSITY measures the variability in the density values across all detected hotspots in a patch. It is the standard deviation of the densities, normalized to cells per 1000 square micrometers, and summarizes the heterogeneity in immune cell distribution. |
| 4zdvx40r0hmt | TPMFE | PLASMA_ADJ_DENSITY | PLASMA_ADJ_DENSITY is a normalized parameter that quantifies the spatial density of plasma cells adjacent to a set of target cell types in a tumor patch. It is calculated as the ratio of plasma cells that are located within a defined distance threshold from tumor cells, macrophages, fibroblasts, or eosinophils to the total number of plasma cells in the patch. This ratio allows for direct comparison across different patient cases or tumor regions by normalizing the count of adjacent plasma cells against the overall plasma cell population in each patch. |
| 50m1a00hsphp | TLPMNEF | CLUSTER_ROUNDNESS_VAR_INDEX | CLUSTER_ROUNDNESS_VAR_INDEX is a normalized metric that represents the mean of the standard deviations of cluster roundness values across all analyzed cell types. Roundness for each cluster is calculated using the formula (4 * π * Area)/(Perimeter^2), ensuring that values range from 0 to 1. This index offers a single, comparable value per patch that captures the variability in cluster shapes, thereby reflecting the heterogeneity in cell cluster morphology across different patient cases. |
| 51z9tupmru9x | T | ATYPICAL_MITOSIS_FRACTION | ATYPICAL_MITOSIS_FRACTION is a normalized metric representing the fraction of mitotic figures identified as atypical within a segmented tumor patch. It is computed by dividing the number of atypical mitoses by the total number of mitotic events in the patch, yielding a ratio that allows for consistent comparison across different patient cases and tumor regions. |
| 5874tj38id16 | TL | TUMOR_MITOSIS_INDEX | TUMOR_MITOSIS_INDEX represents the average number of mitotic figures detected per tumor cell within a patch. This parameter is normalized, as it is computed by dividing the total count of mitotic figures in tumor cells by the total number of tumor cells, making it suitable for comparing across different patient cases. |
| 5874tj38id16 | TL | LYMPHO_MITOSIS_INDEX | LYMPHO_MITOSIS_INDEX represents the average number of mitotic figures detected per lymphocyte within a patch. This normalized metric is calculated by dividing the total mitotic figure count in lymphocytes by the total number of lymphocytes, thereby enabling consistent comparisons across images. |
| 5874tj38id16 | TL | BALANCED_MITOTIC_INDEX | BALANCED_MITOTIC_INDEX is the ratio of the tumor mitosis index to the lymphocyte mitosis index. This parameter reflects the relative proliferation rates of tumor cells versus lymphocytes, offering a balanced view of cell mitotic activity in the tissue and is normalized by nature of being a ratio. |
| 5g20zfr91npg | TFNMP | SKEWNESS_FIBROBLASTS | SKEWNESS_FIBROBLASTS measures the asymmetry in the directional angle distribution of fibroblast infiltration vectors toward tumor cell clusters. It is computed by converting the directional vectors (from fibroblast positions to the centroids of nearest tumor clusters) into angles and then calculating the skewness of these angles. This normalized statistic is suited for comparing infiltration patterns across different tumor regions and patient cases. |
| 5g20zfr91npg | TFNMP | SKEWNESS_MACROPHAGES | SKEWNESS_MACROPHAGES quantifies the asymmetry in the angular distribution of macrophage infiltration vectors toward tumor cell clusters. By calculating the directional angles between macrophage positions and the centroids of nearby tumor clusters and then deriving the skewness, this parameter reflects the bias in infiltration direction. Its normalized nature makes it ideal for cross-patient comparisons in spatial analyses. |
| 5g20zfr91npg | TFNMP | SKEWNESS_PLASMA | SKEWNESS_PLASMA captures the asymmetry of the directional angle distribution of plasma cell infiltration vectors relative to tumor cell clusters. This measure, obtained from the angles defined by the vectors connecting plasma cells to their nearest tumor cluster centroids, delivers a normalized value that enables meaningful comparisons of plasma cell spatial organization across different patient cases. |
| 5gf3vz0hq3nj | TLP | TUMOR_IMMUNE_RATIO | TUMOR_IMMUNE_RATIO: A normalized numeric value calculated as the ratio of tumor cells to the total number of immune cells (lymphocytes and plasma cells) in a patch deemed an immune hotspot. This metric allows for meaningful comparisons between patient cases by providing a relative measure independent of absolute cell counts. |
| 5l3lqo2vfvz6 | FE | FIBROBLAST_MEAN_INTENSITY | FIBROBLAST_MEAN_INTENSITY represents the normalized average nuclear intensity of fibroblasts measured within the central tumor region of each patch. This parameter is computed by extracting the pixel values corresponding to the nuclei of fibroblast cells from H&E stained images and determining their mean intensity, providing a standardized metric for comparing different patient cases. |
| 5l3lqo2vfvz6 | FE | EOSINOPHIL_MEAN_INTENSITY | EOSINOPHIL_MEAN_INTENSITY represents the normalized average nuclear intensity of eosinophils measured within the central tumor region of each patch. It is derived by calculating the mean intensity of nuclear pixels for eosinophil cells, ensuring that comparisons across different tumor samples are based on normalized, quantitative measurements. |
| 5l3lqo2vfvz6 | FE | INTENSITY_DIFFERENCE | INTENSITY_DIFFERENCE is the normalized numeric difference calculated by subtracting the average eosinophil nuclear intensity from the average fibroblast nuclear intensity within the central region of a tumor patch. This parameter provides a direct comparative metric to assess the variation in nuclear intensities between the two cell types, enabling cross-case analysis. |
| 5oleffvht0r7 | T | MEAN_VACUOLATION_SCORE | MEAN_VACUOLATION_SCORE represents the average fraction of a tumor cell's cytoplasm occupied by vacuoles within a patch. It is calculated by first determining the fraction of vacuolated area in each tumor cell and then averaging these values across all tumor cells in the patch. Because it is a ratio, it is normalized and can be used to compare different patient cases. |
| 5oleffvht0r7 | T | SD_VACUOLATION_SCORE | SD_VACUOLATION_SCORE measures the variability (standard deviation) in the vacuolation fractions among tumor cells in a patch. It quantifies the heterogeneity in vacuolation patterns across cells, computed from already normalized fractions, thus allowing for reliable comparisons between different samples. |
| 5oleffvht0r7 | T | MAX_VACUOLATION_SCORE | MAX_VACUOLATION_SCORE indicates the highest vacuolation fraction recorded among tumor cells in the patch. This parameter highlights the extreme level of vacuolation found in a single cell relative to its cytoplasmic area. Being based on normalized fractions, it is suitable for comparative analysis across different patient cases. |
| 5r8jz0wlkoj3 | TMNF | NUCLEAR_SIZE_ENTROPY | NUCLEAR_SIZE_ENTROPY measures the Shannon entropy of the distribution of nuclear sizes within a patch. It is computed from a normalized probability distribution derived from a histogram of nuclear areas, and higher values indicate a higher degree of heterogeneity in cell nuclear sizes. This measure is useful for comparing different patient cases as it normalizes the variability in nuclear features. |
| 5r8jz0wlkoj3 | TMNF | MIN_NUCLEAR_AREA | MIN_NUCLEAR_AREA represents the smallest nuclear area observed among cells within a patch, expressed in square micrometers. It captures the lower bound of nuclear size within the patch and is computed consistently, making it comparable across different patient cases despite being an absolute measure. |
| 5r8jz0wlkoj3 | TMNF | MAX_NUCLEAR_AREA | MAX_NUCLEAR_AREA represents the largest nuclear area observed among cells within a patch, expressed in square micrometers. It provides the upper bound of the nuclear size variation and, as a numerical value derived under standardized conditions, can be used to compare different patient cases. |
| 5r8jz0wlkoj3 | TMNF | MEAN_NUCLEAR_AREA | MEAN_NUCLEAR_AREA is the average nuclear area computed for all detected cells within a patch, expressed in square micrometers. This parameter captures the central tendency of nuclear sizes and is calculated under standardized imaging conditions, ensuring comparability across different patient cases. |
| 5r8jz0wlkoj3 | TMNF | STD_NUCLEAR_AREA | STD_NUCLEAR_AREA quantifies the variability (standard deviation) of the nuclear areas within a patch, expressed in square micrometers. It reflects the spread or dispersion of nuclear sizes and is derived from normalized measurements, thus allowing for meaningful comparisons across patient cases. |
| 5tl5q88tbcbc | T | MEAN_NUCLEAR_IRREGULARITY | MEAN_NUCLEAR_IRREGULARITY quantifies the average proportional deviation of tumor cell nuclear perimeters from an ideal elliptical shape within a patch. This parameter is normalized since it calculates a ratio, making it comparable across different patient cases. |
| 5tl5q88tbcbc | T | MEDIAN_NUCLEAR_IRREGULARITY | MEDIAN_NUCLEAR_IRREGULARITY represents the median value of the nuclear irregularity scores within a patch. It provides a robust measure of central tendency that is normalized and numerical, ensuring comparability across various datasets. |
| 5tl5q88tbcbc | T | SD_NUCLEAR_IRREGULARITY | SD_NUCLEAR_IRREGULARITY measures the standard deviation of the nuclear irregularity scores in a patch. It captures the variability in the deviations of nuclear perimeters, providing insight into the heterogeneity of nuclear shapes in a normalized, numeric form. |
| 5tl5q88tbcbc | T | MIN_NUCLEAR_IRREGULARITY | MIN_NUCLEAR_IRREGULARITY indicates the smallest deviation value among tumor cell nuclei within a patch. It reflects the cell with the closest match to an ideal ellipse and is numerical and normalized, allowing for direct comparisons between patches. |
| 5tl5q88tbcbc | T | MAX_NUCLEAR_IRREGULARITY | MAX_NUCLEAR_IRREGULARITY indicates the highest deviation from the ideal elliptical perimeter found among tumor cell nuclei within a patch. This parameter is numerical and normalized, providing an upper bound measure of nuclear irregularity which is useful for identifying highly irregular, potentially aggressive cells. |
| 5x9n4wqjyhoe | TLPMNEF | MEAN_LOBE_COUNT_TUMOR | MEAN_LOBE_COUNT_TUMOR: Represents the average number of lobes detected in tumor (epithelial) cells per patch. This metric is computed as the mean of lobe counts for each cell in the tumor category, providing a normalized measure of cell border complexity across patches, which facilitates reliable comparisons between different patient cases. |
| 5x9n4wqjyhoe | TLPMNEF | MEAN_LOBE_COUNT_LYMPHO | MEAN_LOBE_COUNT_LYMPHO: Captures the average number of lobes identified in lymphocyte cells per patch. By averaging the detected lobes in these cells, the metric reflects localized morphological features normalized per patch and supports inter-case analysis. |
| 5x9n4wqjyhoe | TLPMNEF | MEAN_LOBE_COUNT_PLASMA | MEAN_LOBE_COUNT_PLASMA: Indicates the average lobe count in plasma cells per patch. The value summarizes the number of distinct protrusions observed on these cells, offering a normalized measurement that can be compared across various tumor regions. |
| 5x9n4wqjyhoe | TLPMNEF | MEAN_LOBE_COUNT_MACRO | MEAN_LOBE_COUNT_MACRO: Measures the mean number of lobes in macrophage cells per patch. This parameter is derived by averaging the lobation across cells in this group, resulting in a normalized and comparable metric of cell shape complexity. |
| 5x9n4wqjyhoe | TLPMNEF | MEAN_LOBE_COUNT_NEUTRO | MEAN_LOBE_COUNT_NEUTRO: Provides the average number of lobes detected in neutrophil cells per patch. It quantifies the morphological features of these immune cells in a standardized manner, enabling comparisons across patient samples. |
| 5x9n4wqjyhoe | TLPMNEF | MEAN_LOBE_COUNT_EOSINO | MEAN_LOBE_COUNT_EOSINO: Represents the average lobation observed in eosinophil cells per patch. By computing the mean number of cell border protrusions, this normalized parameter helps assess cell morphology variations across a patient cohort. |
| 5x9n4wqjyhoe | TLPMNEF | MEAN_LOBE_COUNT_FIBRO | MEAN_LOBE_COUNT_FIBRO: Reflects the average number of lobes in fibroblast (connective tissue) cells per patch. The parameter summarizes cell border irregularities and is normalized to allow consistent comparisons between different cases. |
| 5x9n4wqjyhoe | TLPMNEF | GLOBAL_VARIATION_INDEX | GLOBAL_VARIATION_INDEX: Computes the standard deviation of the average lobe counts across all specified cell types in a patch. This index quantifies the overall variation in cell lobation within a patch and acts as a normalized measure of heterogeneity across tumor regions. |
| 60agf3b1031v | TLPMNEF | COMPOSITE_COARSENESS | COMPOSITE_COARSENESS represents the aggregated mean cytoplasmic texture coarseness across all considered cell types within a tumor patch. It is calculated by averaging the computed contrast values from the Gray Level Co-Occurrence Matrix, reflecting the overall heterogeneity in cell cytoplasmic texture, with higher values indicating greater heterogeneity. |
| 60agf3b1031v | TLPMNEF | TUMOR_CELL_COARSENESS | TUMOR_CELL_COARSENESS is the average cytoplasmic texture coarseness specifically computed for tumor cells. This parameter quantifies the texture heterogeneity in the tumor cell population, which may correlate with metabolic states and tumor aggressiveness. |
| 60agf3b1031v | TLPMNEF | LYMPHOCYTE_COARSENESS | LYMPHOCYTE_COARSENESS measures the average coarseness of the cytoplasmic texture in lymphocytes. Calculated as the mean contrast from the cellular image analysis, it serves as an indicator of lymphocyte activation and potential immune response. |
| 60agf3b1031v | TLPMNEF | PLASMA_CELL_COARSENESS | PLASMA_CELL_COARSENESS quantifies the mean cytoplasmic texture coarseness for plasma cells. This metric, derived from the contrast features of the cell images, reflects aspects of antibody production activity by capturing the granularity of the cytoplasmic texture. |
| 60agf3b1031v | TLPMNEF | MACROPHAGE_COARSENESS | MACROPHAGE_COARSENESS is the average coarseness measurement for macrophages within a patch. It captures subtle variations in cytoplasmic texture that are indicative of the phagocytic activity of these cells. |
| 60agf3b1031v | TLPMNEF | NEUTROPHIL_COARSENESS | NEUTROPHIL_COARSENESS represents the mean cytoplasmic texture coarseness for neutrophils. This parameter reflects their granulation state via a contrast calculation, providing insight into their activation level and granularity. |
| 60agf3b1031v | TLPMNEF | EOSINOPHIL_COARSENESS | EOSINOPHIL_COARSENESS denotes the average coarseness of the cytoplasmic texture for eosinophils. The computed metric assesses the granule content and textural heterogeneity within these cells. |
| 60agf3b1031v | TLPMNEF | FIBROBLAST_COARSENESS | FIBROBLAST_COARSENESS is the mean coarseness value calculated for fibroblasts. This parameter evaluates the texture heterogeneity which provides indirect insight into the cells’ extracellular matrix production activity. |
| 68vixahsk988 | TLPMF | FRACTION_ELONGATED_TUMOR | FRACTION_ELONGATED_TUMOR measures the proportion of tumor nuclei, which are co-localized with nearby immune and stromal cells, that are classified as elongated based on a nuclear aspect ratio exceeding a defined threshold. This parameter is normalized as it represents a fraction, allowing comparison across different patients and patches. |
| 68vixahsk988 | TLPMF | MEAN_ASPECT_RATIO | MEAN_ASPECT_RATIO represents the average aspect ratio of the tumor cell nuclei that are in proximity to immune/stromal cells within a patch. The aspect ratio is calculated as the ratio of the maximum to the minimum dimension of each tumor nucleus, providing a normalized measure of nuclear elongation that facilitates comparisons across different cases. |
| 6asxm10225kh | TLNFM | SLOPE_TUMOR | SLOPE_TUMOR represents the normalized slope of the radial density gradient for tumor cells. It is calculated by binning the cells based on their distance from the tumor centroid, computing cell densities (cells per unit area) in each annular bin, and then fitting a linear regression model. The resulting slope, after conversion to a standardized unit, enables comparison of tumor cell infiltration patterns across different patient cases. |
| 6asxm10225kh | TLNFM | SLOPE_LYMPHOCYTE | SLOPE_LYMPHOCYTE represents the normalized slope of the radial density gradient for lymphocytes. Similar to the tumor cell parameter, lymphocyte densities are determined in concentric annular regions emanating from the tumor center, and a regression model is used to derive the gradient, which reflects changes in lymphocyte distribution within the tumor microenvironment. |
| 6asxm10225kh | TLNFM | SLOPE_FIBROBLAST | SLOPE_FIBROBLAST represents the normalized slope of the radial density gradient for fibroblasts (derived from connective tissue cells). This parameter quantifies how fibroblast density changes with radial distance from the tumor center, with the slope derived through linear regression applied to density values calculated over specified radial bins. |
| 6asxm10225kh | TLNFM | SLOPE_MACROPHAGE | SLOPE_MACROPHAGE represents the normalized slope of the radial density gradient for macrophages. It is obtained by computing macrophage densities in radial bins around the tumor center and fitting a linear regression to these values. The resulting slope, expressed in a normalized scale, facilitates the comparison of macrophage infiltration across different tumor samples. |
| 6asxm10225kh | TLNFM | R2_TUMOR | R2_TUMOR is the R-squared value obtained from the linear regression for tumor cell density gradients. This statistical metric indicates the proportion of variance in tumor cell density explained by the radial distance, thus serving as a measure of the goodness-of-fit of the regression model. |
| 6asxm10225kh | TLNFM | R2_LYMPHOCYTE | R2_LYMPHOCYTE is the R-squared value resulting from the regression analysis of lymphocyte density across radial bins. It measures how well the linear model describes the variation in lymphocyte distribution with respect to the tumor center. |
| 6asxm10225kh | TLNFM | R2_NEUTROPHIL | R2_NEUTROPHIL is the R-squared value derived from the regression model applied to neutrophil density versus radial distance. Although the corresponding slope for neutrophils is not selected for further analysis, the R2 metric remains valid and indicates the strength of the linear trend in neutrophil infiltration. |
| 6asxm10225kh | TLNFM | R2_FIBROBLAST | R2_FIBROBLAST is the R-squared value from the regression of fibroblast (connective tissue cell) density against radial distance from the tumor center. It quantifies the explanatory power of the linear model used to capture the spatial distribution of fibroblasts. |
| 6asxm10225kh | TLNFM | R2_MACROPHAGE | R2_MACROPHAGE is the R-squared value obtained from the regression analysis of macrophage density relative to radial distance. This parameter provides a measure of how well the linear model fits the spatial trend in macrophage infiltration. |
| 6bvjtpe2siin | FPT | TRIAD_AREA_FRACTION | TRIAD_AREA_FRACTION represents the proportion of a standardized 1x1 mm patch area that is occupied by spatial clusters where tumor cells, plasma cells, and fibroblasts are all present. This fraction is obtained by summing the areas of these triad regions and dividing by the total patch area, thus providing a normalized metric that facilitates comparisons across patient cases. |
| 6bvjtpe2siin | FPT | MEAN_TRIAD_SIZE | MEAN_TRIAD_SIZE is the average area of the valid triad regions identified within each patch. Although it is reported in square micrometers, the metric reflects an average figure over a uniform patch size, making it a numerically consistent parameter across different patient samples. |
| 6ex81i326gfd | TLPMNE | WEIGHTED_ECCENTRICITY_SCORE | WEIGHTED_ECCENTRICITY_SCORE represents the final aggregated score for a patch, computed by calculating the eccentricity of individual cell nuclei, grouping them by cell type, and then weighting each cell type's mean eccentricity by its relative local abundance. This normalized metric (ranging from 0 to 1) reflects the overall degree of nuclear elongation in the analyzed region of the tissue, providing a direct measure to compare different patient cases. |
| 6ex81i326gfd | TLPMNE | TUMOR_MEAN_ECCENTRICITY | TUMOR_MEAN_ECCENTRICITY measures the average nuclear eccentricity of tumor cells (coded as ‘Epithelial’ in the processing pipeline). It is calculated as the sum of the eccentricity values of tumor cell nuclei divided by their count within a patch. This parameter is normalized between 0 and 1 and gives an indication of the shape elongation specific to tumor cells in the region. |
| 6ex81i326gfd | TLPMNE | LYMPHO_MEAN_ECCENTRICITY | LYMPHO_MEAN_ECCENTRICITY denotes the mean nuclear eccentricity value for lymphocytes in the patch. It is determined by averaging the computed eccentricity values of lymphocyte nuclei, resulting in a normalized measure between 0 and 1 that serves to characterize the morphological properties of these immune cells in the tumor microenvironment. |
| 6ex81i326gfd | TLPMNE | PLASMA_MEAN_ECCENTRICITY | PLASMA_MEAN_ECCENTRICITY calculates the average nuclear eccentricity for plasma cells by averaging the eccentricity calculations for individual plasma cell nuclei within a patch. The normalized value between 0 and 1 provides insight into the geometric characteristics of plasma cells. |
| 6ex81i326gfd | TLPMNE | MACRO_MEAN_ECCENTRICITY | MACRO_MEAN_ECCENTRICITY represents the mean nuclear eccentricity for macrophages. Derived from averaging the eccentricity values for individual macrophage nuclei, this parameter is normalized between 0 and 1 and reflects the overall elongation pattern of macrophage nuclei in the analyzed patch. |
| 6ex81i326gfd | TLPMNE | NEUTRO_MEAN_ECCENTRICITY | NEUTRO_MEAN_ECCENTRICITY quantifies the average nuclear eccentricity of neutrophils by summing their individual eccentricity values and dividing by their count within the patch. This normalized measurement (ranging from 0 to 1) serves as a comparative metric for neutrophil nuclear shape across different patient cases. |
| 6ex81i326gfd | TLPMNE | EOSINO_MEAN_ECCENTRICITY | EOSINO_MEAN_ECCENTRICITY indicates the mean nuclear eccentricity of eosinophils, calculated by averaging the eccentricity of the individual eosinophil nuclei present in a patch. The parameter is normalized between 0 and 1 and assists in assessing the degree of nuclear elongation among these cells. |
| 6fjjrnvhn16o | T | BLEB_FREQUENCY | BLEB_FREQUENCY represents the average number of blebs per tumor cell within a given patch. It is calculated by dividing the total number of detected blebs in tumor cells by the number of tumor cells in the patch, ensuring a normalized metric that facilitates comparison across different patient cases. |
| 6fjjrnvhn16o | T | MEAN_BLEBS_PER_CELL | MEAN_BLEBS_PER_CELL is the arithmetic mean of bleb counts for all tumor cells in a patch. This parameter provides a normalized value reflecting the typical blebbing activity per tumor cell in the region, making it suitable for comparative analysis between patient cases. |
| 6fjjrnvhn16o | T | SD_BLEBS_PER_CELL | SD_BLEBS_PER_CELL indicates the standard deviation of bleb counts across the tumor cells in a patch. It quantifies the variability or dispersion in blebbing activity among cells, offering insight into the heterogeneity of cell behavior in a normalized manner. |
| 6fpoa2c004jk | LP | MEDIAN_LYMPHO_ROUNDNESS | MEDIAN_LYMPHO_ROUNDNESS: This parameter quantifies the median roundness of lymphocyte nuclei within the epithelial compartment of a tumor patch. The roundness is computed by considering each cell's nucleus area and perimeter, resulting in a dimensionless metric typically ranging from 0 to 1. This normalized value allows for reliable comparison across different patient cases. |
| 6fpoa2c004jk | LP | MEDIAN_PLASMA_ROUNDNESS | MEDIAN_PLASMA_ROUNDNESS: This parameter represents the median roundness of plasma cell nuclei in the epithelial compartment of a tumor patch. The metric is derived by calculating the roundness of individual plasma cells using their polygon area and perimeter, followed by taking the median value. Being dimensionless and normalized, it supports cross-patient comparative analysis. |
| 6fpoa2c004jk | LP | ROUNDNESS_DIFFERENCE | ROUNDNESS_DIFFERENCE: This parameter is the difference between the median lymphocyte roundness and the median plasma cell roundness (MEDIAN_LYMPHO_ROUNDNESS - MEDIAN_PLASMA_ROUNDNESS) within each tumor patch. It reflects the discrepancy in nuclear morphology between the two cell types, which may indicate differences in cell activation or maturation states, and is normalized for comparison across cases. |
| 6hfe6isotuym | TLMN | MEAN_ECCENTRICITY | MEAN_ECCENTRICITY quantifies the average nuclear eccentricity of tumor cells within a patch. This metric represents the normalized average shape deviation of tumor cell nuclei, computed via an ellipse fitting approach to the cell polygons, thus allowing comparisons between different patient cases. |
| 6hfe6isotuym | TLMN | SD_ECCENTRICITY | SD_ECCENTRICITY measures the standard deviation of nuclear eccentricity for tumor cells in a patch. This numeric value reflects the variability in cell nuclear shape within the examined region and is normalized, enabling meaningful inter-case comparisons. |
| 6hfe6isotuym | TLMN | COEF_VARIATION | COEF_VARIATION represents the coefficient of variation, calculated as the ratio of the standard deviation to the mean nuclear eccentricity. This normalized parameter indicates the relative dispersion of tumor cell nuclear shape deviations, making it suitable for comparing tumor heterogeneity across different patient patches. |
| 6ho393hisvgu | TLPMNEF | MEAN_DIVERGENCE_ANGLE | MEAN_DIVERGENCE_ANGLE measures the average angle between consecutive directional segments along curvilinear paths defined within each tissue patch. This parameter is derived by calculating the angle between vectors representing adjacent cell positions, yielding a normalized degree measure that reflects the overall curvature trend in a patch. |
| 6ho393hisvgu | TLPMNEF | STD_DIVERGENCE_ANGLE | STD_DIVERGENCE_ANGLE represents the standard deviation of divergence angles computed along the curvilinear paths within a patch. It quantifies the variability of the turn angles among sequential segments, providing a normalized measure of how consistent or variable the orientation changes are in the local cell arrangement. |
| 6ho393hisvgu | TLPMNEF | MAX_DIVERGENCE_ANGLE | MAX_DIVERGENCE_ANGLE captures the highest divergence angle observed among all segmented parts of the curvilinear paths within a patch. This parameter indicates the sharpest or most extreme deviation in the trajectory, expressed as a degree value, and serves as a normalized measure of the maximum turning intensity in the patch. |
| 6j3i4e4fo1wc | ME | AVG_DISTANCE_UM | AVG_DISTANCE_UM measures the average distance, in micrometers, between macrophages and eosinophils within the stromal compartment of each tissue patch. This parameter is derived by computing the pairwise Euclidean distances between the centroid coordinates of the two cell types (after converting from pixel units to micrometers) and then calculating the average of these distances. It provides a normalized measure of intercellular spacing that can be compared across different patient cases. |
| 6j3i4e4fo1wc | ME | WEIGHTED_INFILTRATION_RATIO | WEIGHTED_INFILTRATION_RATIO integrates cellular density and spatial proximity by first computing the ratio of macrophage count to eosinophil count and then weighting this ratio by the inverse of the average intercellular distance (AVG_DISTANCE_UM). This normalized metric reflects both the relative abundance of these immune cells and the closeness of their interactions within the stromal compartment, making it suitable for comparative analyses among different patient tissues. |
| 6lf4yyiuk0vi | TFME | MEAN_NUCLEAR_SHAPE_FACTOR | MEAN_NUCLEAR_SHAPE_FACTOR represents the average nuclear shape factor of tumor cells within each patch. This parameter is computed using a formula that relates the cell nucleus's area to its perimeter, yielding a dimensionless value ranging from 0 to 1, where 1 indicates a perfectly circular nucleus. It is normalized as it is an average value independent of patch size. |
| 6lf4yyiuk0vi | TFME | MEAN_FIBROBLAST_ORIENTATION | MEAN_FIBROBLAST_ORIENTATION is the average orientation angle of fibroblasts measured per patch. The angle is derived from the major axis of each fibroblast's minimum rotated rectangle, expressed in degrees (0 to 180°). This average value is comparable across patches since it standardizes the orientation measurement. |
| 6lf4yyiuk0vi | TFME | MACROPHAGE_DENSITY | MACROPHAGE_DENSITY quantifies the normalized density of macrophages within a patch. It is calculated by dividing the number of macrophages in the patch by the patch's area (with a fixed patch size), ensuring that the value is normalized and comparable across different patient cases. |
| 6lf4yyiuk0vi | TFME | EOSINOPHIL_DENSITY | EOSINOPHIL_DENSITY quantifies the density of eosinophils in each patch in a normalized manner, as it is also computed by dividing the eosinophil count by the patch area. This normalization facilitates meaningful comparisons across different images and patches. |
| 6n7kinb4sekb | F | MEAN_NUCLEOLAR_INDEX | MEAN_NUCLEOLAR_INDEX is a numeric metric that represents the average ratio of the nucleolar area to the nuclear area across all fibroblast cells within a patch. It is computed by first isolating the nuclei in fibroblast cells using specific image masks and applying noise reduction and intensity normalization. After segmenting the nucleolar regions using a thresholding technique, the ratio is calculated for each cell and then averaged across the patch. This normalization makes the index comparable across different patient cases. |
| 6n7kinb4sekb | F | MEDIAN_NUCLEOLAR_INDEX | MEDIAN_NUCLEOLAR_INDEX is a numeric metric calculated as the median of the nucleolar prominence ratios (nucleolar area divided by nuclear area) for fibroblast cells in the patch. By using the median, this parameter provides a robust central tendency measure that can reduce the influence of extreme values, ensuring that the comparison across patient cases is standardized and reliable. |
| 6n7kinb4sekb | F | SD_NUCLEOLAR_INDEX | SD_NUCLEOLAR_INDEX is a numeric metric that quantifies the variation in the nucleolar prominence index across fibroblast cells in a patch. It is computed as the standard deviation of the individual nucleolar indices, reflecting how much variation exists in the ratio of nucleolar to nuclear area within the patch. This parameter is normalized by the nature of the ratio and serves to highlight the dispersion of cellular characteristics across the analyzed patch. |
| 6waelwi5u4mj | TF | MEAN_MIN_DIST_UM | MEAN_MIN_DIST_UM represents the average of the minimum distances calculated from each tumor cell to its closest fibroblast cell, measured in micrometers. This parameter is normalized because it converts raw pixel measurements into micrometers, allowing for comparisons across different patient cases. |
| 6waelwi5u4mj | TF | MEDIAN_MIN_DIST_UM | MEDIAN_MIN_DIST_UM represents the median of the minimum distances from tumor cells to their nearest fibroblast cell, measured in micrometers. By capturing the central tendency, it provides a robust and comparable metric across different patches and cases. |
| 6waelwi5u4mj | TF | SD_MIN_DIST_UM | SD_MIN_DIST_UM represents the standard deviation of the minimum distances from tumor cells to their closest fibroblast cell, measured in micrometers. This numeric metric quantifies the variability in the spatial relationship between tumor cells and fibroblasts, making it useful for comparative analysis. |
| 6waelwi5u4mj | TF | MIN_MIN_DIST_UM | MIN_MIN_DIST_UM represents the smallest observed minimum distance among tumor cells to the nearest fibroblast cell, measured in micrometers. As a normalized numeric metric, it highlights the most intimate interaction point in the tissue microenvironment. |
| 6waelwi5u4mj | TF | MAX_MIN_DIST_UM | MAX_MIN_DIST_UM represents the largest observed minimum distance among tumor cells to the nearest fibroblast cell, measured in micrometers. This numeric and normalized parameter captures the extent of separation observed, facilitating comparisons between different tissue patches. |
| 6zakt1v3yswb | LN | COLOCALIZATION_RATIO | COLOCALIZATION_RATIO represents the proportion of stromal lymphocytes that have at least one neutrophil within a 50µm radius. This normalized metric quantifies the degree of spatial co-localization between lymphocytes and neutrophils in the stromal compartment, allowing for comparisons across different patient cases. |
| 7cnctx1v21ov | M | GIANT_CELL_PROPORTION | GIANT_CELL_PROPORTION represents the fraction of macrophages classified as giant cells based on the area threshold. It is computed as the ratio of giant cell count to the total number of macrophages in a patch, making it normalized and directly comparable across different patient cases. |
| 7cnctx1v21ov | M | AREA_THRESHOLD | AREA_THRESHOLD is the computed cutoff value (in μm²) used to classify a macrophage as a giant cell. It is determined by calculating the third quartile of macrophage cell areas in a patch and adding 1.5 times the interquartile range. This parameter provides a local statistical standard for distinguishing unusually large cells across different patches. |
| 7cnctx1v21ov | M | MEAN_CELL_AREA | MEAN_CELL_AREA is the average area (in μm²) of all macrophages within a patch. By providing a central tendency measurement for cell area across patches, it offers a numeric metric that is comparable across different patients. |
| 7cnctx1v21ov | M | MEDIAN_CELL_AREA | MEDIAN_CELL_AREA represents the middle value of the macrophage cell area distribution (in μm²) in a patch. It serves as a robust measure of central tendency, minimizing the influence of extreme values, thereby allowing for comparison between different patient cases. |
| 7cvsd6q7oukq | TLPMNEF | T_PACKING_EFFICIENCY | T_PACKING_EFFICIENCY measures the circular packing efficiency of tumor cells by comparing the observed mean inter-centroid distances of tumor cell clusters with an idealized distance derived from an optimal hexagonal packing arrangement. The value is normalized between 0 and 1, where a value closer to 1 indicates near-perfect circular packing efficiency. |
| 7cvsd6q7oukq | TLPMNEF | L_PACKING_EFFICIENCY | L_PACKING_EFFICIENCY measures the circular packing efficiency for lymphocytes using the same approach as for tumor cells. It assesses how closely the spatial distribution of lymphocytes matches an ideal hexagonal arrangement, with values normalized between 0 and 1. |
| 7cvsd6q7oukq | TLPMNEF | P_PACKING_EFFICIENCY | P_PACKING_EFFICIENCY quantifies the circular packing efficiency for plasma cells by comparing the observed distribution of plasma cell centroids with an idealized hexagonal configuration. The resulting normalized score ranges from 0 to 1, where a score closer to 1 reflects more efficient packing. |
| 7cvsd6q7oukq | TLPMNEF | M_PACKING_EFFICIENCY | M_PACKING_EFFICIENCY computes the circular packing efficiency for macrophages by evaluating the observed inter-cell distances against an ideal arrangement based on hexagonal packing principles. The efficiency score is normalized between 0 and 1, allowing for comparison across different patient cases. |
| 7cvsd6q7oukq | TLPMNEF | E_PACKING_EFFICIENCY | E_PACKING_EFFICIENCY assesses the circular packing efficiency for eosinophils by determining the ratio between the ideal hexagonal packing distance and the observed average distance among eosinophil centroids. The score is normalized between 0 and 1, indicating the degree of packing efficiency. |
| 7cvsd6q7oukq | TLPMNEF | F_PACKING_EFFICIENCY | F_PACKING_EFFICIENCY evaluates the circular packing efficiency for fibroblasts by comparing the observed spacing of fibroblast centroids with an idealized, uniform hexagonal pattern. The resulting value is normalized between 0 and 1, where higher values indicate better packing efficiency. |
| 7d725uk5is6o | P | MEAN_NUCLEAR_ELONGATION_RATIO | MEAN_NUCLEAR_ELONGATION_RATIO represents the average ratio of the longest to shortest axes of plasma cell nuclei within a specific tissue patch. This parameter is computed by determining the major and minor axis lengths of each plasma cell nucleus based on the eigenvalues derived from the covariance of the polygon coordinates and subsequently averaging these ratios across all plasma cells in the patch. |
| 7d725uk5is6o | P | STD_NUCLEAR_ELONGATION_RATIO | STD_NUCLEAR_ELONGATION_RATIO quantifies the variability of the nuclear elongation ratios among plasma cells in a given patch. It provides a measure of dispersion around the mean ratio, indicating the consistency or heterogeneity of nuclear shapes across the cells. |
| 7d725uk5is6o | P | MIN_NUCLEAR_ELONGATION_RATIO | MIN_NUCLEAR_ELONGATION_RATIO indicates the smallest observed ratio of the major to minor axis lengths of plasma cell nuclei within a patch. It reflects the minimum degree of nuclear elongation among the plasma cells analyzed in that region. |
| 7d725uk5is6o | P | MAX_NUCLEAR_ELONGATION_RATIO | MAX_NUCLEAR_ELONGATION_RATIO captures the highest observed ratio of the major to minor axis lengths of plasma cell nuclei in a patch. This parameter signifies the maximum extent of nuclear elongation, providing insight into the upper range of morphological variation among the plasma cells. |
| 7dw7hpgjq5fs | E | FOLDING_RATIO_MEAN | FOLDING_RATIO_MEAN measures the average ratio of the actual nuclear perimeter to the fitted ellipse perimeter of eosinophils within a patch. This ratio quantifies the degree of nuclear envelope folding and is normalized, enabling comparisons across different patient cases. |
| 7dw7hpgjq5fs | E | FOLDING_RATIO_SD | FOLDING_RATIO_SD represents the standard deviation of the nuclear envelope folding ratios in a patch, reflecting the variability in folding severity among eosinophils. It is a normalized metric that allows for consistent comparison across patient samples. |
| 7dw7hpgjq5fs | E | FOLDING_RATIO_MIN | FOLDING_RATIO_MIN indicates the smallest folding ratio observed in the patch, capturing the cell with the least folded nuclear envelope relative to the fitted ellipse. As a ratio, it is normalized and suitable for cross-sample analysis. |
| 7dw7hpgjq5fs | E | FOLDING_RATIO_MAX | FOLDING_RATIO_MAX denotes the largest folding ratio in the patch, highlighting the most pronounced nuclear irregularity among eosinophils when compared to the fitted ellipse. This normalized measure can be used to compare the extent of nuclear folding across different cases. |
| 7jw8betv292d | TM | TUMOR_MACROPHAGE_MIXTURE_INDEX | The TUMOR_MACROPHAGE_MIXTURE_INDEX is a normalized metric that quantifies the degree of spatial mixing between tumor cells and macrophages within a given tissue patch. It is calculated as the ratio of edges connecting tumor cells and macrophages to the edges connecting cells of the same type (either tumor-tumor or macrophage-macrophage). By normalizing the count of tumor-macrophage intersections by the number of same-type connections, this parameter facilitates a direct comparison across different patient cases, highlighting variations in local tumor microenvironments that may indicate a tumor-promoting inflammatory or immune-suppressive setting. |
| 7l7dgkhe8uy1 | TLPMNEF | RADIAL_OFFSET_MEAN | RADIAL_OFFSET_MEAN represents the mean value of the radial nucleo-cytoplasmic alignment offset for all cells in a patch after conversion into micrometers. It measures the average displacement of the nucleus relative to the cytoplasmic centroid along the direction from the tumor center to its periphery. A positive value indicates that, on average, the nucleus is displaced toward the tumor periphery. |
| 7l7dgkhe8uy1 | TLPMNEF | RADIAL_OFFSET_SD | RADIAL_OFFSET_SD is the standard deviation of the radial offset values in a patch, reported in micrometers. It quantifies the variability among the individual cell measurements of the radial alignment offset, showing the consistency of the nuclear displacements along the tumor radial direction. |
| 7l7dgkhe8uy1 | TLPMNEF | ABSOLUTE_DISPLACEMENT_MEAN | ABSOLUTE_DISPLACEMENT_MEAN is the average of the Euclidean distances between the nuclear centroid and the computed cytoplasmic centroid of cells within a patch (converted to micrometers). This metric captures the mean magnitude of nuclear-cytoplasmic displacement regardless of direction. |
| 7l7dgkhe8uy1 | TLPMNEF | ABSOLUTE_DISPLACEMENT_SD | ABSOLUTE_DISPLACEMENT_SD refers to the standard deviation of the absolute displacement measurements across cells in the patch. This value reflects the variability in the overall nuclear-cytoplasmic displacement distances among cells. |
| 7l7dgkhe8uy1 | TLPMNEF | INNER_ZONE_OFFSET_MEAN | INNER_ZONE_OFFSET_MEAN is the mean radial offset measured exclusively for cells located in the inner third portion of the patch, based on the radial distance from the tumor center. This parameter provides insight into nuclear alignment behavior in the region closest to the tumor core, with values reported in micrometers. |
| 7l7dgkhe8uy1 | TLPMNEF | MIDDLE_ZONE_OFFSET_MEAN | MIDDLE_ZONE_OFFSET_MEAN is the average radial offset for cells that are situated in the middle one-third of the patch radius from the tumor center. It measures the typical level of nuclear displacement along the radial axis for the intermediate zone of the tumor, with units in micrometers. |
| 7l7dgkhe8uy1 | TLPMNEF | OUTER_ZONE_OFFSET_MEAN | OUTER_ZONE_OFFSET_MEAN is the mean radial offset for cells found in the outer third of the patch radius. This indicator, given in micrometers, reflects the average displacement of the nucleus along the radial direction in the periphery of the tumor, highlighting variations in cellular morphology at the tumor margins. |
| 7pddaht077kg | T | BUDDING_FREQUENCY | BUDDING_FREQUENCY measures the density of tumor buds within a standardized 1x1 mm patch. It is determined by first identifying tumor cells that are near stromal cells, indicating their location at the invasive front. These cells are further grouped using a spatial clustering technique to highlight small cell clusters that meet the criteria for tumor buds. The count of these clusters is then divided by the patch area (in mm²), resulting in a normalized metric that enables comparison across different patient cases. |
| 7wm5k1vwmdw0 | TLPMNEF | TOTAL_DISCONTINUITY_SCORE | TOTAL_DISCONTINUITY_SCORE is a composite metric representing the aggregate magnitude of local density differences across a grid imposed on a tumor patch. It is calculated by summing the absolute differences in normalized cell densities between horizontally and vertically adjacent grid cells, allowing comparison across different patient cases by using density rather than raw cell counts. |
| 7wm5k1vwmdw0 | TLPMNEF | MEAN_HORIZONTAL_GRADIENT | MEAN_HORIZONTAL_GRADIENT is a numeric parameter that measures the average absolute difference in normalized cell density between horizontally adjacent grid cells. This parameter quantifies local horizontal discontinuities in cell distribution, facilitating the assessment of abrupt spatial changes in tumor architecture. |
| 7wm5k1vwmdw0 | TLPMNEF | MEAN_VERTICAL_GRADIENT | MEAN_VERTICAL_GRADIENT is a numeric parameter that captures the average absolute difference in normalized cell density between vertically adjacent grid cells. It provides insight into vertical spatial gradients within the patch, serving as a tool to detect abrupt changes in cell density distribution. |
| 7wm5k1vwmdw0 | TLPMNEF | MAX_DENSITY_GRADIENT | MAX_DENSITY_GRADIENT represents the largest single density difference observed between any two adjacent grid cells, whether horizontally or vertically. As a normalized metric, it pinpoints the most significant local change in cell density, which could indicate critical areas of tissue disruption. |
| 7xn60s48whr8 | TELMF | MEDIAN_EOSINOPHIL_GRADIENT_SD | MEDIAN_EOSINOPHIL_GRADIENT_SD represents the median value of the standard deviation computed from the pixel intensity gradients in the cytoplasmic region of eosinophils, aggregated by patch. It reflects the central tendency of the gradient variability in cell regions that have infiltrated tumor cells, lymphocytes, macrophages, and fibroblasts. |
| 7xn60s48whr8 | TELMF | MEAN_EOSINOPHIL_GRADIENT_SD | MEAN_EOSINOPHIL_GRADIENT_SD is the average standard deviation of the intensity gradients measured in the eosinophil cytoplasmic regions across cells within each patch. This parameter provides an overall quantification of gradient dispersion, which serves as a normalized metric for comparing different patches or patient cases. |
| 7xn60s48whr8 | TELMF | MAX_EOSINOPHIL_GRADIENT_SD | MAX_EOSINOPHIL_GRADIENT_SD indicates the maximum standard deviation of the cytoplasmic intensity gradients among all eosinophils in a given patch, capturing the most pronounced variability in the cytoplasmic staining intensities where immune and tumor cells co-localize. |
| 7xn60s48whr8 | TELMF | MIN_EOSINOPHIL_GRADIENT_SD | MIN_EOSINOPHIL_GRADIENT_SD denotes the minimum standard deviation calculated from the pixel intensity gradients in the cytoplasmic regions of eosinophils across a patch, representing the lowest level of gradient variability detected within the analyzed regions. |
| 80miewc127ti | TLPMNEF | PSEUDO_PROP_TUMOR | PSEUDO_PROP_TUMOR measures the fraction of tumor cells displaying nuclear pseudoinclusions. It is derived by dividing the number of tumor cells with validated pseudoinclusion features by the total tumor cell count in a tissue patch, making it a normalized and numeric indicator of tumor cellular morphology. |
| 80miewc127ti | TLPMNEF | PSEUDO_PROP_LYMPHO | PSEUDO_PROP_LYMPHO quantifies the proportion of lymphocytes that contain nuclear pseudoinclusions. This metric is calculated as the ratio of lymphocytes with detected pseudoinclusions to the total number of lymphocytes in the analyzed patch, ensuring normalization across different samples. |
| 80miewc127ti | TLPMNEF | PSEUDO_PROP_PLASMA | PSEUDO_PROP_PLASMA represents the proportion of plasma cells with nuclear pseudoinclusions. It is computed by taking the count of plasma cells confirmed to have pseudoinclusions and dividing it by the total plasma cell count within the patch, yielding a normalized and numeric parameter. |
| 80miewc127ti | TLPMNEF | PSEUDO_PROP_MACRO | PSEUDO_PROP_MACRO indicates the fraction of macrophages exhibiting nuclear pseudoinclusions. This parameter is determined by dividing the number of macrophages with pseudoinclusion traits by the overall macrophage count present in a patch, thus providing a normalized measure. |
| 80miewc127ti | TLPMNEF | PSEUDO_PROP_NEUTRO | PSEUDO_PROP_NEUTRO measures the proportion of neutrophils that contain nuclear pseudoinclusions. It is calculated as the ratio of neutrophils with pseudoinclusion detection to the total neutrophil count in the patch, offering a normalized numeric value for comparative analysis. |
| 80miewc127ti | TLPMNEF | PSEUDO_PROP_EOSINO | PSEUDO_PROP_EOSINO quantifies the fraction of eosinophils with nuclear pseudoinclusions. The value is derived from dividing the number of eosinophils showing pseudoinclusion features by the total eosinophil count in the region, resulting in a normalized parameter. |
| 80miewc127ti | TLPMNEF | PSEUDO_PROP_FIBRO | PSEUDO_PROP_FIBRO represents the proportion of fibroblast cells displaying nuclear pseudoinclusions. It is determined by the ratio of fibroblasts with identified pseudoinclusions to the overall fibroblast count in the patch, ensuring the parameter is normalized and numeric. |
| 80miewc127ti | TLPMNEF | PREVALENCE_INDEX | PREVALENCE_INDEX is a composite metric that aggregates the pseudoinclusion proportions of all cell types. Calculated as the mean of the valid pseudoinclusion proportions from tumor cells, lymphocytes, plasma cells, macrophages, neutrophils, eosinophils, and fibroblasts, it provides an overall numeric and normalized measure of nuclear pseudoinclusion prevalence within a tissue patch. |
| 893s7fc1rj9u | N | MEAN_HIGH_INTENSITY_AREA | MEAN_HIGH_INTENSITY_AREA measures the average area, expressed in square micrometers (μm²), of the high-intensity regions within the cytoplasm of neutrophils in a given patch. It is computed by first segmenting the cytoplasmic region into high and low intensity sub-compartments using intensity thresholding, and then averaging the area of the high-intensity parts across all neutrophils in the patch. |
| 893s7fc1rj9u | N | MEAN_LOW_INTENSITY_AREA | MEAN_LOW_INTENSITY_AREA represents the average area, in square micrometers (μm²), of the low-intensity regions within the cytoplasm of neutrophils per patch. This parameter is computed by identifying and segmenting the low-intensity sub-compartments after applying an intensity threshold, and then averaging these areas for the patch under analysis. |
| 893s7fc1rj9u | N | MEAN_HIGH_INTENSITY | MEAN_HIGH_INTENSITY quantifies the average pixel intensity in the high-intensity cytoplasmic regions of neutrophils in a patch. It is derived by first partitioning the cytoplasmic area into high and low intensity zones based on Otsu thresholding and then calculating the mean intensity of the pixels exceeding this threshold. |
| 893s7fc1rj9u | N | MEAN_LOW_INTENSITY | MEAN_LOW_INTENSITY calculates the average pixel intensity of the low-intensity regions within the cytoplasm of neutrophils in a patch. This is achieved by isolating the pixels below the threshold obtained via Otsu's method and computing their average intensity. |
| 893s7fc1rj9u | N | MEAN_HIGH_INTENSITY_VAR | MEAN_HIGH_INTENSITY_VAR measures the average variance of pixel intensities in the high-intensity regions of the cytoplasm. By quantifying the variation in intensity values within these regions for each neutrophil and averaging them across the patch, it provides a measure of heterogeneity in the high-intensity compartments. |
| 893s7fc1rj9u | N | MEAN_LOW_INTENSITY_VAR | MEAN_LOW_INTENSITY_VAR assesses the average variance of pixel intensities in the low-intensity regions of the cytoplasm. It is calculated by determining the variability of intensities in the low-intensity areas after segmentation and averaging these variances for the entire patch. |
| 893s7fc1rj9u | N | MEAN_HIGH_LOW_RATIO | MEAN_HIGH_LOW_RATIO denotes the average proportion of the cytoplasmic area occupied by high-intensity regions relative to the total cytoplasmic area. It is computed as the ratio of the high-intensity area to the sum of high- and low-intensity areas, averaged over all neutrophils in the patch, offering a normalized metric for comparing different patches. |
| 8budzj62q1fh | TLPMNEF | MEAN_COLOR_DIFF | MEAN_COLOR_DIFF: This parameter quantifies the average absolute difference in cytoplasmic color intensity between neighboring cell pairs within a defined patch. It is computed by taking the mean of the absolute differences across all cell pairs detected within a fixed Euclidean distance, ensuring the measure is normalized for comparative analysis across patches and patient cases. |
| 8budzj62q1fh | TLPMNEF | MEDIAN_COLOR_DIFF | MEDIAN_COLOR_DIFF: This parameter represents the median of the absolute differences in cytoplasmic color intensity between neighboring cell pairs. It offers a robust, normalized measure of central tendency that reduces the influence of extreme values, allowing for reliable comparisons across different tumor regions. |
| 8budzj62q1fh | TLPMNEF | STD_COLOR_DIFF | STD_COLOR_DIFF: This parameter calculates the standard deviation of the absolute differences in cytoplasmic color intensity among neighboring cell pairs. It reflects the variability or dispersion of these differences, providing insight into the consistency of cell-to-cell interactions within a patch in a normalized and comparable manner. |
| 8budzj62q1fh | TLPMNEF | MAX_COLOR_DIFF | MAX_COLOR_DIFF: This parameter records the maximum absolute difference in cytoplasmic color intensity observed between any pair of neighboring cells. It captures the extreme difference within the patch, serving as a normalized indicator of the upper bound of heterogeneity in cell color differences. |
| 8fbvh6bef9ro | TMNEF | MEAN_FIBROBLAST_RING_THICKNESS_UM | MEAN_FIBROBLAST_RING_THICKNESS_UM measures the average thickness in micrometers of the fibroblast ring surrounding multi-cellular hotspots. These hotspots, defined by clusters of tumor cells, macrophages, neutrophils, and eosinophils, are identified by spatial clustering and their boundaries approximated via convex hulls. The mean thickness is computed by averaging the minimal distances of nearby fibroblast cells (located in the stromal tissue) from the cluster boundary, and then converting that distance to micrometers. |
| 8fbvh6bef9ro | TMNEF | MIN_RING_THICKNESS_UM | MIN_RING_THICKNESS_UM represents the smallest fibroblast ring thickness in micrometers among all identified hotspots in a patch. It captures the minimal distance from the fibroblast cells to the corresponding hotspot boundary, reflecting the lowest observed stromal barrier across the analyzed region. |
| 8fbvh6bef9ro | TMNEF | MAX_RING_THICKNESS_UM | MAX_RING_THICKNESS_UM indicates the largest fibroblast ring thickness in micrometers among the hotspots within a patch. It reflects the widest distance that fibroblasts occupy surrounding any of the multi-cell hotspots, thereby providing insight into the maximum local stromal response. |
| 8fbvh6bef9ro | TMNEF | SD_RING_THICKNESS_UM | SD_RING_THICKNESS_UM quantifies the standard deviation in micrometers of the fibroblast ring thickness measurements across different hotspots in a patch. This parameter reveals the variability and heterogeneity in fibroblast infiltration around the localized cell clusters. |
| 8futdknzv3j2 | E | VACUOLATION_FREQUENCY | VACUOLATION_FREQUENCY is a normalized, numeric parameter that measures the proportion of eosinophils exhibiting vacuolation within a given tissue patch. It is computed by dividing the number of eosinophils with detected vacuolar regions by the total number of eosinophils in that patch, enabling comparisons across different patient cases. |
| 8k6wr72mo91k | PL | PLASMA_GRADIENT_RED | PLASMA_GRADIENT_RED: This parameter represents the mean radial red color gradient for plasma cells. It is calculated as the difference between the average red channel intensity in the outer cytoplasmic region and the average intensity in the inner cytoplasmic zone (closer to the nucleus), aggregated over all plasma cells within a given patch. Its normalized nature allows for comparison between different patient cases. |
| 8k6wr72mo91k | PL | PLASMA_GRADIENT_GREEN | PLASMA_GRADIENT_GREEN: This parameter represents the mean radial green color gradient for plasma cells. It is determined by computing the difference between the mean green channel intensity in the outer and inner cytoplasmic regions, providing a normalized measure that facilitates cross-patient comparison. |
| 8k6wr72mo91k | PL | PLASMA_GRADIENT_BLUE | PLASMA_GRADIENT_BLUE: This parameter quantifies the mean radial blue color gradient for plasma cells. It is derived by subtracting the mean blue channel intensity of the inner cytoplasmic area from that of the outer zone, resulting in a normalized metric suitable for comparative analysis across patient cases. |
| 8k6wr72mo91k | PL | LYMPHO_GRADIENT_RED | LYMPHO_GRADIENT_RED: This parameter captures the mean radial red color gradient for lymphocytes. It is computed as the difference between the average red channel intensity in the outer cytoplasmic region and that in the inner region of lymphocytes, offering a normalized measurement for comparing different patient cases. |
| 8k6wr72mo91k | PL | LYMPHO_GRADIENT_GREEN | LYMPHO_GRADIENT_GREEN: This parameter denotes the mean radial green color gradient for lymphocytes. By calculating the difference between the green channel intensities of the outer and inner cytoplasmic zones, it provides a normalized value that can be compared across various patient samples. |
| 8k6wr72mo91k | PL | LYMPHO_GRADIENT_BLUE | LYMPHO_GRADIENT_BLUE: This parameter measures the mean radial blue color gradient for lymphocytes. It is obtained by subtracting the mean blue intensity in the inner cytoplasmic region from that in the outer region, yielding a normalized gradient suitable for inter-patient analyses. |
| 8k6wr72mo91k | PL | GRADIENT_DIFF_RED | GRADIENT_DIFF_RED: This parameter reflects the difference in the red channel gradients between plasma cells and lymphocytes. It is calculated by subtracting the mean red gradient of lymphocytes from that of plasma cells, thereby providing a normalized measure of differential cytoplasmic color behavior between the two cell types. |
| 8k6wr72mo91k | PL | GRADIENT_DIFF_GREEN | GRADIENT_DIFF_GREEN: This parameter indicates the difference in the green channel gradients between plasma cells and lymphocytes. It is derived from the subtraction of the lymphocyte green gradient from the plasma cell green gradient, offering a normalized metric for assessing cellular differences. |
| 8k6wr72mo91k | PL | GRADIENT_DIFF_BLUE | GRADIENT_DIFF_BLUE: This parameter represents the difference in the blue channel gradients between plasma cells and lymphocytes. It is computed by subtracting the mean blue gradient of lymphocytes from that of plasma cells, yielding a normalized value that highlights the differential cytoplasmic color gradient between the two groups of cells. |
| 8ki4aha857qf | LE | TEXTURE_CORRELATION | TEXTURE_CORRELATION represents the Pearson correlation coefficient computed between paired texture features extracted from individual lymphocytes and eosinophils in the tumor-stroma transition zone. The texture features are derived from nuclear images using a Gray Level Co-occurrence Matrix-based analysis, and the resulting correlation reflects the degree of similarity between the two types of cells. Since this parameter is normalized and based on a statistical measure, it can be used to compare different patient cases reliably. |
| 8rl6mo0npsku | ML | STROMAL_MACROPHAGE_DENSITY | STROMAL_MACROPHAGE_DENSITY quantifies the density of macrophages within the stromal region of a tumor patch. It measures the number of macrophages normalized by the stroma area (converted into mm²), allowing for robust comparisons across different patient cases regardless of variations in patch size. |
| 8rl6mo0npsku | ML | INTRATUMORAL_LYMPHOCYTE_DENSITY | INTRATUMORAL_LYMPHOCYTE_DENSITY measures the density of lymphocytes within the tumor compartment. It is computed by dividing the count of intratumoral lymphocytes by the tumor area in mm², ensuring that comparisons between different cases are valid and meaningful. |
| 8rl6mo0npsku | ML | COMBINED_INFILTRATION_SCORE | COMBINED_INFILTRATION_SCORE is a composite metric that integrates both the stromal macrophage density and intratumoral lymphocyte density. Calculated as a weighted average (equal weighting of 0.5 for each density), this parameter provides a single, normalized score that reflects the combined immune cell infiltration patterns in the tumor microenvironment. |
| 8vs75ivrtn1e | LPN | LYMPHO_BRIGHTNESS | LYMPHO_BRIGHTNESS represents the mean brightness of lymphocyte cells within a patch. This value is obtained by converting the cell's color image to a grayscale brightness measure using a luminosity formula and then averaging the brightness values of all pixels corresponding to the lymphocyte cells, providing a normalized metric that can be compared across different patient cases. |
| 8vs75ivrtn1e | LPN | PLASMA_BRIGHTNESS | PLASMA_BRIGHTNESS measures the mean brightness of plasma cells within the patch. The brightness for each plasma cell is calculated using a similar grayscale conversion and masking approach as used for lymphocytes, and the mean value is computed, ensuring that the metric is normalized and numeric. |
| 8vs75ivrtn1e | LPN | NEUTRO_BRIGHTNESS | NEUTRO_BRIGHTNESS indicates the average brightness of neutrophil cells in the patch. By applying a standardized conversion from RGB to grayscale and averaging over the identified neutrophil cells, this parameter yields a normalized, numeric measure that facilitates inter-case comparisons. |
| 8vs75ivrtn1e | LPN | LYMPHO_PLASMA_RATIO | LYMPHO_PLASMA_RATIO is the ratio of the mean brightness of lymphocytes to the mean brightness of plasma cells within a patch. This ratio is computed by dividing the lymphocyte brightness by the plasma brightness, offering a normalized comparative metric that highlights differences in staining brightness between these two cell types. |
| 8vs75ivrtn1e | LPN | LYMPHO_NEUTRO_RATIO | LYMPHO_NEUTRO_RATIO gives the ratio of lymphocyte brightness to neutrophil brightness. It is calculated by dividing the mean brightness of lymphocytes by that of neutrophils, providing a normalized and numeric parameter that helps in identifying the relative staining differences between these cell types. |
| 8vs75ivrtn1e | LPN | PLASMA_NEUTRO_RATIO | PLASMA_NEUTRO_RATIO denotes the ratio of plasma cell brightness to neutrophil brightness. This parameter is computed by dividing the mean brightness of plasma cells by the mean brightness of neutrophils, yielding a normalized and numeric measure to compare the staining intensities of these immune cells. |
| 8vs75ivrtn1e | LPN | TRICELL_BRIGHTNESS_RATIO | TRICELL_BRIGHTNESS_RATIO is a composite ratio that compares the brightness of lymphocytes to the geometric mean of plasma and neutrophil brightness values. This parameter succinctly encapsulates the relative staining brightness among the three cell types and is normalized and numeric, making it suitable for further comparative analysis across different patient cases. |
| 92s60c891loy | MN | CO_OCCURRENCE_SCORE | CO_OCCURRENCE_SCORE is a normalized metric ranging from 0 to 1 that quantifies the degree of co-occurrence between macrophages and neutrophils within the stromal regions of tumor patches. It is calculated by counting the pairs of these cells that are closer than a defined spatial threshold and then normalizing this count by dividing by the maximum number of either cell type present in the patch. This normalization ensures that the metric is comparable across different cases, independent of variations in absolute cell counts. |
| 9ipxjvg9f0bk | T | CANNIBALISM_INDEX | CANNIBALISM_INDEX measures the normalized number of tumor cells showing cannibalistic events per 1000 tumor cells in a given patch. This index is calculated by dividing the number of tumor cells with additional dark nuclear structures (beyond the primary nucleus) by the total number of tumor cells and then multiplying by 1000. It serves as a normalized metric, allowing for fair comparison across patches and different patient cases by accounting for variations in tumor cell density. |
| 9ul3s96b7tnl | MF | AVG_CIRCULARITY_MACROPHAGES | AVG_CIRCULARITY_MACROPHAGES measures the average nuclear circularity of macrophages within the stroma for each patch. It is calculated from cell segmentation data using the formula that relates the area and perimeter of the cell nucleus, yielding a normalized, dimensionless value (ranging approximately between 0 and 1) that facilitates comparisons across different tumor samples. |
| 9ul3s96b7tnl | MF | AVG_CIRCULARITY_FIBROBLASTS | AVG_CIRCULARITY_FIBROBLASTS measures the mean nuclear circularity of fibroblasts (identified as connective tissue cells) in the stroma per patch. The value is obtained by computing the nuclear shape using the standard circularity formula and averaging over all fibroblast cells in each patch, ensuring normalization for inter-case comparisons. |
| 9ul3s96b7tnl | MF | CIRCULARITY_DIFFERENCE | CIRCULARITY_DIFFERENCE represents the difference in mean nuclear circularity between macrophages and fibroblasts within each patch. It is calculated by subtracting the average circularity of fibroblasts from that of macrophages, thereby quantifying the disparity in nuclear shape between the two cell types in a normalized, dimensionless form. |
| 9zc93d20j4to | TLPMNEF | MEAN_TENSION | MEAN_TENSION: Represents the average membrane tension score computed across all cells within a patch. This metric is derived by summing the hotspot curvature gradients identified on individual cell membranes, then averaging these values. It offers a normalized measure that facilitates comparison of mechanical stress across different tumor regions and patient cases. |
| 9zc93d20j4to | TLPMNEF | MAX_TENSION | MAX_TENSION: Denotes the maximum membrane tension score observed in a patch. It captures the highest value of localized membrane stress caused by extreme curvature variations, serving as an indicator of the most pronounced mechanical deformation within the tissue. |
| 9zc93d20j4to | TLPMNEF | STD_TENSION | STD_TENSION: Provides the standard deviation of the membrane tension scores in the patch. By quantifying the variability of these scores, it highlights the heterogeneity in mechanical stress among cells and reflects differences in localized membrane curvature within the patch. |
| 9zc93d20j4to | TLPMNEF | L_CELL_TENSION | L_CELL_TENSION: Indicates the mean tension score for Lymphocytes within a patch. This parameter is calculated by averaging the tension scores of cells classified as Lymphocytes, offering a normalized measurement of the membrane stress these immune cells experience. |
| 9zc93d20j4to | TLPMNEF | P_CELL_TENSION | P_CELL_TENSION: Represents the mean tension score for Plasma cells in the patch. This value is obtained by averaging the hotspot-derived tension scores for Plasma cells, thereby providing a standardized insight into their mechanical stress profile. |
| 9zc93d20j4to | TLPMNEF | M_CELL_TENSION | M_CELL_TENSION: Reflects the mean tension score for Macrophages, averaging the membrane tension derived from curvature gradients for cells classified as Macrophages. This normalized measure helps in evaluating the mechanical stress specific to these immune cells. |
| 9zc93d20j4to | TLPMNEF | N_CELL_TENSION | N_CELL_TENSION: Denotes the mean membrane tension score for Neutrophils within a patch. It is computed by averaging the tension scores of Neutrophils, offering a standardized metric for assessing their localized mechanical stress. |
| 9zc93d20j4to | TLPMNEF | E_CELL_TENSION | E_CELL_TENSION: Indicates the average membrane tension score for Epithelial cells in the patch. Derived from the curvature-based tension calculations, this parameter gives a normalized measure of the mechanical stress experienced by Epithelial cells due to local membrane deformations. |
| 9zc93d20j4to | TLPMNEF | F_CELL_TENSION | F_CELL_TENSION: Provides the mean tension score for Fibroblasts, calculated by averaging the hotspot-derived tension values for cells classified as Fibroblasts. This normalized metric reflects the level of mechanical stress in the extracellular matrix-associated fibroblasts. |
| a0nmxb7j7r5k | TNFE | MEDIAN_NEUTROPHIL_FIBROBLAST_NUCLEAR_ASPECT_RATIO | This parameter quantifies the relative elongation of neutrophil nuclei compared to fibroblast nuclei within stromal regions that contain both tumor cells and eosinophils. It is derived by computing the nuclear aspect ratio (the ratio of the longer to the shorter side of the minimum rotated rectangle) for each cell and then calculating the median value for both neutrophils and fibroblasts within each patch. The final ratio, obtained by dividing the median neutrophil aspect ratio by the median fibroblast aspect ratio, provides a normalized metric that enables comparison across different patient samples, thereby reflecting differences in cellular morphology associated with tumor progression. |
| a1r4lnoz372o | TF | TUMOR_MEAN_ASPECT_RATIO | TUMOR_MEAN_ASPECT_RATIO represents the average nucleus shape of tumor cells at the epithelial-stromal boundary. The metric is derived by isolating tumor cells at this interface, computing each cell's nucleus aspect ratio using the dimensions of its minimum rotated rectangle, and then averaging these ratios over all qualifying tumor cells. This normalized, unitless value enables reliable comparisons across different patient cases. |
| a1r4lnoz372o | TF | FIBRO_MEAN_ASPECT_RATIO | FIBRO_MEAN_ASPECT_RATIO represents the average nucleus shape of fibroblast cells located at the tumor-stroma interface. It is calculated by selecting fibroblast cells in close proximity to tumor cells, computing the aspect ratio of each cell’s nucleus from its minimum rotated rectangle, and averaging the results. The resulting normalized metric facilitates standardized comparisons across various patient samples. |
| a1r4lnoz372o | TF | ASPECT_RATIO_DIFFERENCE | ASPECT_RATIO_DIFFERENCE is the numerical difference between the mean nucleus aspect ratios of tumor cells and fibroblast cells at the epithelial-stromal boundary. It is calculated by subtracting FIBRO_MEAN_ASPECT_RATIO from TUMOR_MEAN_ASPECT_RATIO. This parameter provides a normalized indicator of the relative morphological contrast between the two cell types, making it valuable for comparative analyses across patients. |
| a69txy8bjd67 | P | CLOCKFACE_UNIFORMITY_SCORE | CLOCKFACE_UNIFORMITY_SCORE quantifies the deviation of the plasma cell’s nuclear chromatin intensity distribution from an ideal uniform clock-face pattern. It is computed by dividing the standard deviation of the dark chromatin intensity values across 12 sectors by their mean intensity, thereby normalizing the variability. Lower values indicate a pattern that closely adheres to the classic clock-face distribution. |
| a69txy8bjd67 | P | CLOCKFACE_MEAN_INTENSITY | CLOCKFACE_MEAN_INTENSITY represents the average intensity of the dark chromatin regions across the 12 angular sectors of the plasma cell nucleus. This parameter provides an overall measure of chromatin density or darkness within the nucleus, and being an average metric, it is normalized to allow comparisons across different patches and patient samples. |
| a69txy8bjd67 | P | CLOCKFACE_SECTOR_STDEV | CLOCKFACE_SECTOR_STDEV measures the variability in dark chromatin intensities across the 12 sectors within the nucleus. It reflects how uniformly the chromatin is distributed, with higher values indicating more variability. As a standard deviation computed within the normalized context of cell sectors, it qualifies for comparative analysis across various cases. |
| a7k4x8wxmdm9 | MFTN | MEDIAN_MACRO_FIBRO_EOSIN_RATIO | This parameter represents the median ratio of cytoplasmic eosin intensity between macrophages and fibroblasts within a 1x1 mm patch of tumor tissue that is enriched for both tumor cells and neutrophils. The median ratio is computed by first calculating the mean red channel intensity of the eosin stain for each macrophage and fibroblast within the patch, then forming all possible valid pairs (skipping pairs where the fibroblast intensity is zero) and finally deriving the median of these ratios. This normalized metric allows for comparison across different patient cases by providing a standardized measure of localized immune-stromal interactions. |
| acfa3zns4k3b | ELNT | MEAN_INFILTRATION_DISTANCE_UM | The parameter 'MEAN_INFILTRATION_DISTANCE_UM' quantifies the average distance in micrometers between tumor cells and immune cells (lymphocytes, eosinophils, and neutrophils) within each analyzed image patch. It is computed by first determining the Euclidean distances between the centroids of tumor cells and those of the immune cells, then converting these distances from pixels to micrometers using a conversion factor, and finally averaging the distances over all tumor-immune cell pairs. This numeric measure is normalized in the sense that it presents a standardized spatial metric, making it comparable across different patient cases. Lower average values indicate that immune cells are in closer proximity to tumor cells, which may suggest a more active or favorable immune surveillance within the tumor microenvironment. |
| agc9cmi0wbtb | TN | MEAN_MIN_DIST_UM | MEAN_MIN_DIST_UM represents the calculated average of the minimum distances from each neutrophil to its nearest tumor cell within a patch. This distance, converted from pixel units to micrometers, provides a normalized measure of spatial proximity between neutrophils and tumor cells, allowing for comparisons across different tumor patches and patient cases. |
| agc9cmi0wbtb | TN | MEDIAN_MIN_DIST_UM | MEDIAN_MIN_DIST_UM represents the median value of the minimum distances from each neutrophil to its closest tumor cell within a patch. By summarizing the distribution of individual cell-to-cell distances, this parameter offers a robust, normalized metric for spatial interactions that is useful for comparing local microenvironment characteristics across different patient cases. |
| aoo3l9l96i47 | N | NET_FORMATION_INDEX | NET_FORMATION_INDEX is a normalized, dimensionless metric calculated as the ratio of the total area of detected NET-like structures to the total tumor area within an image patch. It is derived by first calculating the NET area via image processing techniques on neutrophil image patches (after applying filters and morphological operations), and then normalizing this value by the tumor area obtained from the tissue segmentation mask. This allows for robust comparison across different patient cases, as the index adjusts for variations in absolute tumor size. |
| apektun7edg5 | TLPMNEF | BLEB_FREQ_TUMOR | BLEB_FREQ_TUMOR measures the average number of membrane blebs per cell in regions classified as tumor. It quantifies the blebbing activity in tumor cells by aggregating individual cell bleb counts within a patch and normalizing by the number of tumor cells. |
| apektun7edg5 | TLPMNEF | BLEB_FREQ_STROMA | BLEB_FREQ_STROMA quantifies the average number of membrane blebs per cell in stromal regions. This parameter provides insight into the blebbing behavior of cells in the surrounding non-tumor tissue, normalized by the cell count in those areas. |
| apektun7edg5 | TLPMNEF | BLEB_FREQ_T_CELLS | BLEB_FREQ_T_CELLS represents the average bleb count per epithelial cell, specifically targeting the tumor cell subset. It is calculated by averaging the bleb counts from cells identified as epithelial in patch-level analysis. |
| apektun7edg5 | TLPMNEF | BLEB_FREQ_L_CELLS | BLEB_FREQ_L_CELLS calculates the average number of membrane blebs per lymphocyte. This parameter helps evaluate the level of blebbing in lymphocytes that are part of the immune response within the tumor microenvironment. |
| apektun7edg5 | TLPMNEF | BLEB_FREQ_P_CELLS | BLEB_FREQ_P_CELLS indicates the average number of blebs per plasma cell by averaging the bleb counts for cells categorized as plasma cells. It is used to analyze the blebbing behavior in these specific immune cells. |
| apektun7edg5 | TLPMNEF | BLEB_FREQ_M_CELLS | BLEB_FREQ_M_CELLS measures the average number of blebs per macrophage. It reflects the behavior of macrophages in the tumor environment through a normalized average of individual bleb counts. |
| apektun7edg5 | TLPMNEF | BLEB_FREQ_N_CELLS | BLEB_FREQ_N_CELLS represents the average bleb frequency in neutrophils. This parameter is derived by averaging the count of blebs from neutrophils, providing a normalized metric indicative of their activity. |
| apektun7edg5 | TLPMNEF | BLEB_FREQ_E_CELLS | BLEB_FREQ_E_CELLS quantifies the average number of membrane blebs per eosinophil. It gives insight into the blebbing activity of eosinophils by computing the mean bleb count for these cells. |
| apektun7edg5 | TLPMNEF | BLEB_FREQ_F_CELLS | BLEB_FREQ_F_CELLS assesses the average number of blebs per fibroblast (or connective cell) in a patch. It is determined by averaging the individual bleb counts of fibroblasts, serving as a normalized measure of their blebbing behavior. |
| avcsilmdnxa2 | LPMN | MEAN_LYMPHO_ROUNDNESS | MEAN_LYMPHO_ROUNDNESS is a normalized metric that represents the average nuclear roundness of lymphocytes in immune-rich regions of tumor patches. This parameter is derived by calculating the roundness for each lymphocyte based on the area and perimeter properties of its nucleus, where a perfect circle would yield a value of 1. The resulting mean value provides a normalized measure that can be used to compare different patient cases without being influenced by the number of cells present. |
| b9mdtdnze212 | FE | FIBROBLAST_EOSINOPHIL_RATIO | FIBROBLAST_EOSINOPHIL_RATIO represents the normalized ratio of fibroblasts to eosinophils measured in the 50µm peri-tumoral stroma region. The ratio is computed by dividing the number of fibroblasts by the number of eosinophils present within the peri-tumoral area of each patch extracted from whole-slide images. This parameter is used to compare cellular distributions among different patient cases, and a missing value (or NA) is assigned if no eosinophils are detected to handle cases of division by zero. |
| bb0rcp9nz60b | F | SCALLOPING_FREQ_MEAN | SCALLOPING_FREQ_MEAN measures the average number of significant indentations (membrane scallops) per fibroblast within a patch. This parameter is derived by analyzing each fibroblast cell in the patch, detecting their membrane indentations based on a defined minimal area threshold, and then computing the arithmetic mean of these counts. It is normalized because it reflects a per-cell average, enabling comparison across different patient cases and tumor regions. |
| bb0rcp9nz60b | F | SCALLOPING_FREQ_SD | SCALLOPING_FREQ_SD represents the standard deviation of the number of membrane indentations per fibroblast in the patch. It quantifies the variability in the scalloping frequency among the fibroblast cells, providing insight into the distribution and consistency of this feature within individual patches. Being a measure of dispersion computed on a per-cell basis ensures that this parameter is normalized and suitable for cross-case comparisons. |
| bb40vn0ypt7n | TLMEF | MEAN_CORRIDOR_CURVATURE_SD | MEAN_CORRIDOR_CURVATURE_SD represents the average standard deviation of the curvature computed along each lymphocyte corridor detected within a patch. This metric quantifies how variable the bending of the lymphocyte pathway is across the patch. By averaging these curvature variations, it normalizes the measurement, allowing for direct comparison between different patient cases and regions. |
| bb40vn0ypt7n | TLMEF | MAX_CORRIDOR_CURVATURE_SD | MAX_CORRIDOR_CURVATURE_SD captures the highest standard deviation of curvature among all lymphocyte corridors in a patch. This parameter reflects the corridor that exhibits the greatest variability in its smoothness, indicating potentially complex or irregular infiltration patterns. Its normalized nature makes it suitable for comparing tumor region properties across different patients. |
| blu695zg2zhe | TLPMEF | MEDIAN_EOSINOPHIL_DIST_UM | MEDIAN_EOSINOPHIL_DIST_UM measures the median distance, expressed in micrometers, between eosinophils and the tumor boundary within a standardized patch. The tumor boundary is defined by constructing a convex hull of tumor cell centroids, and distances are calculated from each eosinophil to the nearest point on this boundary. This parameter provides a normalized metric to compare spatial cell distributions across different patient cases. |
| blu695zg2zhe | TLPMEF | MEDIAN_FIBROBLAST_DIST_UM | MEDIAN_FIBROBLAST_DIST_UM represents the median distance, in micrometers, from fibroblasts to the tumor boundary within a given patch. Similar to the eosinophil distance, the measurement is computed using the convex hull of tumor cell centroids as the boundary reference. This normalized spatial measurement supports comparisons across various tumor regions. |
| blu695zg2zhe | TLPMEF | RATIO_EOSIN_FIBRO_DIST | RATIO_EOSIN_FIBRO_DIST is the ratio of the median eosinophil distance to the median fibroblast distance from the tumor boundary in a patch. This parameter normalizes the spatial relationship between the two cell types, providing a unitless measure that facilitates direct comparison between different patient cases. |
| bn6z5yjtblnc | TLPMNEF | RD_PATTERN_INDEX | RD_PATTERN_INDEX represents the overall reaction-diffusion pattern similarity index calculated for each patch. It is derived by subtracting the averaged mean squared errors of the seven cell types from 1, normalizing the measure to a range between 0 and 1. A higher value indicates a better match between the observed spatial density maps and those simulated using reaction-diffusion dynamics. |
| bn6z5yjtblnc | TLPMNEF | RD_ERROR_EPITHELIAL | RD_ERROR_EPITHELIAL quantifies the discrepancy between observed and simulated density maps for Epithelial cells by computing the mean squared error. Since the density maps are normalized over each patch, this parameter provides a standardized metric to assess how well the spatial distribution of Epithelial cells is captured by the simulation. |
| bn6z5yjtblnc | TLPMNEF | RD_ERROR_LYMPHOCYTE | RD_ERROR_LYMPHOCYTE measures the mean squared error between the observed and simulated density maps of Lymphocyte cells. This normalized error score helps evaluate the degree to which the reaction-diffusion simulation successfully replicates the true spatial patterning of these cells. |
| bn6z5yjtblnc | TLPMNEF | RD_ERROR_PLASMA | RD_ERROR_PLASMA assesses the mean squared error between the observed and simulated spatial density maps for Plasma cells. It offers a numerical measure of the accuracy of the simulation in reproducing the normalized spatial distribution of Plasma cells within each tissue patch. |
| bn6z5yjtblnc | TLPMNEF | RD_ERROR_NEUTROPHIL | RD_ERROR_NEUTROPHIL represents the mean squared error calculated for Neutrophil cells by comparing their observed and simulated density maps. This metric, derived from normalized density distributions, indicates how accurately the simulation mimics the spatial dynamics of Neutrophil cells. |
| bn6z5yjtblnc | TLPMNEF | RD_ERROR_EOSINOPHIL | RD_ERROR_EOSINOPHIL quantifies the error for Eosinophil cells by computing the mean squared difference between the observed and simulated density maps. This normalized numerical metric reflects the fidelity of the reaction-diffusion model in capturing the spatial patterns of Eosinophil cells. |
| bn6z5yjtblnc | TLPMNEF | RD_ERROR_MACROPHAGE | RD_ERROR_MACROPHAGE provides a measure of the discrepancy between the observed and simulated density maps for Macrophage cells in the form of mean squared error. The normalized nature of the density maps ensures that this parameter can be reliably compared across different patient patches. |
| bn6z5yjtblnc | TLPMNEF | RD_ERROR_CONNECTIVE | RD_ERROR_CONNECTIVE calculates the mean squared error between the observed and simulated density maps for Connective cells. This error metric, based on normalized density maps, objectively evaluates the simulation’s performance in replicating the spatial organization of Connective tissue cells. |
| bp3ntrumhx3t | TFLME | PINCH_POINT_PROPORTION | PINCH_POINT_PROPORTION is a normalized metric that indicates the density of multi-cell contact zones within each tissue patch. It is calculated as the ratio of the number of valid cross-cell pinch point clusters (where a cluster is defined by the presence of at least two different cell types in close contact) to the total number of relevant cells (tumor cells, fibroblasts, lymphocytes, macrophages, and eosinophils). This parameter is numeric and standardized, allowing for effective comparisons across different patient cases and tumor regions. |
| c525ut90puqm | TLPMNEF | AGGREGATION_PROBABILITY | The AGGREGATION_PROBABILITY parameter measures the proportion of cells in a patch that are part of cell clusters meeting specific criteria: the clusters must contain at least one eosinophil and have a minimum total of three cells. This ratio is calculated by dividing the total number of cells in such qualifying clusters by the total cell count within the patch. Because this parameter expresses a proportion, it normalizes the data and allows for meaningful comparisons across different patient cases and tumor regions. |
| c9a1cel0ln1s | TLPMNEF | HYPERCHROM_FRAC_TUMOR | HYPERCHROM_FRAC_TUMOR measures the normalized fraction of tumor (epithelial) cells in a patch that have hyperchromatic nuclei. This is determined by comparing the average nuclear grayscale intensity to a calibrated threshold, allowing comparison across different patient cases. |
| c9a1cel0ln1s | TLPMNEF | HYPERCHROM_FRAC_LYMPHO | HYPERCHROM_FRAC_LYMPHO represents the normalized proportion of lymphocytes within a patch that exhibit hyperchromatic nuclei by dividing the number of lymphocytes with mean nuclear intensity below the threshold by the total lymphocyte count. |
| c9a1cel0ln1s | TLPMNEF | HYPERCHROM_FRAC_PLASMA | HYPERCHROM_FRAC_PLASMA indicates the fraction of plasma cells with hyperchromatic nuclei in a patch. It is computed as the ratio of hyperchromatic plasma cells to all plasma cells, making it a normalized metric suitable for cross-case analysis. |
| c9a1cel0ln1s | TLPMNEF | HYPERCHROM_FRAC_MACRO | HYPERCHROM_FRAC_MACRO quantifies the normalized fraction of macrophages with hyperchromatic nuclei. It compares the number of hyperchromatic macrophages to the total macrophage count, enabling consistent inter-patient comparisons. |
| c9a1cel0ln1s | TLPMNEF | HYPERCHROM_FRAC_NEUTRO | HYPERCHROM_FRAC_NEUTRO calculates the fraction of neutrophils with hyperchromatic nuclei in a patch by dividing the hyperchromatic neutrophil count by the total neutrophil count, providing a normalized measurement. |
| c9a1cel0ln1s | TLPMNEF | HYPERCHROM_FRAC_EOSINO | HYPERCHROM_FRAC_EOSINO captures the normalized proportion of eosinophils exhibiting hyperchromatic nuclei by taking the ratio of eosinophils with low mean nuclear intensity to the overall eosinophil count. |
| c9a1cel0ln1s | TLPMNEF | HYPERCHROM_FRAC_FIBRO | HYPERCHROM_FRAC_FIBRO measures the fraction of fibroblasts (derived from connective tissue cells) with hyperchromatic nuclei in a patch, calculated as the number of hyperchromatic fibroblasts divided by the total fibroblast count, ensuring standardized values across different cases. |
| cavupnbujvz0 | TLPMNEF | GRANULOCYTE_DENSITY | GRANULOCYTE_DENSITY represents the normalized density (cells per mm²) of granulocytes, specifically neutrophils and eosinophils, found within the tumor invasive margin. This metric is obtained by calculating the number of granulocyte cells located in the defined margin region and then dividing by the area of that region, enabling standardized comparisons across different patches and patient cases. |
| cavupnbujvz0 | TLPMNEF | STROMAL_DENSITY | STROMAL_DENSITY measures the normalized density (cells per mm²) of stromal cells, including lymphocytes, plasma cells, macrophages, and connective tissue cells, within the tumor invasive margin. It is computed by counting the eligible stromal cells present in the invasive margin and normalizing this count by the margin's area, thereby allowing comparisons between different regions and patient samples. |
| cavupnbujvz0 | TLPMNEF | DENSITY_DIFFERENCE | DENSITY_DIFFERENCE is the difference between the granulocyte density and the stromal cell density within the invasive margin (cells per mm²). This parameter quantifies the relative prevalence of granulocytes over other stromal cells, serving as an indicator of the local inflammatory environment and facilitating patient-to-patient comparisons. |
| cavupnbujvz0 | TLPMNEF | MARGIN_AREA | MARGIN_AREA quantifies the area of the invasive margin region (in mm²) within each patch. It is determined by identifying the margin through tissue segmentation and converting the pixel count of the margin into millimeters squared. Although not a direct density metric, its standardized measurement is critical for normalizing the cell count data used in the other parameters. |
| cdn7usdxh93a | TMP | TRIPLET_DENSITY | TRIPLET_DENSITY represents the count of unique tumor-macrophage-plasma cell triplets normalized by the total number of cells analyzed in the patch. This parameter is designed to provide a relative measure of triplet occurrence by accounting for variations in cell numbers across different patches, enabling meaningful comparisons between patient cases. |
| cdn7usdxh93a | TMP | NORMALIZED_TRIPLET_INDEX | NORMALIZED_TRIPLET_INDEX is calculated by dividing the number of unique tumor-macrophage-plasma cell triplets by the theoretical maximum number of such triplets (the product of the counts of tumor cells, macrophages, and plasma cells). This normalization ensures that the measure reflects the degree of triplet-based cell adjacency in a way that is independent of the absolute cell counts, making it suitable for comparative analysis across samples. |
| ci2yirhytzio | ELMT | FRACTION_CLUSTERED | FRACTION_CLUSTERED represents the fraction of eosinophils within a given patch that are clustered with at least one tumor cell, one lymphocyte, and one macrophage within a 50μm radius. This normalized metric is calculated by dividing the number of 'clustered' eosinophils by the total number of eosinophils, ensuring that comparisons across different patient cases and patches are valid and reflective of the microenvironment's cellular interactions. |
| cm84sr2kmet5 | MP | MEDIAN_ROUNDNESS_DIFF | MEDIAN_ROUNDNESS_DIFF is a numeric parameter that represents the difference between the median nuclear roundness of macrophages and that of plasma cells, calculated within intratumoral regions of a tumor. Nuclear roundness, computed using a geometric formula based on cell polygon area and perimeter, yields values between 0 and 1, with 1 representing perfect circularity. This metric captures variations in shape that may reflect different cellular activation states, and because both contributing medians are derived from a normalized scale (0 to 1), the final difference value can be reliably compared across different patient cases. |
| cncgcmauoxfn | T | PYKNOTIC_FRACTION | PYKNOTIC_FRACTION represents the proportion of tumor cells within a patch that exhibit pyknotic features, determined by identifying nuclei with dark, condensed chromatin. This parameter is normalized by dividing the number of tumor cells classified as pyknotic by the total number of tumor cells, allowing for fair comparisons across different patient cases. |
| cncgcmauoxfn | T | MEAN_NUCLEUS_INTENSITY | MEAN_NUCLEUS_INTENSITY reflects the average intensity of the nuclear regions of tumor cells within a patch. It is calculated by averaging the pixel intensities in the nuclear regions, which are isolated using a specific mask. This normalized measure provides insight into the staining characteristics or chromatin condensation of the nuclei, and it enables cross-patient comparisons. |
| coamj9uz943s | L | APOPTOTIC_RATE | APOPTOTIC_RATE is a normalized, numeric metric representing the proportion of apoptotic lymphocytes within a given tumor patch. It is calculated by dividing the number of lymphocytes showing apoptotic features—detected through nuclear fragmentation analysis—by the total lymphocyte count in that patch. This ratio allows for direct and comparable analysis across different patient cases and tissue regions. |
| cpzakh6nv1ya | TLF | FIBROBLAST_PROXIMITY_INDEX_MEAN | FIBROBLAST_PROXIMITY_INDEX_MEAN represents the average of the fibroblast proximity index values calculated across all fibroblast cells in a patch. This index is computed by assessing the distances from a fibroblast's centroid to the nearest tumor cell and lymphocyte, combining these distances using an inverse summation approach. A higher mean indicates that fibroblasts are, on average, positioned closer and more equally between tumor cells and lymphocytes, suggesting a potential regulatory role in intercellular interactions. |
| cpzakh6nv1ya | TLF | FIBROBLAST_PROXIMITY_INDEX_STD | FIBROBLAST_PROXIMITY_INDEX_STD quantifies the standard deviation of the fibroblast proximity index values within a patch. It reflects the variability or heterogeneity of fibroblast positioning relative to tumor cells and lymphocytes. A higher standard deviation indicates a wider range of spatial relationships among fibroblasts, pointing to possible differences in regional microenvironments within the patch. |
| cpzakh6nv1ya | TLF | FIBROBLAST_PROXIMITY_INDEX_MEDIAN | FIBROBLAST_PROXIMITY_INDEX_MEDIAN measures the central tendency of the fibroblast proximity index values in a patch, using the median to reduce the influence of outlier values. A higher median suggests that at least half of the fibroblast cells are positioned in a spatial arrangement that is favorable for influencing interactions between tumor cells and lymphocytes. |
| cr15xd0yr39d | F | FIBROBLAST_DENSITY_GRADIENT | FIBROBLAST_DENSITY_GRADIENT measures the rate of change in fibroblast density as a function of distance from the tumor boundary. It is derived by applying linear regression to fibroblast densities computed within successive concentric annular zones. By converting the slope from a pixel-based measure to micrometers, it provides a normalized metric that facilitates comparisons across different patient cases. Negative values indicate a decline in density with increasing distance from the tumor. |
| cr15xd0yr39d | F | MEAN_FIBROBLAST_DENSITY | MEAN_FIBROBLAST_DENSITY represents the average fibroblast density across all the concentric zones around the tumor boundary. For each zone, the density is calculated as the number of fibroblast cells divided by the area of the zone. This parameter is normalized by area, making it suitable for comparing local stromal characteristics across different tissue patches. |
| cr15xd0yr39d | F | MAX_FIBROBLAST_DENSITY | MAX_FIBROBLAST_DENSITY indicates the highest observed fibroblast density among the analyzed concentric zones. Similar to the mean density, each zone’s density is calculated by normalizing the fibroblast count to the zone's area. This parameter highlights localized regions with peak fibroblast concentration and is normalized for comparative analysis. |
| crxqjsghlna4 | TLNE | FUSION_RATIO | FUSION_RATIO measures the degree of spatial overlap between hotspots formed by different cell types within a tissue patch. It is computed as the ratio of the combined area where hotspots of at least two different cell types intersect to the total hotspot area derived from all the cell types. The ratio is normalized between 0 and 1—where 0 indicates no intersection and 1 indicates complete overlap—which allows for effective comparisons between different patient cases and tumor regions. This parameter is numeric, derived from spatial and density calculations, and it reflects complex immune-tumor interactions that could be linked to prognosis variations. |
| cso3gntgtfnn | NE | REPULSION_INDEX | REPULSION_INDEX: This parameter is a normalized ratio calculated by dividing the observed median distance between neutrophils and eosinophils by the mean median distance derived from multiple random simulations. It quantifies the degree of spatial repulsion between these immune cell types, with values greater than 1 indicating a repulsive interaction. Its normalization ensures that it can be reliably compared across different patient cases and tumor patches. |
| cso3gntgtfnn | NE | OBSERVED_MEDIAN_DIST_UM | OBSERVED_MEDIAN_DIST_UM: This metric represents the median distance in micrometers between neutrophil-eosinophil pairs as measured directly from the spatial coordinates obtained in a tissue patch. The value is normalized by being derived within a standardized patch setting, making it applicable for cross-patient and cross-region comparisons in the analysis of immune cell spatial interactions. |
| cso3gntgtfnn | NE | EXPECTED_MEDIAN_DIST_UM | EXPECTED_MEDIAN_DIST_UM: This parameter measures the mean of the median distances, in micrometers, calculated from a series of simulations that generate random spatial distributions of the cells within a patch. It serves as a baseline for what random cell distribution would produce and is normalized by the underlying patch dimensions, allowing for valid comparisons across different cases. |
| ctk5c7hwx2d2 | TLPMNEF | TUMOR_AXIS_RATIO_IQR | TUMOR_AXIS_RATIO_IQR measures the interquartile range of the major-to-minor axis ratios for tumor cells within a given patch. This metric quantifies the variability in the cell shape, indicating the range between the 25th and 75th percentile values, and is computed by comparing eigenvalue-derived ratios from principal component analysis applied to extracted cell nuclei polygons. |
| ctk5c7hwx2d2 | TLPMNEF | LYMPHO_AXIS_RATIO_IQR | LYMPHO_AXIS_RATIO_IQR measures the interquartile range of the major-to-minor axis ratios for lymphocytes in each patch. This parameter reflects the spread of cell shape variability among lymphocytes, offering insight into morphological heterogeneity within the sampled tumor region. |
| ctk5c7hwx2d2 | TLPMNEF | PLASMA_AXIS_RATIO_IQR | PLASMA_AXIS_RATIO_IQR quantifies the interquartile range of the major-to-minor axis ratios for plasma cells in the patch. It captures the distribution spread of these ratios, ensuring that variability in cell structure for plasma cells is considered in a normalized way across different samples. |
| ctk5c7hwx2d2 | TLPMNEF | MACRO_AXIS_RATIO_IQR | MACRO_AXIS_RATIO_IQR measures the interquartile range of the major-to-minor axis ratios for macrophages. This parameter is calculated by determining the statistical spread (between the 25th and 75th percentiles) of the axis ratios derived from the cell shape analysis, ensuring comparability across different patient cases. |
| ctk5c7hwx2d2 | TLPMNEF | NEUTRO_AXIS_RATIO_IQR | NEUTRO_AXIS_RATIO_IQR quantifies the interquartile range of the major-to-minor axis ratios for neutrophils in a patch. This normalized metric reflects the variability in cell morphology by assessing the distribution range of the computed axis ratios among neutrophils. |
| ctk5c7hwx2d2 | TLPMNEF | EOSINO_AXIS_RATIO_IQR | EOSINO_AXIS_RATIO_IQR measures the interquartile range of the major-to-minor axis ratios for eosinophils within the patch. It provides a normalized statistic that quantifies cell shape variability by computing the difference between the 75th and 25th percentile values of the derived axis ratios. |
| ctk5c7hwx2d2 | TLPMNEF | FIBRO_AXIS_RATIO_IQR | FIBRO_AXIS_RATIO_IQR quantifies the interquartile range of the major-to-minor axis ratios for fibroblasts in each patch. It captures the variability in cell shape by assessing the spread of the measurements, ensuring that any variations are represented in a manner that allows comparisons across different patient cases. |
| d103nxwtuq5y | LNF | BRIDGING_RATIO | BRIDGING_RATIO is a normalized numeric metric that quantifies the degree of three-way interaction among lymphocytes, neutrophils, and fibroblasts in each tumor patch. It is calculated as the ratio of connected clusters containing all three cell types to the total number of connected clusters in the patch, thereby allowing robust comparisons across multiple patient cases by accounting for variations in overall cell clustering. |
| d1yteqtqdw85 | TEFLM | MEDIAN_BRIDGING_DISTANCE_UM | MEDIAN_BRIDGING_DISTANCE_UM represents the median of the shortest distances calculated between eosinophils and their nearest tumor cells within each patch that meets the defined cell density criteria. The measurements are provided in micrometers and offer a normalized metric to compare spatial relationships across different patient cases by capturing the central tendency of these bridging distances. |
| d1yteqtqdw85 | TEFLM | MIN_BRIDGING_DISTANCE_UM | MIN_BRIDGING_DISTANCE_UM denotes the smallest distance in micrometers observed between any eosinophil and its closest tumor cell in qualifying patches. This parameter reflects the closest point of interaction and, being normalized through a consistent conversion factor, can be compared across different tissue samples. |
| d1yteqtqdw85 | TEFLM | MAX_BRIDGING_DISTANCE_UM | MAX_BRIDGING_DISTANCE_UM indicates the largest distance, in micrometers, among the minimum distances computed between eosinophils and their nearest tumor cells within each valid patch. It quantifies the upper range of these bridging distances in a normalized manner, facilitating cross-case comparisons. |
| d1yteqtqdw85 | TEFLM | VALID_PATCH | VALID_PATCH is a binary flag (1 or 0) that indicates whether a given patch meets all the minimum density criteria for fibroblasts, lymphocytes, and macrophages, and contains both eosinophils and tumor cells. This numeric flag ensures that only patches with adequate cellular composition are included in the spatial analysis, thereby serving as a normalized quality control measure. |
| d775va6ryqxx | FLPT | MEAN_FIBROBLAST_ROUNDNESS | The parameter 'MEAN_FIBROBLAST_ROUNDNESS' represents the normalized average nuclear roundness of fibroblast cells located in the stromal regions of patches that have tumor cells, lymphocytes, and plasma cells. It is computed by taking the roundness of each fibroblast nucleus—calculated as a ratio (from 0 to 1, where 1 indicates a perfect circle)—and then averaging these values across the patch. This normalized metric is numeric and enables robust comparisons between different patient cases. |
| d7jamvsj0oyo | LF | STD_TEXTURE_DIFF | STD_TEXTURE_DIFF measures the standard deviation of the absolute differences in nuclear texture contrast between lymphocyte and fibroblast pairs within tumor stroma. This parameter captures the variability in nuclear texture differences, reflecting the heterogeneity in the immune-stromal interface. It is derived from local patch analysis by computing the contrast feature from masked nucleus images, enabling normalized comparison across different patient cases. |
| d7jamvsj0oyo | LF | MEAN_TEXTURE_DIFF | MEAN_TEXTURE_DIFF represents the mean of the absolute differences in nuclear texture contrast between lymphocyte and fibroblast pairs in the tumor stroma. This parameter provides an average measure of the distinction in nuclear texture, indicating typical differences between the two cell types. Its computation across patches allows for normalized comparison across patient cases. |
| dkzv8k4o7j0u | TL | TUMOR_LYMPHO_MIN_DIST_MEAN | TUMOR_LYMPHO_MIN_DIST_MEAN represents the average minimum distance, measured in micrometers, from each tumor cell to its nearest lymphocyte within a patch. This metric is derived by first computing the Euclidean distances between the centroids of tumor cells and lymphocytes, selecting the minimum distance for each tumor cell, and then averaging these distances. It normalizes cell-to-cell spatial interactions across patches. |
| dkzv8k4o7j0u | TL | TUMOR_LYMPHO_MIN_DIST_STD | TUMOR_LYMPHO_MIN_DIST_STD represents the standard deviation of the minimum distances, measured in micrometers, from tumor cells to their nearest lymphocyte in the patch. This metric provides insight into the variability around the average proximity between tumor cells and lymphocytes, thus serving as a normalized measure of spatial dispersion. |
| dkzv8k4o7j0u | TL | TUMOR_LYMPHO_MIN_DIST_MEDIAN | TUMOR_LYMPHO_MIN_DIST_MEDIAN represents the median of the minimum distances, measured in micrometers, from tumor cells to the closest lymphocyte within the patch. This value offers a robust, normalized summary that is less sensitive to outliers compared to the mean, thereby supporting comparisons across patches. |
| dmp42vhjm3j0 | T | MEAN_MUCIN_PROPORTION | MEAN_MUCIN_PROPORTION represents the average fraction of mucin-stained area relative to the total cytoplasmic area of tumor cells within a patch. This parameter is normalized, allowing direct comparison between different patient cases and tissue regions. |
| dmp42vhjm3j0 | T | MEDIAN_MUCIN_PROPORTION | MEDIAN_MUCIN_PROPORTION indicates the middle value in the distribution of mucin-stained proportions among tumor cells in a patch, providing a robust measure of central tendency that is normalized across patient samples. |
| dmp42vhjm3j0 | T | SD_MUCIN_PROPORTION | SD_MUCIN_PROPORTION measures the variability in mucin-stained proportions within a patch by calculating the standard deviation. As a normalized statistical metric, it enables comparison of dispersion in mucin content across different patches and patient cases. |
| dmp42vhjm3j0 | T | MAX_MUCIN_PROPORTION | MAX_MUCIN_PROPORTION captures the highest mucin proportion measured among tumor cells in a patch. Being normalized, it highlights the extreme value of mucin content for comparative analysis across various tissue samples. |
| dn2ncx5w1z8b | TLPMNE | CENTROID_DIST_VAR_UM | CENTROID_DIST_VAR_UM represents the variance of all pairwise Euclidean distances calculated between cell centroids in each patch. The distances are first computed in pixel units and then converted to micrometers, ensuring comparability across different patients. This metric quantifies the overall spread in inter-cell distances, which can indicate the level of spatial heterogeneity and organization of the tissue. |
| dn2ncx5w1z8b | TLPMNE | CENTROID_DIST_STD_UM | CENTROID_DIST_STD_UM represents the standard deviation of all pairwise Euclidean distances between cell centroids in each patch. After converting distances from pixels to micrometers, this parameter measures the dispersion of spatial distances among the cells, offering a normalized metric that reflects the degree of variability in local tissue architecture. |
| do0xj537ecm3 | TFEL | FIBROBLAST_CV | FIBROBLAST_CV quantifies the variability in fibroblast infiltration areas across tumor-lymphocyte clusters by measuring the ratio of the standard deviation to the mean infiltration area. This normalized metric allows for the comparison of immune response variability across different patient cases, independent of the absolute numbers of fibroblast cells. |
| do0xj537ecm3 | TFEL | EOSINOPHIL_CV | EOSINOPHIL_CV quantifies the variation in eosinophil infiltration areas across tumor-lymphocyte clusters. It is calculated as the standard deviation divided by the mean area of eosinophil infiltrations, providing a normalized measure that enables reliable comparison of local immune responses across different tumor patches. |
| do0xj537ecm3 | TFEL | MEAN_FIBROBLAST_AREA | MEAN_FIBROBLAST_AREA represents the average area, in square micrometers, of fibroblast infiltration observed in the buffered regions surrounding tumor-lymphocyte clusters. By averaging the areas across clusters within a standardized patch, this parameter yields a normalized value that is comparable across different patient samples. |
| do0xj537ecm3 | TFEL | MEAN_EOSINOPHIL_AREA | MEAN_EOSINOPHIL_AREA represents the average area, in square micrometers, of eosinophil infiltration in the regions surrounding tumor-lymphocyte clusters. This parameter is computed by averaging the infiltration areas across clusters, ensuring that the value is normalized and suitable for cross-patient analysis. |
| dqfjymymxaro | TLPF | TUMOR_PERIMETER_VARIANCE | TUMOR_PERIMETER_VARIANCE quantifies the variability in the perimeters of tumor cell polygons within each analyzed patch. It is computed by first measuring each tumor cell's perimeter, converting the pixel-based measurement to micrometers, and then calculating the statistical variance of these values. This measure is numeric and expressed in square micrometers, allowing for meaningful comparisons across different patient cases. |
| dqfjymymxaro | TLPF | LYMPHOCYTE_PERIMETER_VARIANCE | LYMPHOCYTE_PERIMETER_VARIANCE represents the variance in the perimeters of lymphocyte cell polygons within a patch. Similar to the tumor cells, the perimeters are measured and converted into micrometers before the variance is calculated. The resulting numeric value provides insights into the diversity of lymphocyte sizes and can be reliably used to compare patches across patients. |
| dqfjymymxaro | TLPF | PLASMA_PERIMETER_VARIANCE | PLASMA_PERIMETER_VARIANCE captures the dispersion in the perimeters of plasma cell polygons in each patch. The metric is derived by measuring and converting the perimeters into micrometers and then computing the variance of these measurements. As a numeric parameter, it is suitable for comparing plasma cell morphological heterogeneity among different patient samples. |
| dsowqid8zsrr | TLPMNEF | RIPPLE_INDEX_TUMOR | RIPPLE_INDEX_TUMOR measures the variability in the density of tumor (epithelial) cells across concentric annular rings that radiate outward from the tumor core. The value is computed by taking the difference between the maximum and minimum cell density in the rings, divided by the mean density, resulting in a normalized ratio that allows comparison across different patient cases. |
| dsowqid8zsrr | TLPMNEF | RIPPLE_INDEX_LYMPHO | RIPPLE_INDEX_LYMPHO measures the variation in lymphocyte density using the same approach as for tumor cells. It evaluates how lymphocyte distribution changes in concentric rings surrounding the tumor core, providing a normalized metric of density fluctuation through the ratio of the difference between extreme density values to the mean density. |
| dsowqid8zsrr | TLPMNEF | RIPPLE_INDEX_PLASMA | RIPPLE_INDEX_PLASMA quantifies the degree of density variation for plasma cells across the concentric rings emanating from the tumor core. The metric is normalized, reflecting the contrast between areas of highest and lowest plasma cell densities relative to the average density in the patch. |
| dsowqid8zsrr | TLPMNEF | RIPPLE_INDEX_MACRO | RIPPLE_INDEX_MACRO assesses the variability in macrophage density through concentric ring analysis. It is computed as the normalized difference between the maximum and minimum densities of macrophages relative to the mean density, thereby enabling standardized inter-patient comparisons. |
| dsowqid8zsrr | TLPMNEF | RIPPLE_INDEX_NEUTRO | RIPPLE_INDEX_NEUTRO calculates the fluctuation in neutrophil density across the designated rings around the tumor center. The process results in a normalized value that reflects local density variations, calculated as the difference between the highest and lowest densities divided by the mean density. |
| dsowqid8zsrr | TLPMNEF | RIPPLE_INDEX_EOSINO | RIPPLE_INDEX_EOSINO gauges the variability in eosinophil density in concentric rings extending from the tumor core. By using the normalized ratio of (max density - min density) to the average density, this metric serves as a consistent measure for comparing eosinophil distribution across samples. |
| dsowqid8zsrr | TLPMNEF | RIPPLE_INDEX_CONNECTIVE | RIPPLE_INDEX_CONNECTIVE measures the variation in the density of connective tissue cells along concentric rings from the tumor core. The computed normalized ratio ((max density - min density) / mean density) allows for meaningful comparisons of connective tissue density changes across different patient cases. |
| e0teoez7asls | TLMPF | MEDIAN_ELONGATION_RATIO | MEDIAN_ELONGATION_RATIO: This parameter represents the median ratio of the longer side to the shorter side of the minimum rotated rectangle for cells that are part of density-based hotspot clusters. It quantifies the morphological elongation of cells in regions with high cell density, reflecting potential invasive processes and immunological pressures. Being a median of ratios, it is inherently normalized and comparable across different tumor patches. |
| e1lt0d8d82px | MF | MACRO_R_MEAN | MACRO_R_MEAN represents the aggregated mean red channel intensity of the cytoplasmic regions extracted from macrophages within a given tissue patch. This value is computed by averaging the red pixel intensities over masked regions corresponding to the macrophage cytoplasm, and it provides a normalized measure of staining intensity that can be compared across different patient cases. |
| e1lt0d8d82px | MF | MACRO_G_MEAN | MACRO_G_MEAN represents the aggregated mean green channel intensity of the cytoplasmic regions extracted from macrophages within a given tissue patch. It is calculated by averaging the green pixel values from the masked cytoplasmic areas, offering a standardized numeric measure that permits inter-patient comparison. |
| e1lt0d8d82px | MF | MACRO_B_MEAN | MACRO_B_MEAN represents the aggregated mean blue channel intensity of the macrophage cytoplasm for the patch. By averaging blue pixel intensities over properly isolated cell regions, this measure provides a normalized value suitable for comparison across different tumor samples. |
| e1lt0d8d82px | MF | MACRO_R_STD | MACRO_R_STD denotes the aggregated standard deviation of the red channel intensities for the macrophage cytoplasmic pixels in a patch. This statistical measure reflects the variability in red staining within the cell cytoplasm and is normalized over the patch, ensuring comparability between patient cases. |
| e1lt0d8d82px | MF | MACRO_G_STD | MACRO_G_STD denotes the aggregated standard deviation of the green channel intensities for the macrophage cytoplasm. This measure quantifies the variation in green staining intensities within the macrophage cell regions, making it a normalized numeric parameter for cross-case analysis. |
| e1lt0d8d82px | MF | MACRO_B_STD | MACRO_B_STD indicates the aggregated standard deviation of the blue channel intensities for macrophage cytoplasmic pixels. It is calculated over the extracted masked regions, providing a numeric measure of the variability in blue staining that is normalized across the patches. |
| e1lt0d8d82px | MF | FIBRO_R_MEAN | FIBRO_R_MEAN represents the aggregated mean red channel intensity for the fibroblast cytoplasmic region within a tissue patch. Derived from averaging the red pixel intensities from masked fibroblast regions, it offers a normalized value that facilitates comparisons across different patient cases. |
| e1lt0d8d82px | MF | FIBRO_G_MEAN | FIBRO_G_MEAN represents the aggregated mean green channel intensity for the fibroblast cytoplasm. This parameter is computed as an average of the green pixel values from masked cytoplasmic regions of fibroblasts, ensuring that the measurement is standardized and comparable between samples. |
| e1lt0d8d82px | MF | FIBRO_B_MEAN | FIBRO_B_MEAN represents the aggregated mean blue channel intensity for the fibroblast cytoplasmic regions. It is calculated by averaging the blue values over the masked areas, offering a normalized metric for assessing staining intensity across various patient cases. |
| e1lt0d8d82px | MF | FIBRO_R_STD | FIBRO_R_STD represents the aggregated standard deviation of the red channel intensities from the fibroblast cytoplasm across a tissue patch. This measure reflects the variability in red staining within the fibroblast cytoplasmic regions and is normalized by being computed as an average over the patch. |
| e1lt0d8d82px | MF | FIBRO_G_STD | FIBRO_G_STD represents the aggregated standard deviation of green channel intensities for the fibroblast cytoplasm. By quantifying the variability in green staining across the masked cell regions, it serves as a numeric and standardized parameter for comparative analysis. |
| e1lt0d8d82px | MF | FIBRO_B_STD | FIBRO_B_STD represents the aggregated standard deviation of the blue channel intensities in the fibroblast cytoplasmic region. This statistic provides a normalized measure of the variability in blue staining and is derived from averaging individual cell-level standard deviations within the patch. |
| e1lt0d8d82px | MF | COLOR_UNIFORMITY_SCORE | COLOR_UNIFORMITY_SCORE is a composite parameter that quantifies the color uniformity between macrophages and fibroblasts within a given patch. It is calculated as the Euclidean distance between the aggregated mean RGB vectors obtained from the macrophage and fibroblast groups. Lower values indicate a higher similarity in cytoplasmic staining between the two cell types, and this normalized metric facilitates direct comparisons across different regions and patient cases. |
| e4poglf4h65x | TLM | SYNERGY_SCORE | SYNERGY_SCORE is a normalized metric that quantifies the tricellular interaction among tumor cells, lymphocytes, and macrophages in the peri-ductal area. It is calculated by summing the pairwise adjacency counts between tumor-lymphocyte, tumor-macrophage, and lymphocyte-macrophage cell pairs, then normalizing this total by the product of the number of peri-ductal tumor cells and the number of peri-ductal immune cells. This normalization allows for meaningful comparisons across different patient cases and patches, as it adjusts for variations in cell densities. |
| eg19h4nw4205 | P | PERCENTAGE_RUSSELL_POSITIVE | PERCENTAGE_RUSSELL_POSITIVE measures the proportion of plasma cells that exhibit at least one Russell body. This parameter is computed as the percentage obtained by dividing the number of plasma cells with Russell bodies by the total number of plasma cells in a patch. As a ratio, it is normalized to allow comparisons across different patient cases. |
| eg19h4nw4205 | P | MEAN_RUSSELL_BODIES_PER_CELL | MEAN_RUSSELL_BODIES_PER_CELL represents the average count of Russell bodies per plasma cell within a given patch. It is calculated by dividing the total number of detected Russell bodies by the number of plasma cells, thereby providing a normalized measure that facilitates comparisons between different patient samples. |
| eo28emi6wsrp | P | MEAN_RED_INTENSITY | MEAN_RED_INTENSITY represents the average red channel intensity measured in the peripheral cytoplasmic region of plasma cells within a patch. This metric quantifies the degree of eosinophilic staining and is calculated across all plasma cells present in the patch, making it suitable for comparing different patient cases as it is normalized. |
| eo28emi6wsrp | P | STD_RED_INTENSITY | STD_RED_INTENSITY reflects the variability or spread of red channel intensities in the peripheral cytoplasmic regions of plasma cells within a patch. This measure indicates the consistency of staining intensity among cells, providing additional insight into the distribution of staining characteristics and is normalized for cross-case analysis. |
| eql5q5i4mabl | TF | MEAN_NUCLEAR_COLOR_DIFF | MEAN_NUCLEAR_COLOR_DIFF measures the average absolute difference in nuclear grayscale intensity between paired tumor cells and fibroblasts within the epithelial-stroma transitional zones of a patch. This parameter summarizes the typical nuclear color difference and is computed by averaging all the differences obtained from the identified transitional pairs, making it a normalized metric that facilitates comparison across different patient cases. |
| eql5q5i4mabl | TF | SD_NUCLEAR_COLOR_DIFF | SD_NUCLEAR_COLOR_DIFF quantifies the variability of the nuclear color differences within each patch by calculating the standard deviation of all computed differences between tumor cells and fibroblasts in the transitional zones. This metric is numeric and normalized per patch, and it provides insight into the heterogeneity of nuclear intensities. |
| eql5q5i4mabl | TF | MIN_NUCLEAR_COLOR_DIFF | MIN_NUCLEAR_COLOR_DIFF represents the smallest observed absolute difference in nuclear grayscale intensity between tumor cells and fibroblasts within the patch’s transitional zones. It is a normalized numeric indicator of the lowest nuclear color difference recorded, offering a baseline measurement for that region. |
| eql5q5i4mabl | TF | MAX_NUCLEAR_COLOR_DIFF | MAX_NUCLEAR_COLOR_DIFF denotes the largest observed absolute difference in nuclear grayscale intensity between tumor cells and fibroblasts within the transitional zones of a patch. As a normalized numeric parameter computed per patch, it highlights the highest contrast in nuclear intensity present, useful for distinguishing variations in staining intensity. |
| erx9arvi9vex | TM | PROP_TUMOR_MACRO_COLOC | PROP_TUMOR_MACRO_COLOC is a normalized metric that measures the proportion of tumor cells at the tumor-stroma interface which are in close proximity to macrophages. It is derived by first identifying tumor cells at the interface (using a distance threshold from stroma cells), then determining which of these interface tumor cells have a macrophage within a predefined proximity threshold. The final value is calculated by dividing the number of interface tumor cells with proximal macrophages by the total number of interface tumor cells, resulting in a normalized, numeric value that allows for comparisons across different tumor regions and patient cases. |
| ez0yjvi6mnpw | MEF | MEF_TRIAD_DENSITY | MEF_TRIAD_DENSITY is a normalized ratio that quantifies the efficiency of tricellular organization. It is calculated by dividing the number of valid macrophage-eosinophil-fibroblast triads—where each triad has all pairwise distances within the specified spatial threshold—by the total number of possible triads within a patch. This metric allows for direct comparisons across different patient cases by accounting for variations in the pool of cells in each analyzed patch. |
| ez0yjvi6mnpw | MEF | MEF_TRIAD_MEAN_DIST | MEF_TRIAD_MEAN_DIST represents the average maximum pairwise distance within each validated triad, with distances converted to micrometers. This parameter summarizes the typical spatial proximity of the triads and provides a standardized measure for comparing local cell-cell interactions across different tumor patches and patient cases. |
| f0imvbbwmal3 | FMN | ORIENTATION_COHERENCE_SCORE | ORIENTATION_COHERENCE_SCORE is a normalized metric ranging from 0 to 1 that quantifies how well aligned the fibroblasts, macrophages, and neutrophils are in local tumor patches. It is computed by first calculating the average angular differences between each cell pair in triplets found within 50μm proximity, then converting this average difference to a coherence score, where 1 signifies perfect alignment and 0 indicates random or perpendicular orientations. This normalization allows for direct comparison across different patient cases. |
| f0imvbbwmal3 | FMN | MEAN_ANGLE_DIFFERENCE | MEAN_ANGLE_DIFFERENCE represents the average angular difference, measured in degrees, between the orientations of each pair of cells in valid triplets of fibroblasts, macrophages, and neutrophils within a local patch. It captures the typical misalignment between cell nuclei, providing a continuous numeric indicator of the overall orientation consistency in a given tissue region. |
| f0imvbbwmal3 | FMN | SD_ANGLE_DIFFERENCE | SD_ANGLE_DIFFERENCE is the standard deviation of the angular differences calculated among the cell pairs in each valid triplet. This parameter measures the variability in cell orientation differences within a patch, where a lower value indicates consistent alignment and a higher value suggests greater dispersion, thus offering insight into the reliability of the alignment measurement. |
| f72d4lasqhj3 | TNP | MEAN_PERIMETER_TUMOR | MEAN_PERIMETER_TUMOR represents the average nuclear perimeter of tumor cells located in the deeper compartment of a tissue patch. This parameter is normalized by taking the mean of perimeters (converted to micrometers) and thus allows comparison across different patient cases. |
| f72d4lasqhj3 | TNP | MEAN_PERIMETER_NEUTROPHILS | MEAN_PERIMETER_NEUTROPHILS represents the average nuclear perimeter of neutrophilic granulocytes in the deeper compartment of a tissue patch. It is a numeric, normalized value computed by averaging the nuclear perimeters measured in micrometers across all neutrophils in the specified region. |
| f72d4lasqhj3 | TNP | MEAN_PERIMETER_PLASMA | MEAN_PERIMETER_PLASMA represents the average nuclear perimeter of plasma cells in the deeper compartment of a tissue patch. This parameter is normalized and calculated by averaging the nuclear perimeter values (in micrometers) of all plasma cells present in that region. |
| f72d4lasqhj3 | TNP | DERIVED_PERIMETER_DIFFERENCE | DERIVED_PERIMETER_DIFFERENCE is the numeric value obtained by subtracting the smallest mean nuclear perimeter from the largest one among the three cell types (tumor cells, neutrophils, and plasma cells) within a patch. This parameter captures the disparity in nuclear sizes among these cells in the deeper compartment and is normalized for effective comparison across different patient cases. |
| f8xeja7mayfv | TNEM | MAX_CLUSTER_SHAPE_COMPLEXITY | The parameter 'MAX_CLUSTER_SHAPE_COMPLEXITY' measures the maximum shape complexity observed among clusters of tumor cells combined with immune cells (neutrophils, eosinophils, and macrophages) within an image patch. This complexity is calculated by first clustering cells based on their spatial proximity, then merging the cell polygons within each cluster, and finally computing the ratio of the cluster's perimeter to two times the square root of (pi multiplied by the cluster's area). A value of 1 indicates a perfect circle (minimal complexity), while values greater than 1 indicate increasingly irregular and complex shapes, making the measure comparable across different samples and patient cases. |
| fmwyw3uxvqko | TFM | PV_TUMOR_DENSITY | PV_TUMOR_DENSITY measures the density of tumor cells in the perivascular zone. It is calculated by dividing the number of tumor cells located within the defined perivascular area by the total area of that region (expressed in μm²), allowing for normalized comparisons between different patches and patient cases. |
| fmwyw3uxvqko | TFM | PV_FIBROBLAST_DENSITY | PV_FIBROBLAST_DENSITY represents the density of fibroblasts in the perivascular zone. This parameter is a normalization of fibroblast counts by the perivascular area, which enables consistent comparisons across different tumor regions and patient samples. |
| fmwyw3uxvqko | TFM | AVG_TUMOR_FIBROBLAST_DIST | AVG_TUMOR_FIBROBLAST_DIST is the average Euclidean distance between tumor cells and fibroblasts within the perivascular region. The distance is calculated by converting pixel distances into micrometers, resulting in a standardized measurement of spatial proximity between these cell types. |
| fmwyw3uxvqko | TFM | AVG_TUMOR_MACROPHAGE_DIST | AVG_TUMOR_MACROPHAGE_DIST quantifies the average distance between tumor cells and macrophages in the perivascular zone. This metric, expressed in micrometers after converting pixel measurements, provides a normalized insight into the spatial relationships between these cell populations. |
| fmwyw3uxvqko | TFM | AVG_FIBROBLAST_MACROPHAGE_DIST | AVG_FIBROBLAST_MACROPHAGE_DIST calculates the mean distance between fibroblasts and macrophages in the perivascular area. By averaging the pairwise Euclidean distances, the parameter offers a comparative and normalized measure of how these two cell types are spatially arranged within the tumor microenvironment. |
| fmwyw3uxvqko | TFM | TRI_CELL_SYNERGY_INDEX | TRI_CELL_SYNERGY_INDEX is a composite metric that integrates cell density and spatial proximity information from tumor cells, fibroblasts, and macrophages in the perivascular region. It is computed as the sum of the normalized densities of these three cell types divided by the sum of their average pairwise distances (with a small constant added to the denominator to prevent division by zero), thereby yielding a normalized index that reflects their combined synergistic behavior. |
| fuq5u2lwticv | MFLE | SPATIAL_ENTROPY | SPATIAL_ENTROPY is a normalized numeric parameter that quantifies the spatial randomness or organization of selected immune and stromal cells—namely macrophages, fibroblasts, lymphocytes, and eosinophils—within a tumor patch. The measure is based on the concept of Shannon entropy, where the tumor region (patch) is divided into smaller non-overlapping windows. In each window, the relative frequency of the target cells is computed, and these frequency values are used to determine the entropy through the formula H = -Σ (p * log(p)). A higher entropy value indicates that the cells are distributed more randomly or uniformly across the patch, while a lower entropy value suggests a more clustered or organized arrangement. This normalized score facilitates comparisons across different patient cases because it is based on probability distributions rather than raw cell counts. |
| fvkmkinmw8u5 | N | MEAN_NUCLEAR_SEGMENTATION | MEAN_NUCLEAR_SEGMENTATION is the average number of detected nuclear lobes across all neutrophils in a patch. This metric is derived by computing the segmentation score (number of lobes identified based on nuclear concavity) for each neutrophil within a standardized tissue patch and then calculating their mean value, making it a normalized parameter suitable for comparing different patient cases. |
| fvkmkinmw8u5 | N | MEDIAN_NUCLEAR_SEGMENTATION | MEDIAN_NUCLEAR_SEGMENTATION represents the median number of lobes detected in neutrophils within a patch. It offers a robust central tendency measure that is less influenced by extreme values, ensuring that the typical nuclear segmentation level is accurately captured across different patches. |
| fvkmkinmw8u5 | N | MAX_NUCLEAR_SEGMENTATION | MAX_NUCLEAR_SEGMENTATION records the maximum nuclear segmentation score observed among all neutrophils in the patch. It highlights the highest complexity of nuclear lobulation detected, providing insight into the most segmented nuclei present in the tissue region. |
| fvkmkinmw8u5 | N | MIN_NUCLEAR_SEGMENTATION | MIN_NUCLEAR_SEGMENTATION captures the minimum nuclear segmentation score among neutrophils in a patch. This parameter indicates the simplest nuclear morphology found, serving as a baseline for comparing variations in nuclear segmentation. |
| fvkmkinmw8u5 | N | STD_NUCLEAR_SEGMENTATION | STD_NUCLEAR_SEGMENTATION is the standard deviation of the nuclear segmentation scores within a patch. It quantifies the variability or dispersion in the segmentation scores, reflecting the consistency or diversity of neutrophil nuclear morphology in the tissue region. |
| fvkwnltk02g6 | L | HYPERFISSION_RATIO | HYPERFISSION_RATIO represents a normalized measure calculated as the ratio of the total number of small punctate mitochondrial fragments detected in lymphocyte cells to the total number of lymphocytes in a given patch. This ratio enables comparison across different patient cases by standardizing the mitochondrial fragmentation data to the cell count, thereby eliminating biases introduced by varying raw cell counts. |
| fwor219jk1ib | TLPMNEF | MEAN_PROTRUSIONS_PER_CELL | This parameter represents the average number of detected protrusions per cell in each patch. It is calculated by dividing the sum of all detected protrusions in the patch by the number of cells present. As a normalized metric, it adjusts for differences in cell count across patches and thus enables reliable comparisons between different patient cases. |
| fwor219jk1ib | TLPMNEF | SD_PROTRUSIONS_PER_CELL | This parameter quantifies the variability in cell protrusion counts in each patch by reporting the standard deviation of protrusions per cell. It provides insight into the consistency of protrusion behavior across cells, and being a normalized statistic, it can be used to compare differences in cell morphology across various tumor regions and patient datasets. |
| g133hv06behu | TLPMNEF | CORRELATION_COEFF | This parameter represents the cross-correlation coefficient between cell perimeter irregularity and chromatic intensity in a given patch. It quantifies the strength and direction of the linear relationship between these two cell-level measurements, with values ranging from -1 to 1. Being a normalized, dimensionless metric, it allows for meaningful comparisons across different patient cases. |
| g133hv06behu | TLPMNEF | MEAN_IRREGULARITY | This parameter represents the average perimeter irregularity index of cells in a patch. The index is calculated by comparing the measured cell perimeter to that of an idealized circle based on cell area, yielding a dimensionless ratio (typically at least 1). This metric reflects deviations from circularity and is normalized to facilitate comparison between different tissue samples and patient cases. |
| g133hv06behu | TLPMNEF | MEAN_CHROMATIC_INTENSITY | This parameter reflects the average chromatic intensity of cells in the patch, derived from the mean of pixel intensities from the RGB channels of cell images. With values typically ranging between 0 and 255, this normalized metric provides a consistent quantitative measure of staining intensity across different patient cases. |
| g15dokr783mt | FTME | PEAK_DIST_FIBRO_TUMOR | PEAK_DIST_FIBRO_TUMOR represents the average distance, expressed in micrometers, from fibroblasts to tumor cells at the point of highest normalized density. This is achieved by computing the radial distribution function (RDF) between fibroblast centroids and tumor cell centroids within patches, binning these distances into fixed annuli, normalizing by the annulus area and number of source fibroblasts, and identifying the bin center with the maximum density. The result is a normalized numeric metric ideal for comparing spatial relationships between different patient tumor patches. |
| g15dokr783mt | FTME | PEAK_DIST_FIBRO_MACRO | PEAK_DIST_FIBRO_MACRO is a numeric measure that quantifies the mean radial distance from fibroblasts to macrophages at which the normalized density is maximal. The parameter is computed by evaluating pairwise Euclidean distances between fibroblast and macrophage centroids, partitioning these distances using predefined radial bins, and normalizing counts by the area of each annulus and the count of fibroblasts. This normalized measure enables the comparison of spatial patterns in different tumor regions. |
| g15dokr783mt | FTME | PEAK_DIST_FIBRO_EOSINO | PEAK_DIST_FIBRO_EOSINO quantifies the mean distance in micrometers at which fibroblasts experience the peak density of eosinophils. The measurement involves calculating the radial distribution function between fibroblasts and eosinophils, binning the computed distances, normalizing by the area of each radial bin and the number of fibroblasts, and then selecting the bin center corresponding to the highest density. As a normalized numeric parameter, it is well-suited for cross-case analyses of tumor microenvironment organization. |
| g2t6jgej14st | TLPMNEF | TEXTURE_CONVERGENCE_SCORE | TEXTURE_CONVERGENCE_SCORE: A primary metric that represents the average absolute value of the Pearson correlation coefficients computed between standardized texture features of different cell types within a local tissue patch. It quantifies the overall degree of convergence or similarity in texture patterns among the cell types, making it suitable for comparing across different patient cases. |
| g2t6jgej14st | TLPMNEF | MEAN_CORRELATION | MEAN_CORRELATION: This parameter provides the average correlation strength between the texture features of various cell types within a patch. It is derived from the individual absolute Pearson correlations computed between pairs of cell type feature sets, offering a normalized measure of texture similarity. |
| g2t6jgej14st | TLPMNEF | MAX_CORRELATION | MAX_CORRELATION: This metric captures the highest absolute correlation value observed among all cell type pairs in a given patch. It indicates the strongest similarity in texture features between any two cell types, facilitating the identification of regions with pronounced textural convergence. |
| g2t6jgej14st | TLPMNEF | MIN_CORRELATION | MIN_CORRELATION: This parameter indicates the lowest absolute correlation value among the evaluated cell type pairs in a patch. It highlights the weakest association in texture features, reflecting areas with minimal texture pattern similarity among the cell types. |
| g2t6jgej14st | TLPMNEF | CORRELATION_STD | CORRELATION_STD: The standard deviation of the absolute correlation coefficients across cell type pairs, this metric quantifies the variability in texture similarity measurements within a patch. A lower standard deviation suggests more uniform convergence patterns, while a higher value indicates greater variability. |
| g30tqbp1b8bm | LP | MEAN_LYMPHO_DIST_UM | MEAN_LYMPHO_DIST_UM measures the average distance, in micrometers, from lymphocyte cells to the boundary of the nearest tumor island. This metric assesses the typical proximity of immune cells relative to tumor regions within a patch, allowing comparisons across different patient cases since distances are scaled. |
| g30tqbp1b8bm | LP | MIN_LYMPHO_DIST_UM | MIN_LYMPHO_DIST_UM represents the smallest distance, in micrometers, observed between any lymphocyte cell and a tumor island. This parameter captures the closest approach of immune cells to tumor clusters and is normalized to offer comparable values between patches. |
| g30tqbp1b8bm | LP | MAX_LYMPHO_DIST_UM | MAX_LYMPHO_DIST_UM records the largest distance, in micrometers, between a lymphocyte and the nearest tumor island. It provides an indication of the furthest spatial separation within the patch, making it suitable for cross-case analysis. |
| g30tqbp1b8bm | LP | STD_LYMPHO_DIST_UM | STD_LYMPHO_DIST_UM is the standard deviation of the distances, in micrometers, of lymphocyte cells from the boundaries of tumor islands. This statistic reflects the variability in lymphocyte positioning relative to tumor clusters and is normalized for comparison across different samples. |
| g30tqbp1b8bm | LP | MEAN_PLASMA_DIST_UM | MEAN_PLASMA_DIST_UM measures the average distance, in micrometers, of plasma cells from the nearest tumor island. It quantifies the general spatial relationship between plasma cells and tumor regions, normalized to enable consistent comparisons. |
| g30tqbp1b8bm | LP | MIN_PLASMA_DIST_UM | MIN_PLASMA_DIST_UM captures the minimum distance, in micrometers, between any plasma cell and tumor island boundary. This parameter identifies the closest interaction between plasma cells and tumor clusters, facilitating standardized evaluation across patches. |
| g30tqbp1b8bm | LP | MAX_PLASMA_DIST_UM | MAX_PLASMA_DIST_UM calculates the maximum distance, in micrometers, observed from plasma cells to their nearest tumor island. This metric reflects the extreme spatial separation and is scaled identically across patient cases. |
| g30tqbp1b8bm | LP | STD_PLASMA_DIST_UM | STD_PLASMA_DIST_UM represents the standard deviation of plasma cell distances to tumor islands, in micrometers. This measure indicates the dispersion in plasma cell positioning relative to tumor areas, supporting comparative analysis by using normalized units. |
| g33cixg6wjh0 | TLPMNEF | MORANS_I | MORANS_I is a spatial autocorrelation coefficient that quantifies the clustering of cellular pleomorphism within a tumor patch. It is computed using a method based on Moran's I, taking into account the spatial relationships among cells through a weight matrix defined by the distance between cell centroids. This parameter provides a normalized measure ranging from -1 to 1, where higher values indicate stronger spatial clustering of similar pleomorphism values. |
| g33cixg6wjh0 | TLPMNEF | MEAN_PLEOMORPHISM | MEAN_PLEOMORPHISM represents the average pleomorphism index across all analyzed cells in a patch. The pleomorphism index is calculated by multiplying a normalized measure of nuclear area deviation (z-score) by a measure of the nuclear shape irregularity. This average value offers a normalized indication of the overall cellular pleomorphism within the patch. |
| g33cixg6wjh0 | TLPMNEF | SD_PLEOMORPHISM | SD_PLEOMORPHISM is the standard deviation of the pleomorphism indices of the cells within a patch. It reflects the heterogeneity of the cellular pleomorphism, providing insight into the variability of nuclear size and shape irregularity among cells. This metric is normalized and suitable for comparing different regions. |
| g45tqhgxrmj0 | TLPMNF | MEAN_ASYMMETRY_RATIO | MEAN_ASYMMETRY_RATIO represents the average deviation from a perfect circular shape for all targeted cells within a specific patch. This measurement is obtained by calculating the individual asymmetry of each cell (derived as 1 minus the circularity, where circularity is computed from the cell’s area and perimeter) and then averaging these values over the cells in the patch. It is normalized across different patches, allowing for comparison between various tumor regions. |
| g45tqhgxrmj0 | TLPMNF | STD_ASYMMETRY_RATIO | STD_ASYMMETRY_RATIO is the standard deviation of the asymmetry measures for cells within a patch. It reflects the variability in cell shape asymmetry among the cells considered and provides insight into the heterogeneity of cell morphology within the patch. Like the mean, this value is normalized to facilitate comparisons across different patient cases and tumor regions. |
| g4v13u4p0ra4 | E | MEAN_EOSINOPHIL_ROUNDNESS | MEAN_EOSINOPHIL_ROUNDNESS measures the average roundness of eosinophil nuclei within a patch. The roundness is computed using the formula (4 * π * Area) / (Perimeter^2), resulting in a normalized value between 0 and 1, where 1 signifies a perfectly round nucleus. |
| g4v13u4p0ra4 | E | STD_EOSINOPHIL_ROUNDNESS | STD_EOSINOPHIL_ROUNDNESS quantifies the variability in nuclear roundness among eosinophils in a patch. It represents the standard deviation of the computed roundness values, providing insight into the heterogeneity of nuclear shapes. |
| g4v13u4p0ra4 | E | MIN_EOSINOPHIL_ROUNDNESS | MIN_EOSINOPHIL_ROUNDNESS represents the lowest nuclear roundness value observed among eosinophils in a patch. This parameter highlights the minimal degree of roundness, indicating the most irregularly shaped nucleus. |
| g4v13u4p0ra4 | E | MAX_EOSINOPHIL_ROUNDNESS | MAX_EOSINOPHIL_ROUNDNESS indicates the highest nuclear roundness value recorded in a patch, reflecting the most circular nucleus among the eosinophils. |
| g7d6fxkxywys | TLPMNEF | MEAN_COMPLEXITY_INDEX | MEAN_COMPLEXITY_INDEX represents the average number of distinct cell types found across all identified clusters within a patch. It is computed by first determining the count of unique cell types in each cluster (the heterocellular complexity index) and then averaging these counts over all clusters. This metric provides a normalized measure of cellular heterogeneity that can be used to compare different patient cases. |
| g7d6fxkxywys | TLPMNEF | MAX_COMPLEXITY_INDEX | MAX_COMPLEXITY_INDEX denotes the highest number of distinct cell types observed in any single cluster within the patch. By identifying the peak heterogeneity present in a cluster, this parameter offers insight into the maximum level of cellular diversity and interaction occurring in the tumor microenvironment. |
| g7d6fxkxywys | TLPMNEF | MEDIAN_COMPLEXITY_INDEX | MEDIAN_COMPLEXITY_INDEX is the median value of the heterocellular complexity indices across all clusters in the patch. It serves as a robust metric of central tendency that summarizes the typical level of cell-type diversity within clusters, minimizing the impact of extreme values. |
| g7d6fxkxywys | TLPMNEF | STD_COMPLEXITY_INDEX | STD_COMPLEXITY_INDEX measures the standard deviation of the heterocellular complexity indices across clusters, reflecting the variability in cluster composition within a patch. A higher standard deviation suggests greater differences in cellular heterogeneity among clusters, providing additional context on the distribution of cell type diversity. |
| ga5hvvnsmng3 | TLPMNF | T_NUCLEAR_GRANULARITY | T_NUCLEAR_GRANULARITY represents the median standard deviation of pixel intensities within the nucleus of tumor cells in each patch. This metric quantifies the texture granularity of tumor cell nuclei, enabling comparison of nuclear variability across patient samples. |
| ga5hvvnsmng3 | TLPMNF | L_NUCLEAR_GRANULARITY | L_NUCLEAR_GRANULARITY measures the median variability in pixel intensities within lymphocyte nuclei per patch. It reflects the texture features of lymphocyte nuclear regions, providing a normalized and comparable measure across different cases. |
| ga5hvvnsmng3 | TLPMNF | P_NUCLEAR_GRANULARITY | P_NUCLEAR_GRANULARITY is the median standard deviation of pixel intensities within the nuclear region of plasma cells in each patch. It captures the inherent granularity of plasma cell nuclei, facilitating cross-sample analysis. |
| ga5hvvnsmng3 | TLPMNF | M_NUCLEAR_GRANULARITY | M_NUCLEAR_GRANULARITY denotes the median standard deviation of pixel intensities calculated from macrophage nuclei per patch. This parameter serves to quantify the typical nuclear texture of macrophages relative to other patches and patient cases. |
| ga5hvvnsmng3 | TLPMNF | N_NUCLEAR_GRANULARITY | N_NUCLEAR_GRANULARITY represents the median standard deviation of pixel intensities within the nuclear region of neutrophils in each patch. It provides a robust measure of nuclear granularity that is useful for comparing tissue samples. |
| ga5hvvnsmng3 | TLPMNF | F_NUCLEAR_GRANULARITY | F_NUCLEAR_GRANULARITY reflects the median variability in pixel intensities computed from eosinophil nuclei per patch, offering insights into the texture and chromatin distribution in these cells, making it a suitable parameter for comparative analysis. |
| ga5hvvnsmng3 | TLPMNF | GRANULARITY_CONTRAST_SCORE | GRANULARITY_CONTRAST_SCORE is calculated as the difference between the highest and lowest median nuclear granularity values among the six cell types within a patch. This parameter captures the degree of contrast in nuclear texture features across different cell populations, serving as an aggregate measure of heterogeneity in tissue samples. |
| gc3787ob1dag | L | LYMPHO_HOTSPOT_DENSITY | LYMPHO_HOTSPOT_DENSITY measures the number of lymphocyte clusters normalized by the patch area (per mm²). This parameter is computed by detecting clusters of lymphocytes based on spatial proximity and then dividing the number of clusters by the area of the patch. Its normalized nature makes it suitable for comparing different patient cases. |
| gc3787ob1dag | L | MEAN_HOTSPOT_SIZE | MEAN_HOTSPOT_SIZE represents the average number of lymphocytes within each detected cluster in a patch. By calculating the mean cluster size, it provides insight into the typical density of cells within clusters. Although it is derived from cell counts, it is a ratio that offers a relative measure, allowing for comparison across different patient cases. |
| gc3787ob1dag | L | STD_HOTSPOT_SIZE | STD_HOTSPOT_SIZE quantifies the variability in the sizes of lymphocyte clusters within a patch. It is calculated as the standard deviation of the number of lymphocytes per cluster, indicating the consistency of cluster sizes. As a measure of dispersion, it is independent of absolute counts and is appropriate for normalized comparisons between cases. |
| gdjsgkvaev88 | TLPMNF | T_CYTO_NUCLEAR_RATIO_MEAN | T_CYTO_NUCLEAR_RATIO_MEAN represents the average ratio of the estimated cytoplasmic area to the nuclear area for tumor cells in a given patch. This normalized metric quantifies cell morphology by comparing the relative sizes of the cytoplasm and nucleus, enabling consistent comparisons across different patient cases. |
| gdjsgkvaev88 | TLPMNF | T_CYTO_NUCLEAR_RATIO_SD | T_CYTO_NUCLEAR_RATIO_SD is the standard deviation of the cytoplasmic-to-nuclear ratios among tumor cells in the patch. This value captures the variability in cell morphology, providing insight into the consistency of tumor cell characteristics across different patches. |
| gdjsgkvaev88 | TLPMNF | L_CYTO_NUCLEAR_RATIO_MEAN | L_CYTO_NUCLEAR_RATIO_MEAN denotes the mean ratio of the cytoplasmic area to nuclear area for lymphocytes in the patch. This normalized value is used to compare cell characteristics by examining the average cell structure of lymphocytes across various patient samples. |
| gdjsgkvaev88 | TLPMNF | L_CYTO_NUCLEAR_RATIO_SD | L_CYTO_NUCLEAR_RATIO_SD measures the spread or variability of the cytoplasmic-to-nuclear ratios among lymphocytes, reflecting the heterogeneity in lymphocyte morphology between different tissue patches. |
| gdjsgkvaev88 | TLPMNF | P_CYTO_NUCLEAR_RATIO_MEAN | P_CYTO_NUCLEAR_RATIO_MEAN is the average ratio of the estimated cytoplasmic area to the nuclear area for plasma cells within the patch. As a normalized measurement, this parameter allows for the comparison of plasma cell morphology across cases. |
| gdjsgkvaev88 | TLPMNF | P_CYTO_NUCLEAR_RATIO_SD | P_CYTO_NUCLEAR_RATIO_SD represents the standard deviation of the cytoplasmic-to-nuclear ratio for plasma cells. This numeric value indicates the degree of variation in plasma cell morphology within the analyzed patch. |
| gdjsgkvaev88 | TLPMNF | M_CYTO_NUCLEAR_RATIO_MEAN | M_CYTO_NUCLEAR_RATIO_MEAN indicates the mean ratio of the estimated cytoplasmic area to nuclear area for macrophages in a patch. Being a normalized metric, it facilitates comparisons of cell structural characteristics across different tumor cases. |
| gdjsgkvaev88 | TLPMNF | M_CYTO_NUCLEAR_RATIO_SD | M_CYTO_NUCLEAR_RATIO_SD is the standard deviation of the cytoplasmic-to-nuclear ratios for macrophages, capturing the variability in macrophage morphology which might be indicative of functional diversity in the tissue microenvironment. |
| gdjsgkvaev88 | TLPMNF | N_CYTO_NUCLEAR_RATIO_MEAN | N_CYTO_NUCLEAR_RATIO_MEAN reflects the average ratio of cytoplasmic area to nuclear area for neutrophils in the patch. This normalized parameter enables uniform comparisons of neutrophil morphology across different tissue regions. |
| gdjsgkvaev88 | TLPMNF | N_CYTO_NUCLEAR_RATIO_SD | N_CYTO_NUCLEAR_RATIO_SD shows the standard deviation of the cytoplasmic-to-nuclear ratios among neutrophils. It provides a measure of dispersion in neutrophil cell structure within a patch. |
| gdjsgkvaev88 | TLPMNF | F_CYTO_NUCLEAR_RATIO_MEAN | F_CYTO_NUCLEAR_RATIO_MEAN represents the average cytoplasmic-to-nuclear ratio for fibroblasts. This value is normalized to allow comparison of fibroblast cellular morphology across multiple patient cases. |
| gdjsgkvaev88 | TLPMNF | F_CYTO_NUCLEAR_RATIO_SD | F_CYTO_NUCLEAR_RATIO_SD is the standard deviation of the cytoplasmic-to-nuclear ratio for fibroblasts, indicating the variability in fibroblast morphology within the tissue patch analyzed. |
| ggnwbvkbyml5 | TLPMNEF | PHASE_SEP_STRENGTH | PHASE_SEP_STRENGTH is a normalized metric that quantifies the phase separation strength within a tissue patch. It is calculated by comparing the mean local segregation score—derived from the fraction of same-type neighboring cells—to the expected ratio under random cell distribution, thereby providing a standardized measure for comparing different tumor cases. |
| ggnwbvkbyml5 | TLPMNEF | MEAN_LOCAL_SEGREGATION | MEAN_LOCAL_SEGREGATION represents the average local segregation score for all cells in a patch. Each local score is computed as the ratio of same-type neighbor count to the total number of neighbors, resulting in a normalized value between 0 and 1. This makes it suitable for comparing cellular segregation across multiple patient cases. |
| ggnwbvkbyml5 | TLPMNEF | STD_LOCAL_SEGREGATION | STD_LOCAL_SEGREGATION is the standard deviation of local segregation scores within a patch. It quantifies the variability in cell segregation, indicating the degree of heterogeneity in local cell-type clustering. Since the underlying scores are normalized, this parameter also serves as a normalized measure for inter-patient comparisons. |
| gjldjt9i4u3g | LM | MACRO_LYMPHO_ASPECT_RATIO_DIFF | MACRO_LYMPHO_ASPECT_RATIO_DIFF quantifies the difference in nuclear aspect ratios between macrophages and lymphocytes in tumor patches. It is derived by first identifying the immune cells located within a defined epithelial boundary, determined by a fixed distance threshold from tumor regions. For each cell, the nuclear aspect ratio is calculated based on the ratio of the longer to the shorter side of its bounding box. The median aspect ratio is then computed separately for macrophages and lymphocytes, and the parameter represents the difference obtained by subtracting the median of lymphocytes from that of macrophages. This numeric value provides a normalized measure suitable for comparisons across different patient cases. |
| gl6ktgrh39ja | L | LYMPHO_ER_EXPANSION_SCORE_MEAN | LYMPHO_ER_EXPANSION_SCORE_MEAN is a normalized parameter that represents the mean ratio of pale cytoplasmic regions, which approximate the endoplasmic reticulum, to the total cytoplasmic area in lymphocytes across a patch. This measurement is computed by dividing the number of pixels identified as pale (ER) by the overall cytoplasmic pixels for each lymphocyte, and then averaging across all lymphocytes in the patch, allowing comparisons between different patient cases. |
| gl6ktgrh39ja | L | LYMPHO_ER_EXPANSION_SCORE_STD | LYMPHO_ER_EXPANSION_SCORE_STD provides a numeric value indicating the variability in the ER expansion scores across lymphocytes within a patch. It calculates the standard deviation of the ratios of pale cytoplasmic (ER) pixels to total cytoplasmic pixels, offering insights into the heterogeneity of immune cell responses across tumor regions. |
| gl6ktgrh39ja | L | LYMPHO_ER_EXPANSION_SCORE_MEDIAN | LYMPHO_ER_EXPANSION_SCORE_MEDIAN is the median value of the ER expansion scores measured as the ratio of pale cytoplasmic (ER) region area to total cytoplasmic area in lymphocytes within a patch. This robust statistic helps mitigate the effect of outliers, supporting reliable comparative analysis across different patient images. |
| gukfxd2coi0z | FNM | NUCLEAR_CONTRAST_STD | NUCLEAR_CONTRAST_STD: This parameter quantifies the variation in nuclear contrast across the three distinct cell types (fibroblasts, neutrophils, and macrophages) within each image patch by calculating the standard deviation of their mean nuclear contrast values. The measure is normalized per patch and reflects differences in cellular behavior that may be associated with tumor progression. |
| gukfxd2coi0z | FNM | MEAN_FIBROBLAST_CONTRAST | MEAN_FIBROBLAST_CONTRAST: This parameter measures the average contrast of the nuclei in fibroblasts within a patch. It is computed as the difference between the average intensity of the nuclear region and its immediate dilated background, providing a normalized metric that allows for comparisons across different patient cases. |
| gukfxd2coi0z | FNM | MEAN_NEUTROPHIL_CONTRAST | MEAN_NEUTROPHIL_CONTRAST: This parameter represents the average nuclear contrast for neutrophils in a patch. It is derived from the difference between nuclear intensity and local background intensity, serving as a normalized measure to facilitate comparison between tumor regions from different cases. |
| gukfxd2coi0z | FNM | MEAN_MACROPHAGE_CONTRAST | MEAN_MACROPHAGE_CONTRAST: This parameter quantifies the average nuclear contrast for macrophages within a patch by computing the difference between the nuclear intensity and the surrounding background. It serves as a normalized metric for comparing cellular characteristics across different patients. |
| gukqrbvah70c | TMFNE | INFILTRATION_SYNERGY_INDEX | INFILTRATION_SYNERGY_INDEX is a normalized metric that quantifies the average alignment of infiltration direction vectors of immune and stromal cells relative to nearby tumor cells within a defined patch. The index, ranging from -1 to 1, is computed by aggregating the pairwise cosine similarity values between unit vectors calculated from immune cell positions to their nearest tumor cells. Values closer to 1 indicate strong directional coordination among cells, values near 0 suggest random alignment, and negative values imply opposing directions. This normalization allows direct comparison across different patient cases and patches. |
| gv72hx3t51nv | ME | MEAN_MACROPHAGE_ANGLE | MEAN_MACROPHAGE_ANGLE represents the average orientation of macrophage nuclei in the invasive margin of tumor patches. It is computed by first identifying macrophages located within a defined proximity to tumor cells, then calculating each cell’s nuclear orientation based on the geometric properties of its nucleus, and finally averaging these angles. This parameter is numeric and normalized, allowing comparisons across different patient cases. |
| gv72hx3t51nv | ME | MEAN_EOSINOPHIL_ANGLE | MEAN_EOSINOPHIL_ANGLE indicates the average orientation angle of eosinophil nuclei in the invasive margin. It is derived by filtering eosinophils that meet the proximity requirement to tumor cells, calculating the orientation of each cell’s nucleus through shape analysis, and taking the mean of these angles. Being an average based on relative measurements, it is also normalized and numeric. |
| gv72hx3t51nv | ME | ORIENTATION_DIFFERENCE | ORIENTATION_DIFFERENCE measures the absolute difference between the mean nuclear orientation angles of macrophages and eosinophils in the invasive margin. It is calculated by taking the computed averages from both cell types and determining the absolute difference, adjusted to reflect the smallest angular difference (within 0-90 degrees). This metric is numeric and normalized, making it useful for comparative analyses across different patient cases. |
| gwpughdydiox | T | APOPTOTIC_PREVALENCE | APOPTOTIC_PREVALENCE is a normalized numeric parameter that represents the percentage of tumor cells within a given tissue patch that exhibit apoptotic bodies. The measurement is derived by first identifying tumor cells based on their location in the epithelial compartment and confirming the presence of apoptotic bodies through morphological analysis of nuclear fragments. This percentage facilitates the comparison of apoptotic phenomena across different patient cases because it is based on the proportion of affected cells relative to the total number of tumor cells in the patch. |
| h1cqngwanmqe | TLPMNF | K_TUMOR_TO_IMMUNE | K_TUMOR_TO_IMMUNE quantifies the spatial clustering of tumor cells relative to the surrounding immune and fibroblast cells within a defined tissue patch. It is calculated by counting the number of immune/fibroblast cell centroids within a fixed distance from tumor cell centroids and normalizing this count by both the product of the number of tumor and immune/fibroblast cells and the area of the patch. This normalization allows for direct comparisons across different patches and patient cases by accounting for variations in cell density and patch size. |
| h1cqngwanmqe | TLPMNF | K_IMMUNE_TO_TUMOR | K_IMMUNE_TO_TUMOR measures the spatial clustering from the perspective of immune and fibroblast cells relative to tumor cells. It is derived by counting tumor cell centroids within a set distance of immune/fibroblast cell centroids, then normalizing by the product of the immune/fibroblast and tumor cell counts and the area of the patch. This normalized metric enables comparison of spatial interactions across different tissue patches, ensuring that differences in cell numbers and patch sizes do not affect the assessment of the spatial relationship. |
| h2e3hdfif0vc | FLME | NUCLEAR_SIZE_SKEWNESS | NUCLEAR_SIZE_SKEWNESS is a normalized, numeric metric that represents the skewness of the distribution of nuclear sizes measured in tumor infiltration zones. It is derived from aggregating the nuclear areas of specific cell types (fibroblasts, macrophages, lymphocytes, and eosinophils) selected from stromal regions. The skewness value captures the asymmetry of the nuclear size distribution, offering a comparative statistical measure across different patient cases and patches. Being a dimensionless descriptor, it is suitable for comparing various samples regardless of the raw cell count differences. |
| h9mbxunm5jp8 | F | BRANCHING_COMPLEXITY_MEAN | BRANCHING_COMPLEXITY_MEAN represents the average normalized branching complexity measured across fibroblast cells within a tumor patch. The branching complexity is derived by first detecting branch points along the fibroblast polygon boundary using an angle threshold, and then normalizing the count by the cell's perimeter. This allows reliable comparisons across different patient cases. |
| h9mbxunm5jp8 | F | BRANCHING_COMPLEXITY_SD | BRANCHING_COMPLEXITY_SD is the standard deviation of the normalized branching complexity values of fibroblast cells in a patch. This parameter quantifies the variability in the branching patterns, which can indicate differences in stromal activation within the tumor microenvironment. |
| h9mbxunm5jp8 | F | BRANCH_POINTS_PER_PERIMETER_MEAN | BRANCH_POINTS_PER_PERIMETER_MEAN is the average value of branch points per unit perimeter across fibroblast cells. Like the mean branching complexity, it is normalized by the cell polygon perimeter, ensuring comparisons across different patches and patient cases. |
| h9mbxunm5jp8 | F | MAX_BRANCHING_SCORE | MAX_BRANCHING_SCORE indicates the maximum normalized branching score observed within a patch. It identifies the most pronounced fibroblast branching complexity in the tumor region, offering insights into the highest level of stromal activation. |
| hcewjex44zyl | F | MEAN_CYTO_NUCLEAR_RATIO | MEAN_CYTO_NUCLEAR_RATIO represents the average value of the cytoplasm-to-nucleus ratios calculated for fibroblasts within a given patch. It provides a normalized metric that reflects the central tendency of cellular morphology, allowing for effective comparisons across different patient cases. |
| hcewjex44zyl | F | VARIANCE_CYTO_NUCLEAR_RATIO | VARIANCE_CYTO_NUCLEAR_RATIO measures the degree of dispersion or variability in the cytoplasm-to-nucleus ratios among fibroblasts in a patch. This parameter is crucial for understanding the heterogeneity of cell morphological states in various tumor regions. |
| hcewjex44zyl | F | MEDIAN_CYTO_NUCLEAR_RATIO | MEDIAN_CYTO_NUCLEAR_RATIO is the middle value of the sorted cytoplasm-to-nucleus ratios in a specific patch. It serves as an alternative measure of central tendency that is less affected by outliers, ensuring robust comparisons between different patient samples. |
| hcewjex44zyl | F | MIN_CYTO_NUCLEAR_RATIO | MIN_CYTO_NUCLEAR_RATIO is the lowest observed cytoplasm-to-nucleus ratio among the fibroblasts in a patch. This parameter helps to identify the lower bound of cell morphological ratios and contributes to a comprehensive understanding of cellular variability. |
| hcewjex44zyl | F | MAX_CYTO_NUCLEAR_RATIO | MAX_CYTO_NUCLEAR_RATIO is the highest observed cytoplasm-to-nucleus ratio in a given patch. It indicates the upper bound of the fibroblast cytoplasm-to-nucleus ratios and, when used alongside other statistical measures, supports the analysis of cell-level morphological extremes across patient cases. |
| hfn2emann5uj | TLMFP | FIBROBLAST_PLASMA_CV | FIBROBLAST_PLASMA_CV quantifies the coefficient of variation for the distances measured between fibroblasts and their nearest plasma cells within the infiltration zone. This zone is defined as the spatial region where the convex hulls of tumor cells, lymphocytes, and macrophages intersect. The parameter is computed as the ratio of the standard deviation to the mean of these distances, providing a normalized metric of variability in cell proximity that is comparable across patient cases. |
| hfn2emann5uj | TLMFP | INFILTRATION_ZONE_AREA | INFILTRATION_ZONE_AREA measures the area of the overlapping region formed by the convex hulls of tumor cells, lymphocytes, and macrophages. Expressed in μm², it captures the spatial extent of the infiltration zone where these cell populations converge. The numeric value is derived using a consistent conversion factor, ensuring that the measurement is comparable between different patient cases. |
| ho3opbnxudic | TFLN | MEDIAN_FRACTAL_DIM | MEDIAN_FRACTAL_DIM measures the overall median fractal dimension across all analyzed cell types within a tissue patch. It is derived by calculating the fractal dimensions of the nuclear contours of the selected cells using a box-counting method and then computing the median of these values. The metric is numeric, dimensionless, and normalized, making it suitable for comparing different patient cases. |
| ho3opbnxudic | TFLN | TUMOR_FRACTAL_DIM | TUMOR_FRACTAL_DIM captures the median fractal dimension specifically for tumor cell nuclear contours within a patch. The parameter is obtained by filtering tumor cells, calculating the fractal dimension of each tumor cell’s nuclear contour through a box-counting approach, and then taking the median value. This numeric and normalized measure is intended to assess nuclear irregularity in tumor cells across various patient samples. |
| ho3opbnxudic | TFLN | FIBROBLAST_FRACTAL_DIM | FIBROBLAST_FRACTAL_DIM represents the median fractal dimension of fibroblast cell nuclear contours within a tissue patch. This parameter is calculated by isolating fibroblast cells, evaluating the complexity of their nuclear shapes using a box-counting method, and then determining the median fractal dimension. Being a numeric and normalized metric, it can be compared across different tumor cases. |
| hr50xnqs5i02 | NF | MEDIAN_FIBROBLAST_AXIS_LENGTH | MEDIAN_FIBROBLAST_AXIS_LENGTH: This parameter measures the median value of the major nuclear axis lengths for fibroblast cells within each patch, expressed in micrometers. It is normalized on a per-patch basis, enabling comparisons across different patient cases. |
| hr50xnqs5i02 | NF | MEDIAN_NEUTROPHIL_AXIS_LENGTH | MEDIAN_NEUTROPHIL_AXIS_LENGTH: This parameter captures the median major nuclear axis length of neutrophil cells in a patch, given in micrometers. It provides a normalized metric that facilitates comparing nuclear morphology across different patches and patient cases. |
| hr50xnqs5i02 | NF | FIBROBLAST_NEUTROPHIL_AXIS_RATIO | FIBROBLAST_NEUTROPHIL_AXIS_RATIO: This parameter represents the ratio of the median fibroblast nuclear axis length to the median neutrophil nuclear axis length for each patch. As a normalized metric, it aids in comparing the relative nuclear sizes between fibroblasts and neutrophils across different samples. |
| hs8n69jvwl3c | LP | BRIDGING_INDEX | BRIDGING_INDEX is the ratio of the number of bridged lymphocyte-plasma cell pairs to the total number of potential pairs within a patch. This parameter quantifies the efficiency of fibrotic bridging, allowing for normalized comparisons across different tumor regions and patient cases. |
| hs8n69jvwl3c | LP | MEAN_BRIDGE_LENGTH_UM | MEAN_BRIDGE_LENGTH_UM represents the average length, measured in micrometers, of the fibrotic bridges between lymphocytes and plasma cells that have been validated by an intersection with connective tissue cells. Being an average measure, it provides a normalized metric for comparing the physical extent of cell-cell bridging across different patches. |
| hs8n69jvwl3c | LP | MEDIAN_BRIDGE_LENGTH_UM | MEDIAN_BRIDGE_LENGTH_UM captures the median length of the fibrotic bridges in micrometers. This central tendency measure is used to mitigate the effects of outliers, offering a robust and normalized comparison of bridging distances across various tumor regions. |
| i1fmddkrjxoz | TFMP | MEDIAN_SYNERGY_INDEX | MEDIAN_SYNERGY_INDEX represents the median value of pairwise cosine similarities calculated between the migration vectors derived from four cell types (tumor cells, fibroblasts, macrophages, and plasma cells) in each patch. This parameter is normalized as the cosine similarity values are scaled from -1 to 1, allowing for direct comparisons between different patient cases by capturing the degree of coordinated movement among these cell types. Higher values indicate stronger synergy and coordinated infiltration behavior within the tumor microenvironment. |
| i8krlu90vzlg | FPMN | STD_ORIENTATION | STD_ORIENTATION measures the circular standard deviation of the mean cytoplasmic gradient orientations among fibroblasts, plasma cells, macrophages, and neutrophils near the tumor-stroma interface. This parameter is derived by first selecting cells within a defined proximity to the interface, ensuring they belong to the specified cell types. For each cell, the cytoplasmic eosin intensity gradient is computed using edge detection techniques to obtain pixel-wise gradient orientations. The mean orientation for each cell is then calculated by applying circular statistics, taking into account the periodic nature of angular data. Finally, the parameter represents the dispersion of these mean orientations across cells in a patch, computed using circular standard deviation. Being a normalized statistical measure, it facilitates reliable comparisons across tumor regions and different patient cases. |
| i952pb8q43kd | MNF | MACRO_DENSITY_BIN1 | MACRO_DENSITY_BIN1 measures the density of macrophages in the first concentric radial bin, calculated as the number of macrophages per square micron. The density is normalized by dividing the count of cells within the bin’s annular area by the computed area, making the value comparable across different samples. |
| i952pb8q43kd | MNF | MACRO_DENSITY_BIN2 | MACRO_DENSITY_BIN2 measures the density of macrophages in the second concentric radial bin, representing cells that are slightly further from the tumor center. The value is obtained by normalizing the cell count with respect to the annular area of the bin, ensuring it is comparable between patches. |
| i952pb8q43kd | MNF | MACRO_DENSITY_BIN3 | MACRO_DENSITY_BIN3 measures the normalized density of macrophages in the third radial bin. It reflects the cell infiltration pattern as one moves outward from the tumor center, with the density calculated using the annular area corresponding to that bin. |
| i952pb8q43kd | MNF | MACRO_DENSITY_BIN4 | MACRO_DENSITY_BIN4 represents the density of macrophages in the fourth concentric bin. It is derived by counting the macrophages in the specified radial range and dividing by the annular area, yielding a normalized metric for inter-case comparison. |
| i952pb8q43kd | MNF | MACRO_DENSITY_BIN5 | MACRO_DENSITY_BIN5 measures the density of macrophages in the fifth and outermost radial bin from the tumor center. The cell count for this bin is normalized by its annular area, making it a comparative metric across different tumor patches. |
| i952pb8q43kd | MNF | NEUTRO_DENSITY_BIN4 | NEUTRO_DENSITY_BIN4 captures the density of neutrophils in the fourth radial bin from the tumor center. The parameter is calculated by dividing the number of neutrophils within the bin’s annulus by the bin’s area, thus normalizing the value for comparison. |
| i952pb8q43kd | MNF | NEUTRO_DENSITY_BIN5 | NEUTRO_DENSITY_BIN5 measures the normalized density of neutrophils in the fifth concentric radial bin. It quantifies neutrophil infiltration in areas further from the tumor center by calculating cells per square micron of the respective annulus. |
| i952pb8q43kd | MNF | FIBRO_DENSITY_BIN1 | FIBRO_DENSITY_BIN1 measures the density of fibroblasts in the innermost radial bin, reflecting the closest proximity to the tumor center. The cell count is normalized by the annular area, offering a standardized metric for cell infiltration assessment. |
| i952pb8q43kd | MNF | FIBRO_DENSITY_BIN2 | FIBRO_DENSITY_BIN2 represents the normalized density of fibroblasts in the second radial bin. It indicates the distribution of fibroblasts in a region slightly removed from the tumor center, with the density computed as cells per square micron of the annulus. |
| i952pb8q43kd | MNF | FIBRO_DENSITY_BIN3 | FIBRO_DENSITY_BIN3 captures the density of fibroblasts in the third radial bin. This metric is normalized using the annular area to reveal trends in fibroblast infiltration as a function of distance from the tumor centroid. |
| i952pb8q43kd | MNF | FIBRO_DENSITY_BIN4 | FIBRO_DENSITY_BIN4 measures the density of fibroblasts in the fourth radial bin, calculated by dividing the count of cells in the bin by the annular area. The normalization ensures the metric is suitable for comparing different patches or patient cases. |
| i952pb8q43kd | MNF | FIBRO_DENSITY_BIN5 | FIBRO_DENSITY_BIN5 reflects the normalized density of fibroblasts in the fifth and outermost radial bin. The parameter is derived by normalizing the cell count with its respective annular area, providing a robust measure of fibroblast distribution relative to the tumor center. |
| ieq1ijotl1ov | TFLN | NC_RATIO_STD | NC_RATIO_STD measures the variability in the nuclear-cytoplasmic ratio of tumor cells within each patch. This parameter is calculated by first determining the ratio of the nuclear area to the cytoplasmic area for each tumor cell and then computing the standard deviation of these ratios. It is computed only in patches that meet specific criteria, including a sufficient density of fibroblasts, the presence of at least one lymphocyte and one neutrophil, and a minimum number of tumor cells. This measure is normalized as it reflects a dispersion of ratios rather than raw counts, allowing for meaningful comparison across different patient cases. |
| iib1gj61twct | TLPMEF | LYMPHOCYTE_FRACTION | LYMPHOCYTE_FRACTION represents the normalized fraction of stromal cells within a tumor patch that are identified as lymphocytes. It is calculated by dividing the count of lymphocytes by the total count of stromal cells, ensuring that results are comparable across different patient cases and patches. |
| iib1gj61twct | TLPMEF | NEUTROPHIL_FRACTION | NEUTROPHIL_FRACTION represents the normalized fraction of stromal cells within a tumor patch that are identified as neutrophils. It is calculated by dividing the count of neutrophils by the total count of stromal cells, allowing for consistent comparison between different tumors and patches. |
| iib1gj61twct | TLPMEF | ABSOLUTE_DIFFERENCE | ABSOLUTE_DIFFERENCE is the absolute difference between the LYMPHOCYTE_FRACTION and NEUTROPHIL_FRACTION within a tumor patch. This parameter captures the magnitude of divergence in the proportions of lymphocytes and neutrophils, facilitating a comparative analysis across different patient cases with normalized values. |
| ijw7ae22b1px | M | TRANSFORMATION_INDEX | TRANSFORMATION_INDEX measures the proportion of epithelioid macrophages within a patch by calculating the ratio of macrophages with a shape factor greater than 0.7 to the total number of macrophages. This yields a normalized value between 0 and 1, allowing for comparison across different patient cases. |
| ik8qe8kxy4s6 | TLPMNEF | MEAN_STROMAL_INTENSITY | MEAN_STROMAL_INTENSITY measures the average grayscale intensity of stromal cells within each patch. It is computed by extracting the cell region from a grayscale image using a corresponding mask, and then averaging the pixel intensities. The value, which ranges from 0 to 255, indicates how pale the staining is, with higher values suggesting the paler staining characteristic of myxoid stromal changes. Since it is an average computed over valid cells in each patch, the parameter is normalized and can be compared across different patient cases. |
| ik8qe8kxy4s6 | TLPMNEF | MEAN_TEXTURE_CONTRAST | MEAN_TEXTURE_CONTRAST quantifies the average texture contrast of stromal cells within each patch. This parameter is derived by calculating texture features using a gray-level co-occurrence matrix (GLCM) on the masked regions of cells. The contrast reflects the variation in grayscale intensity within the cell regions, where lower values indicate a more homogeneous or looser matrix typical of myxoid stroma. Being computed as an average over the stromal cells in a given patch, it is normalized for inter-case comparisons. |
| inwt107ba24a | M | MEAN_NUCLEAR_IRREGULARITY | The 'MEAN_NUCLEAR_IRREGULARITY' parameter quantifies the average Fourier-based nuclear contour irregularity score for macrophages within a tissue patch. This score is computed by measuring the radial distances from the nucleus centroid to its boundary, resampling these distances uniformly, and applying a Fourier transform. The high-frequency components, normalized by the direct current component, offer a standardized measure of nuclear shape irregularity that allows for meaningful comparisons between different patient cases. |
| inwt107ba24a | M | STD_NUCLEAR_IRREGULARITY | The 'STD_NUCLEAR_IRREGULARITY' parameter represents the standard deviation of the nuclear irregularity scores among macrophages in a patch. This metric assesses the variability of the Fourier-based irregularity measurements, indicating the degree of heterogeneity in nuclear shape irregularity within the patch. It is derived from normalized scores, ensuring that comparisons across patches and patients remain valid. |
| ipdc69iz50h8 | TLPMNEF | TUMOR_FRAG_SCORE | TUMOR_FRAG_SCORE is a normalized measure representing the fragmentation of tumor cell colonies. It quantifies how fragmented the tumor cells are within a given patch by computing the ratio of the number of contiguous tumor clusters (minus one) to the total number of tumor cells (minus one), yielding a value between 0 and 1 where higher scores indicate more fragmentation. |
| ipdc69iz50h8 | TLPMNEF | LYMPHO_FRAG_SCORE | LYMPHO_FRAG_SCORE is a normalized metric that measures the fragmentation of lymphocyte colonies within a patch. It is calculated similarly by determining the ratio between the number of contiguous lymphocyte clusters and the total lymphocyte count, providing a standardized score where higher values denote increased fragmentation. |
| ipdc69iz50h8 | TLPMNEF | PLASMA_FRAG_SCORE | PLASMA_FRAG_SCORE quantifies the fragmentation of plasma cell colonies. This score is derived from the ratio of distinct plasma cell clusters relative to the total number of plasma cells in the patch, normalized to a range from 0 to 1, with higher scores indicating more fragmented colony formations. |
| ipdc69iz50h8 | TLPMNEF | MACRO_FRAG_SCORE | MACRO_FRAG_SCORE provides a standardized assessment of the fragmentation in macrophage colonies. The scoring involves assessing the number of individual clusters of macrophages formed within a patch compared to the overall cell count, ensuring the score is normalized between 0 and 1 so that greater values correlate with a higher degree of fragmentation. |
| ipdc69iz50h8 | TLPMNEF | NEUTRO_FRAG_SCORE | NEUTRO_FRAG_SCORE measures the spatial fragmentation of neutrophil colonies by computing a normalized ratio that reflects the degree of cluster separation among neutrophils in the analyzed patch. Scores closer to 1 reveal increased cluster fragmentation. |
| ipdc69iz50h8 | TLPMNEF | EOSINO_FRAG_SCORE | EOSINO_FRAG_SCORE is a normalized parameter that captures how fragmented eosinophil colonies are within a patch. The measure is derived from the ratio of distinct eosinophil clusters to the total eosinophil count, with higher values representing more fragmentation. |
| ipdc69iz50h8 | TLPMNEF | FIBRO_FRAG_SCORE | FIBRO_FRAG_SCORE quantifies the fragmentation of fibroblast colonies. It is normalized by comparing the number of contiguous fibroblast clusters against the total fibroblast population within the patch, ensuring comparability across different cases with a score range of 0 to 1. |
| ipy2a4r7ntfx | LNF | SYNERGY_INDEX | SYNERGY_INDEX is a normalized measure of the overall tricellular infiltration synergy within a tumor patch. It quantifies the proportion of actual close-proximity interactions among lymphocytes, neutrophils, and fibroblasts relative to the total possible number of interactions. This parameter is derived by summing the counts of interactions between lymphocytes and neutrophils, lymphocytes and fibroblasts, and neutrophils and fibroblasts, then dividing by the total number of possible pairs computed based on the counts of these cell types. Its normalized value, ranging between 0 and 1, ensures that it is comparable across different patient cases and patches. |
| isw6w97iy0po | TLPMNEF | DEGENERATION_PROB_MEAN | DEGENERATION_PROB_MEAN is the average probability score, computed on a patch level, that indicates the likelihood of early cytoplasmic degeneration. The score is derived by applying a logistic transformation to a linear combination of texture contrast, texture homogeneity, and normalized intensity metrics, and it is normalized between 0 and 1 so that it can be directly compared across different patient cases. |
| isw6w97iy0po | TLPMNEF | DEGENERATION_PROB_STD | DEGENERATION_PROB_STD represents the standard deviation of the degeneration probability scores calculated for individual cells within a patch. This parameter quantifies the variability or heterogeneity in cytoplasmic degeneration risk across cells and is inherently normalized by virtue of the underlying probability scores being in a comparable scale. |
| isw6w97iy0po | TLPMNEF | TEXTURE_CONTRAST_MEAN | TEXTURE_CONTRAST_MEAN measures the average contrast extracted from the gray level co-occurrence matrix of the normalized grayscale cell images. It reflects the degree of cytoplasmic granularity and variation in texture among cells, allowing for meaningful comparisons between patches though it is computed using a standardized method. |
| isw6w97iy0po | TLPMNEF | TEXTURE_HOMOGENEITY_MEAN | TEXTURE_HOMOGENEITY_MEAN is the average homogeneity measure computed from the GLCM of the normalized grayscale images. This parameter assesses the uniformity of cytoplasmic texture in cells, with higher values indicating smoother textures. Its computation is based on normalized matrices making it applicable for cross-case comparisons. |
| isw6w97iy0po | TLPMNEF | INTENSITY_MEAN | INTENSITY_MEAN is the mean normalized intensity derived from the cell images after converting them to grayscale and applying a normalization by dividing by 255. This metric captures the degree of cytoplasmic clearing, ensuring that intensity values are on a comparable scale across different patches. |
| itqudz8fop31 | TFLM | MEDIAN_DIST_TO_TRIPLE_CENTER_UM | MEDIAN_DIST_TO_TRIPLE_CENTER_UM represents the median Euclidean distance, measured in micrometers, from the centroids of tumor cell clusters to their closest triple infiltration center within a tissue patch. A triple infiltration center is determined by averaging the positions of one fibroblast, one lymphocyte, and one macrophage that are spatially co-located within a pre-defined threshold. Tumor cell clusters are formed using a density-based clustering method, and only clusters meeting a minimum cell count are considered. The distance, originally computed in pixel units, is converted to micrometers to allow for consistent comparisons across different patient cases. This normalized metric is used to assess the spatial relationship between tumor regions and immune cell infiltration sites, providing insight into the tumor microenvironment. |
| iug1rhficsc1 | PN | PLASMA_NEUTROPHIL_RATIO | PLASMA_NEUTROPHIL_RATIO represents the normalized ratio of plasma cells to neutrophils within the entire patch. It is calculated as the count of plasma cells divided by the count of neutrophils (with a small epsilon added to avoid division by zero), providing a comparative measure of immune cell distribution across different cases. |
| iug1rhficsc1 | PN | PLASMA_NEUTROPHIL_RATIO_TUMOR | PLASMA_NEUTROPHIL_RATIO_TUMOR is the normalized metric calculated specifically for the tumor region within a patch. This parameter reflects the balance of plasma cells to neutrophils in the tumor compartment, computed similarly by dividing the plasma cell count by the neutrophil count (with an epsilon adjustment) to ensure comparability between cases. |
| iug1rhficsc1 | PN | PLASMA_NEUTROPHIL_RATIO_STROMA | PLASMA_NEUTROPHIL_RATIO_STROMA is the normalized ratio of plasma cells to neutrophils observed in the stroma region of the patch. It is derived by dividing the number of plasma cells by the number of neutrophils in the stroma, with appropriate handling of zero counts to facilitate consistent comparisons across patient samples. |
| j5mk3xdd8udl | TLNPF | MEDIAN_ALIGNMENT_ANGLE | MEDIAN_ALIGNMENT_ANGLE represents the median value of the absolute differences in nuclear orientation angles between neighboring cells within a patch. These alignment angles are normalized to fall between 0 and 90 degrees, allowing for comparison across different patient cases. A lower median value indicates that the nuclei are more consistently aligned. |
| j5mk3xdd8udl | TLNPF | MAX_ALIGNMENT_ANGLE | MAX_ALIGNMENT_ANGLE captures the highest observed alignment angle difference among all valid cell pairs in a patch. Since the differences are normalized to a range of 0 to 90 degrees, this parameter reflects the maximum deviation in nuclear orientation and can be compared across different patient samples. |
| j5mk3xdd8udl | TLNPF | STD_ALIGNMENT_ANGLE | STD_ALIGNMENT_ANGLE measures the standard deviation of the alignment angle differences among neighboring cells within a patch. With values normalized between 0 and 90 degrees, it indicates the variability in nuclear orientation alignment, providing insights into the consistency of cellular alignment in different tumor regions. |
| j783uunl8wxh | TLPMNEF | MEAN_APOPTOTIC_SCORE | MEAN_APOPTOTIC_SCORE measures the average apoptotic feature score across all cells in a tumor patch. It is computed by assessing specific apoptotic features—nuclear fragmentation, crescent-shaped chromatin, and cytoplasmic shrinkage—at the cell level, normalizing each cell's score to a 0-1 range, and then averaging these scores. This normalization enables direct comparisons across different patient cases. |
| j783uunl8wxh | TLPMNEF | MEDIAN_APOPTOTIC_SCORE | MEDIAN_APOPTOTIC_SCORE represents the median value of the normalized apoptotic feature scores for all cells within a patch. It provides a robust measure of central tendency that is less affected by extreme values, thereby offering an alternative to the mean for comparing apoptotic activity across different patients. |
| j783uunl8wxh | TLPMNEF | STD_APOPTOTIC_SCORE | STD_APOPTOTIC_SCORE quantifies the variability in the apoptotic feature scores among cells in a patch. By calculating the standard deviation of scores that range from 0 to 1, it indicates the consistency or heterogeneity of apoptotic activity within the patch, which aids in comparative analysis across patient cases. |
| j9rmcueww10d | TLPMNEF | MEAN_NICHE_OVERLAP | MEAN_NICHE_OVERLAP measures the average proportion of cell boundaries shared with cells of different types within a patch. It is calculated by aggregating the individual cell-level ratios of heterotypic shared edges (obtained from Voronoi tessellation) and then averaged across all cells in the patch. The resulting metric, normalized between 0 and 1, allows for direct comparison among different patient cases. |
| j9rmcueww10d | TLPMNEF | STD_NICHE_OVERLAP | STD_NICHE_OVERLAP quantifies the variability in cell boundary sharing proportions across a patch. By computing the standard deviation of the individual cell heterotypic overlap ratios, it indicates how consistently cell interactions are distributed within the patch. Like the mean, it is normalized between 0 and 1, facilitating comparisons across various tumor samples. |
| j9rmcueww10d | TLPMNEF | MAX_NICHE_OVERLAP | MAX_NICHE_OVERLAP identifies the highest proportion of heterotypic cell boundary overlap found among all cells in a patch. This maximum value pinpoints areas within the tissue that exhibit the most intense mixing of different cell types. Being normalized between 0 and 1, it serves as a valuable metric for comparing regions and patient cases. |
| jd74w2s8s3c0 | M | MEAN_HEMOSIDERIN_INDEX | The MEAN_HEMOSIDERIN_INDEX represents the average proportion of brown pixels relative to the total cell area within macrophage cells in each patch. It is derived by calculating, for each macrophage cell, the ratio of pixels identified as brown (using defined RGB thresholds) to the total number of pixels in the cell area, and then averaging these ratios over all macrophages in the patch. This normalized metric enables comparison of iron deposition levels across different patient cases. |
| jd74w2s8s3c0 | M | MAX_HEMOSIDERIN_INDEX | The MAX_HEMOSIDERIN_INDEX indicates the highest hemosiderin score observed among all macrophage cells within a patch. It reflects the maximum portion of brown pixels in any individual cell, thus highlighting the cell with the most significant iron accumulation. This metric serves as a normalized indicator of the extreme manifestation of iron deposition within the tissue section. |
| jd74w2s8s3c0 | M | SD_HEMOSIDERIN_INDEX | The SD_HEMOSIDERIN_INDEX calculates the standard deviation of hemosiderin scores among macrophage cells in a patch. It measures the variability or dispersion in iron deposition, providing insight into the consistency or heterogeneity of the iron accumulation across the analyzed cells. This parameter is normalized and aids in understanding the distribution of iron staining intensities within the patch. |
| jtzhp327hy9v | TLF | ZONE1_T_MEAN_AREA | ZONE1_T_MEAN_AREA: This parameter represents the average area (in μm²) of tumor cells (T) located in Zone 1, the innermost concentric region around the tumor centroid. It is calculated by averaging the areas derived from the cell polygon geometry with appropriate scaling. |
| jtzhp327hy9v | TLF | ZONE1_T_STD_AREA | ZONE1_T_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of tumor cells in Zone 1, providing a measure of the variability in cell sizes within this central region. |
| jtzhp327hy9v | TLF | ZONE1_T_MEAN_PERIM | ZONE1_T_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of tumor cells in Zone 1, reflecting the typical cell boundary length calculated from the polygon properties of the cells. |
| jtzhp327hy9v | TLF | ZONE1_T_STD_PERIM | ZONE1_T_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of tumor cells in Zone 1, indicating the dispersion in the cell boundary lengths. |
| jtzhp327hy9v | TLF | ZONE1_T_MEAN_CIRC | ZONE1_T_MEAN_CIRC: This parameter measures the average circularity of tumor cells in Zone 1. Circularity is computed using the formula (4π * area) / (perimeter²), with values approaching 1 indicating a more circular shape. |
| jtzhp327hy9v | TLF | ZONE1_T_STD_CIRC | ZONE1_T_STD_CIRC: This parameter indicates the standard deviation of the circularity of tumor cells in Zone 1, reflecting variability in the shape of the tumor cells. |
| jtzhp327hy9v | TLF | ZONE1_L_MEAN_AREA | ZONE1_L_MEAN_AREA: This parameter represents the average area (in μm²) of lymphocyte cells (L) in Zone 1, calculated by averaging the cell polygon areas after appropriate conversion. |
| jtzhp327hy9v | TLF | ZONE1_L_STD_AREA | ZONE1_L_STD_AREA: This parameter indicates the standard deviation of the areas (in μm²) of lymphocyte cells in Zone 1, providing insights into the variability of lymphocyte cell sizes in this zone. |
| jtzhp327hy9v | TLF | ZONE1_L_MEAN_PERIM | ZONE1_L_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of lymphocyte cells in Zone 1, reflecting the typical cell boundary length measured from their polygon properties. |
| jtzhp327hy9v | TLF | ZONE1_L_STD_PERIM | ZONE1_L_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of lymphocyte cells in Zone 1, reflecting the variation in boundary lengths among the cells. |
| jtzhp327hy9v | TLF | ZONE1_L_MEAN_CIRC | ZONE1_L_MEAN_CIRC: This parameter measures the average circularity of lymphocyte cells in Zone 1, where circularity is calculated using the formula that compares area to the square of the perimeter, indicating cell shape uniformity. |
| jtzhp327hy9v | TLF | ZONE1_L_STD_CIRC | ZONE1_L_STD_CIRC: This parameter indicates the standard deviation of the circularity of lymphocyte cells in Zone 1, capturing the variability in the shapes of these cells. |
| jtzhp327hy9v | TLF | ZONE1_F_MEAN_AREA | ZONE1_F_MEAN_AREA: This parameter represents the average area (in μm²) of fibroblast cells (F) in Zone 1, derived by averaging the areas calculated from cell polygons after unit conversion. |
| jtzhp327hy9v | TLF | ZONE1_F_STD_AREA | ZONE1_F_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of fibroblast cells in Zone 1, giving a measure of the variability in their cell sizes. |
| jtzhp327hy9v | TLF | ZONE1_F_MEAN_PERIM | ZONE1_F_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of fibroblast cells in Zone 1, computed from the polygon lengths and corresponding scaling factors. |
| jtzhp327hy9v | TLF | ZONE1_F_STD_PERIM | ZONE1_F_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of fibroblast cells in Zone 1, indicating the spread in cell boundary lengths. |
| jtzhp327hy9v | TLF | ZONE1_F_MEAN_CIRC | ZONE1_F_MEAN_CIRC: This parameter measures the average circularity of fibroblast cells in Zone 1. It is derived using the formula (4π*area)/(perimeter²) to assess how circular the cells are. |
| jtzhp327hy9v | TLF | ZONE1_F_STD_CIRC | ZONE1_F_STD_CIRC: This parameter indicates the standard deviation of the circularity of fibroblast cells in Zone 1, reflecting the variability in cell shapes in this zone. |
| jtzhp327hy9v | TLF | ZONE2_T_MEAN_AREA | ZONE2_T_MEAN_AREA: This parameter represents the average area (in μm²) of tumor cells located in Zone 2, the second concentric region from the tumor centroid, reflecting a regional morphological characteristic. |
| jtzhp327hy9v | TLF | ZONE2_T_STD_AREA | ZONE2_T_STD_AREA: This parameter represents the standard deviation of the cell areas (in μm²) of tumor cells in Zone 2, capturing the extent of variation in cell sizes. |
| jtzhp327hy9v | TLF | ZONE2_T_MEAN_PERIM | ZONE2_T_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of tumor cells in Zone 2, computed from the cell boundary lengths measured in that zone. |
| jtzhp327hy9v | TLF | ZONE2_T_STD_PERIM | ZONE2_T_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of tumor cells in Zone 2, indicating variability in the cell boundaries. |
| jtzhp327hy9v | TLF | ZONE2_T_MEAN_CIRC | ZONE2_T_MEAN_CIRC: This parameter measures the average circularity of tumor cells in Zone 2, where circularity (calculated as 4π*area/perimeter²) indicates the degree of roundness in cell shape. |
| jtzhp327hy9v | TLF | ZONE2_T_STD_CIRC | ZONE2_T_STD_CIRC: This parameter indicates the standard deviation of the circularity of tumor cells in Zone 2, reflecting the dispersion in cell shape uniformity. |
| jtzhp327hy9v | TLF | ZONE2_L_MEAN_AREA | ZONE2_L_MEAN_AREA: This parameter represents the average area (in μm²) of lymphocyte cells in Zone 2, derived from the mean of the cell polygon areas after proper conversion. |
| jtzhp327hy9v | TLF | ZONE2_L_STD_AREA | ZONE2_L_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of lymphocyte cells in Zone 2, showing the variability in their sizes. |
| jtzhp327hy9v | TLF | ZONE2_L_MEAN_PERIM | ZONE2_L_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of lymphocyte cells in Zone 2, reflecting the typical cell boundary length in this mid-region. |
| jtzhp327hy9v | TLF | ZONE2_L_STD_PERIM | ZONE2_L_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of lymphocyte cells in Zone 2, capturing the spread of boundary measurements. |
| jtzhp327hy9v | TLF | ZONE2_L_MEAN_CIRC | ZONE2_L_MEAN_CIRC: This parameter measures the average circularity of lymphocyte cells in Zone 2, indicating the typical shape roundness by comparing area and perimeter. |
| jtzhp327hy9v | TLF | ZONE2_L_STD_CIRC | ZONE2_L_STD_CIRC: This parameter indicates the standard deviation of the circularity of lymphocyte cells in Zone 2, reflecting the variability in cell shape among lymphocytes. |
| jtzhp327hy9v | TLF | ZONE2_F_MEAN_AREA | ZONE2_F_MEAN_AREA: This parameter represents the average area (in μm²) of fibroblast cells in Zone 2, computed by averaging cell polygon areas with the proper scaling. |
| jtzhp327hy9v | TLF | ZONE2_F_STD_AREA | ZONE2_F_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of fibroblast cells in Zone 2, providing insights into the heterogeneity of cell sizes. |
| jtzhp327hy9v | TLF | ZONE2_F_MEAN_PERIM | ZONE2_F_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of fibroblast cells in Zone 2, reflecting typical boundary measurements in this zone. |
| jtzhp327hy9v | TLF | ZONE2_F_STD_PERIM | ZONE2_F_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of fibroblast cells in Zone 2, indicating the degree of variation in their boundary lengths. |
| jtzhp327hy9v | TLF | ZONE2_F_MEAN_CIRC | ZONE2_F_MEAN_CIRC: This parameter measures the average circularity of fibroblast cells in Zone 2, using the ratio of area to perimeter squared to assess shape uniformity. |
| jtzhp327hy9v | TLF | ZONE2_F_STD_CIRC | ZONE2_F_STD_CIRC: This parameter indicates the standard deviation of the circularity of fibroblast cells in Zone 2, reflecting how cell shape varies in this region. |
| jtzhp327hy9v | TLF | ZONE3_T_MEAN_AREA | ZONE3_T_MEAN_AREA: This parameter represents the average area (in μm²) of tumor cells in Zone 3, a mid-range concentric region from the tumor centroid, summarizing typical cell size. |
| jtzhp327hy9v | TLF | ZONE3_T_STD_AREA | ZONE3_T_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of tumor cells in Zone 3, capturing the diversity in cell sizes within this zone. |
| jtzhp327hy9v | TLF | ZONE3_T_MEAN_PERIM | ZONE3_T_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of tumor cells in Zone 3, reflecting the typical boundary length derived from cell morphology. |
| jtzhp327hy9v | TLF | ZONE3_T_STD_PERIM | ZONE3_T_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of tumor cells in Zone 3, indicating variability in cell edge lengths. |
| jtzhp327hy9v | TLF | ZONE3_T_MEAN_CIRC | ZONE3_T_MEAN_CIRC: This parameter measures the average circularity of tumor cells in Zone 3, with values calculated based on the relationship between area and perimeter to denote shape roundness. |
| jtzhp327hy9v | TLF | ZONE3_T_STD_CIRC | ZONE3_T_STD_CIRC: This parameter indicates the standard deviation of the circularity of tumor cells in Zone 3, showing the dispersion in cell shape metrics. |
| jtzhp327hy9v | TLF | ZONE3_L_MEAN_AREA | ZONE3_L_MEAN_AREA: This parameter represents the average area (in μm²) of lymphocyte cells in Zone 3, determined by averaging the areas derived from their polygon representation. |
| jtzhp327hy9v | TLF | ZONE3_L_STD_AREA | ZONE3_L_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of lymphocyte cells in Zone 3, highlighting variability in their sizes. |
| jtzhp327hy9v | TLF | ZONE3_L_MEAN_PERIM | ZONE3_L_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of lymphocyte cells in Zone 3, reflecting the typical cell boundary lengths in this region. |
| jtzhp327hy9v | TLF | ZONE3_L_STD_PERIM | ZONE3_L_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of lymphocyte cells in Zone 3, indicating the diversity of the perimeter measurements. |
| jtzhp327hy9v | TLF | ZONE3_L_MEAN_CIRC | ZONE3_L_MEAN_CIRC: This parameter measures the average circularity of lymphocyte cells in Zone 3, with the metric derived from the formula comparing area and perimeter squared to indicate shape consistency. |
| jtzhp327hy9v | TLF | ZONE3_L_STD_CIRC | ZONE3_L_STD_CIRC: This parameter indicates the standard deviation of the circularity for lymphocyte cells in Zone 3, representing the variability in their shape roundness. |
| jtzhp327hy9v | TLF | ZONE3_F_MEAN_AREA | ZONE3_F_MEAN_AREA: This parameter represents the average area (in μm²) of fibroblast cells in Zone 3, obtained by averaging the scaled polygon areas of these cells. |
| jtzhp327hy9v | TLF | ZONE3_F_STD_AREA | ZONE3_F_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of fibroblast cells in Zone 3, describing the heterogeneity in cell sizes. |
| jtzhp327hy9v | TLF | ZONE3_F_MEAN_PERIM | ZONE3_F_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of fibroblast cells in Zone 3, reflecting the typical measurement of cell boundaries. |
| jtzhp327hy9v | TLF | ZONE3_F_STD_PERIM | ZONE3_F_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of fibroblast cells in Zone 3, indicating the spread in cell boundary lengths. |
| jtzhp327hy9v | TLF | ZONE3_F_MEAN_CIRC | ZONE3_F_MEAN_CIRC: This parameter measures the average circularity of fibroblast cells in Zone 3, computed from the area-to-perimeter relationship to assess shape properties. |
| jtzhp327hy9v | TLF | ZONE3_F_STD_CIRC | ZONE3_F_STD_CIRC: This parameter indicates the standard deviation of circularity for fibroblast cells in Zone 3, reflecting the variability in cell shape. |
| jtzhp327hy9v | TLF | ZONE4_T_MEAN_AREA | ZONE4_T_MEAN_AREA: This parameter represents the average area (in μm²) of tumor cells in Zone 4, the outer region closer to the tumor periphery, summarizing cell size in that zone. |
| jtzhp327hy9v | TLF | ZONE4_T_STD_AREA | ZONE4_T_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of tumor cells in Zone 4, highlighting variability in their sizes. |
| jtzhp327hy9v | TLF | ZONE4_T_MEAN_PERIM | ZONE4_T_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of tumor cells in Zone 4, illustrating the typical cell boundary measurement in this outer zone. |
| jtzhp327hy9v | TLF | ZONE4_T_STD_PERIM | ZONE4_T_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of tumor cells in Zone 4, capturing variation in cell edge lengths. |
| jtzhp327hy9v | TLF | ZONE4_T_MEAN_CIRC | ZONE4_T_MEAN_CIRC: This parameter measures the average circularity of tumor cells in Zone 4, using the standard formula to assess the roundness of the cell shapes in this region. |
| jtzhp327hy9v | TLF | ZONE4_T_STD_CIRC | ZONE4_T_STD_CIRC: This parameter indicates the standard deviation of the circularity of tumor cells in Zone 4, reflecting the diversity in cell shape. |
| jtzhp327hy9v | TLF | ZONE4_L_MEAN_AREA | ZONE4_L_MEAN_AREA: This parameter represents the average area (in μm²) of lymphocyte cells in Zone 4, calculated from the mean cell polygon area after proper scaling. |
| jtzhp327hy9v | TLF | ZONE4_L_STD_AREA | ZONE4_L_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of lymphocyte cells in Zone 4, indicating how variable cell sizes are in this zone. |
| jtzhp327hy9v | TLF | ZONE4_L_MEAN_PERIM | ZONE4_L_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of lymphocyte cells in Zone 4, reflecting the typical extent of the cell borders. |
| jtzhp327hy9v | TLF | ZONE4_L_STD_PERIM | ZONE4_L_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of lymphocyte cells in Zone 4, capturing the variation in their boundary lengths. |
| jtzhp327hy9v | TLF | ZONE4_L_MEAN_CIRC | ZONE4_L_MEAN_CIRC: This parameter measures the average circularity of lymphocyte cells in Zone 4, computed to reflect the typical cell shape roundness in this region. |
| jtzhp327hy9v | TLF | ZONE4_L_STD_CIRC | ZONE4_L_STD_CIRC: This parameter indicates the standard deviation of the circularity of lymphocyte cells in Zone 4, illustrating the spread in shape uniformity. |
| jtzhp327hy9v | TLF | ZONE4_F_MEAN_AREA | ZONE4_F_MEAN_AREA: This parameter represents the average area (in μm²) of fibroblast cells in Zone 4, determined by averaging the scaled polygon areas of these cells in the outer region. |
| jtzhp327hy9v | TLF | ZONE4_F_STD_AREA | ZONE4_F_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of fibroblast cells in Zone 4, providing a measure of size variability. |
| jtzhp327hy9v | TLF | ZONE4_F_MEAN_PERIM | ZONE4_F_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of fibroblast cells in Zone 4, reflecting the typical measurement of cell boundary lengths in this zone. |
| jtzhp327hy9v | TLF | ZONE4_F_STD_PERIM | ZONE4_F_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of fibroblast cells in Zone 4, indicating the degree of variation in cell border lengths. |
| jtzhp327hy9v | TLF | ZONE4_F_MEAN_CIRC | ZONE4_F_MEAN_CIRC: This parameter measures the average circularity of fibroblast cells in Zone 4, computed using the area and perimeter relationship to denote shape consistency. |
| jtzhp327hy9v | TLF | ZONE4_F_STD_CIRC | ZONE4_F_STD_CIRC: This parameter indicates the standard deviation of the circularity for fibroblast cells in Zone 4, reflecting variability in their shape. |
| jtzhp327hy9v | TLF | ZONE5_T_MEAN_AREA | ZONE5_T_MEAN_AREA: This parameter represents the average area (in μm²) of tumor cells in Zone 5, the outermost concentric region near the tumor periphery, summarizing typical cell size in that area. |
| jtzhp327hy9v | TLF | ZONE5_T_STD_AREA | ZONE5_T_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of tumor cells in Zone 5, highlighting the variability in cell sizes at the tumor edge. |
| jtzhp327hy9v | TLF | ZONE5_T_MEAN_PERIM | ZONE5_T_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of tumor cells in Zone 5, reflecting the typical cell boundary length in this peripheral region. |
| jtzhp327hy9v | TLF | ZONE5_T_STD_PERIM | ZONE5_T_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of tumor cells in Zone 5, capturing the dispersion in boundary lengths. |
| jtzhp327hy9v | TLF | ZONE5_T_MEAN_CIRC | ZONE5_T_MEAN_CIRC: This parameter measures the average circularity of tumor cells in Zone 5, using the ratio of area to perimeter squared to assess cell shape roundness in the outer zone. |
| jtzhp327hy9v | TLF | ZONE5_T_STD_CIRC | ZONE5_T_STD_CIRC: This parameter indicates the standard deviation of the circularity of tumor cells in Zone 5, illustrating the variability in cell shape at the periphery. |
| jtzhp327hy9v | TLF | ZONE5_L_MEAN_AREA | ZONE5_L_MEAN_AREA: This parameter represents the average area (in μm²) of lymphocyte cells in Zone 5, derived from the mean of the cell polygon areas in the peripheral region. |
| jtzhp327hy9v | TLF | ZONE5_L_STD_AREA | ZONE5_L_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of lymphocyte cells in Zone 5, indicating variability in their sizes near the tumor edge. |
| jtzhp327hy9v | TLF | ZONE5_L_MEAN_PERIM | ZONE5_L_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of lymphocyte cells in Zone 5, reflecting the typical cell boundary length in this outermost zone. |
| jtzhp327hy9v | TLF | ZONE5_L_STD_PERIM | ZONE5_L_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of lymphocyte cells in Zone 5, capturing the variation in their boundary lengths. |
| jtzhp327hy9v | TLF | ZONE5_L_MEAN_CIRC | ZONE5_L_MEAN_CIRC: This parameter measures the average circularity of lymphocyte cells in Zone 5, computed to reflect the degree of roundness of these cells at the tumor periphery. |
| jtzhp327hy9v | TLF | ZONE5_L_STD_CIRC | ZONE5_L_STD_CIRC: This parameter indicates the standard deviation of the circularity of lymphocyte cells in Zone 5, showing the variability in their shape. |
| jtzhp327hy9v | TLF | ZONE5_F_MEAN_AREA | ZONE5_F_MEAN_AREA: This parameter represents the average area (in μm²) of fibroblast cells in Zone 5, calculated from the averaged and scaled cell polygon areas in the peripheral region. |
| jtzhp327hy9v | TLF | ZONE5_F_STD_AREA | ZONE5_F_STD_AREA: This parameter represents the standard deviation of the areas (in μm²) of fibroblast cells in Zone 5, indicating the heterogeneity in cell size near the tumor edge. |
| jtzhp327hy9v | TLF | ZONE5_F_MEAN_PERIM | ZONE5_F_MEAN_PERIM: This parameter denotes the average perimeter (in μm) of fibroblast cells in Zone 5, reflecting the typical measurement of cell boundaries in this outer zone. |
| jtzhp327hy9v | TLF | ZONE5_F_STD_PERIM | ZONE5_F_STD_PERIM: This parameter represents the standard deviation of the perimeters (in μm) of fibroblast cells in Zone 5, capturing the spread in their cell boundary lengths. |
| jtzhp327hy9v | TLF | ZONE5_F_MEAN_CIRC | ZONE5_F_MEAN_CIRC: This parameter measures the average circularity of fibroblast cells in Zone 5, using the standard shape metric to assess how round the cells are at the periphery. |
| jtzhp327hy9v | TLF | ZONE5_F_STD_CIRC | ZONE5_F_STD_CIRC: This parameter indicates the standard deviation of the circularity of fibroblast cells in Zone 5, reflecting the variability in cell shape in this outermost region. |
| jv8lbzwmp6og | TP | MEAN_TUMOR_ECCENTRICITY | MEAN_TUMOR_ECCENTRICITY measures the average nuclear eccentricity of tumor cells within a defined patch. It is computed from the cell nucleus shape using principal component analysis to derive the major and minor axes, and then normalizing the eccentricity value to fall between 0 and 1, where 1 indicates maximum elongation. |
| jv8lbzwmp6og | TP | MEAN_PLASMA_ECCENTRICITY | MEAN_PLASMA_ECCENTRICITY measures the average nuclear eccentricity of plasma cells in a patch following the same geometric analysis method. This parameter is normalized between 0 and 1, ensuring that comparisons across different patient cases and patches are valid. |
| jv8lbzwmp6og | TP | ECCENTRICITY_DIFFERENCE | ECCENTRICITY_DIFFERENCE represents the numerical difference between the mean nuclear eccentricity of tumor cells and plasma cells in each patch. It is calculated by subtracting MEAN_PLASMA_ECCENTRICITY from MEAN_TUMOR_ECCENTRICITY, providing a normalized metric to assess how much more elongated the tumor nuclei are compared to the plasma nuclei. |
| jv9pcvk403yl | TLFP | LYMPHO_PLASMA_DIST_CV | LYMPHO_PLASMA_DIST_CV represents the coefficient of variation of the pairwise distances between lymphocytes and plasma cells. This parameter quantifies the variability in the spatial separation between these two immune cell types computed in patches that also contain tumor cells and fibroblasts. Since it is a normalized measure (calculated as the standard deviation divided by the mean distance), it enables comparisons across different patient cases by reflecting relative dispersion in local spatial clustering patterns. |
| jy99902eq0up | TLPMNEF | TRIANGULARITY_TUMOR | TRIANGULARITY_TUMOR measures the average triangularity index of tumor (epithelial) cells in a patch. It is computed as the ratio of each cell's polygonal area to the maximal area of an equilateral triangle with the same perimeter. This normalized metric allows for comparison across different patches and patients, with values closer to 1 indicating a shape approaching a perfect equilateral triangle. |
| jy99902eq0up | TLPMNEF | TRIANGULARITY_LYMPHO | TRIANGULARITY_LYMPHO represents the mean triangularity index of lymphocytes in the patch. It quantifies the extent to which lymphocyte cell shapes approximate an equilateral triangle by averaging the computed ratios for all lymphocyte cells, ensuring the parameter is normalized for inter-patient or inter-region comparisons. |
| jy99902eq0up | TLPMNEF | TRIANGULARITY_PLASMA | TRIANGULARITY_PLASMA quantifies the mean triangularity index of plasma cells within a patch. By calculating the ratio of each plasma cell's polygon area to the maximum area for an equilateral triangle with the same perimeter, and then averaging these values, this metric provides a standardized measure of cell shape deviation. |
| jy99902eq0up | TLPMNEF | TRIANGULARITY_MACRO | TRIANGULARITY_MACRO captures the mean triangularity index of macrophages in the patch. It is derived by measuring each macrophage's shape based on the ratio of its area to that of an optimal equilateral triangle with the same perimeter, resulting in a normalized parameter that facilitates comparison across different cases. |
| jy99902eq0up | TLPMNEF | TRIANGULARITY_NEUTRO | TRIANGULARITY_NEUTRO calculates the mean triangularity index of neutrophils by averaging the computed ratios of each neutrophil's measured area to the maximum possible area of an equilateral triangle with the same perimeter. This numeric and normalized metric provides insight into cell morphology consistency. |
| jy99902eq0up | TLPMNEF | TRIANGULARITY_EOSINO | TRIANGULARITY_EOSINO reflects the average triangularity index for eosinophils in the patch. It standardizes the measurement of eosinophil cell shapes by averaging the ratio-based triangularity indices, ensuring that the values are normalized and comparable across patches. |
| jy99902eq0up | TLPMNEF | TRIANGULARITY_FIBRO | TRIANGULARITY_FIBRO measures the mean triangularity index of fibroblasts (connective cells) in the patch. This parameter is computed by evaluating each fibroblast's shape using the area-to-maximum equilateral triangle area ratio and then averaging across the type, providing a normalized indicator of cell shape. |
| jy99902eq0up | TLPMNEF | TRIANGULARITY_GLOBAL | TRIANGULARITY_GLOBAL is the global mean triangularity index for the patch, calculated by averaging the mean indices of all considered cell types. This parameter provides an overall normalized metric of cell shape triangularity that integrates the contributions from tumor cells, lymphocytes, plasma cells, macrophages, neutrophils, eosinophils, and fibroblasts. |
| k1y77ejt9l47 | MNF | NORMALIZED_MACROPHAGE_FRACTION | NORMALIZED_MACROPHAGE_FRACTION represents the proportion of macrophages among the total cells in the stromal compartment. It is calculated by dividing the count of macrophages by the total number of stromal cells in each tissue patch, providing a normalized value that allows for comparison across different patient cases. |
| k1y77ejt9l47 | MNF | NORMALIZED_NEUTROPHIL_FRACTION | NORMALIZED_NEUTROPHIL_FRACTION indicates the fraction of neutrophils in the stroma. It is derived by dividing the count of neutrophils by the total stromal cell count, ensuring that the measurement is normalized and comparable across various patches and patients. |
| k1y77ejt9l47 | MNF | NORMALIZED_FIBROBLAST_FRACTION | NORMALIZED_FIBROBLAST_FRACTION measures the fraction of fibroblasts (or connective tissue cells) within the stromal compartment. This value is obtained by dividing the fibroblast count by the total number of cells in the stroma, offering a normalized metric suitable for cross-case analysis. |
| k1y77ejt9l47 | MNF | MNF_INFILTRATION_RATIO | MNF_INFILTRATION_RATIO is a composite metric that combines the normalized fractions of macrophages, neutrophils, and fibroblasts by multiplying them together. This ratio reflects the balanced or unbalanced presence of these cell types in the stromal compartment and provides a normalized scalar value useful for comparing different patient cases. |
| k2jqrv3jhh7o | M | AVG_PSEUDOPODS_PER_CELL | AVG_PSEUDOPODS_PER_CELL measures the average number of pseudopod extensions per macrophage cell within a patch. This value is obtained by summing the pseudopod counts from each individual macrophage and then dividing by the total number of macrophages present, making it normalized and directly comparable across different tumor regions. |
| k2jqrv3jhh7o | M | MAX_PSEUDOPODS_PER_CELL | MAX_PSEUDOPODS_PER_CELL captures the highest number of pseudopod extensions observed in any single macrophage within a patch. This parameter provides insight into the peak morphological activity at a cellular level and is normalized since it reflects an individual cell’s measurement rather than an overall raw count. |
| k2jqrv3jhh7o | M | STD_PSEUDOPODS_PER_CELL | STD_PSEUDOPODS_PER_CELL represents the standard deviation of the pseudopod counts among macrophage cells in a patch. It quantifies the variability in the number of pseudopod extensions per cell, thus providing a normalized measure of dispersion that can be compared across different patient cases. |
| k9rduimyw7uu | TPMNF | TUMOR_WAVE_AMPLITUDE | TUMOR_WAVE_AMPLITUDE quantifies the spatial fluctuation of tumor cells by measuring the amplitude of their radial distribution from the tumor centroid. This measurement is derived by first computing the Euclidean distances of tumor cells from a calculated tumor centroid, binning these distances uniformly over a fixed number of radial bins, and then calculating the difference between the maximum and minimum values in the histogram. The resulting amplitude reflects the local spatial heterogeneity of tumor cell distribution and is comparable across different patient cases due to the consistent patch size and binning strategy. |
| k9rduimyw7uu | TPMNF | PLASMA_WAVE_AMPLITUDE | PLASMA_WAVE_AMPLITUDE measures the variation in the spatial distribution of plasma cells relative to the tumor centroid. By calculating the distances of plasma cells from the tumor center and categorizing them into consistent radial bins, the amplitude is determined as the difference between the highest and lowest bin counts. This parameter serves as an indicator of the fluctuating density of plasma cells around the tumor and is normalized by a uniform analytical framework, ensuring comparability across different samples. |
| k9rduimyw7uu | TPMNF | MACROPHAGE_WAVE_AMPLITUDE | MACROPHAGE_WAVE_AMPLITUDE captures the extent of fluctuation in the macrophage spatial distribution with respect to the tumor centroid. The procedure involves calculating distances from the tumor center for macrophages, organizing these distances into fixed radial bins, and then computing the amplitude as the difference between the maximum and minimum bin counts. This measure offers insights into the irregular infiltration patterns of macrophages in tumor regions and is rendered comparable across samples through standardized processing. |
| k9rduimyw7uu | TPMNF | NEUTROPHIL_WAVE_AMPLITUDE | NEUTROPHIL_WAVE_AMPLITUDE is an indicator of the spatial heterogeneity in the distribution of neutrophils around the tumor centroid. It is obtained by determining the radial distances of neutrophils from the tumor center, binning these distances into a predefined number of intervals, and then calculating the amplitude as the difference between the peak and trough counts in the histogram. This parameter is numeric and normalized by adopting a uniform analysis method, making it suitable for comparative studies. |
| k9rduimyw7uu | TPMNF | FIBROBLAST_WAVE_AMPLITUDE | FIBROBLAST_WAVE_AMPLITUDE measures the fluctuation in the radial distribution of fibroblasts (serving as a proxy for cancer‐associated fibroblasts) from the tumor centroid. The amplitude is computed by determining the distances of fibroblasts from a common tumor center, distributing them evenly into radial bins, and then finding the difference between the highest and lowest count among these bins. This parameter reflects the spatial variability of fibroblast infiltration and is properly normalized through the use of consistent binning, allowing meaningful comparisons across different patient cases. |
| k9rduimyw7uu | TPMNF | ENSEMBLE_AMPLITUDE_VARIABILITY | ENSEMBLE_AMPLITUDE_VARIABILITY represents the overall variability in the amplitude of the radial distributions across the five distinct cell types (tumor cells, plasma cells, macrophages, neutrophils, and fibroblasts). Calculated as the standard deviation of the individual cell type amplitudes, it provides a single numerical metric that quantifies the consistency or disparity of spatial fluctuations among the different cell populations. This composite measure is normalized and numeric, ensuring its utility for comparative analyses across different tumor patches. |
| ka5y1lv14ug8 | LPMNEF | LYMPHO_NUCLEAR_ROUNDNESS | LYMPHO_NUCLEAR_ROUNDNESS represents the mean nuclear roundness of lymphocytes in stromal regions. This parameter is derived by calculating the circularity of each lymphocyte's nucleus using the formula based on the cell's area and perimeter, and then averaging these values within each patch. The resulting values, ranging from 0 to 1, allow for normalized comparisons across different patient samples. |
| ka5y1lv14ug8 | LPMNEF | PLASMA_NUCLEAR_ROUNDNESS | PLASMA_NUCLEAR_ROUNDNESS represents the mean nuclear roundness of plasma cells in the stroma. Each plasma cell's nuclear shape is quantified using a circularity metric computed from area and perimeter measurements, and the mean value is calculated per patch, providing a normalized numeric indicator for comparing across different patients. |
| ka5y1lv14ug8 | LPMNEF | MACRO_NUCLEAR_ROUNDNESS | MACRO_NUCLEAR_ROUNDNESS represents the mean nuclear roundness of macrophages located in stromal regions. The nuclear shape of macrophages is quantified using a standard roundness calculation based on their area and perimeter, and the mean roundness per patch is computed. This numeric metric, ranging from 0 to 1, facilitates comparisons across patient cases. |
| ka5y1lv14ug8 | LPMNEF | NEUTRO_NUCLEAR_ROUNDNESS | NEUTRO_NUCLEAR_ROUNDNESS represents the mean nuclear roundness of neutrophils in the stroma. The parameter is obtained by calculating the roundness of each neutrophil’s nucleus from its area and perimeter measurements using a defined formula, and then averaging these values across the patch. The result is a normalized numeric value that supports inter-sample comparisons. |
| ka5y1lv14ug8 | LPMNEF | EOSINO_NUCLEAR_ROUNDNESS | EOSINO_NUCLEAR_ROUNDNESS represents the mean nuclear roundness of eosinophils in stromal regions. Similar to the other cell types, the nuclear roundness is calculated from area and perimeter measures and then averaged for the patch, yielding a numeric metric between 0 and 1 for use in comparative analyses. |
| ka5y1lv14ug8 | LPMNEF | FIBRO_NUCLEAR_ROUNDNESS | FIBRO_NUCLEAR_ROUNDNESS represents the mean nuclear roundness of fibroblasts (identified as connective tissue cells) in the stroma. The roundness of each fibroblast nucleus is calculated based on its geometric properties, and these values are averaged over the cells present in each patch. This results in a normalized numeric parameter that helps in comparing nuclear morphology across different cases. |
| kd1mop5gtij0 | L | MITOTIC_FRACTION | MITOTIC_FRACTION represents the ratio of lymphocytes exhibiting mitotic features relative to the total number of lymphocytes in a patch. This metric is obtained by first detecting mitotic signatures through edge detection on the grayscale images of lymphocyte nuclei and then computing the fraction of cells meeting the mitotic criteria. Being a relative measure, it is normalized, making it suitable for comparing different patient cases, and it directly relates to the biological hypothesis that higher mitotic activity in lymphocytes may indicate a more active immune response influencing tumor outcomes. |
| kd1mop5gtij0 | L | MEAN_EDGE_INTENSITY | MEAN_EDGE_INTENSITY measures the average intensity of the edge features identified in lymphocytes within a patch. It is calculated by averaging the edge intensity values obtained from applying an edge detection filter to the grayscale images of lymphocyte nuclei. This average value is a normalized metric that provides supplementary information on the overall morphological changes of the cells, thereby supporting the assessment of mitotic activity. |
| kgdicz1tb42w | FP | DUAL_INFILTRATION_RATIO | DUAL_INFILTRATION_RATIO is a normalized metric that quantifies the balance between fibroblast and plasma cell infiltration within the invasive margin of a tumor. It is calculated by dividing the number of fibroblasts by the number of plasma cells in the defined invasive margin area, which allows for meaningful comparisons across different patient cases. The metric includes proper handling for cases with zero plasma cells, ensuring its numeric reliability across the cohort. |
| kih77xvmq4w6 | MNP | MACROPHAGE_INFILTRATION_IRREGULARITY | MACROPHAGE_INFILTRATION_IRREGULARITY quantifies the shape irregularity of macrophage clusters within the stroma. It is obtained by clustering spatially proximal macrophages, forming a representative infiltration zone by uniting their cell boundaries, and then computing the ratio of the perimeter to the square root of the area. The metric is normalized to allow comparison across different patches and patients, reflecting edge complexity in the immune cell distribution. |
| kih77xvmq4w6 | MNP | PLASMA_INFILTRATION_IRREGULARITY | PLASMA_INFILTRATION_IRREGULARITY assesses the irregularity in the shape of plasma cell infiltration zones in the stroma. It is calculated by grouping plasma cells based on spatial proximity, creating a composite boundary that represents the infiltration zone, and then measuring the ratio of the perimeter of this boundary to the square root of its area. This numeric, normalized score highlights the complexity of plasma cell dispersal within the tumor microenvironment. |
| kih77xvmq4w6 | MNP | COMBINED_INFILTRATION_IRREGULARITY | COMBINED_INFILTRATION_IRREGULARITY is an aggregated metric that averages the individual irregularity scores from different immune cell types (including macrophages, neutrophils, and plasma cells) present in the stroma. By integrating the separate measures into a single score, it provides a comprehensive, normalized assessment of the overall complexity and heterogeneity of immune cell infiltration patterns within each patch. |
| kj07wrpc76e3 | T | MEAN_CHROMOCENTER_COUNT | MEAN_CHROMOCENTER_COUNT measures the average number of chromocenters per tumor cell within a patch. It is calculated by counting the number of chromocenters identified in each tumor cell, and then computing the average across all tumor cells in that patch. This normalized metric allows for comparison across different patient cases by summarizing the typical chromocenter organization within tumor regions. |
| kj07wrpc76e3 | T | SD_CHROMOCENTER_COUNT | SD_CHROMOCENTER_COUNT quantifies the standard deviation of chromocenter counts across tumor cells within a patch. By assessing the variability in the number of chromocenters from cell to cell, this metric provides insight into the heterogeneity of nuclear architecture among tumor cells. It is a numeric measure that, despite being in the same units as the raw counts, becomes meaningful when compared in the context of the averaged chromocenter count. |
| kj07wrpc76e3 | T | CV_CHROMOCENTER_COUNT | CV_CHROMOCENTER_COUNT represents the coefficient of variation of chromocenter counts, calculated as the ratio of the standard deviation to the mean chromocenter count (with a small offset to avoid division by zero). This parameter offers a normalized measure of relative variability, enabling direct comparison of chromocenter distribution uniformity between patches or across different patient cases. |
| kte4j9jyotrv | TLPMNEF | Epithelial_MEAN_ALIGNMENT | Epithelial_MEAN_ALIGNMENT measures the average angle in degrees between the major axis of epithelial cells and the radial direction from the tumor center within a patch. A lower value indicates that epithelial cells are more consistently aligned toward or away from the tumor center, suggesting a potential structured response in tissue organization. |
| kte4j9jyotrv | TLPMNEF | Epithelial_STD_ALIGNMENT | Epithelial_STD_ALIGNMENT quantifies the variation (standard deviation) in alignment angles of epithelial cells. It reflects how consistently the cells follow the directional alignment pattern, with lower values indicating uniformity in cell orientation relative to the tumor center. |
| kte4j9jyotrv | TLPMNEF | Lymphocyte_MEAN_ALIGNMENT | Lymphocyte_MEAN_ALIGNMENT measures the average angular difference (in degrees) between the major axis of lymphocytes and the radially-oriented vector from the tumor centroid. This normalized measurement, confined between 0 and 90 degrees, indicates the degree of directional consistency among lymphocytes in the tumor vicinity. |
| kte4j9jyotrv | TLPMNEF | Lymphocyte_STD_ALIGNMENT | Lymphocyte_STD_ALIGNMENT captures the standard deviation of the lymphocyte angular differences. It evaluates the consistency of the alignment among lymphocytes, where lower standard deviation values imply more uniform radial alignment within each patch. |
| kte4j9jyotrv | TLPMNEF | Plasma_MEAN_ALIGNMENT | Plasma_MEAN_ALIGNMENT calculates the mean angle (in degrees) between the primary axis of plasma cells and the radial axis extending from the tumor center. This normalized parameter assesses the tendency of plasma cells to orient in a specific manner around the tumor region. |
| kte4j9jyotrv | TLPMNEF | Plasma_STD_ALIGNMENT | Plasma_STD_ALIGNMENT provides the standard deviation of the angular differences for plasma cells. It offers insight into the variability of plasma cell orientations, where lower variance supports a more consistent directional alignment in the local tumor environment. |
| kte4j9jyotrv | TLPMNEF | Neutrophil_MEAN_ALIGNMENT | Neutrophil_MEAN_ALIGNMENT represents the average angle (in degrees) between the principal axis of neutrophils and the radial vector from the tumor center. This metric, normalized to a range of 0 to 90 degrees, serves as an indicator of the degree of radial alignment exhibited by neutrophils. |
| kte4j9jyotrv | TLPMNEF | Neutrophil_STD_ALIGNMENT | Neutrophil_STD_ALIGNMENT measures the dispersion (standard deviation) of the neutrophil alignment angles. A lower value suggests that neutrophils within the patch maintain a consistent alignment pattern relative to the tumor center. |
| kte4j9jyotrv | TLPMNEF | Eosinophil_MEAN_ALIGNMENT | Eosinophil_MEAN_ALIGNMENT quantifies the mean angular difference in degrees between the main axis of eosinophils and the radial line from the tumor center. The normalization ensures the value lies between 0 and 90 degrees, with smaller angles indicating better radial alignment. |
| kte4j9jyotrv | TLPMNEF | Eosinophil_STD_ALIGNMENT | Eosinophil_STD_ALIGNMENT quantifies the variability (standard deviation) in the angular differences for eosinophils. It assesses the consistency of directional alignment among eosinophils, with lower values indicating more homogeneous orientation patterns. |
| kte4j9jyotrv | TLPMNEF | Macrophage_MEAN_ALIGNMENT | Macrophage_MEAN_ALIGNMENT computes the average angle (in degrees) between the main orientation of macrophages and the radial direction emanating from the tumor center. This normalized measurement helps in comparing epithelial response patterns by providing a common metric across patches. |
| kte4j9jyotrv | TLPMNEF | Macrophage_STD_ALIGNMENT | Macrophage_STD_ALIGNMENT captures the standard deviation of the alignment angles for macrophages. It reflects the consistency of macrophage orientation with respect to the tumor center, where lower standard deviation indicates more uniform alignment. |
| kte4j9jyotrv | TLPMNEF | Connective_MEAN_ALIGNMENT | Connective_MEAN_ALIGNMENT determines the mean angle (in degrees) between the major axis of connective tissue cells and the radial vector from the tumor center. With normalization to the range of 0 to 90 degrees, it provides a standardized measure of directional alignment in the tumor microenvironment. |
| kte4j9jyotrv | TLPMNEF | Connective_STD_ALIGNMENT | Connective_STD_ALIGNMENT assesses the spread (standard deviation) of the alignment angles for connective tissue cells, offering insight into the variability and consistency of their alignment relative to the tumor center. |
| ky8lzpau1sb6 | TFME | MEDIAN_TUMOR_FIBRO_IRREGULARITY_RATIO | The MEDIAN_TUMOR_FIBRO_IRREGULARITY_RATIO parameter measures the median ratio of tumor cell perimeter irregularity to fibroblast perimeter irregularity in small tumor patches that demonstrate significant immune cell infiltration. Only patches containing both at least 5 macrophages and 5 eosinophils are considered. Within these patches, the perimeter irregularity for each tumor cell and fibroblast is computed as the ratio of the cell's boundary length to a theoretical value derived from the cell's area. The median value of these irregularities is calculated separately for tumor cells and fibroblasts, and then the ratio of these medians is derived. Values greater than one indicate that, on average, tumor cells exhibit more irregular perimeters than fibroblasts. The ratio is numeric in nature and normalized by using median values, making it suitable for comparative analysis across different tumor cases. |
| kzargm5wbse9 | TMNP | CROSS_CELL_CURVATURE_VARIANCE | CROSS_CELL_CURVATURE_VARIANCE measures the variability in the average boundary curvature values computed for four distinct cell types in each patch. This parameter reflects the heterogeneity in cell boundary properties, indicating differences in local cell-cell interactions and potential mechanical constraints within tumor regions. |
| kzargm5wbse9 | TMNP | MEAN_TUMOR_CURVATURE | MEAN_TUMOR_CURVATURE represents the average curvature of the boundaries of tumor cells (derived from epithelial cell data) within a patch. This metric is calculated by averaging the curvature measurements along the cell perimeter, providing insight into the geometric characteristics of tumor cell morphology. |
| kzargm5wbse9 | TMNP | MEAN_MACROPHAGE_CURVATURE | MEAN_MACROPHAGE_CURVATURE denotes the mean boundary curvature calculated for macrophages in a given patch. It quantifies the average curvature value from the polygon boundaries of macrophage cells, which may be linked to their role in tumor microenvironments. |
| kzargm5wbse9 | TMNP | MEAN_NEUTROPHIL_CURVATURE | MEAN_NEUTROPHIL_CURVATURE indicates the average curvature of the cell boundaries for neutrophils within a patch. This parameter is derived from the local curvature values computed along the cell edges and reflects neutrophil morphology. |
| kzargm5wbse9 | TMNP | MEAN_PLASMA_CURVATURE | MEAN_PLASMA_CURVATURE captures the average boundary curvature of plasma cells present in the patch. It is computed by averaging the curvature measurements along the segmented cell boundaries, highlighting the shape characteristics of plasma cells. |
| kzwbiyl7ta2g | TFEP | AVG_CHROMATIN_CONDENSATION_RATIO | AVG_CHROMATIN_CONDENSATION_RATIO quantifies the average proportion of chromatin within tumor cell nuclei that shows condensed features. This metric is derived by applying a threshold to the gray-level intensity of the nucleus, identifying pixels that fall below this threshold as representing condensed chromatin. The ratio is normalized by considering the fraction relative to the total number of pixels representing the nucleus, thereby enabling valid comparisons across different patches and patient cases. |
| l0u3sr2tkhcj | TLPMNEF | MEAN_SAT_TUMOR | MEAN_SAT_TUMOR represents the average saturation value calculated from tumor (epithelial) cells within a patch. This measure is derived by converting cell images to the HSV color space and computing the average of the saturation channel for each tumor cell. It serves as a normalized metric reflecting the metabolic and structural features of tumor cells across different patient cases. |
| l0u3sr2tkhcj | TLPMNEF | MEAN_SAT_LYMPHO | MEAN_SAT_LYMPHO denotes the average saturation value of lymphocyte cells within a patch. The saturation is computed in a way that is normalized to allow comparison between different cases, and it reflects key immune cell characteristics that might influence the tumor microenvironment. |
| l0u3sr2tkhcj | TLPMNEF | MEAN_SAT_PLASMA | MEAN_SAT_PLASMA is the normalized average saturation value measured in plasma cells. Plasma cells are evaluated through the same method of masking and color deconvolution, providing an insight into features driven by B-cell biology within the tumor context. |
| l0u3sr2tkhcj | TLPMNEF | MEAN_SAT_MACRO | MEAN_SAT_MACRO captures the average saturation value of macrophage cells in a patch. By analyzing the saturation from each macrophage through a normalized approach, this parameter reflects phagocytic and metabolic properties that may be correlated with tumor behavior. |
| l0u3sr2tkhcj | TLPMNEF | MEAN_SAT_NEUTRO | MEAN_SAT_NEUTRO represents the average saturation value of neutrophil cells. The calculation is based on the average reading of saturation extracted from the masked cell areas, offering a standardized quantitative feature for these granulocytes. |
| l0u3sr2tkhcj | TLPMNEF | MEAN_SAT_EOSINO | MEAN_SAT_EOSINO denotes the normalized average saturation value of eosinophil cells within the tissue patch. This parameter quantifies cell-specific saturation to shed light on the specialized characteristics of eosinophils in the tumor microenvironment. |
| l0u3sr2tkhcj | TLPMNEF | MEAN_SAT_FIBRO | MEAN_SAT_FIBRO measures the average saturation value of fibroblast cells (connective tissue cells). It is derived by averaging the cell-specific saturation levels and provides a normalized assessment of stromal cell properties essential for interpreting local tissue architecture. |
| l0u3sr2tkhcj | TLPMNEF | SAT_VARIATION_SCORE | SAT_VARIATION_SCORE is a metric that quantifies the heterogeneity of color saturation across the various cell types in each patch. It is calculated as the standard deviation of the mean saturation values for tumor, lymphocyte, plasma, macrophage, neutrophil, eosinophil, and fibroblast cells. This normalized score helps capture variation in tissue composition and cellular characteristics. |
| l16pvjvurctr | TLN | DENSITY_TUMOR | DENSITY_TUMOR measures the number of tumor cells per unit area (cells/μm²) within the central tumor region. This density is computed by normalizing the count of tumor cells using the area of a specifically defined central region, ensuring comparability across different patient patches. |
| l16pvjvurctr | TLN | DENSITY_LYMPHOCYTE | DENSITY_LYMPHOCYTE quantifies the number of lymphocytes per unit area (cells/μm²) in the central tumor region. The metric is normalized by the central area, making it suitable for cross-patient comparisons of immune cell infiltration. |
| l16pvjvurctr | TLN | DENSITY_NEUTROPHIL | DENSITY_NEUTROPHIL represents the number of neutrophils per unit area (cells/μm²) within the central tumor region. Normalization by the computed area allows for standardized comparisons between different tumor patches. |
| l16pvjvurctr | TLN | AVG_DIST_TUMOR_LYMPHO | AVG_DIST_TUMOR_LYMPHO calculates the average Euclidean distance (in micrometers) between tumor cells and lymphocytes in the central tumor region. This metric quantifies the spatial proximity between these two cell types, providing insight into their interaction dynamics. |
| l16pvjvurctr | TLN | AVG_DIST_TUMOR_NEUTRO | AVG_DIST_TUMOR_NEUTRO measures the average Euclidean distance (in micrometers) between tumor cells and neutrophils within the central tumor region. It serves as an indicator of the spatial relationship and potential interaction between tumor cells and neutrophils. |
| l16pvjvurctr | TLN | AVG_DIST_LYMPHO_NEUTRO | AVG_DIST_LYMPHO_NEUTRO determines the average Euclidean distance (in micrometers) between lymphocytes and neutrophils in the central tumor region. This spatial metric helps assess how closely these two distinct immune cell types are positioned relative to one another. |
| l16pvjvurctr | TLN | OVERALL_AVG_DISTANCE | OVERALL_AVG_DISTANCE is the mean of all pairwise average distances computed among tumor cells, lymphocytes, and neutrophils in the central tumor region. This single metric provides a summary of the overall spatial interactions among the different cell types. |
| l16pvjvurctr | TLN | SYNERGY_INDEX | SYNERGY_INDEX is a composite metric that multiplies the densities of tumor cells, lymphocytes, and neutrophils and then divides the product by the overall average pairwise distance. This index integrates both cell density and spatial proximity, offering a measure of the synergistic interactions among the cell types, which may have prognostic relevance. |
| l2xu8qrn46gt | TLPMNEF | EOSIN_INTENSITY_STD | EOSIN_INTENSITY_STD measures the variability of the weighted mean eosin intensities calculated across multiple small neighborhood windows within a patch. It reflects local tissue heterogeneity by quantifying the standard deviation of these intensity values, thereby providing insight into the fluctuations in eosin staining across small regions. |
| l2xu8qrn46gt | TLPMNEF | EOSIN_INTENSITY_MEAN | EOSIN_INTENSITY_MEAN represents the overall average of weighted mean eosin intensities computed from all valid neighborhood windows in a patch. This parameter summarizes the central tendency of eosin staining intensity after cells have been weighted by their cell type frequency, facilitating comparison across different patient cases. |
| l2xu8qrn46gt | TLPMNEF | EOSIN_INTENSITY_MIN | EOSIN_INTENSITY_MIN captures the minimum weighted mean eosin intensity observed across the neighborhood windows in a patch. It highlights the lowest regional intensity, which may indicate areas of low eosin staining or subtle tissue alterations. |
| l2xu8qrn46gt | TLPMNEF | EOSIN_INTENSITY_MAX | EOSIN_INTENSITY_MAX captures the maximum weighted mean eosin intensity observed across the neighborhood windows in a patch. It indicates the highest regional intensity, reflecting areas of elevated eosin staining that may correlate with distinct tissue characteristics. |
| l36s63hxu4wc | LMPF | TRI_CELL_FIBROBLAST_FRACTION | This parameter measures the proportion of tri-cell clusters, consisting of a lymphocyte, a macrophage, and a plasma cell, that also have at least one fibroblast in close proximity (within a 50μm radius). It is normalized by computing the ratio of clusters with adjacent fibroblasts over the total number of tri-cell clusters, making it comparable across different patient cases and spatial patches. |
| l3gt6imebf6x | TLPMNEF | BIFURCATION_DENSITY | BIFURCATION_DENSITY is a normalized spatial metric that quantifies the density of branch points in stromal fiber networks per square millimeter of stromal tissue. It is derived by first isolating the stromal regions, enhancing the continuity of fibrous structures, and then performing skeletonization to identify the fibers. Branch points, defined as pixels in the skeleton with more than two connected neighbors, are counted and the total is normalized by the area of the stromal tissue in the patch (converted into mm²). This normalization process ensures that the metric is comparable across patient samples and patches, accounting for variations in tissue area. |
| lctk2tqclu7q | E | MEAN_GRANULE_INTEGRITY_SCORE | MEAN_GRANULE_INTEGRITY_SCORE: This parameter represents the average ratio of the area of intact eosinophil granule cores (which have been filtered by size and circularity criteria) to the total cell area across all eosinophils within a patch. This ratio is normalized, enabling comparison across different patient cases and tumor regions. |
| lctk2tqclu7q | E | SD_GRANULE_INTEGRITY_SCORE | SD_GRANULE_INTEGRITY_SCORE: This parameter measures the standard deviation of the granule integrity scores for eosinophils within a patch. It provides insight into the variability of granule preservation among cells in the patch and is also a normalized value. |
| lctk2tqclu7q | E | MAX_GRANULE_INTEGRITY_SCORE | MAX_GRANULE_INTEGRITY_SCORE: This parameter captures the maximum granule integrity score observed among the eosinophil cells in a patch. It indicates the highest proportion of intact granule area relative to total cell area in that patch, offering a normalized benchmark for comparison across cases. |
| ldtn3zkviq9n | NFMTP | MEDIAN_NEUTROPHIL_FIBROTIC_CENTROID_DIST | This parameter measures the median distance from neutrophils to the nearest fibrotic area centroid within a tissue patch. Fibrotic areas are defined by clusters of fibroblasts, tumor cells, macrophages, and plasma cells, which are identified using a clustering approach based on spatial proximity. The process involves computing the centroid for each fibrotic cluster and then determining the Euclidean distance from each neutrophil to its closest fibrotic centroid. The resulting median value, expressed in micrometers, provides a standardized, numeric metric that allows for direct comparisons between different patient cases and patches. |
| leze3737y656 | P | MEAN_NUCLEUS_SHIFT | MEAN_NUCLEUS_SHIFT measures the average distance in micrometers between the nucleus centroid and the overall cell centroid for plasma cells within a patch. This average is computed from individual nucleus shift distances, and it provides a normalized metric that can be used to compare cellular characteristics across different patient cases. |
| leze3737y656 | P | STD_NUCLEUS_SHIFT | STD_NUCLEUS_SHIFT represents the standard deviation of the nucleus shift distances in micrometers among plasma cells in a patch. It quantifies the variability in the displacement measurements, reflecting the consistency of the nucleus positioning within the cells. |
| leze3737y656 | P | MEDIAN_NUCLEUS_SHIFT | MEDIAN_NUCLEUS_SHIFT is the median value of the nucleus shift distances in micrometers for plasma cells within a patch. This measure offers a robust indicator of the central tendency of the displacement values, mitigating the influence of outlier measurements. |
| leze3737y656 | P | MAX_NUCLEUS_SHIFT | MAX_NUCLEUS_SHIFT indicates the maximum nucleus shift distance observed in a patch, measured in micrometers. It reflects the highest displacement between the nucleus centroid and the cell centroid among all plasma cells in that patch. |
| leze3737y656 | P | MIN_NUCLEUS_SHIFT | MIN_NUCLEUS_SHIFT indicates the minimum nucleus shift distance observed in a patch, measured in micrometers. It reflects the smallest displacement between the nucleus centroid and the cell centroid among the plasma cells in that patch. |
| lgmnl29vpyck | LME | ENERGY_DIFFERENCE | ENERGY_DIFFERENCE is a numeric metric that represents the maximum difference in the energy measure across three immune cell types. Energy quantifies texture uniformity derived from the Gray-Level Co-occurrence Matrix (GLCM) computed on processed cell images. Because the energy values are normalized averages from GLCM properties, the difference between the highest and the lowest values among cell types provides a robust, normalized indicator of texture uniformity variation. |
| lgmnl29vpyck | LME | CONTRAST_DIFFERENCE | CONTRAST_DIFFERENCE is a numeric parameter capturing the maximum difference in the contrast metric among immune cell types. Contrast, a GLCM-based feature, reflects local intensity variations within the cell images. The computed difference, based on averaged values across several angles, is normalized and allows for the comparison of intensity variation differences across different patient patches. |
| lgmnl29vpyck | LME | HOMOGENEITY_DIFFERENCE | HOMOGENEITY_DIFFERENCE is defined as the maximum difference in homogeneity values among the three immune cell types. Homogeneity measures the closeness of the distribution of elements in the GLCM to its diagonal, indicating texture smoothness. This normalized, numeric parameter reflects the degree of variation in texture smoothness across the cell types within a given patch. |
| lgmnl29vpyck | LME | DISSIMILARITY_DIFFERENCE | DISSIMILARITY_DIFFERENCE is a numeric metric representing the maximum variation in the dissimilarity values derived from the GLCM analysis of cell images. Dissimilarity measures local intensity differences between neighboring pixels. By computing the difference between the highest and lowest average dissimilarity values among the cell types, this parameter provides a normalized indicator of heterogeneity in local texture patterns. |
| lgmnl29vpyck | LME | CORRELATION_DIFFERENCE | CORRELATION_DIFFERENCE is a numeric parameter that measures the maximum difference in correlation values across immune cell types. The correlation metric, obtained from the GLCM, indicates the linear dependency of gray levels within cell images. The difference between the highest and lowest normalized correlation averages across cell types serves as an indicator of variations in texture structural relationships. |
| lnmu3hgrl40l | TLPMNEF | PROLIFERATION_INDEX | PROLIFERATION_INDEX represents the overall proliferation score computed by normalizing the total detected mitosis-like events by the total number of cells in a patch. This normalization allows for comparison across patches and patient cases regardless of variations in cell counts. |
| lnmu3hgrl40l | TLPMNEF | MITOSIS_FREQ_LYMPHOCYTE | MITOSIS_FREQ_LYMPHOCYTE quantifies the frequency of mitosis-like events specifically in lymphocytes by dividing the number of such events by the total count of lymphocytes in the patch, providing a normalized metric for inter-case comparisons. |
| lnmu3hgrl40l | TLPMNEF | MITOSIS_FREQ_PLASMA | MITOSIS_FREQ_PLASMA denotes the normalized frequency of mitosis-like events within plasma cells, calculated as the ratio of detected events to the overall plasma cell number, thereby allowing consistent comparison across different patches. |
| lnmu3hgrl40l | TLPMNEF | MITOSIS_FREQ_MACROPHAGE | MITOSIS_FREQ_MACROPHAGE indicates the frequency of mitosis-like events in macrophages normalized by the total macrophage count. This measurement offers a standardized assessment of these events independent of raw cell numbers. |
| lnmu3hgrl40l | TLPMNEF | MITOSIS_FREQ_EOSINOPHIL | MITOSIS_FREQ_EOSINOPHIL provides a normalized measure of mitosis-like events in eosinophils by dividing the event count by the overall eosinophil count, ensuring comparability across different patient samples. |
| lnmu3hgrl40l | TLPMNEF | MITOSIS_FREQ_EPITHELIAL | MITOSIS_FREQ_EPITHELIAL reflects the frequency of mitosis-like events in epithelial cells, normalized by the epithelial cell count in the patch. This parameter serves as a comparable metric across various patches. |
| lnmu3hgrl40l | TLPMNEF | MITOSIS_FREQ_CONNECTIVE | MITOSIS_FREQ_CONNECTIVE captures the normalized frequency of mitosis-like events in connective tissue cells by relating the detected events to the total count of these cells, facilitating comparison across cases. |
| lof81ylybiqz | ME | DENSITY_GRADIENT_SLOPE | DENSITY_GRADIENT_SLOPE measures the rate of change in the difference between macrophage and eosinophil densities as a function of distance from the tumor edge. It is obtained by performing a linear regression on the density differences calculated within defined radial bins around the tumor boundary. A positive slope indicates an increase in macrophage density relative to eosinophils, making this measure comparable across different patient cases. |
| lof81ylybiqz | ME | MEAN_DENSITY_DIFFERENCE | MEAN_DENSITY_DIFFERENCE represents the average difference between the densities of macrophages and eosinophils across all defined radial bins. By calculating an average value, this parameter normalizes spatial variations and provides a robust metric for comparing the overall relative distribution of these immune cells across multiple patches. |
| lof81ylybiqz | ME | MAX_DENSITY_DIFFERENCE | MAX_DENSITY_DIFFERENCE captures the maximum absolute difference between the macrophage and eosinophil densities observed within the radial bins. It identifies the peak disparity in immune cell distribution near the tumor boundary, ensuring that localized variations are normalized and can be compared across different tumor regions. |
| lpoqfe9559yd | LMF | RATIO_LYMPHOCYTE_TO_MACROPHAGE | RATIO_LYMPHOCYTE_TO_MACROPHAGE represents the normalized ratio of lymphocytes to macrophages within the tumor stroma. This parameter is computed by dividing the number of lymphocytes by the number of macrophages plus an offset of one to avoid division by zero, making it a comparable metric across different patient cases and patches. |
| lpoqfe9559yd | LMF | RATIO_LYMPHOCYTE_TO_FIBROBLAST | RATIO_LYMPHOCYTE_TO_FIBROBLAST quantifies the normalized ratio of lymphocytes to fibroblasts (using connective tissue cells as a proxy) in the tumor stroma. It is calculated by dividing the lymphocyte count by the fibroblast count plus one, enabling consistent comparisons between different tumor patches and patients. |
| lpoqfe9559yd | LMF | RATIO_MACROPHAGE_TO_FIBROBLAST | RATIO_MACROPHAGE_TO_FIBROBLAST denotes the normalized ratio of macrophages to fibroblasts in the tumor stroma. It is derived by dividing the macrophage count by the fibroblast count plus one, ensuring the metric is properly scaled and comparable across varying conditions. |
| lpoqfe9559yd | LMF | COMPOSITE_IMMUNE_STROMAL_SCORE | COMPOSITE_IMMUNE_STROMAL_SCORE is an integrative metric that provides an overall score of immune-stromal interactions within the tumor. It is computed as the arithmetic mean of the three normalized ratios (lymphocyte-to-macrophage, lymphocyte-to-fibroblast, and macrophage-to-fibroblast), offering a consolidated numerical value to assess the balance between immune cells and stromal components. |
| lzph0woaokto | TPLME | MEDIAN_GRANULARITY_DIFFERENCE | MEDIAN_GRANULARITY_DIFFERENCE is a numeric, normalized parameter that quantifies the heterogeneity of cytoplasmic granularity among key cell types within a tissue patch. It is derived by first computing a texture-based granularity metric for each cell based on the mean absolute Laplacian response computed within the cell’s mask. For each of the five cell types—Tumor, Plasma, Lymphocyte, Macrophage, and Eosinophil—the median granularity is calculated, and then all pairwise absolute differences between these medians are computed. The final parameter is the median of these differences, reflecting the intra-patch heterogeneity in granularity that can be compared across different patient cases. |
| m9koi1aqp24n | TLPMNEF | MEAN_SPINDLE_DEV_TUMOR | MEAN_SPINDLE_DEV_TUMOR represents the average absolute deviation (in degrees) of the spindle orientations of mitosis-like cells from the median spindle orientation within tumor regions. This metric is computed per patch, capturing the typical deviation in orientation and thus allows for direct comparison across different patients by standardizing measurements over local tumor areas. |
| m9koi1aqp24n | TLPMNEF | STD_SPINDLE_DEV_TUMOR | STD_SPINDLE_DEV_TUMOR quantifies the variability or spread of the spindle orientation deviations among mitosis-like cells in tumor regions. As the standard deviation of these deviations is computed per patch, it indicates the consistency of spindle orientation within the tumor compartment and is a normalized metric that enables comparisons between patient cases. |
| m9koi1aqp24n | TLPMNEF | MEAN_SPINDLE_DEV_STROMA | MEAN_SPINDLE_DEV_STROMA is the average absolute deviation (in degrees) of the spindle orientations from the median orientation in stroma regions. By aggregating the measured deviations of cells within each stroma compartment per patch, this parameter reflects local orientation variability and provides a normalized measure to compare different patient samples. |
| m9koi1aqp24n | TLPMNEF | STD_SPINDLE_DEV_STROMA | STD_SPINDLE_DEV_STROMA measures the dispersion of spindle orientation deviations of cells in stroma regions. This normalized statistical parameter, computed for each patch, reveals the level of heterogeneity in cell orientation within the stroma and is suitable for cross-patient comparisons. |
| ma5xh94j58ag | TE | AVG_TUMOR_NUCLEAR_AREA | AVG_TUMOR_NUCLEAR_AREA measures the average nuclear area of tumor cells within a 1x1 mm patch. The value is derived by computing the nuclear area (converted into square micrometers) for each tumor cell and then averaging these values. This normalization allows comparison across different patients. |
| ma5xh94j58ag | TE | AVG_EOS_NUCLEAR_AREA | AVG_EOS_NUCLEAR_AREA measures the average nuclear area of eosinophils within a patch. It is calculated by converting the nuclear area of each eosinophil into square micrometers and then obtaining the mean. This averaged metric enables reliable comparisons between different samples. |
| ma5xh94j58ag | TE | NUCLEAR_AREA_DIFFERENCE | NUCLEAR_AREA_DIFFERENCE represents the difference between the average nuclear area of tumor cells and that of eosinophils within a patch. It is determined by subtracting AVG_EOS_NUCLEAR_AREA from AVG_TUMOR_NUCLEAR_AREA. This normalized metric is useful for comparing cellular behavior across different patient cases. |
| mzyyd3bipg2e | TPF | NUCLEAR_SHAPE_ROUNDNESS_VARIANCE | NUCLEAR_SHAPE_ROUNDNESS_VARIANCE quantifies the heterogeneity in nuclear circularity across three cell populations within a tumor patch. It is computed by extracting the roundness (a shape descriptor defined as 4π*(area)/(perimeter^2)) for each cell type and then calculating the variance of the mean roundness values between these groups. This variance index is numeric and normalized, allowing for direct comparison across different patient cases and patches. |
| mzyyd3bipg2e | TPF | NUCLEAR_SHAPE_ELONGATION_VARIANCE | NUCLEAR_SHAPE_ELONGATION_VARIANCE measures the heterogeneity in the nuclear elongation among three cell populations within a tumor patch. Elongation is evaluated as the ratio derived from the principal axes of the cell nucleus shape, reflecting the degree of elongation for each cell. The parameter is calculated as the variance of the average elongation values across the three groups, providing a normalized, numeric indicator of shape heterogeneity between cell types. |
| mzzkv9m2h3np | TFN | NUCLEAR_AREA_VARIANCE | NUCLEAR_AREA_VARIANCE measures the variation in nuclear areas (in square micrometers) among tumor cells, fibroblasts, and neutrophils within each analyzed tumor patch. This parameter is computed by first identifying cells of interest based on their cell type and location within the tumor or stroma compartments, then calculating each cell’s nuclear area using a scaled conversion from the polygon area. The variance is then derived from all these values, providing an indication of heterogeneity in nuclear morphology that may be associated with tumor aggressiveness and a disordered tumor microenvironment. |
| n1c89xlzvnye | TLFNP | FIBROBLAST_HUE_SKEWNESS | FIBROBLAST_HUE_SKEWNESS measures the skewness of the hue distribution of fibroblast cytoplasmic regions in tumor patches. It is computed by converting fibroblast cell images from RGB to HSV, extracting the hue channel, aggregating the hue values across fibroblasts, and calculating the skewness of this distribution. This parameter is normalized as it represents a dimensionless measure of distribution asymmetry, making it suitable for inter-patient comparisons. |
| n661htp4m7a9 | TLNMF | CONTRAST_VARIANCE_ALL | CONTRAST_VARIANCE_ALL measures the variance of GLCM contrast values calculated from the nuclear regions of all analyzed target cells within a patch. It reflects the heterogeneity of nuclear texture across the entire cellular composition of a tumor region. |
| n661htp4m7a9 | TLNMF | CONTRAST_VARIANCE_TUMOR | CONTRAST_VARIANCE_TUMOR measures the variance in GLCM contrast values computed exclusively from tumor cells. This parameter indicates the variability of nuclear textural features within tumor regions. |
| n661htp4m7a9 | TLNMF | CONTRAST_VARIANCE_STROMA | CONTRAST_VARIANCE_STROMA measures the variance of GLCM contrast values for cells located in stromal regions. It provides insight into the heterogeneity of nuclear texture among non-tumor supportive cells. |
| n661htp4m7a9 | TLNMF | MEAN_CONTRAST_TUMOR | MEAN_CONTRAST_TUMOR represents the average GLCM contrast value for tumor cells' nuclear regions, offering a summary metric of the typical nuclear texture characteristic of these cells. |
| n661htp4m7a9 | TLNMF | MEAN_CONTRAST_LYMPHO | MEAN_CONTRAST_LYMPHO represents the average GLCM contrast value calculated from lymphocyte nuclear regions, providing insights into the nuclear texture features of immune cells infiltrating the tumor microenvironment. |
| n661htp4m7a9 | TLNMF | MEAN_CONTRAST_NEUTRO | MEAN_CONTRAST_NEUTRO represents the mean GLCM contrast value determined for neutrophils, reflecting the overall nuclear textural characteristic of these inflammatory cells. |
| n661htp4m7a9 | TLNMF | MEAN_CONTRAST_MACRO | MEAN_CONTRAST_MACRO represents the average GLCM contrast value derived from macrophage nuclei, summarizing the typical textural features associated with these immune cells. |
| n661htp4m7a9 | TLNMF | MEAN_CONTRAST_FIBRO | MEAN_CONTRAST_FIBRO represents the mean GLCM contrast value computed from fibroblast nuclei, providing a measure of nuclear texture that is characteristic of the stromal connective tissue cells involved in the tumor microenvironment. |
| nvagkfyscxur | E | PYKNOTIC_FREQUENCY | PYKNOTIC_FREQUENCY represents the fraction of eosinophils with pyknotic nuclei relative to the total number of eosinophils in a tumor patch. This parameter is derived by identifying pyknotic nuclei based on a set intensity threshold (with lower mean pixel intensity indicating increased nuclear condensation), and computing the ratio of cells meeting this criterion. As a normalized value, it allows comparison across different patient cases by mitigating the influence of raw cell counts. |
| nvagkfyscxur | E | MEAN_NUCLEUS_INTENSITY | MEAN_NUCLEUS_INTENSITY is a numeric measure that reflects the average intensity of the eosinophil nuclei within a patch. The intensity is calculated from grey-scale pixel values using a nucleus mask, providing an average measure of nuclear staining. Lower mean intensity values indicate darker staining, which can be associated with nuclear condensation. This parameter is normalized in that it represents an average value, making it suitable for cross-case comparisons. |
| nwywe3c7kipv | EL | EOS_MEAN_R | EOS_MEAN_R is the average intensity value of the red color channel extracted from eosinophils located in the intratumoral compartment of a patch. This value is calculated by selecting the valid cell pixels using the cell mask and averaging their red channel intensities, making it a normalized metric that can be compared across different patient cases. |
| nwywe3c7kipv | EL | EOS_MEAN_G | EOS_MEAN_G is the average intensity value of the green color channel for eosinophils within the tumor region of a patch. It is derived by averaging the green channel values of the pixels identified within each eosinophil, ensuring a normalized and numeric measurement. |
| nwywe3c7kipv | EL | EOS_MEAN_B | EOS_MEAN_B is the computed average intensity of the blue color channel for eosinophils in the intratumoral area of a patch. The calculation involves selecting valid pixels using the cell mask and averaging their blue channel intensities, yielding a standardized numeric value. |
| nwywe3c7kipv | EL | LYM_MEAN_R | LYM_MEAN_R is the average red channel intensity value for lymphocytes found in the intratumoral compartment of a patch. It is determined by averaging the red values of the valid cellular pixels from lymphocytes, resulting in a normalized numeric parameter. |
| nwywe3c7kipv | EL | LYM_MEAN_G | LYM_MEAN_G is the average green channel intensity value for lymphocytes within the tumor region of a patch. The value is obtained by averaging the green channel pixel intensities for the valid parts of lymphocyte cells, ensuring it is a normalized and comparable measurement. |
| nwywe3c7kipv | EL | LYM_MEAN_B | LYM_MEAN_B is the computed average blue channel intensity for lymphocytes in the intratumoral region of a patch. The measurement is made by averaging the blue channel values from valid cell pixel data, providing a standardized and numeric metric. |
| nwywe3c7kipv | EL | COLOR_DIFFERENCE | COLOR_DIFFERENCE is the Euclidean distance between the average RGB color values of eosinophils and lymphocytes within each patch. It is calculated by taking the differences of the individual red, green, and blue channel means for the two cell types, squaring these differences, summing them, and taking the square root. This parameter quantifies the color difference in a normalized numeric value that can be used for cross-patient comparison. |
| nxs44adea7hb | T | JUNCTION_FRAGMENTATION_INDEX | JUNCTION_FRAGMENTATION_INDEX: This normalized metric measures the ratio of discontinuous cell-cell junctions to the total number of potential junction contacts within each tumor patch. It provides a value between 0 and 1, allowing for comparisons across different patient cases by reflecting the degree of fragmentation in tumor cell boundaries. |
| nxs44adea7hb | T | MEAN_JUNCTION_LENGTH_UM | MEAN_JUNCTION_LENGTH_UM: This parameter represents the average length of the detected cell-cell junctions in micrometers for each patch. It is a normalized measure since it averages the junction lengths, making it suitable for comparing different tumor regions across patient cases. |
| o5858rxcdxxd | LE | TUMOR_CENTER_X | TUMOR_CENTER_X is the computed x-coordinate of the tumor center within a given patch. It is determined by averaging the x-coordinates of tumor cells, which are selected based on their inclusion in the tumor mask and classification as epithelial. This coordinate serves as a spatial reference for defining concentric rings for further radial analysis. |
| o5858rxcdxxd | LE | TUMOR_CENTER_Y | TUMOR_CENTER_Y is the corresponding y-coordinate of the tumor center calculated from the average of the y-coordinates of the tumor cells in the patch. This parameter establishes the vertical spatial location of the tumor center, essential for anchoring the radial binning process used in subsequent analyses. |
| o5858rxcdxxd | LE | LYMPHO_GRADIENT | LYMPHO_GRADIENT represents the infiltration velocity gradient for lymphocytes. It is derived by partitioning the patch into concentric rings, calculating the density of lymphocytes in each ring (by normalizing cell counts by the area of the ring), and fitting a linear regression to the density profile. The resulting slope, converted to cells per cubic micrometer, quantifies how lymphocyte density changes from the tumor center to the periphery. |
| o5858rxcdxxd | LE | EOSINO_GRADIENT | EOSINO_GRADIENT quantifies the infiltration velocity gradient for eosinophils using a similar process. Concentric rings are used to compute eosinophil densities, and a linear regression is fitted to these densities. The slope of this regression, also scaled to cells per cubic micrometer, indicates the rate at which eosinophil density decreases or increases radially. |
| o5858rxcdxxd | LE | LYMPHO_R2 | LYMPHO_R2 is the R-squared value obtained from the linear regression analysis for lymphocyte density. This parameter measures the goodness-of-fit of the regression model, indicating how well the linear gradient reflects the variation in lymphocyte density across the defined radial bins. |
| o5858rxcdxxd | LE | EOSINO_R2 | EOSINO_R2 is the R-squared value for the linear regression model applied to the eosinophil density profile. It serves as a statistical measure of the fit's quality, reflecting the extent to which the linear model captures the radial distribution of eosinophil densities in the patch. |
| ob5oi7kex13w | TLPMNEF | MEAN_PROANGIO_SCORE | MEAN_PROANGIO_SCORE represents the average composite pro-angiogenic morphology score calculated for a given patch. It is derived by combining two normalized shape descriptors (elongation and perimeter irregularity) from fibroblast and macrophage cells to produce a score on a 0-1 scale, thereby enabling robust comparison across different patient cases. |
| ob5oi7kex13w | TLPMNEF | STD_PROANGIO_SCORE | STD_PROANGIO_SCORE indicates the standard deviation of the composite pro-angiogenic scores within a patch. It measures the variability in the pro-angiogenic morphologic configurations among the cell population, reflecting the heterogeneity in cell shape characteristics on a normalized scale. |
| ob5oi7kex13w | TLPMNEF | MAX_PROANGIO_SCORE | MAX_PROANGIO_SCORE is the highest observed composite pro-angiogenic score within a patch. It identifies the cell in the patch that exhibits the most pronounced pro-angiogenic morphological features, based on normalized shape descriptors. |
| ob5oi7kex13w | TLPMNEF | MIN_PROANGIO_SCORE | MIN_PROANGIO_SCORE is the lowest observed composite pro-angiogenic score within a patch. It highlights the cell with the least pro-angiogenic morphological characteristics among fibroblasts and macrophages, ensuring a normalized measurement for comparative analysis. |
| onaljturxh0j | F | NORMALIZED_MITOTIC_RATE | NORMALIZED_MITOTIC_RATE is a normalized numeric parameter representing the ratio of mitotic figures detected in fibroblast nuclei to the total number of fibroblasts within a tissue patch. This parameter is calculated by dividing the total count of mitotic figures by the total count of fibroblasts, allowing for meaningful comparisons across different patient cases and patches as it adjusts for variations in cell numbers. It reflects the rate of mitotic activity in the fibroblasts, which is significant in analyzing tumor stroma activation and potential tumor progression. |
| oua58wg9o0gs | TLPMNF | ASYMMETRY_SCORE | ASYMMETRY_SCORE is a normalized metric that quantifies the asymmetry in the immune-to-fibroblast cell ratios across defined radial sectors surrounding the tumor boundary. It is calculated by taking the difference between the maximum and minimum sector ratios divided by the mean sector ratio, thereby providing a comparative measure across different patient cases. |
| oua58wg9o0gs | TLPMNF | MAX_SECTOR_RATIO | MAX_SECTOR_RATIO represents the highest calculated immune-to-fibroblast ratio among all the sectors around the tumor. It serves to identify the region with the most pronounced relative immune cell infiltration compared to fibroblast presence. |
| oua58wg9o0gs | TLPMNF | MIN_SECTOR_RATIO | MIN_SECTOR_RATIO indicates the lowest immune-to-fibroblast ratio observed among the sectors. This metric highlights the sector where the relative contribution of immune cells is the smallest in comparison to fibroblasts, offering insight into local variations in cell distribution. |
| oua58wg9o0gs | TLPMNF | MEAN_SECTOR_RATIO | MEAN_SECTOR_RATIO is the average value of the immune-to-fibroblast ratios calculated over all sectors. It provides an overall balanced measure of immune and fibroblast cell distribution around the tumor, making it useful for comparing different patches across patient cases. |
| p200sfpkjyu7 | P | KARYORRHEXIS_INDEX | The parameter 'KARYORRHEXIS_INDEX' quantifies the ratio of plasma cells that exhibit nuclear fragmentation within a tumor patch. It is calculated by dividing the count of plasma cells with fragmented nuclei (as determined by a solidity measure below a fixed threshold) by the total number of plasma cells in that patch. This normalized value, which ranges from 0 to 1, allows for meaningful comparisons across different patient cases and tumor regions, making it a robust metric for assessing nuclear fragmentation in plasma cells. |
| p3lr4g8ah9je | L | LYMPHO_CIRC_STD | LYMPHO_CIRC_STD measures the variability in the nuclear circularity of lymphocytes within each patch. It is computed as the standard deviation of the circularity values, which are derived from the geometry of each lymphocyte's nucleus. This metric is normalized in that it reflects relative dispersion and allows for comparison across different patches and patient cases. |
| p3lr4g8ah9je | L | LYMPHO_CIRC_MEAN | LYMPHO_CIRC_MEAN represents the average nuclear circularity of lymphocytes in a patch. By calculating the mean value of the circularity measurements, this parameter provides a baseline indicator of the typical nuclear shape of lymphocytes that can be compared across different tumor samples. |
| pdrbf8rthyas | N | MEAN_RUFFLE_INDEX | MEAN_RUFFLE_INDEX represents the average ruffling index calculated from individual neutrophils in a patch. The ruffling index is a normalized, dimensionless value computed as the ratio of the actual cell membrane contour length to its convex hull length minus one. This parameter quantifies the extent of membrane ruffling and is useful for comparing neutrophil activation states across different tumor patches. |
| pdrbf8rthyas | N | SD_RUFFLE_INDEX | SD_RUFFLE_INDEX represents the standard deviation of the ruffling indices within a patch, reflecting the variability in neutrophil membrane ruffling. Being a normalized numeric measure, it allows for comparisons of the consistency or diversity of cell morphological features across different patches and patient cases. |
| pq8ltdtkoyvy | LPNE | PROXIMAL_FRACTION | PROXIMAL_FRACTION measures the fraction of lymphocytes that are within a defined proximity (50μm) of immune effector cells (plasma cells, neutrophils, and eosinophils) within each tumor patch. This ratio, derived from the number of proximal lymphocytes divided by the total lymphocyte count, is normalized and thus allows for meaningful comparisons across different patient cases. |
| pq8ltdtkoyvy | LPNE | RIPLEY_K | RIPLEY_K quantifies the spatial clustering of proximal lymphocytes using a normalized Ripley’s K function. By incorporating the constant patch area and adjusting for the number of proximal lymphocytes, this metric provides a scale-invariant measure of cell clustering behavior, enabling comparisons across tumor patches and patient samples. |
| pshmp2v0um9b | PF | INFILTRATION_GRADIENT | INFILTRATION_GRADIENT measures the rate at which the combined density of fibroblasts and plasma cells decreases with distance from the tumor center. It is computed by fitting a linear regression line to the cell density values calculated in concentric annular bins around the tumor center. This slope provides insight into the spatial gradient of cell infiltration, helping to compare tumor patches across different patient cases. |
| pshmp2v0um9b | PF | MEAN_COMBINED_DENSITY | MEAN_COMBINED_DENSITY is the average density of fibroblasts and plasma cells across all defined radial bins within a tumor patch. It is calculated by normalizing the cell counts in each annular ring by the area of that ring, ensuring that the metric is comparable across different patches. This parameter provides a summary metric of the overall cell density in the examined region. |
| pshmp2v0um9b | PF | MAX_DISTANCE_UM | MAX_DISTANCE_UM represents the maximum distance, measured in micrometers, from the tumor center to the farthest detected fibroblast or plasma cell in a patch. By determining the furthest radial extent of cell presence, this parameter quantifies the spatial spread of the examined cells and is normalized to a physical length scale, enabling comparisons between patches. |
| pthkkxs1bzg6 | T | NUCLEAR_GRANULARITY_MEAN | NUCLEAR_GRANULARITY_MEAN represents the average chromatin granularity across tumor cell nuclei within a patch. It is computed by first quantifying local intensity variations in the nucleus using a Laplacian filter applied to the hematoxylin channel, and then taking the mean of the standard deviation values from each cell. This process yields a normalized, numeric metric suitable for comparing tumor patches across different patient cases. |
| pthkkxs1bzg6 | T | NUCLEAR_GRANULARITY_STD | NUCLEAR_GRANULARITY_STD measures the variability or heterogeneity of chromatin granularity among tumor cell nuclei within a patch. It is derived by calculating the standard deviation of the individual granularity scores from the cells, reflecting the degree of dispersion in nuclear texture features. As a normalized numeric value, it ensures that differences between patches are comparable. |
| pthkkxs1bzg6 | T | NUCLEAR_GRANULARITY_MAX | NUCLEAR_GRANULARITY_MAX captures the highest observed granularity score in a patch, indicating the presence of tumor cells with exceptionally heterogeneous chromatin patterns. This parameter is computed as the maximum value from the set of granularity scores of all tumor cells in the patch, providing a normalized numeric metric for identifying extreme cellular characteristics. |
| pv87oyavj70f | TN | TUMOR_NEUTRO_INTENSITY_DIFF | TUMOR_NEUTRO_INTENSITY_DIFF measures the difference between the average color intensities of tumor cells and neutrophils in invasive areas, providing a normalized metric that can be compared across different patient cases. It reflects the contrast in staining within the invasion fronts which may be indicative of active immune-tumor interactions. |
| pv87oyavj70f | TN | MEAN_TUMOR_INTENSITY | MEAN_TUMOR_INTENSITY represents the average pixel intensity of tumor cells within the invasive front regions. This value is computed by averaging normalized intensity measurements from the cell nuclei, allowing for cross-sample comparisons. |
| pv87oyavj70f | TN | MEAN_NEUTRO_INTENSITY | MEAN_NEUTRO_INTENSITY represents the average pixel intensity of neutrophils within the invasive front regions. It is calculated by averaging normalized intensity measurements from the neutrophils’ nuclei, ensuring the metric is comparable across different cases. |
| pwgu5xkrz79g | P | NUCLEOLAR_VARIATION_INDEX | NUCLEOLAR_VARIATION_INDEX measures the heterogeneity of nucleolar sizes within a tumor patch by computing the ratio of the standard deviation to the mean of the nucleolar areas detected in plasma cells. This normalized metric is designed to allow comparisons across different patient cases by accounting for variations relative to the average nucleolar size. |
| pwgu5xkrz79g | P | MEAN_NUCLEOLAR_AREA_UM2 | MEAN_NUCLEOLAR_AREA_UM2 represents the average area of the segmented nucleoli in square micrometers within a tumor patch. This measurement is derived from processing the segmented nucleoli areas after image enhancement and morphological operations, ensuring that the metric is normalized and comparable across varying samples. |
| pwgu5xkrz79g | P | STD_NUCLEOLAR_AREA_UM2 | STD_NUCLEOLAR_AREA_UM2 quantifies the spread or variability in the sizes of nucleoli within a tumor patch by calculating the standard deviation of nucleolar areas, measured in square micrometers. As a normalized parameter, it reflects the degree of dispersion in nucleolar sizes which may be linked to heterogeneity in cellular function. |
| pwj08z9nojkn | MTFNP | SKEWNESS_ASPECT_RATIO | SKEWNESS_ASPECT_RATIO represents the statistical skewness of the distribution of macrophage nuclear aspect ratios within a patch. It quantifies the asymmetry in the measurements of the ratio of the major axis to the minor axis of macrophage nuclei, offering insights into variability in nuclear shape that may be linked to tumor progression. This parameter is computed only for patches that have sufficient macrophages and relevant cell type enrichment, ensuring its comparability across different patient cases. |
| pwj08z9nojkn | MTFNP | MEAN_ASPECT_RATIO | MEAN_ASPECT_RATIO is the average of the individual aspect ratios computed from the segmented macrophage nuclei in a patch. It provides a summary statistic for the typical nuclear shape within that patch. Since the aspect ratio is calculated as the ratio of the longer edge to the shorter edge of the minimum rotated rectangle enclosing each nucleus, the resulting mean offers a normalized measure that can be compared across different patient cases. |
| py4v9uchwd72 | TE | DENSITY_GRADIENT | DENSITY_GRADIENT quantifies the change in eosinophil density from the tumor center towards its periphery. This parameter is calculated by performing a linear regression on the eosinophil densities measured in concentric rings around the tumor center, where the regression slope (adjusted for proper spatial scaling) represents the gradient. It is normalized with respect to distance and area, making it comparable across different patient cases. |
| py4v9uchwd72 | TE | MEAN_DENSITY | MEAN_DENSITY represents the average eosinophil density across all concentric rings defined within a tumor patch. The density in each ring is determined by dividing the count of eosinophils by the ring's area, and the overall mean is computed by averaging these normalized values. This parameter is numeric and normalized, allowing uniform comparison among different tumor patches. |
| py4v9uchwd72 | TE | OUTER_RING_DENSITY | OUTER_RING_DENSITY measures the normalized eosinophil density in the outermost ring of a tumor patch. It is calculated by dividing the number of eosinophils found in the periphery by the area of this ring. This normalization by area ensures that the parameter is quantitative and comparable across varying patch sizes and patient cases. |
| pysqdv5wqqpw | TLPMNEF | MEAN_CONVOLUTION_SCORE | MEAN_CONVOLUTION_SCORE: This parameter represents the average membrane convolution score of cells within a patch. It is calculated by first computing the convolution score for each individual cell—derived from the standard deviation of local curvature fluctuations along the cell's polygonal outline—and then averaging these scores over all cells in the patch. This normalized metric allows for comparison across different patient cases. |
| pysqdv5wqqpw | TLPMNEF | MEDIAN_CONVOLUTION_SCORE | MEDIAN_CONVOLUTION_SCORE: This metric denotes the median value of the membrane convolution scores for cells in a patch. Like the mean score, individual convolution scores are computed based on the variability of the local curvature along the cell membrane, but the median is used here to provide a robust central tendency measure that is less affected by outliers. It is a normalized parameter suitable for cross-patient analysis. |
| pysqdv5wqqpw | TLPMNEF | STD_CONVOLUTION_SCORE | STD_CONVOLUTION_SCORE: This parameter quantifies the variability or dispersion of the cell membrane convolution scores within a patch. It is computed as the standard deviation of the individual cell convolution scores, which are obtained from the fluctuations in curvature along the cell boundary. This normalized measure helps in understanding the heterogeneity of membrane convolution characteristics across different cells within the same patch. |
| pzm4xozmu5qc | TLPMNEF | OVERLAP_SCORE | OVERLAP_SCORE represents the overall spatial overlap integral of normalized density distributions across all cell types within a patch. It is derived by normalizing the density heat maps computed for individual cell types using kernel density estimation, then calculating the element-wise multiplication of these normalized maps over a defined grid. The final score, obtained by summing the values of the resulting product matrix, provides a comprehensive measure of the spatial concomitance of cell-type densities. |
| pzm4xozmu5qc | TLPMNEF | MAX_DENSITY_OVERLAP | MAX_DENSITY_OVERLAP indicates the highest local density overlap value within the patch. It identifies the grid location where the simultaneous high densities of different cell types reach their peak, reflecting a probable hotspot of cellular interaction. |
| pzm4xozmu5qc | TLPMNEF | MEAN_DENSITY_OVERLAP | MEAN_DENSITY_OVERLAP captures the average density overlap over the entire patch. This parameter summarizes the typical level of overlapping density across the grid, offering insight into the general spatial co-localization pattern of the cell types. |
| pzm4xozmu5qc | TLPMNEF | STD_DENSITY_OVERLAP | STD_DENSITY_OVERLAP reflects the variability or dispersion in the density overlap values across the patch. By computing the standard deviation of the overlap matrix, it quantifies how consistently the cell-type densities overlap throughout the patch. |
| q3nz9he4lwbf | TLPMNEF | TOTAL_ENERGY_SCORE | TOTAL_ENERGY_SCORE is the aggregated surrogate energy score computed for each patch. It quantifies the total deviation of the observed cell-cell distances from an idealized packing configuration (where each pair-wise distance ideally equals the sum of the effective cell radii). Because the analysis is conducted on standardized patches, this cumulative energy measurement is normalized for comparing tumor regions from different patients. |
| q3nz9he4lwbf | TLPMNEF | MEAN_CELL_RADIUS_UM | MEAN_CELL_RADIUS_UM represents the average effective radius of the cells within a patch, computed from each cell's polygonal area and then converted to micrometers. This parameter gives a normalized measure of typical cell size across different tumor regions, facilitating inter-patient comparisons. |
| q3nz9he4lwbf | TLPMNEF | MEAN_PAIR_ENERGY | MEAN_PAIR_ENERGY is the average energy contribution per cell pair in the patch, calculated as the average squared deviation between the observed inter-cell distances and the ideal contact distances (the sum of the cell radii). It provides a normalized metric that reflects the typical local deviation from perfect cellular packing. |
| q3nz9he4lwbf | TLPMNEF | STD_PAIR_ENERGY | STD_PAIR_ENERGY captures the variability in the energy contributions of cell pairs within the patch. It measures the dispersion around the mean pair energy, indicating the heterogeneity in local cell packing deviations. This variability, calculated consistently across standardized patches, is suitable for comparative analyses between patient cases. |
| q6af3xhnfium | TE | TUMOR_NUCLEAR_KURTOSIS_MEAN | TUMOR_NUCLEAR_KURTOSIS_MEAN represents the mean kurtosis calculated from the distribution of radial distances from each vertex of tumor cell nuclei contours to their centroids in the peritumoral region. This metric provides a normalized measure of nuclear shape irregularity that can be compared across different patient cases. |
| q6af3xhnfium | TE | EOS_NUCLEAR_KURTOSIS_MEAN | EOS_NUCLEAR_KURTOSIS_MEAN represents the mean kurtosis calculated from the radial distance distribution of eosinophil nuclear boundaries in the peritumoral region. It quantifies the irregularity in the nuclear shape of eosinophils and serves as a normalized parameter for comparative analysis. |
| q6af3xhnfium | TE | COMBINED_KURTOSIS_METRIC | COMBINED_KURTOSIS_METRIC is the arithmetic average of the tumor nuclear kurtosis mean and the eosinophil nuclear kurtosis mean. By integrating these two normalized measurements, it offers a singular metric that reflects the overall irregularity in nuclear boundaries within the peritumoral microenvironment. |
| qgesprtxleri | TLPMNEF | MEMBRANE_HOMOGENEITY_TUMOR | MEMBRANE_HOMOGENEITY_TUMOR quantifies the average standard deviation of membrane pixel intensities for tumor (epithelial) cells within a given patch. This metric is derived by isolating the membrane region through dilating the nucleus boundary and subtracting the original nucleus mask, then computing the texture homogeneity via the standard deviation of the pixel values, and finally averaging these values across all tumor cells in the patch. It is normalized at the patch level, allowing comparisons between different patient cases. |
| qgesprtxleri | TLPMNEF | MEMBRANE_HOMOGENEITY_LYMPHO | MEMBRANE_HOMOGENEITY_LYMPHO measures the average variability in membrane pixel intensity for lymphocyte cells in a patch. The analysis involves extracting the membrane region from each lymphocyte, computing a standard deviation for its pixel intensities, and then calculating the mean of these deviations over all lymphocytes in the patch. This provides a normalized metric that can be contrasted across different samples. |
| qgesprtxleri | TLPMNEF | MEMBRANE_HOMOGENEITY_PLASMA | MEMBRANE_HOMOGENEITY_PLASMA reflects the mean standard deviation of the membrane pixel intensities for plasma cells in the patch. The process involves segmenting the membrane region from each plasma cell, measuring the standard deviation as an indicator of homogeneity, and averaging these measurements over all plasma cells. The result is a normalized parameter useful for comparative analysis. |
| qgesprtxleri | TLPMNEF | MEMBRANE_HOMOGENEITY_NEUTRO | MEMBRANE_HOMOGENEITY_NEUTRO determines the mean variability in membrane intensity for neutrophil cells within a patch by computing the standard deviation of pixel values in the delineated membrane region for each cell and averaging the values across all neutrophils. This normalized metric allows reliable comparison between patches from different patient cases. |
| qgesprtxleri | TLPMNEF | MEMBRANE_HOMOGENEITY_MACRO | MEMBRANE_HOMOGENEITY_MACRO computes the average standard deviation of membrane pixel intensities for macrophage cells in the patch. The homogeneity measure is obtained by isolating the membrane region via a dilation approach, calculating the standard deviation of pixel intensities for each macrophage, and then averaging these values. The normalization at the patch level ensures comparability across different samples. |
| qgesprtxleri | TLPMNEF | MEMBRANE_HOMOGENEITY_EOSINO | MEMBRANE_HOMOGENEITY_EOSINO quantifies the mean standard deviation of membrane pixel intensities for eosinophil cells within a patch. By extracting the membrane region from each eosinophil, computing its standard deviation, and averaging these values, this parameter provides a normalized measure of membrane homogeneity that can be used to compare tissue patches. |
| qgesprtxleri | TLPMNEF | MEMBRANE_HOMOGENEITY_FIBRO | MEMBRANE_HOMOGENEITY_FIBRO represents the mean standard deviation of membrane pixel intensities for fibroblast (connective) cells in the patch. It is calculated by identifying the membrane region for each fibroblast cell, determining the texture variability through the standard deviation of pixel intensities, and then averaging these results across the patch. This normalized parameter is designed for comparative analysis across different patient tissue samples. |
| qkgk1z0gwe4v | LM | LYMPHOCYTE_PAIRED_FRACTION | LYMPHOCYTE_PAIRED_FRACTION represents the fraction of lymphocytes in a tissue patch that are paired with at least one macrophage. This pairing is determined by checking if a lymphocyte’s centroid is within a maximum distance threshold of 50 micrometers from any macrophage’s centroid. The resulting fraction, calculated as the number of paired lymphocytes divided by the total lymphocytes in the patch, provides a normalized measure suitable for comparing cell-cell interactions across different patient cases. |
| qkgk1z0gwe4v | LM | MACROPHAGE_PAIRED_FRACTION | MACROPHAGE_PAIRED_FRACTION represents the fraction of macrophages in a tissue patch that are paired with at least one lymphocyte. Similar to the lymphocyte parameter, pairing is defined when the distance between a macrophage and a lymphocyte is less than or equal to 50 micrometers. This fraction, obtained by dividing the number of paired macrophages by the total number of macrophages in the patch, is normalized and numeric, allowing for comparative analyses across patient samples. |
| qkgk1z0gwe4v | LM | MEAN_PAIR_DISTANCE_UM | MEAN_PAIR_DISTANCE_UM measures the average distance in micrometers between paired lymphocytes and macrophages within a tissue patch. Only distances that meet the maximum pairing threshold (50 micrometers) are considered. The distances are initially computed in pixels and then converted into micrometers using a predefined conversion factor. This numeric measure reflects the typical spatial separation between interacting immune cells, providing a standardized metric for comparison. |
| qpck58ubf34g | M | MACROPHAGE_STROMA_TO_TUMOR_RATIO | MACROPHAGE_STROMA_TO_TUMOR_RATIO represents the normalized measure that compares the number of macrophages in the stromal compartment to the number in the tumor compartment for each patch. It is calculated by dividing the stromal macrophage count by the tumor macrophage count, thereby adjusting for differences in raw cell counts and allowing direct comparison across different patient cases. This metric provides insights into the balance of macrophage distribution between compartments and its potential implications on the immunosuppressive microenvironment. |
| qpem375t1s28 | TN | WAVELET_ENTROPY_MEAN | WAVELET_ENTROPY_MEAN measures the average wavelet entropy across all analyzed tumor-neutrophil interfaces within a patch. This parameter quantifies the complexity and chaotic nature of intensity patterns along the cellular interface, derived from the energy distribution obtained via a discrete wavelet transform of the intensity profiles extracted from the image patches. Its normalization allows for a fair comparison across different patient cases. |
| qpem375t1s28 | TN | WAVELET_ENTROPY_STD | WAVELET_ENTROPY_STD quantifies the variability in the wavelet entropy measurements among the tumor-neutrophil interfaces within a patch. It serves as an indicator of the heterogeneity in interface complexity, reflecting differences in the chaotic invasion patterns across different interfaces. This measure is also normalized, making it suitable to compare across various patient samples. |
| qpka6hikt12c | TMNE | NUCLEOLUS_SIZE_SKEWNESS | NUCLEOLUS_SIZE_SKEWNESS represents the skewness of the distribution of nucleolus sizes measured in tumor cells within immune-rich patches. The calculation involves extracting nucleolus area measurements from a set of tumor cells, applying image processing to isolate the nucleolar region, and then using statistical methods to compute the skewness (a dimensionless measure of asymmetry) of the distribution. This normalized metric enables comparison across different patient cases. |
| r24ti8bsf915 | LF | MEAN_INFILTRATION_DISTANCE_UM | The MEAN_INFILTRATION_DISTANCE_UM parameter quantifies the mean Euclidean distance in micrometers between fibroblasts and lymphocytes located at the tumor's invasive front. It is calculated by identifying cells in the stromal compartment that are within a defined proximity (12.5 μm) to tumor cells, computing pairwise distances among these filtered cells, and averaging the results. This parameter is normalized, allowing for comparison across different patient cases. |
| r45nttzftmh3 | TLMNEF | SHANNON_DIVERSITY_INDEX | The parameter 'SHANNON_DIVERSITY_INDEX' measures the diversity of six different cell types (Tumor cells, Lymphocytes, Macrophages, Neutrophils, Eosinophils, and Connective tissue cells) within the tumor region. It is computed on a per-patch basis, where each patch is a defined region (e.g., 1x1 mm), and is derived by first calculating the proportions of each cell type from their counts, and then applying the Shannon diversity index formula. This procedure normalizes the diversity measurement, allowing for comparisons across different patient cases, and results in a numeric value that indicates higher diversity with increased index values. |
| r569qsgntxcr | TLPMNEF | MEAN_NUCLEAR_CYTO_SHIFT | MEAN_NUCLEAR_CYTO_SHIFT represents the average distance (in micrometers) between the centers-of-mass of nuclear and cytoplasmic intensity distributions across all cells in a patch. This measurement is normalized by converting pixel distances to micrometers, ensuring its comparability across different patient cases and tumor regions. |
| r569qsgntxcr | TLPMNEF | MAX_NUCLEAR_CYTO_SHIFT | MAX_NUCLEAR_CYTO_SHIFT is defined as the highest observed distance (in micrometers) between the nuclear and cytoplasmic intensity centers-of-mass among the cells within a patch. It highlights the extreme shift value and helps identify outlier behavior in the cellular spatial organization. |
| r569qsgntxcr | TLPMNEF | STD_NUCLEAR_CYTO_SHIFT | STD_NUCLEAR_CYTO_SHIFT quantifies the variability (standard deviation in micrometers) of the distances between the nuclear and cytoplasmic centers-of-mass within a patch. This parameter serves to reveal the degree of heterogeneity in subcellular spatial shifts across the cell population. |
| r6mdz63uhhcx | MN | AVG_MAC_NEUT_ORIENTATION_DIFF | AVG_MAC_NEUT_ORIENTATION_DIFF represents the average angular difference (in degrees) calculated between the mean nuclear orientation of macrophages and that of neutrophils within the stromal compartment. This metric is normalized within a 0 to 90 degree range, ensuring comparability across different patient cases and tumor regions. It is derived by fitting ellipses to the nuclear outlines of cells and computing the absolute difference between the average orientations of these two cell types in each patch. |
| r6mdz63uhhcx | MN | STD_MAC_NEUT_ORIENTATION_DIFF | STD_MAC_NEUT_ORIENTATION_DIFF measures the standard deviation of all pairwise differences in nuclear orientation between macrophages and neutrophils in the stroma. This parameter quantifies the variability or heterogeneity in the alignment of these cells, providing insight into the consistency of their directional attributes across the analyzed patches. Like the average orientation difference, it is a normalized metric, facilitating comparisons among different patient cases. |
| rb1jlud8h82a | TLPMNEF | MEAN_CRISPNESS_T | MEAN_CRISPNESS_T represents the average gradient magnitude computed along the nuclear envelope boundary of tumor (epithelial) cells within a patch. It is a normalized metric obtained by averaging per-cell crispness values, making it comparable across different patches and patient cases. |
| rb1jlud8h82a | TLPMNEF | MEAN_CRISPNESS_L | MEAN_CRISPNESS_L indicates the average nuclear envelope crispness of lymphocytes in the patch. This value is derived from the mean of the gradient magnitudes measured at the boundaries of lymphocyte nuclei, ensuring a normalized comparison across patches. |
| rb1jlud8h82a | TLPMNEF | MEAN_CRISPNESS_P | MEAN_CRISPNESS_P quantifies the average intensity of edge sharpness (crispness) at the nuclear boundary of plasma cells in a patch. The value is computed by averaging gradient magnitude values along the nuclear perimeter, providing a normalized metric for analysis. |
| rb1jlud8h82a | TLPMNEF | MEAN_CRISPNESS_M | MEAN_CRISPNESS_M is the average crispness measure of the nuclear envelope for macrophages in the patch. Each cell’s crispness is calculated from its boundary gradient values, and the patch-level value is normalized by taking the mean across all macrophages. |
| rb1jlud8h82a | TLPMNEF | MEAN_CRISPNESS_N | MEAN_CRISPNESS_N reflects the mean nuclear envelope crispness for neutrophils in the patch. It results from averaging the gradient magnitude values obtained from edge-detection of the neutrophil nuclei, thereby providing a normalized comparison across different slides. |
| rb1jlud8h82a | TLPMNEF | MEAN_CRISPNESS_E | MEAN_CRISPNESS_E represents the normalized average crispness of eosinophils' nuclear envelopes. It is computed as the mean of the gradient magnitudes along the nuclear boundaries, ensuring consistency and comparability between patches. |
| rb1jlud8h82a | TLPMNEF | MEAN_CRISPNESS_F | MEAN_CRISPNESS_F provides the average nuclear envelope crispness of fibroblast cells within a patch. This is a normalized metric derived by averaging the gradient magnitudes measured along the boundaries of fibroblast nuclei. |
| rc4003zafb9k | E | EOSINOPHIL_MEAN_INTENSITY | EOSINOPHIL_MEAN_INTENSITY measures the average red-channel intensity of the cytoplasm in eosinophils within a given patch. It is derived by isolating the red channel of the cell image through a pixel mask that identifies the cell region, then calculating the mean value across all eligible eosinophilic cells in the patch. This normalized measure allows comparison of the cytoplasmic staining intensity across different tumor cases. |
| rc4003zafb9k | E | EOSINOPHIL_STD_INTENSITY | EOSINOPHIL_STD_INTENSITY quantifies the variability in red-channel intensity among eosinophils in the patch. By calculating the standard deviation of the red-channel intensity values derived from each individual eosinophil's masked cytoplasmic region, this parameter assesses the uniformity or heterogeneity in staining, offering insights into the range of activation states within the sample. |
| rc4003zafb9k | E | EOSINOPHIL_MIN_INTENSITY | EOSINOPHIL_MIN_INTENSITY indicates the lowest red-channel intensity observed among all eosinophils in the patch. This value is obtained by identifying the minimum intensity from the set of average intensities computed for each eosinophil, and can help pinpoint areas with lower activation or reduced eosin staining. |
| rc4003zafb9k | E | EOSINOPHIL_MAX_INTENSITY | EOSINOPHIL_MAX_INTENSITY reflects the highest red-channel intensity among the eosinophils in the patch. It is calculated by determining the maximum averaged intensity value from the evaluated cells, providing a measure of the most strongly stained, potentially highly activated eosinophils in the tumor region. |
| ricudkps5egg | TE | MEAN_TUMOR_BRIGHTNESS_IF | MEAN_TUMOR_BRIGHTNESS_IF: This parameter represents the average nuclear brightness of tumor cells located at the invasive front of the tumor. It is calculated by identifying the tumor cells that are in close spatial proximity to eosinophils (based on a defined distance threshold), extracting their nuclear brightness values from the designated image areas, and averaging these intensities. This average value is normalized, making it suitable for comparing different patient cases. |
| ricudkps5egg | TE | MEAN_EOSINOPHIL_BRIGHTNESS_IF | MEAN_EOSINOPHIL_BRIGHTNESS_IF: This parameter reflects the average nuclear brightness of eosinophils at the invasive front. It is computed by selecting eosinophil cells that are near tumor cells using a spatial threshold, extracting their nuclear brightness from the image using a binary mask, and subsequently averaging these values. The use of an average ensures that the parameter is normalized across different patches and patient cases. |
| ricudkps5egg | TE | BRIGHTNESS_DIFFERENCE_IF | BRIGHTNESS_DIFFERENCE_IF: This parameter quantifies the absolute difference between the average nuclear brightness of tumor cells and the average nuclear brightness of eosinophils at the invasive front. This difference captures the contrast in nuclear brightness between the two cell types, a contrast that may indicate underlying biological processes related to tumor aggression. Since it is derived from normalized averages, it is comparable across different patients. |
| rmjlvntymsav | N | MEAN_DIST_TO_MARGIN_UM | MEAN_DIST_TO_MARGIN_UM measures the average distance of neutrophils to the nearest tumor margin in micrometers. This parameter summarizes the overall proximity of neutrophils to the tumor border and is normalized to allow comparison across different patient cases. |
| rmjlvntymsav | N | MEDIAN_DIST_TO_MARGIN_UM | MEDIAN_DIST_TO_MARGIN_UM indicates the middle value of the distances from neutrophils to the tumor margin, offering a robust central value that is less affected by outliers. This normalization ensures it can be compared across multiple cases. |
| rmjlvntymsav | N | MIN_DIST_TO_MARGIN_UM | MIN_DIST_TO_MARGIN_UM represents the smallest distance recorded between any neutrophil and the tumor margin in micrometers. It identifies the closest approach of neutrophils to the tumor edge, providing a normalized measure of cell infiltration. |
| rmjlvntymsav | N | MAX_DIST_TO_MARGIN_UM | MAX_DIST_TO_MARGIN_UM captures the largest distance among the neutrophils to the tumor margin, measuring the farthest instance of cell-tumor proximity in micrometers. This normalized metric aids in evaluating the spread of neutrophil spatial distribution. |
| rmjlvntymsav | N | STD_DIST_TO_MARGIN_UM | STD_DIST_TO_MARGIN_UM computes the standard deviation of the distances from neutrophils to the tumor margin. It quantifies the variability or dispersion of these distances in micrometers, allowing for standardized comparisons across patches. |
| rn3vnz96b5bt | PM | TUMOR_MEDIAN_PM_DIST | TUMOR_MEDIAN_PM_DIST measures the median of the minimum distances between plasma cells and macrophages within the tumor compartment. For each plasma cell in the tumor region, the closest macrophage's distance is calculated, and the median of these distances—after scaling to micrometers—is used as a summary statistic. This parameter is numeric and normalized, allowing comparison across different patient cases and tissue patches. |
| rn3vnz96b5bt | PM | STROMA_MEDIAN_PM_DIST | STROMA_MEDIAN_PM_DIST measures the median of the minimum distances between plasma cells and macrophages within the stroma compartment. For each plasma cell in the stroma, the distance to its nearest macrophage is computed, and the median of these distances—converted to micrometers—is recorded. As a normalized and numeric metric, it facilitates consistent inter-sample comparisons. |
| rnx1itqtkbyq | EMNT | MEAN_DISTANCE_TO_NEAREST_TARGET_UM | MEAN_DISTANCE_TO_NEAREST_TARGET_UM represents the average distance in micrometers from each macrophage to its nearest target cell, which can be a tumor cell, neutrophil, or eosinophil. This metric is normalized across different patches and patient cases, as it measures a relative spatial distance based on computed centroids and includes a conversion factor from pixels to micrometers. |
| rnx1itqtkbyq | EMNT | MIN_DISTANCE_TO_NEAREST_TARGET_UM | MIN_DISTANCE_TO_NEAREST_TARGET_UM denotes the smallest distance in micrometers recorded between any macrophage and its closest target cell within a patch. This parameter gives insight into the minimum spatial proximity between these cell types and is normalized and used for comparative analysis. |
| rnx1itqtkbyq | EMNT | MAX_DISTANCE_TO_NEAREST_TARGET_UM | MAX_DISTANCE_TO_NEAREST_TARGET_UM indicates the largest distance in micrometers observed between a macrophage and its nearest target cell in a patch. It reflects the maximum spatial separation encountered, normalized for use in comparisons across different patches and patient cases. |
| rnx1itqtkbyq | EMNT | STD_DISTANCE_TO_NEAREST_TARGET_UM | STD_DISTANCE_TO_NEAREST_TARGET_UM is the standard deviation of the distances in micrometers from macrophages to their respective nearest target cells. This parameter provides information on the variability in cell-to-cell distance within a patch, and as such, it is a normalized statistical measure suitable for comparative assessment. |
| s1hvoq70d3dh | M | MEAN_CLUSTER_COMPACTNESS | MEAN_CLUSTER_COMPACTNESS measures the average compactness ratio across all identified macrophage clusters in a patch. The ratio is computed by dividing the area of the merged cell polygons by the area of their convex hull, providing a normalized metric that reflects how closely grouped the cells are relative to their overall spread. |
| s1hvoq70d3dh | M | MEDIAN_CLUSTER_COMPACTNESS | MEDIAN_CLUSTER_COMPACTNESS represents the median value of the compactness ratios calculated for the clusters within a patch. This measure offers a robust central tendency of cluster compactness, mitigating the effect of extreme values and allowing for comparison across patient cases. |
| s1hvoq70d3dh | M | SD_CLUSTER_COMPACTNESS | SD_CLUSTER_COMPACTNESS is the standard deviation of the compactness ratios for all clusters in a patch. It quantifies the variability or dispersion in cluster compactness, aiding in the assessment of heterogeneity in cell cluster formations. |
| s1hvoq70d3dh | M | MIN_CLUSTER_COMPACTNESS | MIN_CLUSTER_COMPACTNESS indicates the lowest compactness ratio observed among the clusters in a patch. This normalized measure helps identify the cluster with the most irregular or spread-out cell arrangement relative to its convex hull. |
| s1hvoq70d3dh | M | MAX_CLUSTER_COMPACTNESS | MAX_CLUSTER_COMPACTNESS shows the highest compactness ratio among the identified clusters in a patch. As a normalized metric, it highlights the cluster with the most condensed formation relative to its covering convex area. |
| s8d1407xf0tu | MPF | TRICELLULAR_SCALING_FACTOR | TRICELLULAR_SCALING_FACTOR is a numeric composite metric that captures the relative relationships between the nucleus sizes and shapes (aspect ratios) of macrophages, plasma cells, and fibroblasts within fibrotic stromal regions of tumor samples. It is derived by first computing the average nucleus area and average nucleus aspect ratio for each cell type from 1x1 mm patches. Then, two scaling factors are calculated: one based on the geometric mean of the ratios of the average nucleus areas from these cell types, and another based on the geometric mean of the ratios of the average nucleus aspect ratios. The final scaling factor is the product of these two measures, making it a normalized parameter that allows comparison across different patient cases. Higher values indicate greater morphological differences among the three cell types, which may reflect significant immune-stromal interactions relevant to tumor progression. |
| sah01iv8gqkj | TLPMNE | Epithelial_SLOPE | Epithelial_SLOPE represents the rate at which the average nuclear area of epithelial cells changes from the tumor center to the margin. It is obtained by computing the mean nuclear area within each of several concentric layers in a tumor patch and applying a linear regression, resulting in a normalized gradient value expressed in μm² per layer that allows for comparison across different patient cases. |
| sah01iv8gqkj | TLPMNE | Epithelial_R2 | Epithelial_R2 indicates the goodness-of-fit of the linear regression model used to relate the average nuclear area of epithelial cells to their corresponding layer indices. This coefficient of determination, normalized between 0 and 1, quantifies how well the linear model explains the variation in nuclear size from the tumor center to the margin. |
| sah01iv8gqkj | TLPMNE | Lymphocyte_SLOPE | Lymphocyte_SLOPE measures the gradient of nuclear size changes in lymphocyte cells as one moves from the center of the tumor outward. Derived from a linear regression on the average nuclear area calculated for multiple concentric layers, it provides a numeric value that is normalized for comparative analysis among different tumor patches. |
| sah01iv8gqkj | TLPMNE | Lymphocyte_R2 | Lymphocyte_R2 represents the R-squared value from the regression analysis performed on lymphocyte cells, indicating the degree to which the linear model explains the variability in nuclear size across the concentric layers. This numeric value is normalized and serves as a quality metric for the fitting procedure. |
| sah01iv8gqkj | TLPMNE | Plasma_SLOPE | Plasma_SLOPE captures the trend in nuclear area change for plasma cells across the concentric layers delineated within a tumor region. Calculated through a linear regression model on the average nuclear areas per layer, this parameter is numeric and normalized, making it suitable for comparing different patches. |
| sah01iv8gqkj | TLPMNE | Plasma_R2 | Plasma_R2 provides the coefficient of determination obtained from fitting a linear regression model to the plasma cell nuclear area data across the tumor layers. It quantitatively assesses the model's fit and is normalized to support standardized interpretation across patient cases. |
| sah01iv8gqkj | TLPMNE | Macrophage_SLOPE | Macrophage_SLOPE quantifies the change in the average nuclear area of macrophage cells from the center of a tumor patch to its periphery. This parameter is derived via linear regression applied to the average nucleus sizes across defined concentric layers, yielding a normalized and numeric slope value. |
| sah01iv8gqkj | TLPMNE | Macrophage_R2 | Macrophage_R2 indicates the goodness-of-fit of the linear regression applied to macrophage nuclear area data distributed across the concentric layers, expressed as a coefficient of determination. This normalized metric evaluates how effectively the model captures the trend in nuclear size changes and facilitates comparisons between patient cases. |
| sawjquw9p4u0 | MF | MACRO_NORM_ANGLE | MACRO_NORM_ANGLE measures the average infiltration angle of macrophages at the tumor-stroma boundary, normalized by the local curvature of the boundary. This parameter is computed by projecting each macrophage's centroid onto the boundary, calculating the angle between the cell's infiltration vector and the boundary's outward normal, and then dividing by the absolute curvature value at that point. The result, expressed in radians, enables comparison across different patient cases by accounting for local geometric variations. |
| sawjquw9p4u0 | MF | FIBRO_NORM_ANGLE | FIBRO_NORM_ANGLE quantifies the mean infiltration angle of fibroblasts (or connective tissue cells) at the tumor-stroma interface, with the angle normalized by the local curvature. This is achieved by projecting the fibroblast cell centroids onto the boundary, determining the angle between their infiltration vectors and the outward normal of the boundary, and normalizing using the local curvature value. Expressed in radians, it provides a standardized measure for assessing fibroblast behavior across varied tissue patches. |
| sawjquw9p4u0 | MF | NORM_ANGLE_DIFF | NORM_ANGLE_DIFF represents the difference between the normalized infiltration angles of macrophages and fibroblasts (MACRO_NORM_ANGLE minus FIBRO_NORM_ANGLE). This parameter captures the comparative directional infiltration behavior between the two cell types at the tumor-stroma boundary, delivering a normalized metric (in radians) that allows for direct comparisons across different patient samples and patch regions. |
| sbmx2ltx6tp5 | TFMN | TRI_CELL_CLUSTER_DENSITY | TRI_CELL_CLUSTER_DENSITY represents the number of valid tri-cell clusters per square millimeter of tissue area. It is normalized by dividing the absolute count of valid tri-cell clusters by the patch area, ensuring that differences in tissue area across patches and patients are accounted for, and allowing for direct comparisons. |
| sbmx2ltx6tp5 | TFMN | TRI_CELL_PROPORTION | TRI_CELL_PROPORTION is the ratio of valid tri-cell clusters to the total number of spatial clusters identified within a tissue patch. This parameter is normalized as a fraction, which makes it suitable for comparing the prevalence of tri-cell interactions across different samples and patients. |
| scv3a111q812 | T | NUCLEAR_IRREG_MEAN | NUCLEAR_IRREG_MEAN represents the mean nuclear irregularity score for a given patch. This score is calculated by comparing the actual perimeter of a tumor cell nucleus to the perimeter of an ideal ellipse fitted to the nucleus, yielding a normalized value that reflects how much the cell’s shape deviates from an ideal elliptical form. |
| scv3a111q812 | T | NUCLEAR_IRREG_STD | NUCLEAR_IRREG_STD denotes the standard deviation of the nuclear irregularity scores within the patch. It is a numeric metric that reflects the variability in nuclear shape irregularity among the tumor cells in that particular patch, allowing for comparison across different patients and regions. |
| scv3a111q812 | T | NUCLEAR_IRREG_MIN | NUCLEAR_IRREG_MIN indicates the minimum nuclear irregularity score observed in the patch. It captures the tumor cell with the nucleus that is closest to having a regular elliptical shape, and because it is derived from a normalized calculation, it facilitates meaningful comparisons between different patches. |
| scv3a111q812 | T | NUCLEAR_IRREG_MAX | NUCLEAR_IRREG_MAX designates the maximum nuclear irregularity score within the patch. This parameter identifies the tumor cell with the highest deviation from the ideal elliptical shape, providing an extreme measure of cellular irregularity that can be normalized across different patient samples. |
| sid40d3wm1ao | TLME | CROSS_CELL_TEXTURE_SKEWNESS | CROSS_CELL_TEXTURE_SKEWNESS quantifies the asymmetry in the distribution of cytoplasmic texture features across a set of cells (tumor cells, lymphocytes, macrophages, and eosinophils) within a tissue patch. This parameter is computed by first deriving individual cell skewness from the intensity distribution of cytoplasmic regions, and then aggregating these values at the patch level. The resulting skewness is a normalized statistical measure that can be compared across different patient cases, as it reflects the degree of extreme variations in cytoplasmic texture patterns rather than absolute counts or raw measurements. |
| swd68862p79f | TLFNE | LYMPHO_BOUNDARY_SMOOTHNESS_VAR | LYMPHO_BOUNDARY_SMOOTHNESS_VAR measures the variability in the smoothness of the lymphocyte nuclear boundary within each patch. The smoothness is calculated as the ratio between the actual perimeter of the nucleus and the perimeter of its convex hull, a normalized metric that reflects cell morphology. The variance of this ratio across lymphocytes in the patch is computed to capture local heterogeneity and can be reliably compared across different patient cases. |
| t15os4h4jrig | N | PERCENT_BARR_POSITIVE | PERCENT_BARR_POSITIVE quantifies the proportion of neutrophils in a tumor patch that have a detectable Barr body. It is calculated as the ratio of neutrophils with a positively identified Barr body to the total number of neutrophils in the patch, multiplied by 100. This percentage allows for normalized comparisons across different patient cases. |
| t15os4h4jrig | N | MEAN_BARR_SCORE | MEAN_BARR_SCORE represents the average Barr body visibility score across neutrophils in a patch. Each neutrophil is assigned a binary score—1 if a candidate Barr body is identified based on criteria of small area and peripheral location, and 0 otherwise. Averaging these scores provides a normalized parameter ranging from 0 to 1, facilitating standardized comparisons between patches. |
| t4bmenq2xctu | TLMF | CV_AREA | CV_AREA is the coefficient of variation of the mean cell areas calculated for the four distinct cell types (Tumor cells, Lymphocytes, Macrophages, and Fibroblasts) within each patch. This metric is dimensionless and is computed by determining the standard deviation of the mean areas of each cell type and dividing it by their arithmetic mean. The process involves extracting the polygon area for each cell, converting it into micrometer squared, and then aggregating these values to reflect the heterogeneity in cell sizes. |
| t4bmenq2xctu | TLMF | CV_PERIMETER | CV_PERIMETER is the coefficient of variation of the mean cell perimeters across the four cell types within each patch. This dimensionless parameter is obtained by computing the mean perimeter for each cell type—after converting the perimeter measurements into micrometers—and then calculating the relative variation among them. It reflects the variability in the cell boundaries, indicating differences in cell shape attributes across the various cell types. |
| t4bmenq2xctu | TLMF | CV_CIRCULARITY | CV_CIRCULARITY represents the coefficient of variation for the mean cell circularity values across the four designated cell types within a patch. Circularity is computed using the relationship between the area and perimeter of the cell, where the metric indicates how close the shape is to a perfect circle. The variation in these mean circularity values (standard deviation divided by the mean) provides insight into the diversity in cell shape regularity across the cell types. |
| t4bmenq2xctu | TLMF | MEAN_AREA | MEAN_AREA is the average cell area across all four cell types within each patch, expressed in micrometer squared. It serves as a contextual reference for comparing overall cell size and is calculated by averaging the mean areas computed for Tumor cells, Lymphocytes, Macrophages, and Fibroblasts after their areas have been appropriately converted to the same unit. |
| t5ypmtxuf51r | L | LYMPHO_NUCLEAR_BASOPHILIA_SCORE | LYMPHO_NUCLEAR_BASOPHILIA_SCORE measures the average absolute deviation of lymphocyte nuclear staining intensities from a global baseline intensity computed over all lymphocytes across patches. This parameter provides a normalized indication of how much the nuclear staining in individual patches deviates from the overall median, reflecting potential variations in immune response that could be associated with patient outcomes. |
| t5ypmtxuf51r | L | MEAN_NUCLEAR_INTENSITY | MEAN_NUCLEAR_INTENSITY captures the average nuclear staining intensity of lymphocytes within each patch. By computing the mean intensity from the selected nuclear pixels, this numeric metric offers a standardized measure that can be compared across different patient cases and tissue patches. |
| t5ypmtxuf51r | L | SD_NUCLEAR_INTENSITY | SD_NUCLEAR_INTENSITY represents the standard deviation of nuclear staining intensities within a patch. This parameter quantifies the variability in staining intensity among lymphocytes, providing insights into the heterogeneity within the tumor microenvironment. Its numeric and continuous nature allows for normalized comparisons across different samples. |
| t65b63e9owrh | LNPT | MEAN_NEUTROPHIL_SHAPE_FACTOR | MEAN_NEUTROPHIL_SHAPE_FACTOR measures the average circularity of neutrophil nuclei within a patch. Circularity is calculated based on the area and perimeter of each nucleus, providing a normalized value between 0 and 1, where a value of 1 indicates a perfect circle. |
| t65b63e9owrh | LNPT | SD_NEUTROPHIL_SHAPE_FACTOR | SD_NEUTROPHIL_SHAPE_FACTOR represents the standard deviation of the circularity values for neutrophil nuclei in the patch, reflecting the variability in nuclear shape among the analyzed neutrophils. |
| t65b63e9owrh | LNPT | MIN_NEUTROPHIL_SHAPE_FACTOR | MIN_NEUTROPHIL_SHAPE_FACTOR identifies the smallest circularity value among neutrophils in the patch, highlighting the most irregular nuclear shape observed. |
| t65b63e9owrh | LNPT | MAX_NEUTROPHIL_SHAPE_FACTOR | MAX_NEUTROPHIL_SHAPE_FACTOR identifies the largest circularity value among neutrophils in the patch, indicating the nucleus with the shape closest to a perfect circle. |
| t745w8xqb0ck | TMN | VARIANCE_TUMOR_NUCLEAR_AREA | VARIANCE_TUMOR_NUCLEAR_AREA represents the variance in the nuclear areas of tumor cells located in the tumor periphery. This metric is calculated from the polygonal representation of the tumor cell nuclei and is scaled to square micrometers (μm²), making it comparable across different patient samples and patches. It reflects the heterogeneity in the tumor nuclear morphology and is normalized for reliable cross-case comparisons. |
| t745w8xqb0ck | TMN | VARIANCE_MACROPHAGE_NUCLEAR_AREA | VARIANCE_MACROPHAGE_NUCLEAR_AREA denotes the variance in nuclear areas of macrophages present in the tumor tissue. Derived from morphometric analysis using polygon data, this measure is converted to μm² and captures the diversity in macrophage nuclear sizes. The normalization inherent in the variance calculation renders it apt for comparative studies across diverse patient cases. |
| t83rjwqxvsf2 | TPE | PLASMA_FRACTION | PLASMA_FRACTION is a normalized metric that quantifies the relative fraction of plasma cells among all stromal cells within an image patch. It is computed by dividing the count of plasma cells by the total count of stromal cells, thereby allowing for valid comparisons across different patient cases and image patches. |
| t83rjwqxvsf2 | TPE | EOSINOPHIL_FRACTION | EOSINOPHIL_FRACTION is a normalized metric that measures the relative proportion of eosinophil cells among the total stromal cells in an image patch. This metric is calculated by dividing the number of eosinophilic granulocytes by the overall stromal cell count, enabling consistent and comparable evaluations across different patients. |
| tbwdlrbxxmtm | FE | EOS_GRADIENT_MEAN | EOS_GRADIENT_MEAN represents the average gradient magnitude calculated from the regions corresponding to eosinophils in each patch. This metric is derived by applying a Sobel filter to the masked grayscale image of eosinophils, resulting in a normalized numeric measure of the cell-level texture intensity that can be compared across different patient cases. |
| tbwdlrbxxmtm | FE | FIB_GRADIENT_MEAN | FIB_GRADIENT_MEAN represents the average gradient magnitude computed from the fibroblast regions within each patch. Similar to the eosinophil metric, this value is generated by isolating fibroblast cells, applying a Sobel filter to the masked image regions, and computing the mean gradient. It provides a normalized numeric measurement for comparative analysis. |
| tbwdlrbxxmtm | FE | GRADIENT_DIFFERENCE | GRADIENT_DIFFERENCE is the numeric differential metric obtained by subtracting the mean gradient of fibroblasts (FIB_GRADIENT_MEAN) from that of eosinophils (EOS_GRADIENT_MEAN) within a patch. This parameter captures the relative difference in cellular texture between the two cell types and is normalized for comparison across patches. |
| tbwdlrbxxmtm | FE | EOS_GRADIENT_STD | EOS_GRADIENT_STD is the standard deviation of gradient magnitudes across all eosinophil cells in a patch. It quantifies the variability in gradient intensity within the eosinophil population and serves as a normalized statistic representing dispersion. |
| tbwdlrbxxmtm | FE | FIB_GRADIENT_STD | FIB_GRADIENT_STD is the standard deviation of gradient magnitudes computed for fibroblast cells in a patch. This metric provides a normalized measure of variability in texture intensity among fibroblasts and is useful for comparative analysis across patient cases. |
| tdgekh2gwmh2 | TFML | Fraction_Rosettes | Fraction_Rosettes is a normalized, numeric metric that quantifies the proportion of tumor cells in a given patch that are engaged in rosette structures. A rosette is defined by the presence of a central tumor cell surrounded by at least one fibroblast, one macrophage, and one lymphocyte, all located within a 50μm radius. This fraction is calculated by dividing the number of rosette-positive tumor cells by the total number of tumor cells in the patch, allowing for effective comparison across different patient cases and conditions. |
| tdpuagjcij9c | TEF | INFILTRATION_GRADIENT_MEAN | INFILTRATION_GRADIENT_MEAN represents the average infiltration velocity gradient calculated across the three cell types (tumor cells, eosinophils, and fibroblasts). This parameter is obtained by first computing individual gradients from the regression of cell density (cells per unit area in annular bins) versus radial distance from the tumor center, and then averaging these gradients. It provides a normalized metric, allowing comparison across different patient cases and patches. |
| tdpuagjcij9c | TEF | INFILTRATION_GRADIENT_TUMOR | INFILTRATION_GRADIENT_TUMOR measures the infiltration velocity gradient specifically for tumor cells. It is derived by calculating the change in tumor cell density with respect to the distance from a computed tumor center. After binning the distances into uniform annular regions and normalizing by the area of these bins, a linear regression is performed, and the slope from this regression defines the tumor infiltration gradient. |
| tdpuagjcij9c | TEF | INFILTRATION_GRADIENT_EOSINOPHIL | INFILTRATION_GRADIENT_EOSINOPHIL quantifies the infiltration velocity gradient for eosinophils. Similar to the tumor cells, it is computed by first determining the distance of eosinophils from the tumor center, binning these distances, calculating the corresponding cell densities, and finally performing a linear regression on these density measures. The resulting slope of this regression indicates the rate at which eosinophil density changes radially, providing a normalized and numeric measure. |
| tdpuagjcij9c | TEF | INFILTRATION_GRADIENT_FIBROBLAST | INFILTRATION_GRADIENT_FIBROBLAST is the infiltration velocity gradient calculated for fibroblasts. The process involves computing the fibroblast cell density in radial bins around the tumor center and determining the slope from a linear regression of these densities versus the bin center distances. This slope reflects how fibroblast density varies with distance, serving as a normalized parameter for comparison across different patches and patient cases. |
| tf2osnyy4qbz | TLF | TUMOR_BRIGHTNESS | TUMOR_BRIGHTNESS is the mean nuclear brightness computed from tumor cells within a given patch. This parameter is derived by averaging the grayscale intensity values of cell nuclei in tumor core regions, providing a normalized measurement that facilitates comparison across different patient cases. |
| tf2osnyy4qbz | TLF | LYMPHOCYTE_BRIGHTNESS | LYMPHOCYTE_BRIGHTNESS represents the mean nuclear brightness of lymphocytes within a patch. It is calculated by averaging the grayscale intensities of lymphocyte nuclei, ensuring the parameter remains numeric and normalized for direct comparisons between different patients and regions. |
| tf2osnyy4qbz | TLF | FIBROBLAST_BRIGHTNESS | FIBROBLAST_BRIGHTNESS denotes the mean nuclear brightness of fibroblasts (specifically connective tissue cells in the stroma regions) within a patch. The absolute average brightness values are computed to yield a normalized metric suitable for inter-case analytical comparisons. |
| tf2osnyy4qbz | TLF | TUMOR_LYMPHOCYTE_RATIO | TUMOR_LYMPHOCYTE_RATIO is a normalized ratio obtained by dividing the mean nuclear brightness of tumor cells by that of lymphocytes. This parameter quantitatively captures the relative difference in nuclear brightness between these two cell types, facilitating comparisons across diverse patient samples. |
| tf2osnyy4qbz | TLF | TUMOR_FIBROBLAST_RATIO | TUMOR_FIBROBLAST_RATIO is a computed ratio that compares the mean nuclear brightness of tumor cells with that of fibroblasts. This ratio is normalized and reflects the relative brightness differences, making it a reliable parameter for inter-patient analysis. |
| tf2osnyy4qbz | TLF | LYMPHOCYTE_FIBROBLAST_RATIO | LYMPHOCYTE_FIBROBLAST_RATIO is the ratio of the mean nuclear brightness of lymphocytes to that of fibroblasts within a patch. Being a normalized metric derived from averaged brightness values, it effectively represents the relative brightness differences to support comparative studies. |
| tgxmxlh1ygl6 | TLNEF | MEDIAN_DIST_TO_CLUSTER | MEDIAN_DIST_TO_CLUSTER represents the median distance, measured in micrometers, from each fibroblast cell within a patch to the centroid of its nearest mixed cell cluster consisting of tumor cells and immune cells. This parameter is computed by first identifying fibroblasts and candidate cluster cells, applying a clustering algorithm to these candidate cells, calculating cluster centroids, and then determining each fibroblast's nearest distance. The median of these distances is used to provide a robust measure of the typical spatial separation between fibroblasts and nearby clusters, enabling effective comparison across different patient cases. |
| tgxmxlh1ygl6 | TLNEF | MIN_DIST_TO_CLUSTER | MIN_DIST_TO_CLUSTER indicates the smallest distance, in micrometers, from any fibroblast cell in a patch to the nearest mixed cell cluster. By calculating the minimum distance for each fibroblast and then selecting the overall minimum across all fibroblasts, this parameter quantifies the closest spatial interaction between fibroblasts and their proximal clusters, offering insight into the most immediate cell-cell contacts in the tumor microenvironment. |
| tgxmxlh1ygl6 | TLNEF | MAX_DIST_TO_CLUSTER | MAX_DIST_TO_CLUSTER quantifies the largest distance, measured in micrometers, from a fibroblast cell to the nearest mixed cell cluster within a patch. It captures the upper bound of fibroblast-to-cluster proximity, reflecting the extent of spatial dispersion of fibroblasts relative to clustered regions, which can inform on heterogeneity across different tissue regions. |
| tgxmxlh1ygl6 | TLNEF | MEAN_DIST_TO_CLUSTER | MEAN_DIST_TO_CLUSTER is the arithmetic average of the distances, expressed in micrometers, from fibroblast cells to their respective nearest mixed cell clusters within a patch. This parameter provides a general measure of spatial proximity by aggregating individual minimal distances, which facilitates consistent comparisons of cell spatial relationships across various patient cases. |
| tgxmxlh1ygl6 | TLNEF | STD_DIST_TO_CLUSTER | STD_DIST_TO_CLUSTER measures the standard deviation of the distances, in micrometers, from fibroblasts to their nearest mixed cell clusters within a patch. This statistical measure captures the variability or dispersion in the fibroblast-to-cluster distances, highlighting how uniform or varied the spatial distribution is within each analyzed region and supporting comparisons between different patient samples. |
| tu6v2xvu307e | TL | TUMOR_LYMPHO_RATIO | TUMOR_LYMPHO_RATIO represents the ratio of tumor cells to lymphocytes within the epithelial region of each patch. This parameter has been computed by dividing the count of tumor cells by the count of lymphocytes, thereby normalizing the measurement across different patient cases. It serves as an indicator of the immune landscape within tumor patches, where a higher ratio may correlate with immune evasion and poorer patient outcomes. |
| u4ja0qhzcwx3 | TLPMNEF | S_PHASE_FRACTION_TUMOR | S_PHASE_FRACTION_TUMOR: This parameter measures the proportion of tumor cells (classified based on epithelial characteristics) within a patch that are in the S-phase of the cell cycle. It is determined by comparing each cell's nuclear intensity to a calculated threshold, providing a normalized ratio that enables cross-case comparisons. |
| u4ja0qhzcwx3 | TLPMNEF | S_PHASE_FRACTION_LYMPHO | S_PHASE_FRACTION_LYMPHO: This metric represents the fraction of lymphocytes within a patch that are actively in S-phase. The determination is based on nuclear intensity compared against an adaptive threshold, resulting in a normalized ratio that reflects proliferative activity in lymphocytes across different patient samples. |
| u4ja0qhzcwx3 | TLPMNEF | S_PHASE_FRACTION_PLASMA | S_PHASE_FRACTION_PLASMA: This parameter quantifies the fraction of plasma cells in the S-phase, using nuclear intensity measurements relative to a standardized threshold. Its normalized value facilitates comparisons between different tissue patches and patient cases. |
| u4ja0qhzcwx3 | TLPMNEF | S_PHASE_FRACTION_MACRO | S_PHASE_FRACTION_MACRO: This parameter calculates the proportion of macrophages that are in S-phase within a patch. By measuring nuclear intensity and establishing a threshold for S-phase classification, it produces a normalized ratio that compares the proliferative state of macrophages across various samples. |
| u4ja0qhzcwx3 | TLPMNEF | S_PHASE_FRACTION_EOSINO | S_PHASE_FRACTION_EOSINO: This parameter determines the fraction of eosinophils in S-phase based on their nuclear intensity relative to an established threshold. The resulting normalized ratio allows for consistent comparisons of eosinophil proliferation between different tumor cases. |
| u4ja0qhzcwx3 | TLPMNEF | S_PHASE_FRACTION_FIBRO | S_PHASE_FRACTION_FIBRO: This metric measures the fraction of fibroblasts classified as being in S-phase by comparing their nuclear intensities to a computed threshold. As a normalized ratio, it offers a standardized measure of proliferation that can be compared across patches and patient cases. |
| u4nlrf59wvg9 | TLPMNEF | TUMOR_MEAN_GRANULES | TUMOR_MEAN_GRANULES represents the average number of granules identified in tumor cells (epithelial cells) within a patch. This value is calculated by averaging the granule counts of individual tumor cells and is used to understand the typical granule abundance in this cell type across different tumor regions. |
| u4nlrf59wvg9 | TLPMNEF | TUMOR_STD_GRANULES | TUMOR_STD_GRANULES denotes the standard deviation of granule counts among tumor cells within a patch, indicating the degree of variability or dispersion in granule numbers among these cells. |
| u4nlrf59wvg9 | TLPMNEF | LYMPHO_MEAN_GRANULES | LYMPHO_MEAN_GRANULES is the average granule count across lymphocyte cells in a patch. It reflects the typical quantity of cytoplasmic granules observed in lymphocytes, providing a normalized measure for immune cell activity in different tumor regions. |
| u4nlrf59wvg9 | TLPMNEF | LYMPHO_STD_GRANULES | LYMPHO_STD_GRANULES captures the variability in granule counts among lymphocytes within a patch by computing the standard deviation, highlighting differences in granule distribution among these cells. |
| u4nlrf59wvg9 | TLPMNEF | PLASMA_MEAN_GRANULES | PLASMA_MEAN_GRANULES measures the mean number of granules detected in plasma cells per patch. This metric is derived by averaging the granule counts from all plasma cells, offering a normalized indicator of granule abundance in these cells. |
| u4nlrf59wvg9 | TLPMNEF | PLASMA_STD_GRANULES | PLASMA_STD_GRANULES indicates the dispersion of granule counts among plasma cells in a patch by calculating the standard deviation, offering insights into the consistency of granule detection in this cell type. |
| u4nlrf59wvg9 | TLPMNEF | MACRO_MEAN_GRANULES | MACRO_MEAN_GRANULES represents the mean granule count in macrophages per patch. It is calculated by taking the average of granule counts from individual macrophages, thereby providing a normalized metric that can be compared between patient cases. |
| u4nlrf59wvg9 | TLPMNEF | MACRO_STD_GRANULES | MACRO_STD_GRANULES is the standard deviation of granule counts in macrophages, reflecting the variability in granule distribution among these cells within a patch. |
| u4nlrf59wvg9 | TLPMNEF | NEUTRO_MEAN_GRANULES | NEUTRO_MEAN_GRANULES reflects the average number of granules detected in neutrophils per patch, computed by averaging the granule counts measured in these cells. |
| u4nlrf59wvg9 | TLPMNEF | NEUTRO_STD_GRANULES | NEUTRO_STD_GRANULES captures the spread or variability in the granule counts of neutrophils within a patch through the computation of their standard deviation. |
| u4nlrf59wvg9 | TLPMNEF | EOSINO_MEAN_GRANULES | EOSINO_MEAN_GRANULES is the average granule count in eosinophils per patch, providing a normalized value that indicates the typical granule abundance in this specific cell type across different regions. |
| u4nlrf59wvg9 | TLPMNEF | EOSINO_STD_GRANULES | EOSINO_STD_GRANULES represents the standard deviation of granule counts in eosinophils, showing the extent of variability in granule distribution among these cells within a patch. |
| u4nlrf59wvg9 | TLPMNEF | FIBRO_MEAN_GRANULES | FIBRO_MEAN_GRANULES calculates the average number of granules present in fibroblasts (connective tissue cells) per patch by averaging the granule counts, offering a normalized metric to compare cellular features across cases. |
| u4nlrf59wvg9 | TLPMNEF | FIBRO_STD_GRANULES | FIBRO_STD_GRANULES is the standard deviation of granule counts in fibroblasts, which provides an indication of the variability in granule numbers among these cells within a patch. |
| u4nlrf59wvg9 | TLPMNEF | OVERALL_GRANULE_VARIATION | OVERALL_GRANULE_VARIATION is a composite metric calculated as the coefficient of variation (the ratio of the standard deviation to the mean) of the mean granule counts across all analyzed cell types in a patch. This parameter provides a normalized measure of the overall heterogeneity in granule distribution across different cell types within tumor regions. |
| u6nyfpqfrqda | L | TEXTURE_HETEROGENEITY_INDEX | The TEXTURE_HETEROGENEITY_INDEX parameter quantifies the variability in nuclear texture among lymphocytes within a tumor patch by computing the standard deviation of each cell's pixel intensity distribution. This measure reflects differences in chromatin pattern that may be associated with immune cell activation, providing a normalized numeric value that can be compared across different patient cases. |
| uaqsh60wp9kg | TLPMNEF | T_CELL_PROPORTION | T_CELL_PROPORTION quantifies the fraction of T cells relative to the total number of immune cells (T cells, B cells, macrophages, and granulocytes) within a given patch of stromal tissue. This normalized value enables comparison across different samples by converting raw counts into a standardized proportion. |
| uaqsh60wp9kg | TLPMNEF | B_CELL_PROPORTION | B_CELL_PROPORTION represents the normalized fraction of B cells among all immune cells in the stromal compartment of a patch. It is calculated by dividing the B cell count by the total immune cell count, ensuring comparability across different patient cases and areas. |
| uaqsh60wp9kg | TLPMNEF | MACROPHAGE_PROPORTION | MACROPHAGE_PROPORTION indicates the proportion of macrophages among the overall immune cells in a tissue patch's stroma. By normalizing the raw macrophage count with the total immune cell count, this metric allows for consistent comparisons between different analyses. |
| uaqsh60wp9kg | TLPMNEF | GRANULOCYTE_PROPORTION | GRANULOCYTE_PROPORTION measures the relative proportion of granulocytes (including both neutrophils and eosinophils) in the patch. The value is derived by dividing the granulocyte count by the total immune cell count, and it is essential for comparing local immune profiles across samples. |
| uaqsh60wp9kg | TLPMNEF | IMMUNE_IMBALANCE_SCORE | IMMUNE_IMBALANCE_SCORE is a composite index that quantifies the deviation of the observed immune cell proportions from a predefined reference distribution. This score is the sum of the absolute differences between the actual proportions of T cells, B cells, macrophages, and granulocytes and their corresponding reference values. It provides a normalized metric of immune imbalance that facilitates comparison between different tumor cases. |
| ui6vjfrk3fpi | TLMEF | CV_TUMOR_TEXTURE | CV_TUMOR_TEXTURE represents the coefficient of variation of the gray-level co-occurrence matrix (GLCM) derived contrast feature from tumor cell cytoplasmic texture. It is calculated as the standard deviation divided by the mean of the contrast values across tumor cells in infiltration zones. This parameter is normalized, allowing for comparison across different patient cases, as it measures relative variability in texture rather than raw counts. |
| umj95ridy1vm | T | MEAN_NC_RATIO | MEAN_NC_RATIO: This parameter represents the average nuclear-to-cytoplasm ratio calculated from all tumor cells within a patch. It is derived by computing individual ratios for each tumor cell (using the nucleus area divided by the approximated cytoplasmic area) and then averaging these values across the patch. This metric is normalized, allowing for meaningful comparisons across different patient cases. |
| umj95ridy1vm | T | MEDIAN_NC_RATIO | MEDIAN_NC_RATIO: This parameter captures the median value of the nuclear-to-cytoplasm ratios among tumor cells in the patch. It provides a robust measure of central tendency that minimizes the influence of outliers, ensuring that the typical cellular morphology is accurately represented for inter-patient comparisons. |
| umj95ridy1vm | T | SD_NC_RATIO | SD_NC_RATIO: This parameter reflects the standard deviation of the nuclear-to-cytoplasm ratios in a patch, indicating the variability or dispersion in cellular morphology. It is calculated from the spread of individual cell ratios and is normalized, enabling direct comparison of heterogeneity across different patches and patient cases. |
| umj95ridy1vm | T | MIN_NC_RATIO | MIN_NC_RATIO: This parameter records the smallest nuclear-to-cytoplasm ratio observed in the patch. It highlights the lowest level of nuclear prominence relative to the cytoplasm among tumor cells, providing insight into the range of cellular features in a normalized manner. |
| umj95ridy1vm | T | MAX_NC_RATIO | MAX_NC_RATIO: This parameter captures the highest nuclear-to-cytoplasm ratio found in the patch. It indicates the maximum level of nuclear prominence relative to the cytoplasm, which can signal more aggressive cellular behavior. The normalized nature of this metric allows for effective comparisons across different patient images. |
| uoyb355lvhaa | TLMFN | PROPORTION_ELONGATED | PROPORTION_ELONGATED quantifies the fraction of macrophages in a given patch that display an elongated morphology. This metric is computed as the ratio of the number of macrophages with an elongation ratio above a set threshold to the total number of macrophages, providing a normalized measure for comparing different patient cases. |
| uoyb355lvhaa | TLMFN | MEAN_ELONGATION_RATIO | MEAN_ELONGATION_RATIO represents the average elongation ratio of macrophages within a patch. The elongation ratio is derived by assessing the proportion between the major and minor axes of the minimum rotated rectangle encompassing each macrophage, yielding a numeric value that reflects the overall tendency towards elongated cell shapes. |
| uoyb355lvhaa | TLMFN | MEDIAN_ELONGATION_RATIO | MEDIAN_ELONGATION_RATIO is the median value of the elongation ratios calculated for all macrophages in the patch. This measure reduces the impact of extreme values and provides a robust indicator of the central tendency of macrophage elongation in a normalized format. |
| uoyb355lvhaa | TLMFN | SD_ELONGATION_RATIO | SD_ELONGATION_RATIO denotes the standard deviation of the elongation ratios among macrophages in a patch. It quantifies the variability in morphological elongation across the cell population, offering insight into the heterogeneity of cell shape within different tissue regions. |
| ut3ouncvbs3g | TFE | TUMOR_CONTRAST | TUMOR_CONTRAST: This parameter measures the average contrast computed from the Haralick texture features in tumor cell nuclei that are located near necrotic regions. The contrast quantifies local intensity variations from the normalized grayscale images of these cells, and is aggregated across cells in a patch for comparative analysis across patient cases. |
| ut3ouncvbs3g | TFE | TUMOR_CORRELATION | TUMOR_CORRELATION: This parameter represents the mean correlation value, reflecting the linear dependency of gray-level intensities in tumor cell nuclei near necrosis. It is derived from Haralick texture analysis of normalized grayscale images and serves to capture directional intensity relationships among tumor cells. |
| ut3ouncvbs3g | TFE | TUMOR_ENERGY | TUMOR_ENERGY: This parameter measures the average energy, indicating the uniformity of intensity distributions within the tumor cell nuclei near necrotic regions. It is computed from normalized grayscale images using Haralick features and aggregated across the tumor cells within a patch. |
| ut3ouncvbs3g | TFE | TUMOR_HOMOGENEITY | TUMOR_HOMOGENEITY: This parameter quantifies the mean homogeneity of tumor cell nuclei near necrotic zones, measuring how closely the texture feature distribution aligns to a uniform pattern. It is derived from the analysis of normalized gray-scale images and is calculated using Haralick feature extraction. |
| ut3ouncvbs3g | TFE | TUMOR_GRADIENT | TUMOR_GRADIENT: This parameter represents the average intensity gradient magnitude in tumor cell nuclei near necrosis. It reflects the rate of change in intensity within the cellular nucleus, computed after Gaussian smoothing of the normalized grayscale image and aggregated over tumor cells within a patch. |
| ut3ouncvbs3g | TFE | FIBRO_CONTRAST | FIBRO_CONTRAST: This parameter computes the mean contrast for fibroblast (connective tissue) cell nuclei located near necrotic regions. It reflects the local variations in intensity in normalized images and is obtained by averaging Haralick-derived contrast values across fibroblasts in a patch. |
| ut3ouncvbs3g | TFE | FIBRO_CORRELATION | FIBRO_CORRELATION: This parameter represents the average correlation value for fibroblast nuclei near necrosis, capturing the linear dependence in pixel intensity patterns derived from normalized grayscale images. It summarizes the texture coherence among fibroblast cells. |
| ut3ouncvbs3g | TFE | FIBRO_ENERGY | FIBRO_ENERGY: This parameter measures the mean energy of fibroblast cell nuclei near necrotic zones, indicating the uniformity or consistency in the intensity pattern of the normalized grayscale image. It is calculated using Haralick features and averaged over the relevant cells. |
| ut3ouncvbs3g | TFE | FIBRO_HOMOGENEITY | FIBRO_HOMOGENEITY: This parameter quantifies the mean homogeneity in fibroblast nuclei situated near necrotic areas. It assesses the smoothness or uniformity in texture, based on normalized grayscale images and Haralick feature extraction. |
| ut3ouncvbs3g | TFE | FIBRO_GRADIENT | FIBRO_GRADIENT: This parameter reflects the mean intensity gradient magnitude in fibroblast nuclei near necrosis. It measures the average change in intensity computed from the gradient of smoothed normalized grayscale images, summarizing local structural transitions. |
| ut3ouncvbs3g | TFE | EOS_CONTRAST | EOS_CONTRAST: This parameter measures the mean contrast in eosinophil nuclei found near necrotic regions. Derived from Haralick texture features of normalized grayscale images, it captures the local intensity variations in these immune cells. |
| ut3ouncvbs3g | TFE | EOS_CORRELATION | EOS_CORRELATION: This parameter represents the average correlation value for eosinophil nuclei near necrosis. It reflects the degree of linear dependency between neighboring pixel intensities as extracted from normalized images using Haralick texture analysis. |
| ut3ouncvbs3g | TFE | EOS_ENERGY | EOS_ENERGY: This parameter quantifies the mean energy in eosinophil nuclei near necrotic zones, indicating the uniformity of their intensity distributions based on normalized grayscale images and Haralick feature computation. |
| ut3ouncvbs3g | TFE | EOS_HOMOGENEITY | EOS_HOMOGENEITY: This parameter calculates the average homogeneity in the texture of eosinophil nuclei adjacent to necrosis. It measures how evenly distributed the gray-level intensities are in the normalized image and is computed via Haralick feature analysis. |
| ut3ouncvbs3g | TFE | EOS_GRADIENT | EOS_GRADIENT: This parameter measures the mean intensity gradient magnitude in eosinophil nuclei near necrotic zones. It reflects the average rate of intensity change within these cells, computed from smoothed normalized grayscale images and aggregated to provide a comparative metric. |
| uujazv48trht | PM | MEAN_PM_CLUSTER_SIZE | MEAN_PM_CLUSTER_SIZE represents the average number of cells within plasma cell-macrophage clusters in a patch. These clusters are defined as groups of cells that are identified using a clustering algorithm and are considered valid only if they contain at least one plasma cell and one macrophage. Since each patch is of a fixed size (1000x1000 μm), this average value serves as a normalized metric, allowing for comparison across different patient cases. |
| uujazv48trht | PM | NUM_PM_CLUSTERS | NUM_PM_CLUSTERS is the total count of valid plasma cell-macrophage clusters found within a patch. A valid cluster, again, must include both a plasma cell and a macrophage. Although it is a count, because all patches have the same defined area, this metric is normalized in the sense that it can be compared across patches and patient cases. |
| uujazv48trht | PM | MAX_PM_CLUSTER_SIZE | MAX_PM_CLUSTER_SIZE indicates the size (in terms of number of cells) of the largest valid plasma cell-macrophage cluster within a patch. This parameter highlights the maximum cluster size observed, and since the analysis is performed on patches of equal dimension, it also represents a normalized metric suitable for comparing different patient cases. |
| uwqz2819mtvl | L | MEAN_VESICLES_PER_LYMPH | MEAN_VESICLES_PER_LYMPH represents the average number of cytoplasmic vesicles detected per lymphocyte within a tissue patch. This metric is computed by identifying lymphocyte cells in the patch, segmenting their cytoplasmic regions to detect vesicle-like structures through intensity-based thresholding and morphological cleaning, and then calculating the mean count from all the lymphocytes in that patch. Its cell-level normalization enables effective comparison across different patient cases. |
| uwqz2819mtvl | L | SD_VESICLES_PER_LYMPH | SD_VESICLES_PER_LYMPH is the standard deviation of the vesicle counts across lymphocytes in a given tissue patch. After counting the vesicles detected by applying a segmentation approach on the cytoplasmic area for each lymphocyte, this parameter quantifies the variability in vesicle production among the cells. This standardized measure ensures that the metric is comparative across different patients. |
| uwqz2819mtvl | L | MAX_VESICLES_PER_LYMPH | MAX_VESICLES_PER_LYMPH indicates the highest vesicle count observed in any single lymphocyte within a patch. By analyzing each lymphocyte using intensity-based segmentation to identify vesicle-like structures and then taking the maximum count across cells, this parameter provides an upper bound of cellular vesicle formation in the patch. The use of a per-cell basis normalization enables its application across various patient cases. |
| uzv2jf4m08lo | TLPMNE | SYNERGY_RATIO | SYNERGY_RATIO is a normalized metric that quantifies the degree of mixed-cell clustering in a tissue patch by computing the ratio of clusters containing multiple cell types to clusters containing a single cell type. This parameter integrates information about spatial co-localization of heterogeneous cells, allowing for comparative analysis between different patient cases, since it reflects the relative prevalence of heterogeneous interactions rather than absolute counts. |
| v14mdscyir5h | LFNT | LYMPHO_FIBRO_EOSIN_RATIO | LYMPHO_FIBRO_EOSIN_RATIO represents the ratio calculated for each tumor patch by dividing the mean eosin intensity of lymphocyte cytoplasms by the mean eosin intensity of fibroblast (connective tissue) cytoplasms. This metric is computed only in patches that contain neutrophils within tumor compartments and is a normalized, dimensionless value that allows comparison between different tumor samples. |
| v14mdscyir5h | LFNT | OVERALL_MEDIAN_RATIO | OVERALL_MEDIAN_RATIO is the median value of the LYMPHO_FIBRO_EOSIN_RATIO values obtained from all eligible patches within a given tumor region. This parameter provides a robust summary of the lymphocyte-to-fibroblast eosin intensity ratios across patches and, being normalized and numeric, is suitable for comparative analysis across patient cases. |
| v46nagdgd2tj | TFLME | MEDIAN_INFILTRATION_SHIFT | MEDIAN_INFILTRATION_SHIFT is a normalized and numeric parameter representing the median value of the shifts in cell type infiltration fractions between consecutive spatial bins within a tumor patch. The parameter is computed by first calculating the infiltration fraction for each cell type in multiple spatial bins, based on the relative counts of tumor cells, fibroblasts, lymphocytes, macrophages, and eosinophils. The shifts between consecutive bins are then obtained by computing the absolute differences in these fractions. The final parameter value is the median of these shifts, providing a robust measure of how dramatically cellular composition changes across different regions of the tumor. A higher value indicates more significant spatial variation in cell infiltration patterns, which might correlate with heterogeneous tumor microenvironments. |
| v5kcps1hr5ye | FL | FIBROBLAST_LYMPHOCYTE_ADJACENCY_RATIO | FIBROBLAST_LYMPHOCYTE_ADJACENCY_RATIO measures the proportion of fibroblasts that are directly adjacent to lymphocytes within a defined proximity threshold of 10 micrometers. This normalized ratio, obtained by dividing the number of fibroblasts with at least one lymphocyte neighbor by the total number of fibroblasts in the analyzed patch, allows for consistent comparisons across different patient cases. |
| v5kcps1hr5ye | FL | MEAN_MIN_DISTANCE_UM | MEAN_MIN_DISTANCE_UM quantifies the average minimum distance from each fibroblast to its nearest lymphocyte, expressed in micrometers. It reflects the typical spatial proximity between fibroblasts and lymphocytes in the stromal compartment, providing a normalized metric useful for comparing immune–stromal interaction patterns across tumor samples. |
| v76jrvgovoon | F | MEAN_MYOFIBRO_INDEX | MEAN_MYOFIBRO_INDEX: This parameter measures the average fibroblast myofibroblast-like morphology within a patch. It is computed by first calculating an individual index for each fibroblast cell, where the index is defined as the aspect ratio of the cell's nucleus (obtained via its minimum rotated rectangle) minus one. A value of zero indicates a circular cell, while higher values denote increased elongation, suggesting a transition towards a myofibroblast-like phenotype. The mean value is then taken across all fibroblasts within the patch, providing a normalized metric suitable for comparing different patient cases. |
| v76jrvgovoon | F | MEDIAN_MYOFIBRO_INDEX | MEDIAN_MYOFIBRO_INDEX: This metric represents the median value of the fibroblast myofibroblast-like morphology index within a patch. Like the mean index, each cell's index is determined by the aspect ratio minus one. By taking the median, this parameter captures the central tendency of the fibroblast morphological changes, offering a robust measure that is less influenced by extreme values, and remains normalized for cross-patient comparisons. |
| v76jrvgovoon | F | SD_MYOFIBRO_INDEX | SD_MYOFIBRO_INDEX: This parameter quantifies the variability in fibroblast morphology within a patch by calculating the standard deviation of the myofibroblast-like morphology indices. It reflects the dispersion or spread in the elongation measurements of individual fibroblast cells. The standard deviation provides insight into the heterogeneity of cell morphology, remains numeric, and is based on a normalized index, making it appropriate for comparative analysis across patient cases. |
| v76jrvgovoon | F | MAX_MYOFIBRO_INDEX | MAX_MYOFIBRO_INDEX: This metric identifies the highest fibroblast myofibroblast-like morphology index within a patch. It indicates the most pronounced elongation observed among the fibroblast cells in that area. By using the computed index that normalizes the cell's shape (aspect ratio minus one), this parameter captures the peak deviation from a round shape, thus facilitating comparisons between different patient patches. |
| v8abqltm7k3u | N | MEAN_EOSIN_INTENSITY | MEAN_EOSIN_INTENSITY measures the average red channel intensity of the neutrophil cytoplasms in each patch. This parameter is computed by extracting the red channel values corresponding to eosin staining within the cell regions, calculating the average intensity for each neutrophil, and then aggregating these averages at the patch level. Being an average intensity metric, it is normalized and can be compared across different patient cases. |
| v8abqltm7k3u | N | EOSIN_INTENSITY_CV | EOSIN_INTENSITY_CV quantifies the relative variability of eosin staining across neutrophils in a patch. It is calculated by dividing the standard deviation of the neutrophil eosin intensity values by their mean intensity and then multiplying by 100, resulting in a percentage. This normalized parameter reflects the heterogeneity of neutrophil activation states and can be used to compare different tumor regions or patient cases. |
| vc9mzweu6u11 | ENL | INFILTRATION_INDEX | INFILTRATION_INDEX represents a normalized metric that quantifies the combined infiltration of three immune cell types—eosinophils, neutrophils, and lymphocytes—within the stromal compartment of a tumor patch. This parameter is calculated by first filtering for stromal cells in a given patch and counting the number of each immune cell type. Each count is then divided by the total number of stromal cells to obtain individual proportions. The final infiltration index is the sum of these proportions, yielding a numeric value between 0 and 1. This normalization allows for comparing immune cell presence across different patient cases and tumor regions. |
| vhkrdsb9oai4 | L | TUMOR_LYMPH_CV | TUMOR_LYMPH_CV measures the coefficient of variation of the nearest-neighbor distances among lymphocytes in the tumor (epithelial) compartment. This normalized metric, computed as the ratio of the standard deviation to the mean distance, captures the heterogeneity in cell spacing and allows for comparison across patient cases. |
| vhkrdsb9oai4 | L | STROMA_LYMPH_CV | STROMA_LYMPH_CV quantifies the relative variability of the nearest-neighbor distances among lymphocytes in the stromal compartment. Like its tumor counterpart, it is computed as the standard deviation divided by the mean distance, offering a normalized measure of spatial dispersion useful for comparative analysis. |
| vhkrdsb9oai4 | L | TUMOR_MEAN_DIST_UM | TUMOR_MEAN_DIST_UM represents the average nearest-neighbor distance between lymphocytes in the tumor compartment, expressed in micrometers. It is obtained by converting pixel distances to physical units and reflects local spatial organization in the tumor region. |
| vhkrdsb9oai4 | L | STROMA_MEAN_DIST_UM | STROMA_MEAN_DIST_UM calculates the mean nearest-neighbor distance between lymphocytes in the stroma, with the distances converted to micrometers. This parameter provides insight into the spatial separation of lymphocytes within the stromal area. |
| vhkrdsb9oai4 | L | TUMOR_STD_DIST_UM | TUMOR_STD_DIST_UM indicates the standard deviation of the nearest-neighbor distances among lymphocytes in the tumor compartment, offering a measure of variability in cell-to-cell distances. It complements the mean distance to depict the spread in spatial organization. |
| vhkrdsb9oai4 | L | STROMA_STD_DIST_UM | STROMA_STD_DIST_UM denotes the standard deviation of the nearest-neighbor distances among lymphocytes in the stromal compartment, reflecting the dispersion in intercellular spacing and providing an assessment of variability in the stroma. |
| vls7yf1jt9c1 | TLPMNEF | T_CELLS_PERCENT | T_CELLS_PERCENT measures the average percentage of epithelial tumor cells present within a defined circular region around each tumor bud in a patch. The values are normalized, as each bud's tumor cell percentage is calculated relative to the total cells in the bud’s microenvironment and then averaged across all buds present in the patch. |
| vls7yf1jt9c1 | TLPMNEF | L_CELLS_PERCENT | L_CELLS_PERCENT quantifies the average percentage of lymphocytes within the circular microenvironment surrounding tumor buds in a tissue patch. This parameter is normalized by converting the raw lymphocyte counts to percentages of total cell counts in each local region, enabling comparisons across different patient cases. |
| vls7yf1jt9c1 | TLPMNEF | P_CELLS_PERCENT | P_CELLS_PERCENT represents the average percentage of plasma cells in the vicinity of tumor buds. The percentage is derived by taking the ratio of plasma cells to the total number of cells within each bud's analysis region, then averaging these percentages over the patch, ensuring a normalized metric. |
| vls7yf1jt9c1 | TLPMNEF | M_CELLS_PERCENT | M_CELLS_PERCENT indicates the average percentage of macrophages located within the defined circular regions around tumor buds in a patch. It is calculated as a percentage of the total cell count in each region and then averaged across all identified buds, making it suitable for cross-case comparisons. |
| vls7yf1jt9c1 | TLPMNEF | N_CELLS_PERCENT | N_CELLS_PERCENT assesses the average percentage of neutrophils in the analyzed microenvironment around tumor buds. Each disk’s neutrophil count is made relative to the total cells present in the region, and the mean of these normalized percentages provides a patch-level metric. |
| vls7yf1jt9c1 | TLPMNEF | E_CELLS_PERCENT | E_CELLS_PERCENT measures the average percentage of eosinophils encountered in the cellular neighborhood of tumor buds. By calculating the percentage of eosinophils relative to the total cells in each circular analysis region and then averaging across all buds, the parameter becomes normalized and comparable across patches. |
| vls7yf1jt9c1 | TLPMNEF | F_CELLS_PERCENT | F_CELLS_PERCENT quantifies the average percentage of connective tissue cells (fibroblasts) in the local area surrounding tumor buds. The value is obtained by determining the fibroblast percentage within the localized region around each tumor bud and then averaging these values over the patch, ensuring normalization. |
| vn6sezfpgcny | LN | NEUTRO_CONTRAST | NEUTRO_CONTRAST is the average contrast value, derived from the gray-level co-occurrence matrix, computed from the gray-scale images of neutrophils located in necrotic-adjacent stroma. This metric quantifies the intensity variation and is normalized through averaging across cells. |
| vn6sezfpgcny | LN | NEUTRO_CORRELATION | NEUTRO_CORRELATION is the mean correlation value from the gray-level co-occurrence matrix for neutrophils in necrotic-adjacent stroma, measuring the linear dependency between pixel pairs in the image. The value is normalized over all neutrophils in the patch. |
| vn6sezfpgcny | LN | NEUTRO_ENERGY | NEUTRO_ENERGY represents the average energy (or uniformity) extracted from the gray-level co-occurrence matrix of neutrophils in the necrotic-adjacent stroma. It reflects the level of order and is computed as an averaged, normalized metric. |
| vn6sezfpgcny | LN | NEUTRO_HOMOGENEITY | NEUTRO_HOMOGENEITY quantifies the average homogeneity from the gray-level co-occurrence matrix for neutrophils in necrotic-adjacent stroma. This parameter indicates how uniform the image texture is, with values normalized across the cell population. |
| vn6sezfpgcny | LN | LYMPHO_CONTRAST | LYMPHO_CONTRAST is the average contrast value calculated from the gray-level co-occurrence matrix for lymphocytes in necrotic-adjacent stroma. It measures intensity differences and is normalized through averaging, making it suitable for comparative analysis. |
| vn6sezfpgcny | LN | LYMPHO_CORRELATION | LYMPHO_CORRELATION is the mean correlation value derived from the gray-level co-occurrence matrix for lymphocytes in necrotic-adjacent stroma. This metric reflects the degree of linear dependency between neighboring pixels, with normalization enabled by averaging. |
| vn6sezfpgcny | LN | LYMPHO_ENERGY | LYMPHO_ENERGY represents the average energy collected from the gray-level co-occurrence matrix of lymphocytes in necrotic-adjacent stroma. This metric assesses texture uniformity and is normalized via averaging over cells. |
| vn6sezfpgcny | LN | LYMPHO_HOMOGENEITY | LYMPHO_HOMOGENEITY measures the average homogeneity from the gray-level co-occurrence matrix for lymphocytes in necrotic-adjacent stroma. This parameter assesses the closeness of pixel distribution, normalized by averaging in each patch. |
| vn6sezfpgcny | LN | CONTRAST_DIFF | CONTRAST_DIFF is the computed difference between the average GLCM contrast values of neutrophils and lymphocytes in necrotic-adjacent stroma. It provides a comparative metric that highlights the variation in texture contrast between the two cell types. |
| vn6sezfpgcny | LN | CORRELATION_DIFF | CORRELATION_DIFF represents the difference between the average GLCM correlation values of neutrophils and lymphocytes. This metric compares the linear dependency of pixel intensities between the two groups, normalized by averaging their individual values. |
| vn6sezfpgcny | LN | ENERGY_DIFF | ENERGY_DIFF is the difference between the average GLCM energy values of neutrophils and lymphocytes in necrotic-adjacent stroma. This difference highlights variations in texture uniformity between the two cell populations, with both values being normalized averages. |
| vn6sezfpgcny | LN | HOMOGENEITY_DIFF | HOMOGENEITY_DIFF is the computed difference between the average GLCM homogeneity values of neutrophils and lymphocytes. By comparing the uniformity of pixel distributions between the two groups, this normalized metric aids in understanding texture differences. |
| vsoekqz63qvt | MTFPN | M_T_INTERFACE_RATIO | M_T_INTERFACE_RATIO: This parameter quantifies the balance between interface macrophages and interface tumor cells by computing a ratio for each patch. It is derived by first identifying tumor cells at the tumor-stroma interface (those in proximity to stroma cells) and macrophages adjacent to tumor cells, then calculating the ratio of these macrophages to tumor cells. This normalization allows for consistent comparisons across different patient cases despite variations in cell counts. |
| vsum9dpmxnbw | TLPMNEF | PSEUDOSTRAT_SCORE | PSEUDOSTRAT_SCORE is a normalized metric representing the ratio of stratified clusters to the total number of valid clusters in a tumor patch. It quantifies the extent of multi-layered organization by assessing whether spatial clusters of cells exhibit distinct consecutive dominant cell types when divided into vertical bins. This ratio allows for direct comparison across different patient cases by accounting for variations in the total number of clusters detected in each patch. |
| vuk9zt48rjc4 | TLPMNEF | GRADIENT_SLOPE_T | GRADIENT_SLOPE_T measures the linear regression slope of the hematoxylin intensity across radial bins from the tumor center to the margin specifically for tumor cells. It quantifies how the nuclear staining changes spatially within the tumor region, providing a normalized measure that can be compared across different patient cases. |
| vuk9zt48rjc4 | TLPMNEF | GRADIENT_SLOPE_L | GRADIENT_SLOPE_L measures the slope of the hematoxylin intensity gradient for lymphocytes. It is derived by computing the average nuclear intensity of lymphocytes in segmented radial bins and applying linear regression, thus providing a numeric and normalized indicator of intensity transition from the tumor center to the margin. |
| vuk9zt48rjc4 | TLPMNEF | GRADIENT_SLOPE_P | GRADIENT_SLOPE_P measures the gradient of hematoxylin intensity for plasma cells. It is computed using the average staining intensity in each radial bin for plasma cells and normalizing the transition from the tumor center to the margin, which enables comparison among various patient cases. |
| vuk9zt48rjc4 | TLPMNEF | GRADIENT_SLOPE_M | GRADIENT_SLOPE_M determines the slope of the change in hematoxylin intensity for macrophages. By dividing the tumor region into normalized radial bins and averaging the nuclear intensity in these bins, a linear regression is performed to yield a numeric parameter reflecting spatial intensity variation. |
| vuk9zt48rjc4 | TLPMNEF | GRADIENT_SLOPE_N | GRADIENT_SLOPE_N calculates the linear regression slope of the hematoxylin intensity gradient for neutrophils, using normalized radial bins that extend from the tumor center to its margin. This parameter is numeric and normalized, allowing for inter-patient comparisons of spatial staining intensity changes. |
| vuk9zt48rjc4 | TLPMNEF | GRADIENT_SLOPE_E | GRADIENT_SLOPE_E represents the gradient slope of hematoxylin intensity for eosinophils. It is derived by averaging the nuclear intensity across normalized radial bins and applying linear regression, providing a normalized, quantitative measure of staining intensity changes from the center to the margin of the tumor. |
| vuk9zt48rjc4 | TLPMNEF | GRADIENT_SLOPE_F | GRADIENT_SLOPE_F quantifies the slope of the hematoxylin intensity gradient for fibroblasts. This parameter is obtained by dividing the tumor region into normalized radial segments and calculating the average nuclear intensity within each segment, with linear regression used to compute a numeric slope that reflects spatial variation. |
| vxhfk51ceavd | TLPE | MEDIAN_NUCLEAR_TEXTURE_CORRELATION | MEDIAN_NUCLEAR_TEXTURE_CORRELATION quantifies the median of Pearson correlation coefficients computed from pairwise comparisons between average texture feature vectors of different cell types (Epithelial, Lymphocyte, Plasma, Eosinophil) that are identified at the infiltration interfaces within tumor patches. The texture features are derived using standard gray-level co-occurrence matrix methods, resulting in normalized values in the range [-1, 1] that allow for direct comparisons across different tumor cases. This parameter is numeric and normalized, making it suitable for comparing local cellular architecture in a standardized way. |
| vxoir1x0dl36 | MN | MACRO_MEAN_GRADIENT | MACRO_MEAN_GRADIENT represents the average intensity of the nuclear color gradient for macrophages within tumor regions. This measurement uses image processing techniques to convert cell nucleus images into grayscale, applies gradient detection to compute local gradient magnitudes, and then calculates the mean value within the segmented nucleus. The resulting value is normalized, enabling comparisons across different patient cases. |
| vxoir1x0dl36 | MN | NEUTRO_MEAN_GRADIENT | NEUTRO_MEAN_GRADIENT represents the average intensity of the nuclear color gradient for neutrophils within tumor regions. The process mirrors that of macrophages by converting the nucleus image to grayscale, computing the gradient magnitudes through edge detection methods, and averaging these values over areas defined by the cell's nucleus mask. This yields a numeric and normalized feature for comparative analysis. |
| vxoir1x0dl36 | MN | GRADIENT_DIFFERENCE | GRADIENT_DIFFERENCE is the numeric difference between the mean nuclear color gradients of macrophages and neutrophils (MACRO_MEAN_GRADIENT minus NEUTRO_MEAN_GRADIENT). It highlights the disparity in nuclear texture between the two cell types, providing insights into their potentially distinct functional or activation states within the tumor environment. |
| w32auhfpn2g8 | TLPMNEF | IMMUNE_DENSITY_PER_MM2 | IMMUNE_DENSITY_PER_MM2 is a normalized metric that quantifies the density of immune cells within a 1 mm² patch of tissue. It is computed by filtering the cells in a patch to include only those classified as immune cells (such as lymphocytes, plasma cells, macrophages, neutrophils, and eosinophils), counting them, and then dividing the count by the constant area of the patch (1 mm²). This results in a numeric value that allows for direct comparisons between different patient cases by providing a standardized measurement of immune cell infiltration. |
| w4s33ckv3y3g | L | LYMPHO_BASOPHILIA_MEAN | LYMPHO_BASOPHILIA_MEAN: This parameter measures the average intensity of the blue channel in the cytoplasm of lymphocyte cells within a patch. The blue channel intensity is used as a proxy for cytoplasmic basophilia, providing a normalized value that reflects the overall staining intensity and allows for comparison across different tumor patches and patient cases. |
| w4s33ckv3y3g | L | LYMPHO_BASOPHILIA_SD | LYMPHO_BASOPHILIA_SD: This parameter captures the standard deviation of the blue channel intensity values of lymphocyte cytoplasm within a patch. It quantifies the heterogeneity in basophilic staining among lymphocytes, offering insight into the variability of immune cell activation and cytoplasmic properties across the patch in a numeric format suitable for inter-case comparisons. |
| w4s33ckv3y3g | L | LYMPHO_BASOPHILIA_MIN | LYMPHO_BASOPHILIA_MIN: This parameter identifies the minimum blue channel intensity among the lymphocyte cells in a given patch. It represents the lowest level of cytoplasmic basophilia observed, ensuring that even the minimal expression values are captured in a normalized, numeric form for comparative analysis. |
| w4s33ckv3y3g | L | LYMPHO_BASOPHILIA_MAX | LYMPHO_BASOPHILIA_MAX: This parameter records the maximum blue channel intensity observed among lymphocyte cells in a patch. It indicates the highest degree of cytoplasmic basophilia detected, serving as a numeric marker that is normalized and can be reliably compared across different patches and patient cases. |
| wbxjim33m700 | L | LYMPHO_NUCLEAR_SIZE_CV | LYMPHO_NUCLEAR_SIZE_CV is the coefficient of variation of lymphocyte nuclear sizes within a patch. It is computed as the ratio of the standard deviation to the mean of nuclear areas of lymphocytes, providing a normalized, dimensionless measure of variability that facilitates robust comparisons across different patient cases. |
| wbxjim33m700 | L | LYMPHO_NUCLEAR_SIZE_MEAN | LYMPHO_NUCLEAR_SIZE_MEAN represents the average nuclear area of lymphocytes in the patch. The value is calculated by converting the nuclear area from pixel measurements to square micrometers, resulting in a standardized metric that can be compared across different samples. |
| wbxjim33m700 | L | LYMPHO_NUCLEAR_SIZE_STD | LYMPHO_NUCLEAR_SIZE_STD indicates the standard deviation of lymphocyte nuclear areas in the patch. Expressed in square micrometers, this parameter quantifies the degree of dispersion around the mean nuclear size, aiding in the evaluation of the heterogeneity of lymphocyte nuclei across patient samples. |
| wfv5j8t7bi37 | TLPMNEF | NON_TUMOR_TO_TUMOR_RATIO | The NON_TUMOR_TO_TUMOR_RATIO parameter quantifies the normalized relationship between non-tumor cells and tumor cells within the epithelial tumor compartment of a patch. It is calculated by dividing the count of non-tumor cells (which include lymphocytes, plasma cells, macrophages, neutrophils, eosinophils, and connective tissue cells) by the count of tumor cells (specifically epithelial tumor cells). This normalized metric allows for meaningful comparisons across different patches and patient cases, as it provides insight into the relative cellular composition of the tumor microenvironment. The parameter is numeric and handles cases with zero tumor cells by assigning a value such as NaN to avoid division errors. |
| wfy8j0vvyvfw | TLMF | OD_SKEWNESS | OD_SKEWNESS measures the skewness of the optical density distribution aggregated from selected cell types in a patch. This parameter quantifies the asymmetry of the distribution, providing insights into the irregularities in tissue architecture, and is normalized for comparison across different patient cases. |
| wfy8j0vvyvfw | TLMF | MEAN_OD | MEAN_OD represents the average optical density calculated over all selected cell regions within a patch. By summarizing the central tendency of the stain intensity, it serves as a normalized metric that facilitates direct comparisons across patches and patient cases. |
| wfy8j0vvyvfw | TLMF | STD_OD | STD_OD quantifies the dispersion or variability of the optical density values in a patch. This standard deviation is derived from the aggregated cell measurements and is normalized, thereby allowing consistent comparison of heterogeneity in tissue properties between different patient samples. |
| whgod6yxdsrv | T | MEAN_NUCLEAR_AREA | MEAN_NUCLEAR_AREA represents the average area of tumor cell nuclei within each patch, converted to square micrometers using a defined pixel-to-micrometer scale. It reflects the typical size of tumor cell nuclei and is normalized for comparison across different patient cases. |
| whgod6yxdsrv | T | VARIANCE_NUCLEAR_AREA | VARIANCE_NUCLEAR_AREA quantifies the variability in the nuclear areas of tumor cells within each patch. By calculating the statistical variance of these areas, it indicates the heterogeneity in nuclear sizes, providing insights into tumor cell diversity that can be compared across patient samples. |
| wliv93z8ljdk | M | MEAN_VACUOLATION_SCORE | MEAN_VACUOLATION_SCORE measures the average percentage of the cell area that is occupied by vacuoles in macrophages within a given patch. This parameter is derived by computing the vacuolated area for each macrophage using a fixed intensity threshold to identify vacuoles, normalizing it by the total cell area, and then averaging these percentage values across all macrophages in the patch to allow for comparison across different patient cases. |
| wliv93z8ljdk | M | STD_VACUOLATION_SCORE | STD_VACUOLATION_SCORE quantifies the variability in the vacuolation scores among macrophages in a patch. It represents the standard deviation of the percentage of the cell area occupied by vacuoles, providing insight into the heterogeneity of vacuolation across macrophages. Being a normalized measure, it allows for meaningful comparisons across different patches and patient cases. |
| wnd16c8s473q | LE | LYMPHO_KURTOSIS_R | LYMPHO_KURTOSIS_R: This parameter represents the mean kurtosis of the red channel intensity distribution in the cytoplasmic regions of lymphocytes, measured per patch. It quantifies the tailedness and peakedness of the red pixel values, providing a normalized metric that allows for comparison across different patient cases. |
| wnd16c8s473q | LE | LYMPHO_KURTOSIS_G | LYMPHO_KURTOSIS_G: This parameter provides the mean kurtosis of the green channel intensity distribution within the lymphocyte cytoplasmic regions at the patch level. It reflects the shape of the pixel intensity distribution in the green spectrum, enabling normalized comparisons between different samples. |
| wnd16c8s473q | LE | LYMPHO_KURTOSIS_B | LYMPHO_KURTOSIS_B: This parameter captures the mean kurtosis of the blue channel intensity distribution in the cytoplasmic regions of lymphocytes, aggregated per patch. It assesses the distribution characteristics of blue pixel values and offers a normalized measure for comparing different patient cases. |
| wnd16c8s473q | LE | EOS_KURTOSIS_R | EOS_KURTOSIS_R: This parameter indicates the mean kurtosis of the red channel intensity distribution in the cytoplasmic regions of eosinophils at the patch level. It objectively measures the tailedness of the red intensity values, serving as a normalized metric for inter-case comparisons. |
| wnd16c8s473q | LE | EOS_KURTOSIS_G | EOS_KURTOSIS_G: This parameter reflects the mean kurtosis of the green channel intensity distribution in the eosinophil cytoplasmic regions per patch. It describes the distribution of green pixel intensities and is normalized to facilitate comparisons across different patient samples. |
| wnd16c8s473q | LE | EOS_KURTOSIS_B | EOS_KURTOSIS_B: This parameter represents the mean kurtosis of the blue channel intensity distribution in the eosinophil cytoplasmic regions, computed per patch. It provides a normalized metric of the distribution shape of blue pixel values, enabling consistent comparisons across different cases. |
| ws0s4cphw8j6 | TLPMNEF | VACUOLIZATION_INDEX | VACUOLIZATION_INDEX measures the average fraction of a cell's area that is occupied by vacuoles across all cells in a patch. This index is normalized because it is computed as a ratio relative to the total cell area, making it comparable across different tumor cases. |
| ws0s4cphw8j6 | TLPMNEF | MEAN_VACUOLE_FRACTION | MEAN_VACUOLE_FRACTION represents the average vacuolization calculated by determining the ratio of vacuole (clear) pixel area to the total cell area for each cell. This parameter is numerically derived and normalized, allowing for comparisons across patient cases. |
| ws0s4cphw8j6 | TLPMNEF | SD_VACUOLE_FRACTION | SD_VACUOLE_FRACTION is the standard deviation of the vacuolization fractions of cells in a patch. It quantifies the heterogeneity in vacuolization across cells and is normalized since each cell’s fraction is computed relative to its own area. |
| ws0s4cphw8j6 | TLPMNEF | MAX_VACUOLE_FRACTION | MAX_VACUOLE_FRACTION captures the maximum vacuolization fraction observed among all cells in a patch. This measurement indicates the cell with the highest proportion of vacuole area relative to its total cell area and is normalized, enabling cross-patient comparisons. |
| wvde31q509vs | NMF | AVG_CURVATURE_NEUTROPHILS | AVG_CURVATURE_NEUTROPHILS represents the average nuclear boundary curvature of neutrophils in a patch. It quantifies the mean curvature computed along the closed polygonal boundary of neutrophil nuclei, normalized by the length of the segments and scaled to micrometers using the conversion factor. This normalized metric allows for comparison across different patient cases. |
| wvde31q509vs | NMF | AVG_CURVATURE_MACROPHAGES | AVG_CURVATURE_MACROPHAGES represents the average nuclear boundary curvature of macrophages in a patch. It is determined by computing local curvature at each vertex along the segmented nuclear polygon, averaging these values, and converting the metric to micrometers. This parameter is normalized and numeric, making it suitable for cross-case analysis. |
| wvde31q509vs | NMF | AVG_CURVATURE_FIBROBLASTS | AVG_CURVATURE_FIBROBLASTS represents the average nuclear boundary curvature of fibroblasts (identified as connective tissue cells) in a patch. The curvature is computed at each vertex along the polygon defining the nucleus, with the average curvature then scaled to micrometers. This normalized numeric metric enables consistent comparison between patient samples. |
| wvde31q509vs | NMF | TRICELLULAR_CURVATURE_INDEX | TRICELLULAR_CURVATURE_INDEX is a combined index that reflects the coordinated morphological curvature of neutrophils, macrophages, and fibroblasts within a patch. It is calculated as the arithmetic mean of the three individual average curvatures when all are available, and it is scaled using the conversion factor to yield a normalized value in micrometers. This parameter provides an integrated measure of tumor-immune-stromal interactions. |
| x99hm7cu7afp | L | MEAN_IRREG_INDEX | MEAN_IRREG_INDEX represents the average value of the membrane irregularity index for lymphocytes within a patch. This index is computed using the ratio derived from the cell perimeter and area, normalized against the expected value for a perfect circle. Higher mean values indicate more irregular lymphocyte membrane shapes, making it useful for comparing cellular characteristics across different patient cases. |
| x99hm7cu7afp | L | MEDIAN_IRREG_INDEX | MEDIAN_IRREG_INDEX indicates the median irregularity index of lymphocyte membranes in the patch. By using the median, this parameter provides a robust measure of the typical cell membrane irregularity that is less sensitive to extreme values or outliers, thus facilitating reliable comparisons across samples. |
| x99hm7cu7afp | L | SD_IRREG_INDEX | SD_IRREG_INDEX is the standard deviation of the lymphocyte membrane irregularity indices within a patch. It measures the variability or spread in the irregularity values, offering insights into the heterogeneity of lymphocyte membrane structures among different patches. |
| x99hm7cu7afp | L | MIN_IRREG_INDEX | MIN_IRREG_INDEX is the lowest value in the set of membrane irregularity indices calculated for lymphocytes within a patch. This parameter highlights the most regular cell membrane observed, providing a comparative reference point against more irregular cell shapes. |
| x99hm7cu7afp | L | MAX_IRREG_INDEX | MAX_IRREG_INDEX is the highest irregularity index recorded in a patch, reflecting the most irregular lymphocyte membrane shape. This measure can be informative in identifying patches where lymphocytes display extreme morphological deviations. |
| x9hmumsjfsx5 | TLM | MEAN_IRREGULARITY | MEAN_IRREGULARITY: This parameter represents the average tumor cell perimeter irregularity score within a patch. It quantifies how much the shape of tumor cell boundaries deviates from a perfect circle (which would score 1) by aggregating the normalized ratio of each cell's perimeter squared to its area. Being a dimensionless metric, it facilitates meaningful comparisons across different patient cases. |
| x9hmumsjfsx5 | TLM | SD_IRREGULARITY | SD_IRREGULARITY: This parameter measures the standard deviation of the tumor cell perimeter irregularity scores within the patch. It provides insight into the variability or heterogeneity in tumor cell morphology, reflecting how consistently or diversely the tumor cells deviate from the ideal shape. It is a numeric, normalized measure enabling cross-case comparisons. |
| x9hmumsjfsx5 | TLM | IMMUNE_DENSITY_MM2 | IMMUNE_DENSITY_MM2: This parameter calculates the density of immune cells (lymphocytes and macrophages) per square millimeter in the patch. It is derived by normalizing the raw count of immune cells by the patch area and scaling to cells per mm², ensuring that it is a normalized and comparable metric across different samples. |
| xa78hq6hi66x | F | FIBROBLAST_ALIGNMENT_VARIANCE | FIBROBLAST_ALIGNMENT_VARIANCE quantifies the circular variance of fibroblast nuclear orientations within a tumor patch. This parameter is computed by aggregating the orientation angles of fibroblast nuclei, calculating the mean cosine and sine, and then determining the mean resultant length. The circular variance, defined as 1 minus the mean resultant length, provides a normalized metric (ranging from 0 to 1) where lower values indicate greater alignment among nuclei. This normalized measure can be compared across different patient cases and patch analyses. |
| xa78hq6hi66x | F | MEAN_ORIENTATION_RAD | MEAN_ORIENTATION_RAD represents the average orientation angle of fibroblast nuclei within a patch, calculated using circular statistics by averaging the sine and cosine components of the individual angles and then computing the arctan of these averages. The result, given in radians, offers a normalized summary of the predominant directionality of fibroblast alignment in the tumor microenvironment and is suitable for comparisons across different patient cases. |
| xg3efe69ec50 | F | FIBROBLAST_VESICLE_RATE | FIBROBLAST_VESICLE_RATE is a normalized, numeric metric that measures the fraction of fibroblast cells in a tissue patch that display cytoplasmic microvesicles. This parameter is calculated by detecting vesicle-like structures within the cytoplasm of fibroblasts through image intensity thresholding and blob detection algorithms. The measure, bounded between 0 and 1, allows for comparative analysis across different patient cases and tissue patches, as it represents the ratio of fibroblast cells exhibiting microvesicles to the total number of fibroblast cells analyzed in each patch. |
| xhp9qziiako5 | TLPMNEF | ENTROPY_MEAN_ALL | ENTROPY_MEAN_ALL: This parameter represents the average cytoplasmic intensity entropy measured across all cells in a patch. It quantifies the overall complexity and randomness of the grayscale intensity distribution within the cell cytoplasm, providing a normalized value that facilitates comparison across different patient samples. |
| xhp9qziiako5 | TLPMNEF | ENTROPY_MEDIAN_ALL | ENTROPY_MEDIAN_ALL: This parameter indicates the median value of the cytoplasmic intensity entropy for all cells in a patch. It captures the central tendency of the entropy distribution and offers a robust, normalized measure for cross-sample comparisons. |
| xhp9qziiako5 | TLPMNEF | ENTROPY_STD_ALL | ENTROPY_STD_ALL: This parameter measures the standard deviation of cytoplasmic intensity entropy for all cells in a patch. It reflects the variability or heterogeneity in the intensity distribution among cells, thus providing a normalized metric for assessing intra-patch diversity. |
| xhp9qziiako5 | TLPMNEF | ENTROPY_MEAN_EPITHELIAL | ENTROPY_MEAN_EPITHELIAL: This parameter calculates the mean cytoplasmic entropy specifically for epithelial cells. It provides a normalized measure of the average complexity in cytoplasmic intensity distribution for this cell type, aiding in comparative analysis between different cases. |
| xhp9qziiako5 | TLPMNEF | ENTROPY_MEAN_LYMPHOCYTE | ENTROPY_MEAN_LYMPHOCYTE: This parameter represents the mean cytoplasmic entropy for lymphocyte cells, summarizing the typical entropy value from these cells. It is normalized to allow effective comparisons across different patient samples. |
| xhp9qziiako5 | TLPMNEF | ENTROPY_MEAN_PLASMA | ENTROPY_MEAN_PLASMA: This parameter provides the mean cytoplasmic intensity entropy for plasma cells, indicating the average level of complexity in their cytoplasmic intensity patterns. This numeric value is normalized for cross-sample assessments. |
| xhp9qziiako5 | TLPMNEF | ENTROPY_MEAN_MACROPHAGE | ENTROPY_MEAN_MACROPHAGE: This parameter reflects the mean cytoplasmic entropy calculated for macrophage cells. It offers a normalized measure of cytoplasmic intensity complexity specific to these cells, making it useful for comparative analysis. |
| xhp9qziiako5 | TLPMNEF | ENTROPY_MEAN_NEUTROPHIL | ENTROPY_MEAN_NEUTROPHIL: This parameter denotes the mean cytoplasmic entropy for neutrophil cells, capturing the average heterogeneity in the cytoplasmic intensity distribution. It is normalized to ensure validity across various patient samples. |
| xhp9qziiako5 | TLPMNEF | ENTROPY_MEAN_EOSINOPHIL | ENTROPY_MEAN_EOSINOPHIL: This parameter measures the mean cytoplasmic entropy for eosinophil cells, summarizing the average degree of randomness in their cytoplasmic intensity patterns. The normalized value supports robust cross-sample comparisons. |
| xhp9qziiako5 | TLPMNEF | ENTROPY_MEAN_CONNECTIVE | ENTROPY_MEAN_CONNECTIVE: This parameter calculates the mean cytoplasmic entropy for connective tissue cells. It provides a normalized measure of the average complexity in the cytoplasmic intensity of these cells, facilitating reliable comparisons between different patient cases. |
| xpj61k7w08hr | TFMLN | STD_MEMBRANE_CURVATURE | STD_MEMBRANE_CURVATURE quantifies the variability in the membrane curvature of cells within a tumor patch by computing the standard deviation of the mean curvature values measured per cell. The process involves first calculating local curvature at multiple points along each cell's outlined membrane and then averaging these to obtain a single curvature value per cell. Finally, the dispersion of these values across all relevant cells (tumor cells, fibroblasts, macrophages, lymphocytes, and neutrophils) in the patch is computed, reflecting the heterogeneity of the mechanical interactions in the tumor microenvironment. |
| xpj61k7w08hr | TFMLN | MEAN_MEMBRANE_CURVATURE | MEAN_MEMBRANE_CURVATURE represents the average of the mean curvature values of cell membranes across all analyzed cells within a tumor patch. The parameter is derived by computing local curvature values along the cell boundary, averaging these values for each cell, and then calculating the mean of these averages within the patch. This value provides insight into the overall membrane curvature characteristic of the cell population in the region, offering a central measure to complement the variability indicated by the standard deviation. |
| xqv6hfpwql0m | TLPMNEF | UNIMODALITY_SCORE | UNIMODALITY_SCORE represents the p-value obtained from Hartigan’s dip test applied to the distribution of pairwise cell distances. A higher p-value indicates stronger evidence for unimodality, suggesting a more homogeneous cell organization in the patch. This metric is statistically derived and normalized, making it suitable for comparing different patient cases. |
| xqv6hfpwql0m | TLPMNEF | DIP_STATISTIC | DIP_STATISTIC is a value produced by Hartigan’s dip test that quantifies the deviation of the cell distance distribution from a unimodal profile. Lower dip statistic values indicate a distribution that is more unimodal. As a normalized metric, it facilitates objective comparisons across tumor regions. |
| xqv6hfpwql0m | TLPMNEF | MEAN_DISTANCE_UM | MEAN_DISTANCE_UM measures the average pairwise Euclidean distance between cell centroids within a patch, after converting pixel distances to micrometers. This normalized metric reflects the typical separation between cells, allowing meaningful comparisons between different patient samples. |
| xqv6hfpwql0m | TLPMNEF | STD_DISTANCE_UM | STD_DISTANCE_UM represents the standard deviation of the pairwise distances between cells in a patch, expressed in micrometers. It quantifies the variability in cell spacing, and its normalized nature ensures that comparisons across patches and different patient cases are valid. |
| xug1xdoxpgm6 | ELMNP | MEAN_MIN_DIST_IMMUNE | MEAN_MIN_DIST_IMMUNE quantifies the average minimum distance, measured in micrometers, from each immune cell to the nearest immune cell of a different type within a tumor patch. The metric is derived by first selecting immune cells located in the tumor stroma, computing Euclidean distances between their centroids (converted from pixels to micrometers), and then averaging the smallest distance for each immune cell that has a neighbor of a different type. This normalization makes the parameter suitable for comparing spatial immune cell distributions across different patient cases. |
| xuq2szzx6okl | TLPMNE | CHROMATIN_REGULARITY_INDEX | CHROMATIN_REGULARITY_INDEX represents the overall measure of chromatin distribution regularity across all cell types in a patch. It is computed by averaging the mean gray-level co-occurrence matrix (GLCM) energy values from each cell type, providing a normalized metric that facilitates comparison of chromatin uniformity between different tumor patches and patient cases. |
| xuq2szzx6okl | TLPMNE | Epithelial_MEAN_ENERGY | Epithelial_MEAN_ENERGY quantifies the average GLCM energy specifically for tumor (epithelial) cells, indicating the regularity and uniformity of their chromatin patterns. A higher value suggests more homogeneous chromatin texture in epithelial nuclei, and because it is an average measure, it is normalized to allow inter-case comparability. |
| xuq2szzx6okl | TLPMNE | Lymphocyte_MEAN_ENERGY | Lymphocyte_MEAN_ENERGY reflects the average chromatin texture regularity in lymphocyte nuclei by calculating the mean GLCM energy. This normalized metric summarizes the uniformity of chromatin distribution in lymphocytes and supports comparative analysis across multiple patient images. |
| xuq2szzx6okl | TLPMNE | Plasma_MEAN_ENERGY | Plasma_MEAN_ENERGY measures the average chromatin energy derived from plasma cell nuclei, indicating the degree of uniformity in their chromatin distribution. As a normalized value, it is useful for comparing chromatin patterns across different tumor patches. |
| xuq2szzx6okl | TLPMNE | Macrophage_MEAN_ENERGY | Macrophage_MEAN_ENERGY computes the mean gray-level co-occurrence matrix energy for macrophage nuclei, serving as an indicator of chromatin regularity in these cells. This normalized parameter assists in cross-case comparisons of macrophage chromatin texture. |
| xuq2szzx6okl | TLPMNE | Neutrophil_MEAN_ENERGY | Neutrophil_MEAN_ENERGY calculates the average GLCM energy for neutrophil nuclei, capturing the chromatin regularity within these cells. Its normalized nature allows it to be used reliably in comparing results from different patient cases. |
| xuq2szzx6okl | TLPMNE | Eosinophil_MEAN_ENERGY | Eosinophil_MEAN_ENERGY represents the mean chromatin texture regularity for eosinophil nuclei measured through GLCM energy. This normalized measure facilitates the comparison of chromatin distribution uniformity in eosinophils across various patches and patient samples. |
| y4jb5ouacurt | FN | MEAN_ORIENTATION_DIFFERENCE | MEAN_ORIENTATION_DIFFERENCE represents the normalized absolute angular difference in degrees between the mean orientation of fibroblast nuclei and that of neutrophil nuclei. Its calculation accounts for the circular nature of orientation data (ensuring the difference is within a defined range), making it useful for comparing orientation patterns across different patient cases. |
| y4jb5ouacurt | FN | FIBROBLAST_MEAN_ANGLE | FIBROBLAST_MEAN_ANGLE represents the normalized average orientation angle of fibroblast nuclei within a patch. This parameter is computed using circular statistics to accurately capture the predominant nuclear alignment, thus facilitating comparisons across various cases and patches. |
| y4jb5ouacurt | FN | NEUTROPHIL_MEAN_ANGLE | NEUTROPHIL_MEAN_ANGLE represents the normalized average orientation angle of neutrophil nuclei within a patch. Similar to the fibroblast measure, this parameter is derived using methods that account for the circular qualities of angular data, making it a reliable metric for cross-case comparison of neutrophil nuclear alignment. |
| y6zzjvoa41dk | TLMNE | INFILTRATION_THICKNESS_TUMOR_UM | INFILTRATION_THICKNESS_TUMOR_UM measures the radial thickness of tumor cell infiltration. It is computed by determining the maximum and minimum distances of tumor cell centroids from a calculated tumor center in each patch, and then taking the difference. The resulting value, given in micrometers, reflects the spatial spread of tumor cells and is normalized for comparison across patient cases. |
| y6zzjvoa41dk | TLMNE | INFILTRATION_THICKNESS_LYMPHO_UM | INFILTRATION_THICKNESS_LYMPHO_UM quantifies the radial thickness of lymphocyte infiltration in a patch. This parameter is derived by calculating the range (difference between maximum and minimum distances) of lymphocyte cell positions from the tumor center, providing a normalized, numeric measure of lymphocyte dispersion in the tumor microenvironment. |
| y6zzjvoa41dk | TLMNE | INFILTRATION_THICKNESS_MACRO_UM | INFILTRATION_THICKNESS_MACRO_UM represents the radial thickness of macrophage infiltration. It is calculated as the difference between the furthest and closest macrophage positions relative to the tumor center in a given patch, offering a normalized metric that reflects the spatial distribution of macrophages. |
| y6zzjvoa41dk | TLMNE | INFILTRATION_THICKNESS_NEUTRO_UM | INFILTRATION_THICKNESS_NEUTRO_UM measures the radial thickness of neutrophil infiltration. By determining the range of neutrophil distances from the tumor center within each patch, this numeric parameter provides insight into the extent of neutrophil dispersion in the tumor region. |
| y6zzjvoa41dk | TLMNE | INFILTRATION_THICKNESS_EOSINO_UM | INFILTRATION_THICKNESS_EOSINO_UM captures the radial thickness of eosinophil infiltration. It computes the difference between the maximum and minimum distances of eosinophil positions from the tumor center in micrometers, yielding a normalized numeric value that indicates how broadly eosinophils are distributed. |
| y6zzjvoa41dk | TLMNE | INFILTRATION_THICKNESS_STD_UM | INFILTRATION_THICKNESS_STD_UM is the standard deviation of the infiltration thicknesses across the five cell types (tumor, lymphocyte, macrophage, neutrophil, and eosinophil). This parameter quantifies the variability or heterogeneity of immune cell infiltration within a patch, offering a normalized measure to compare variations in spatial dispersion across different patient cases. |
| ybk4sdb8e58r | L | LYMPHO_FD_MEAN | LYMPHO_FD_MEAN: This parameter represents the average fractal dimension of lymphocyte nuclei within a tissue patch. It is derived from analyzing the binary nuclear masks using a box counting method, which assesses the complexity of the nuclear boundary. As a normalized measurement, it facilitates comparisons of nuclear shape complexity across different patient cases. |
| ybk4sdb8e58r | L | LYMPHO_FD_SD | LYMPHO_FD_SD: This parameter quantifies the standard deviation of the fractal dimensions calculated for lymphocyte nuclei in a tissue patch. It captures the variability in nuclear complexity among lymphocytes and helps evaluate the consistency or heterogeneity of nuclear morphology. Being a normalized statistical metric, it supports comparative analysis across tumor patches. |
| ybk4sdb8e58r | L | LYMPHO_FD_MAX | LYMPHO_FD_MAX: This parameter indicates the maximum fractal dimension observed among lymphocyte nuclei in a tissue patch. It highlights the highest level of nuclear boundary complexity detected within the patch, reflecting potential cellular activation or irregularity. Like the other fractal dimension metrics, it is normalized to allow for meaningful comparisons between different cases. |
| ykx5i2sg4lm3 | L | MEAN_LYMPHO_NUCLEAR_INDENT | MEAN_LYMPHO_NUCLEAR_INDENT is a normalized metric that represents the average deviation of lymphocyte nuclei from a convex shape, calculated by comparing the area of a nucleus to its convex hull area. This value, which ranges from 0 to 1, provides a standardized measure of nuclear indentation, enabling comparison across different patient cases and tissue patches. |
| ykx5i2sg4lm3 | L | STD_LYMPHO_NUCLEAR_INDENT | STD_LYMPHO_NUCLEAR_INDENT is a normalized metric that quantifies the variability in the nuclear indentation scores of lymphocytes within a tissue patch. By measuring the standard deviation of these indentation scores, it offers insight into the consistency of nuclear shape alterations in the sample, making it useful for comparative analysis across different patient cases. |
| yn308lmfjnrk | T | LIPID_DROPLET_SCORE_MEAN | LIPID_DROPLET_SCORE_MEAN is a normalized metric that represents the mean ratio of vacuole area to cytoplasm area across tumor cells within a patch. This parameter facilitates comparisons across different patient cases by eliminating the bias of raw cell counts. |
| yn308lmfjnrk | T | LIPID_DROPLET_SCORE_SD | LIPID_DROPLET_SCORE_SD is a numeric measure indicating the standard deviation of the lipid droplet scores computed in the patch. It captures the variability or consistency in the lipid accumulation ratio among tumor cells, serving as a normalized indicator for heterogeneity in tumor cell metabolism. |
| yn308lmfjnrk | T | LIPID_DROPLET_SCORE_MAX | LIPID_DROPLET_SCORE_MAX is a normalized value that reflects the highest lipid droplet score observed among tumor cells in a patch. This measurement identifies the cell with maximum lipid accumulation relative to its cytoplasm area, allowing for comparative analysis across different patches and cases. |
| yq1euq150vyd | TLPMNEF | TOTAL_SHIFT_SCORE | TOTAL_SHIFT_SCORE represents the aggregate of compositional changes between adjacent radial bins within a tumor patch. It is calculated by summing the absolute differences in the proportions of various cell types (normalized by cell counts in each bin) observed between each pair of adjacent bins. This score quantifies how drastically the cellular composition shifts from the tumor center to its margin, thereby indicating spatial heterogeneity in the tumor microenvironment. |
| yq1euq150vyd | TLPMNEF | MEAN_BIN_TRANSITION | MEAN_BIN_TRANSITION measures the average magnitude of change in cell type proportions between consecutive radial bins. This parameter is derived by first computing the absolute differences in cell type proportions between each adjacent pair of radial bins, and then calculating the average of these differences. It offers a normalized and comparable metric of the gradual compositional shifts across tumor regions. |
| yumobeuw65ji | T | TUMOR_CYTOPLASM_HOMOGENEITY_MEAN | TUMOR_CYTOPLASM_HOMOGENEITY_MEAN represents the average texture homogeneity score derived from the gray-level co-occurrence matrix (GLCM) computed on normalized cytoplasmic pixel intensities of tumor cells within each patch. This metric is normalized to a range between 0 and 1, allowing direct comparison between different patient cases by reflecting the overall uniformity of tumor cell cytoplasmic staining, with higher values indicating more uniformity. |
| yumobeuw65ji | T | TUMOR_CYTOPLASM_HOMOGENEITY_SD | TUMOR_CYTOPLASM_HOMOGENEITY_SD quantifies the variability in the cytoplasmic homogeneity scores of tumor cells within a patch by calculating the standard deviation of these scores. This metric, also derived from GLCM analysis on normalized pixel intensities, provides insight into the heterogeneity of tumor cell textures in the patch. A lower standard deviation suggests a more consistently homogeneous texture among the cells, while a higher value indicates greater variability. |
| yyklyg0bgzm5 | TFE | TUMOR_DENSITY_PER_1000UM2 | TUMOR_DENSITY_PER_1000UM2 represents the density of tumor cells within a defined patch, normalized to a standard area of 1000 square micrometers. This parameter is computed by dividing the total number of tumor cells identified in the patch by the patch area and then scaling the result, allowing for meaningful comparisons across different patient cases. |
| yyklyg0bgzm5 | TFE | FIBROBLAST_DENSITY_PER_1000UM2 | FIBROBLAST_DENSITY_PER_1000UM2 quantifies the density of fibroblast cells (connective tissue cells) within a patch. The metric is normalized to a 1000 square micrometer area by dividing the number of fibroblasts by the patch area, thereby facilitating comparison of fibroblast presence across various tumor regions. |
| yyklyg0bgzm5 | TFE | EOSINOPHIL_DENSITY_PER_1000UM2 | EOSINOPHIL_DENSITY_PER_1000UM2 measures the density of eosinophil cells in a patch, normalized to a standard area of 1000 square micrometers. This parameter is calculated by taking the number of eosinophils present and scaling by the patch area, enabling consistent evaluation of eosinophil distribution among different samples. |
| z0qnz1mt6b4h | TLPMNEF | TOTAL_CLUSTER_PERIMETER_UM | TOTAL_CLUSTER_PERIMETER_UM quantifies the aggregated outer boundary lengths of all detected cell clusters within a standardized tumor patch. These perimeters are computed by first clustering cell types based on their spatial coordinates, then merging individual cell polygons into union polygons to represent clusters, and finally summing the perimeters of these clusters. The resulting value, converted into micrometers, is normalized across patches due to the consistent patch size, allowing for meaningful comparisons between different patient cases. |
| z7eozkt9lknh | TLNF | CLUSTER_COUNT | CLUSTER_COUNT represents the number of valid cell clusters, where each cluster contains at least 3 cells. It is computed for each tumor patch based on a graph‐based spatial clustering of selected cell types. Since each patch is of a standard size, this count is normalized for cross-patient comparisons. |
| z7eozkt9lknh | TLNF | MEAN_COMPLEXITY_RATIO | MEAN_COMPLEXITY_RATIO is the average of the ratios between the total area occupied by cells in a cluster and the corresponding convex hull area that encloses those cells. This ratio quantifies the shape complexity of cell clusters, with lower values indicating more irregular and intricate cluster shapes. Being a ratio, it is properly normalized across patches. |
| z7eozkt9lknh | TLNF | STD_COMPLEXITY_RATIO | STD_COMPLEXITY_RATIO is the standard deviation of the complexity ratios across all valid clusters in a patch. It provides a measure of the variability of cluster shape complexities. As a dimensionless value computed from normalized ratios, it is suitable for comparing different patient cases. |
| z7eozkt9lknh | TLNF | MEAN_CLUSTER_SIZE | MEAN_CLUSTER_SIZE indicates the average number of cells in each valid cluster within a patch. Although it represents raw counts, the clusters are identified from patches of a fixed, standardized size, ensuring that this metric can be reliably compared across different cases. |
| z7eozkt9lknh | TLNF | MAX_CLUSTER_SIZE | MAX_CLUSTER_SIZE denotes the size of the largest cluster in the patch, measured by the number of cells it contains. Similar to MEAN_CLUSTER_SIZE, because the extraction is performed on uniformly sized patches, this parameter is considered normalized for comparative analysis. |
| zi9b3r8lm8si | TLPMNF | MEAN_LACUNARITY | MEAN_LACUNARITY represents the average lacunarity index computed over all sliding windows within a tissue patch. This metric aggregates the local spatial heterogeneity by averaging the lacunarity values calculated over different grid subregions, providing an overall measure of how irregularly the cell distributions are spaced across the patch. As a dimensionless and normalized value, it facilitates comparison between different patient cases. |
| zi9b3r8lm8si | TLPMNF | STD_LACUNARITY | STD_LACUNARITY quantifies the standard deviation of the lacunarity indices across the patch's sliding windows. By measuring the variability of local heterogeneity, it reflects the consistency or fluctuation in cell distribution patterns within the tissue patch, making it a valuable normalized metric for comparing spatial complexity between different cases. |
| zi9b3r8lm8si | TLPMNF | MIN_LACUNARITY | MIN_LACUNARITY is the smallest lacunarity index observed among all windows in the patch, highlighting the most homogeneous subregion within the tissue area. This parameter, being normalized and derived from the ratio of variance to mean squared count, assists in identifying regions with minimal cellular irregularity compared to other areas in the same patch. |
| zi9b3r8lm8si | TLPMNF | MAX_LACUNARITY | MAX_LACUNARITY is the highest lacunarity index found during the sliding window analysis of a patch. This parameter indicates the area with the greatest degree of spatial heterogeneity and irregularity in cell distribution, serving as a normalized measure to contrast highly heterogeneous regions with more uniformly organized tissue areas. |
| zj72es5dbty1 | TLMF | SYNERGY_RATIO | SYNERGY_RATIO is a derived spatial metric computed for each patch that quantifies the balanced presence of tumor cells, lymphocytes, macrophages, and fibroblasts in perivascular areas. It is calculated as the ratio of the geometric mean to the arithmetic mean of the perivascular cell counts. This yields a value that is close to 1 when the cell types are present in similar numbers and decreases when one cell type predominates, making it a normalized metric suitable for comparing different patient cases. |
| zpt9hmkd307n | TLPMNEF | JS_DIVERGENCE | JS_DIVERGENCE is a normalized numeric measure that quantifies the divergence between the cell type distributions in tumor and stroma compartments. The metric is computed using the Jensen-Shannon divergence method, which first converts the absolute cell counts from each compartment into probability distributions. It then measures the similarity between these two distributions, yielding a value between 0 and 1, where a higher value indicates a greater difference between the compartments. This normalization makes the parameter comparable across different patient cases and patches. |
| zuqi0p6z9r0v | LP | LYMPHO_DENSITY | LYMPHO_DENSITY measures the density of lymphocytes within the tumor compartment. It is derived by dividing the number of lymphocytes detected in a patch by the area of the tumor region (expressed in square micrometers), ensuring that the measurement is normalized for different patch sizes and can be compared across patients. |
| zuqi0p6z9r0v | LP | PLASMA_DENSITY | PLASMA_DENSITY measures the density of plasma cells in the tumor area. It is calculated by dividing the count of plasma cells by the tumor compartment area, making the value normalized and allowing for comparison between different patches and cases. |
| zuqi0p6z9r0v | LP | DUAL_INFILTRATION_SCORE | DUAL_INFILTRATION_SCORE quantifies the simultaneous infiltration of both lymphocytes and plasma cells by computing the harmonic mean of their respective densities. This parameter provides a balanced assessment of dual immune cell presence within the tumor region and is normalized for tumor area, making it suitable for comparative analysis across different patients. |
| 051edvdwzu6d | TM | GIANT_MAC_FREQUENCY | GIANT_MAC_FREQUENCY measures the fraction of macrophages that are tumor-associated giant cells in a given patch. By normalizing the count of tumor-associated giant macrophages by the total number of macrophages, this parameter allows for comparisons across patches and different patient cases. |
| 051edvdwzu6d | TM | AVG_GIANT_MAC_AREA_UM2 | AVG_GIANT_MAC_AREA_UM2 represents the average area of the identified giant macrophages, expressed in square micrometers. This parameter is computed by averaging the converted cell areas of giant macrophages and provides a normalized metric for cell size that can be compared across different patches and patients. |
| 051edvdwzu6d | TM | AVG_TUMOR_DISTANCE_UM | AVG_TUMOR_DISTANCE_UM indicates the average distance, in micrometers, from each giant macrophage to its nearest tumor cell. This parameter is determined by calculating the Euclidean distance for each giant macrophage to the closest tumor cell and normalizing the results through averaging, making it suitable for comparative analysis across different cases. |
| 0gxw2j7vlde8 | P | MEAN_ALIGNMENT_ANGLE | MEAN_ALIGNMENT_ANGLE measures the average degree of alignment between the orientation of individual plasma cells and the direction towards their nearest tumor cluster. The angle is computed using the principal axis of each plasma cell (derived from its shape) and the vector pointing from the cell to the tumor cluster centroid. This metric provides a normalized, comparable value across patches that reflects the typical spatial orientation of plasma cells in relation to tumor clusters. |
| 0gxw2j7vlde8 | P | STD_ALIGNMENT_ANGLE | STD_ALIGNMENT_ANGLE represents the standard deviation of the alignment angles in a given patch. It quantifies the variability in the directional alignment of plasma cells relative to their nearest tumor cluster, offering insights into the heterogeneity of cell orientation across different patches. This parameter is numeric and normalized, making it useful for comparative analysis. |
| 0gxw2j7vlde8 | P | MEAN_DISTANCE_TO_CLUSTER | MEAN_DISTANCE_TO_CLUSTER calculates the average physical distance (in micrometers) from plasma cells to the centroid of their nearest tumor cell cluster. This metric is derived by first determining the nearest tumor cluster for each plasma cell and then computing the average of these distances. It is normalized relative to each patch and aids in assessing the spatial proximity of plasma cells to tumor clusters. |
| 0gxw2j7vlde8 | P | FRACTION_ALIGNED_CELLS | FRACTION_ALIGNED_CELLS denotes the fraction of plasma cells in a patch that have an alignment angle (relative to the vector pointing to the nearest tumor cluster) less than or equal to 30 degrees. By providing a proportion, this parameter offers a normalized measure of how many plasma cells display a strong directional alignment with tumor clusters, which may indicate specific biological interactions in the tumor microenvironment. |
| 0hcs9p05obsh | TLPMNEF | TRANSITION_DENSITY | TRANSITION_DENSITY is a normalized parameter representing the ratio of tumor-to-stroma cell transitions to the total number of valid grid cell boundaries. This ratio reflects the density of transitions along the tumor edge, making it useful for comparing different tumor patches and cases. |
| 0hcs9p05obsh | TLPMNEF | TUMOR_EDGE_TRANSITION_INDEX | TUMOR_EDGE_TRANSITION_INDEX is a normalized metric where the transition density is scaled to a percentage (0-100). It quantitatively measures the abruptness of transitions from tumor to stromal cell types at the tumor boundary, with higher values indicating more pronounced transitions. This facilitates direct comparisons across different patient cases. |
| 0nafkdj0yzad | T | PLEOMORPHISM_SCORE | PLEOMORPHISM_SCORE is a composite metric that reflects the degree of variation in tumor cell nuclear attributes. It is calculated by determining the standard deviations of nuclear areas and nuclear circularity within each patch, normalizing them by dividing by their respective means, and then summing these normalized values. This process results in a score that can be compared across different patient cases and tumor regions. |
| 0nafkdj0yzad | T | STD_AREA_UM2 | STD_AREA_UM2 represents the variability in the areas of tumor cell nuclei within a patch. The nuclear areas, originally measured in pixels, are converted into square micrometers, and the standard deviation is computed from these values. This parameter provides insight into how diverse the sizes of the tumor cell nuclei are throughout a specific region. |
| 0nafkdj0yzad | T | STD_CIRCULARITY | STD_CIRCULARITY quantifies the variability in the shape of tumor cell nuclei across a patch. Nuclear circularity is derived from the area and perimeter of each nucleus using a standard formula. The standard deviation of these circularity values highlights the degree of variation in nuclear shape among tumor cells, allowing for comparability between patches. |
| 0nafkdj0yzad | T | MEAN_AREA_UM2 | MEAN_AREA_UM2 is the average nuclear area of tumor cells within a patch, expressed in square micrometers. By aggregating the converted area measurements of all tumor nuclei, it provides a central value that reflects the overall size of tumor cell nuclei in the analyzed region. |
| 0nafkdj0yzad | T | MEAN_CIRCULARITY | MEAN_CIRCULARITY is the average circularity of tumor cell nuclei within a patch. Circularity values are computed for each nucleus based on its area and perimeter, and the mean of these values serves as an indicator of the typical nuclear shape in the region. This measure is useful for comparing the overall nuclear morphology across different patches. |
| 0ol12y8sm2fg | T | NUCLEAR_IRREGULARITY_SCORE_MEAN | NUCLEAR_IRREGULARITY_SCORE_MEAN measures the average deviation of tumor cell nuclear shapes from a perfect circle within each patch. This value is calculated from the individual nuclear irregularity scores, with 0 indicating a perfect circular shape and higher values signifying increased irregularity. |
| 0ol12y8sm2fg | T | NUCLEAR_IRREGULARITY_SCORE_SD | NUCLEAR_IRREGULARITY_SCORE_SD represents the standard deviation of nuclear irregularity scores in a patch. It highlights the variability in nuclear shape irregularity among tumor cells within that region. |
| 0ol12y8sm2fg | T | NUCLEAR_IRREGULARITY_MAX | NUCLEAR_IRREGULARITY_MAX identifies the maximum nuclear irregularity score observed in a patch, indicating the tumor cell nucleus that deviates most from the ideal circular form. |
| 0ol12y8sm2fg | T | NUCLEAR_IRREGULARITY_MIN | NUCLEAR_IRREGULARITY_MIN captures the minimum nuclear irregularity score found in a patch, reflecting the tumor cell nucleus that is closest to the ideal circular shape. |
| 0ol12y8sm2fg | T | NUCLEAR_PERIMETER_ACTUAL_MEAN_UM | NUCLEAR_PERIMETER_ACTUAL_MEAN_UM is the mean of the actual measured perimeters of tumor cell nuclei in micrometers. It provides an assessment of the average size and contour of the nuclei as observed. |
| 0ol12y8sm2fg | T | NUCLEAR_PERIMETER_IDEAL_MEAN_UM | NUCLEAR_PERIMETER_IDEAL_MEAN_UM is the mean of the ideal perimeters computed for tumor cell nuclei, assuming a perfect circular shape with the same area. This parameter serves as a baseline to compare against the actual perimeter measurements. |
| 0ol12y8sm2fg | T | NUCLEAR_AREA_MEAN_UM2 | NUCLEAR_AREA_MEAN_UM2 calculates the average area of tumor cell nuclei in square micrometers within a patch, offering insight into the typical nuclear size across the tumor region. |
| 0xy34rnm3utq | EL | AVG_DISTANCE_UM | AVG_DISTANCE_UM represents the average Euclidean distance between all eosinophil-lymphocyte pairs within a patch, converted into micrometers. It is a normalized spatial metric calculated by averaging the pairwise distances from the centroids of eosinophils and lymphocytes, which allows for the comparison of spatial distributions across different patient cases. |
| 0xy34rnm3utq | EL | COLOCALIZATION_INDEX | COLOCALIZATION_INDEX is a normalized ratio that quantifies the degree of co-localization between eosinophils and lymphocytes. It is computed by dividing the number of cell pairs with a Euclidean distance less than or equal to a defined threshold by the total number of possible eosinophil-lymphocyte pairs. The resulting value, which ranges from 0 to 1, enables direct comparison of cell-cell interaction levels across different patches and patient cases. |
| 0zfplwl3g9vm | P | PLASMA_NUCLEAR_ECCENTRICITY | The PLASMA_NUCLEAR_ECCENTRICITY parameter measures the average eccentricity ratio of plasma cells in each patch. It quantifies how offset the nucleus is from the cell's geometric center relative to the maximum distance from the center to the cell boundary. Values range from 0 (no offset) to 1 (maximum offset), indicating the degree of nuclear displacement relative to the cellular shape. This normalized, numeric metric allows for consistent comparisons across different patient cases and tissue regions. |
| 14eonjjj3d4i | TF | MEAN_RING_COMPLETENESS | MEAN_RING_COMPLETENESS measures the average proportion, ranging from 0 to 1, of uniformly sampled points along the expected fibroblast ring boundary that are actually covered by fibroblasts. This metric quantifies how continuous the fibroblast ring is around tumor clusters in each patch, making it a normalized indicator suitable for comparing different patient cases. |
| 14eonjjj3d4i | TF | MEAN_RING_THICKNESS_MICRONS | MEAN_RING_THICKNESS_MICRONS represents the average thickness of the fibroblast ring around tumor clusters, expressed in microns. It is derived by calculating the average distance from fibroblast centroids to the tumor cluster’s convex hull, thereby providing a standardized measure of the ring's physical extent. |
| 14eonjjj3d4i | TF | MEAN_CORRALLING_FACTOR | MEAN_CORRALLING_FACTOR is a composite score obtained by multiplying the completeness metric by the average thickness of the fibroblast ring. This factor combines both the spatial continuity and the physical breadth of the fibroblast arrangement, offering a normalized index of the overall corralling efficiency. |
| 1u7mqo8v1sg3 | L | SEQUESTRATION_SCORE | SEQUESTRATION_SCORE is a composite metric that quantifies lymphocyte sequestration by integrating both the area of lymphocyte clusters and the number of lymphocytes within each cluster in the stromal region of a tumor patch. It is computed as the sum of the products of each cluster's area and its lymphocyte count, making it a normalized measure suitable for comparative analysis across uniformly sized patches. |
| 1u7mqo8v1sg3 | L | MEAN_POCKET_AREA_UM2 | MEAN_POCKET_AREA_UM2 represents the average area of the lymphocyte clusters (pockets) within the stromal compartment of a tumor patch. By calculating the mean pocket area, it provides a normalized metric that reflects the typical size of sequestration zones and is comparable across different patient cases due to the consistency of patch dimensions. |
| 1u7mqo8v1sg3 | L | MEAN_LYMPHS_PER_POCKET | MEAN_LYMPHS_PER_POCKET quantifies the average number of lymphocytes per identified lymphocyte cluster in the stromal region of a patch. As an average value derived over clusters, it normalizes the data and facilitates reliable comparisons between tumor patches from different patient cases. |
| 2me4gcbvcwz1 | F | TUNNELING_SCORE | TUNNELING_SCORE is a normalized metric that quantifies the potential invasion conduit in a tissue patch by measuring the maximum length of a continuous chain of elongated fibroblast cells connected to a tumor border, divided by the total number of fibroblasts in the patch and converted to micrometers. It allows comparison across different patient cases by accounting for variations in fibroblast abundance. |
| 2me4gcbvcwz1 | F | MAX_CHAIN_LENGTH | MAX_CHAIN_LENGTH represents the physical extent, in micrometers, of the longest connected chain of fibroblast cells that meets the defined proximity and orientation criteria and is connected to a tumor border. It is measured by summing Euclidean distances between consecutively aligned cell centroids along the principal direction of the chain, providing a spatial metric of fibroblast alignment related to potential tumor invasion. |
| 2sybw8e945mw | LM | IMMUNE_BOUNDARY_DENSITY | IMMUNE_BOUNDARY_DENSITY represents the normalized density of immune cells at the tumor-stroma boundary, calculated by counting the number of immune cells (both lymphocytes and macrophages) within a 50μm proximity to the tumor boundary and then normalizing this count by the patch area, resulting in a value expressed in cells per mm². This normalization allows for comparison across different patient cases and imaging patches. |
| 2sybw8e945mw | LM | LYMPHO_BOUNDARY_RATIO | LYMPHO_BOUNDARY_RATIO is a normalized parameter measuring the proportion of lymphocytes among the immune cells located at the tumor boundary. It reflects the relative contribution of lymphocytes compared to the total interface immune cell population, facilitating comparisons of the immune cell composition across different tumor regions. |
| 2sybw8e945mw | LM | MACRO_BOUNDARY_RATIO | MACRO_BOUNDARY_RATIO is a normalized parameter that quantifies the proportion of macrophages of the immune cells at the tumor-stroma boundary. By calculating the ratio of macrophages relative to the overall interface immune cell count, it provides insight into the characteristic involvement of macrophages in the tumor microenvironment, allowing for direct comparisons between different cases. |
| 35jjuutate91 | T | NUCLEAR_POLARITY_UNIFORMITY_SCORE | NUCLEAR_POLARITY_UNIFORMITY_SCORE measures the circular uniformity of tumor nuclear orientations by computing the mean resultant length (MRL). This numeric score, ranging from 0 to 1, reflects the consistency of nuclear alignment within a patch; values near 1 indicate highly aligned nuclei while lower values suggest a more random orientation. |
| 35jjuutate91 | T | MEAN_NUCLEAR_ORIENTATION | MEAN_NUCLEAR_ORIENTATION represents the average angle of nuclear orientation (in degrees, within the 0 to 180 range) in a given patch. This value is computed from the individual nuclear angles of tumor cells and provides a normalized measure for comparing the predominant nuclear alignment across different tumor regions. |
| 35jjuutate91 | T | STD_NUCLEAR_ORIENTATION | STD_NUCLEAR_ORIENTATION captures the standard deviation of the nuclear orientation angles (in degrees) among tumor cells in each patch. This parameter quantifies the variability or spread of nuclear alignment, offering insights into the heterogeneity of nuclear polarity within the tumor region. |
| 3eq1mpo2lp5t | F | FIBROBLAST_FRACTAL_DIM_MEAN | FIBROBLAST_FRACTAL_DIM_MEAN represents the average fractal dimension calculated from the boundaries of fibroblast cells in each tumor patch. This parameter quantifies the average morphological complexity of these cells, offering a normalized measure that allows for comparison across different patient cases. |
| 3eq1mpo2lp5t | F | FIBROBLAST_FRACTAL_DIM_STD | FIBROBLAST_FRACTAL_DIM_STD indicates the standard deviation of the fractal dimensions among the fibroblasts in a patch. It reflects the variability in cell boundary complexity, ensuring a robust, normalized statistical representation of fibroblast morphological heterogeneity. |
| 3eq1mpo2lp5t | F | FIBROBLAST_FRACTAL_DIM_MIN | FIBROBLAST_FRACTAL_DIM_MIN captures the minimum fractal dimension observed among fibroblast cells within a patch. This parameter highlights the least complex cell boundaries, serving as a normalized measure that helps identify the lower bound of fibroblast morphological complexity. |
| 3eq1mpo2lp5t | F | FIBROBLAST_FRACTAL_DIM_MAX | FIBROBLAST_FRACTAL_DIM_MAX denotes the maximum fractal dimension recorded for fibroblast cell boundaries in the patch. It identifies the most complex cellular structures, providing a normalized metric for assessing extreme variations in cell morphology. |
| 3x2v3ngl61c3 | T | MEAN_EOSIN_INTENSITY | MEAN_EOSIN_INTENSITY measures the average intensity of the red channel, representing eosinophilic staining in tumor cell cytoplasms within a patch. This metric is obtained by averaging the pixel intensities from the cytoplasm regions identified by cell masks, providing a normalized measure that can be compared across different patient cases. |
| 3x2v3ngl61c3 | T | STD_EOSIN_INTENSITY | STD_EOSIN_INTENSITY quantifies the variability of eosin intensity values among tumor cell cytoplasms in a patch by computing the standard deviation. This statistic reflects the heterogeneity of staining patterns and serves as a normalized parameter for comparing differences across patches and patient cases. |
| 3x2v3ngl61c3 | T | EOSIN_INTENSITY_SKEWNESS | EOSIN_INTENSITY_SKEWNESS assesses the asymmetry of the distribution of eosin staining intensities in tumor cell cytoplasms within a patch. It indicates whether the intensity values are symmetrically distributed or biased toward higher or lower values, making it a normalized measure useful for comparative analysis. |
| 4omp9vbqcdzp | L | LYMPHO_GRADIENT_INDEX | LYMPHO_GRADIENT_INDEX quantifies the steepness of the change in lymphocyte density from the tumor boundary into the surrounding stroma. It is determined by performing a linear regression on the lymphocyte densities measured in incremental distance bins from the tumor edge. The regression slope, after being scaled (multiplied by 1000), provides insight into how rapidly lymphocyte concentration decreases (or increases) with distance, making it comparable across different patient cases. |
| 4omp9vbqcdzp | L | MEAN_LYMPHO_DENSITY | MEAN_LYMPHO_DENSITY represents the average density of lymphocytes across all valid distance bins within a patch. This metric is calculated by averaging the lymphocyte densities (normalized by area, using lymphocytes per pixel²) and serves as a normalized measure of overall immune cell presence in the tissue region. |
| 4omp9vbqcdzp | L | MAX_LYMPHO_DENSITY | MAX_LYMPHO_DENSITY captures the maximum lymphocyte density observed in any valid distance bin. This parameter is derived by identifying the bin with the highest density, normalized by the corresponding annular area, and indicates the region with the most intense lymphocyte infiltration. |
| 4omp9vbqcdzp | L | MIN_LYMPHO_DENSITY | MIN_LYMPHO_DENSITY reflects the lowest lymphocyte density measured across the evaluated distance bins. By normalizing the count with the area of each bin (pixels²), this parameter highlights the region within the patch where lymphocyte infiltration is minimal. |
| 4omp9vbqcdzp | L | STD_LYMPHO_DENSITY | STD_LYMPHO_DENSITY measures the variability in lymphocyte densities across the valid distance bins. Calculated as the standard deviation of the densities (normalized per pixel²), it provides insight into the heterogeneity of lymphocyte distribution within the tissue patch. |
| 4qgb2yghxde5 | E | WTI | WTI: The Wavefront Trajectory Index is a spatial surrogate measure that represents the average distance in micrometers of eosinophils from the tumor boundary. It is computed by averaging the distances from each eosinophil's centroid to the computed convex hull of the tumor area, ensuring that it can be normalized and compared across different patient cases. |
| 4qgb2yghxde5 | E | MAX_DISTANCE | MAX_DISTANCE: This parameter measures the maximum distance, in micrometers, from any eosinophil to the tumor boundary within each analyzed patch. It captures the furthest extent of eosinophil infiltration, providing a normalized metric that facilitates comparisons between different tumor samples. |
| 546dlu4693m9 | F | BLEB_FREQUENCY | BLEB_FREQUENCY measures the average number of membrane protrusions (blebs) per fibroblast within each tissue patch. It is derived by dividing the total number of blebs detected by the total number of fibroblasts in the patch, resulting in a normalized value that allows comparison across different patient cases. |
| 546dlu4693m9 | F | MEAN_BLEBS_PER_CELL | MEAN_BLEBS_PER_CELL represents the arithmetic mean of bleb counts across individual fibroblasts in the patch. This parameter quantifies, on average, how many blebs each fibroblast exhibits, and is normalized to facilitate comparisons between different samples. |
| 546dlu4693m9 | F | STD_BLEBS_PER_CELL | STD_BLEBS_PER_CELL is the standard deviation of bleb counts among fibroblasts in a patch. This parameter indicates the variability in bleb formation across cells, providing insight into heterogeneity in cellular behavior, and is normalized for cross-case analysis. |
| 5hiiozwsvdli | T | MEAN_TEXTURE_DIVERGENCE | MEAN_TEXTURE_DIVERGENCE represents the average absolute difference between the chromatin granularity values measured in the core and periphery regions of tumor subclones within a patch. This parameter quantifies the overall level of divergence in nuclear texture, providing insights into intratumoral heterogeneity that may reflect differences in invasive potential. |
| 5hiiozwsvdli | T | MAX_TEXTURE_DIVERGENCE | MAX_TEXTURE_DIVERGENCE is defined as the highest absolute difference in chromatin granularity between the core and periphery across all valid subclones in a patch. This parameter highlights the most extreme case of nuclear texture divergence, which could be indicative of localized aggressive behavior in tumor regions. |
| 5hiiozwsvdli | T | MEDIAN_CORE_GRANULARITY | MEDIAN_CORE_GRANULARITY measures the central tendency of chromatin granularity values for the core regions of the subclones in a patch. This parameter provides a normalized value that facilitates comparisons between different patches by summarizing the typical nuclear texture characteristics of core cells. |
| 5hiiozwsvdli | T | MEDIAN_PERIPHERY_GRANULARITY | MEDIAN_PERIPHERY_GRANULARITY captures the central tendency of chromatin granularity measurements for the periphery regions of tumor subclones within a patch. It offers a normalized summary of the typical nuclear texture features in the periphery, aiding in the comparative analysis of tumor morphology. |
| 6f74bdc3jgw0 | P | FRACTAL_DIM_MEAN | FRACTAL_DIM_MEAN measures the average fractal dimension of plasma cell nuclear boundaries within a patch. This parameter quantifies the complexity of nuclear boundaries using a regression analysis based on the box-counting method, yielding a normalized, dimensionless value suitable for comparisons across different patient cases. |
| 6f74bdc3jgw0 | P | FRACTAL_DIM_MEDIAN | FRACTAL_DIM_MEDIAN represents the median value of the fractal dimensions calculated from plasma cell nuclear boundaries in a patch. This statistic offers a robust central measure that minimizes the impact of outliers, reflecting the typical complexity level of cell nuclei. |
| 6f74bdc3jgw0 | P | FRACTAL_DIM_STD | FRACTAL_DIM_STD quantifies the standard deviation of fractal dimensions among plasma cell nuclei within a patch. It describes the variability in nuclear boundary complexity, thereby providing insight into the heterogeneity of the cellular features in a normalized manner. |
| 6f74bdc3jgw0 | P | FRACTAL_DIM_MIN | FRACTAL_DIM_MIN indicates the minimum fractal dimension observed among the plasma cell nuclei within a patch. This parameter reflects the lower bound of nuclear boundary complexity in the analyzed area, supporting comparative analysis between different cases. |
| 6f74bdc3jgw0 | P | FRACTAL_DIM_MAX | FRACTAL_DIM_MAX indicates the maximum fractal dimension calculated from plasma cell nuclear boundaries in a patch. It reflects the highest complexity observed in nuclear morphology, facilitating effective normalization and comparison across patient samples. |
| 7kok4ofq4bbd | T | CLUSTER1_MEAN_INTENSITY | CLUSTER1_MEAN_INTENSITY captures the average nuclear staining intensity of the tumor cell subclone with lower intensity values within a patch. This measurement reflects one of the subpopulations identified by clustering intensity metrics on tumor cells. |
| 7kok4ofq4bbd | T | CLUSTER2_MEAN_INTENSITY | CLUSTER2_MEAN_INTENSITY captures the average nuclear staining intensity of the tumor cell subclone with higher intensity values within the patch. It provides a contrast to the lower intensity cluster and helps assess variations in nuclear staining. |
| 7kok4ofq4bbd | T | CLUSTER_INTENSITY_STD | CLUSTER_INTENSITY_STD is the standard deviation between the two cluster mean intensities. It quantifies the variability in nuclear staining between the two delineated tumor subclones, offering insight into the heterogeneity of the tumor's nuclear features. |
| 7wi2iaqu07di | E | EOSINOPHIL_BILOBATION_DEVIATION_MEAN | EOSINOPHIL_BILOBATION_DEVIATION_MEAN: This parameter measures the average deviation index across all eosinophils within a tissue patch. It quantifies the degree to which the nuclear shapes of eosinophils diverge from the typical bilobed configuration, using normalized metrics that compare lobe area differences and center separations relative to nuclear size. |
| 7wi2iaqu07di | E | EOSINOPHIL_BILOBATION_DEVIATION_STD | EOSINOPHIL_BILOBATION_DEVIATION_STD: This parameter captures the standard deviation of the deviation indices in a patch, indicating the variability in the morphological deviations of eosinophil nuclei. A higher value reflects greater heterogeneity in nuclear shape deviations among cells. |
| 7wi2iaqu07di | E | EOSINOPHIL_BILOBATION_DEVIATION_MIN | EOSINOPHIL_BILOBATION_DEVIATION_MIN: This parameter represents the minimum deviation index observed in the patch. It identifies the cell with the most typical or closest-to-normal bilobed nuclear configuration, serving as a baseline for comparison. |
| 7wi2iaqu07di | E | EOSINOPHIL_BILOBATION_DEVIATION_MAX | EOSINOPHIL_BILOBATION_DEVIATION_MAX: This parameter reflects the maximum deviation index found in the patch, highlighting the cell with the most atypical nuclear shape. It signals the extent of abnormal morphology relative to the normative bilobed pattern. |
| 7zhu7bue40a7 | F | LATTICE_FORMATION_INDEX | LATTICE_FORMATION_INDEX quantifies the extent to which fibroblast cells form a highly regular and interconnected lattice structure in a tumor patch. It is derived from combining measures of spatial regularity (calculated as the ratio of the standard deviation to the mean of the edge lengths from the Delaunay triangulation) and connectivity deviation (the absolute difference between the observed average cell connectivity and an ideal hexagonal arrangement). A higher value indicates a more organized and potentially biologically significant lattice formation within the tumor environment. |
| 7zhu7bue40a7 | F | REGULARITY_SCORE | REGULARITY_SCORE measures the consistency of the edge lengths between fibroblast cells derived from spatial triangulation. It is calculated as the ratio of the standard deviation of these edge lengths to their mean, thereby providing a normalized metric for assessing how uniformly the fibroblasts are arranged. Lower values indicate a more consistent, lattice-like structure. |
| 7zhu7bue40a7 | F | CONNECTIVITY_SCORE | CONNECTIVITY_SCORE assesses the deviation of the actual average number of connections per fibroblast cell from an ideal target value (typically six for a hexagonal pattern). This parameter reflects how closely the connectivity among the fibroblasts approximates an idealized, organized structure; lower values represent closer adherence to the ideal connectivity. |
| 7zhu7bue40a7 | F | MEAN_EDGE_LENGTH_UM | MEAN_EDGE_LENGTH_UM represents the average length of the edges connecting fibroblast cell centroids, measured in micrometers. This metric reflects the typical spatial distance between neighbouring cells within a patch and is standardized across cases due to the consistent analysis of similarly sized patches. |
| 7zhu7bue40a7 | F | STD_EDGE_LENGTH_UM | STD_EDGE_LENGTH_UM computes the standard deviation of the edge lengths in micrometers, providing a measure of variability in the distances between neighbouring fibroblast cells. As a normalized metric, it helps in understanding the degree of uniformity or dispersion in the spatial arrangement of the cells. |
| 7zhu7bue40a7 | F | AVG_NODE_DEGREE | AVG_NODE_DEGREE is the average number of connections per fibroblast cell within the spatial network constructed from their centroids. This metric reflects the overall connectivity pattern of the cells, with a value closer to the ideal degree (six, indicative of a hexagonal organization) implying a more regular lattice formation. |
| 7zl55ylus53w | TF | TUMOR_FIBRO_CONTACT_PROP | This parameter quantifies the proportion of tumor cells within a given patch that are in direct contact with at least one fibroblast, based on a distance threshold corresponding to 10μm. The value is normalized between 0 and 1, allowing for comparative analysis across different patient cases and patches. |
| 8g7t4mgy4su3 | TLM | TUMOR_DENSITY | TUMOR_DENSITY represents the number of tumor cells per mm² in the patch. It is calculated by counting epithelial cells within the patch and normalizing by the patch area, yielding a measure that permits comparisons across different patient cases. |
| 8g7t4mgy4su3 | TLM | LYMPHOCYTE_DENSITY | LYMPHOCYTE_DENSITY denotes the number of lymphocyte cells per mm² in the patch. This metric is obtained by counting the lymphocytes and dividing by the patch area, which standardizes the measure for analyses among various whole-slide images. |
| 8g7t4mgy4su3 | TLM | MACROPHAGE_DENSITY | MACROPHAGE_DENSITY provides the number of macrophages per mm² in the patch. The value is computed by normalizing the count of macrophage cells by the patch area, making it suitable for cross-patient comparisons. |
| 8g7t4mgy4su3 | TLM | AVG_TUMOR_LYMPHO_DIST | AVG_TUMOR_LYMPHO_DIST measures the average distance, in micrometers, from tumor cells to their nearest lymphocyte. By computing the minimal distances from each tumor cell to a lymphocyte and averaging them, it serves as a robust metric to describe spatial proximity in different tumor regions. |
| 8g7t4mgy4su3 | TLM | AVG_TUMOR_MACRO_DIST | AVG_TUMOR_MACRO_DIST reflects the average distance, in micrometers, from tumor cells to their nearest macrophage. This parameter is determined by finding the closest macrophage to each tumor cell and averaging these distances, providing insights into cell spatial interactions. |
| 8g7t4mgy4su3 | TLM | IMMUNE_EVASION_SCORE | IMMUNE_EVASION_SCORE is a composite metric integrating both cell density and spatial proximity measures. It combines the tumor density with the ratios of lymphocyte and macrophage densities to their corresponding average distances, resulting in a normalized score that quantifies the coordinated immune evasion capacity across different patches. |
| 8t7xnfi97rw0 | LM | CONVERGENCE_SCORE | CONVERGENCE_SCORE quantifies the directional alignment of immune cells (lymphocytes and macrophages) toward the nearest tumor cell cluster. It is calculated as the magnitude of the average of normalized vectors from each immune cell to the centroid of its nearest tumor cell cluster. A score close to 1 indicates strong, coherent alignment, while a score near 0 suggests random orientation, making it a normalized metric suitable for comparing different patient cases. |
| 8t7xnfi97rw0 | LM | MEAN_VECTOR_X | MEAN_VECTOR_X represents the x-component of the average normalized direction vector of immune cells in a patch. It is computed by averaging the x-components of the unit vectors (representing the direction from each immune cell to its nearest tumor cell cluster) across all immune cells in the patch. This numeric parameter helps illustrate the net horizontal directional tendency and is normalized based on the unit vector computations. |
| 8t7xnfi97rw0 | LM | MEAN_VECTOR_Y | MEAN_VECTOR_Y represents the y-component of the average normalized direction vector of immune cells in a patch. Similar to MEAN_VECTOR_X, it is derived by averaging the y-components of the normalized direction vectors from immune cells to the nearest tumor cluster centroids. This parameter captures the vertical component of the net directional trend and is numeric and comparable across different cases. |
| 9dbi53ksda9q | F | FIBROBLAST_STROMAL_COMPACTION_INDEX | FIBROBLAST_STROMAL_COMPACTION_INDEX quantifies the degree of fibroblast clustering within the stroma by calculating the ratio of the total area covered by fibroblast clusters (determined via convex hulls over spatially clustered fibroblast centroids) to the overall stromal area in a given patch. This ratio is normalized, ensuring comparability across different patient cases. |
| 9dbi53ksda9q | F | FIBROBLAST_DENSITY_PER_UM2 | FIBROBLAST_DENSITY_PER_UM2 measures the number of fibroblast cells per square micrometer of stromal area. It is calculated by counting the fibroblast cells identified within the stroma and dividing by the stromal area, making it a normalized parameter for comparing cellular density across different patches. |
| 9dbi53ksda9q | F | MEAN_CLUSTER_AREA_UM2 | MEAN_CLUSTER_AREA_UM2 represents the average area of fibroblast clusters within a patch. The areas are derived from the convex hulls of clusters formed by closely spaced fibroblast centroids, and the average is computed in standardized square micrometers, thereby ensuring the metric is normalized for cross-case comparisons. |
| 9jvro5r6lyat | T | TUMOR_AREA_UM2 | TUMOR_AREA_UM2 quantifies the total tumor area within a patch by assessing the tissue segmentation mask. Specifically, it counts the number of pixels classified as tumor and converts this count into square micrometers using a fixed scaling factor. This measurement, expressed in standardized physical units, allows for direct comparisons across different patient cases and patches. |
| 9vnmzkmoss74 | M | ENGULFMENT_INDEX | ENGULFMENT_INDEX is a normalized numeric metric that quantifies the phagocytic activity of macrophages within tumor regions. It is calculated as the ratio of macrophages exhibiting intracytoplasmic debris — indicative of active engulfment of cellular fragments — to the total number of macrophages present in the patch. The parameter is derived by analyzing segmented patches of tumor tissue, applying thresholding to detect debris within the cell mask, and excluding the main nuclear mass by removing the largest connected component when it comprises more than 50% of the cell area. This resulting ratio, which ranges from 0 to 1, allows for direct comparison across different patient cases by standardizing the measurement. |
| a5p69n9ja6ap | M | MACROPHAGE_ELONGATION_MEAN | MACROPHAGE_ELONGATION_MEAN represents the average ratio of the major axis to the minor axis of macrophage cell polygons within a patch. This measure normalizes cell shape elongation in each patch, allowing for comparison across different patient cases. |
| a5p69n9ja6ap | M | MACROPHAGE_ELONGATION_SD | MACROPHAGE_ELONGATION_SD quantifies the variability in the elongation indices of macrophages within a patch. It is a numeric representation that describes the spread around the mean elongation, enabling standardized assessments across samples. |
| a5p69n9ja6ap | M | MACROPHAGE_ELONGATION_MIN | MACROPHAGE_ELONGATION_MIN indicates the smallest elongation index observed in a given patch. This numeric parameter provides a baseline measure of the least elongated macrophages, ensuring normalization across patient cases. |
| a5p69n9ja6ap | M | MACROPHAGE_ELONGATION_MAX | MACROPHAGE_ELONGATION_MAX displays the highest elongation index of macrophages in a patch. It serves as a normalized numerical value reflecting the extreme elongation potential across different patches. |
| aj3ptmksxe7y | T | DISCONTINUITY_INDEX | DISCONTINUITY_INDEX quantifies tumor fragmentation by computing the ratio of the number of distinct tumor cell clusters in a patch to the total tumor area in that patch, with the tumor area measured in square micrometers. By normalizing the raw cluster count by the tumor area, this parameter provides a dimensionless metric that allows comparison across different patient cases. |
| anh6n7rzeolv | P | GOLGI_PROMINENCE_SCORE | GOLGI_PROMINENCE_SCORE measures the relative prominence of the Golgi zone within plasma cells in each tumor patch. The score is derived by extracting a standardized subregion adjacent to the nucleus from each plasma cell image, applying intensity thresholding to identify areas of Golgi clearing, and calculating the ratio of cleared pixels to the total pixels in that region. These individual scores are then averaged across all plasma cells in the patch, yielding a normalized score on a scale from 0 to 1 that allows for direct comparison across different patient cases. |
| asoy6n1j7u4c | M | ENTANGLEMENT_INDEX | ENTANGLEMENT_INDEX is a composite metric that quantifies the complexity of macrophage infiltration within tumor stroma. It is computed per patch by combining the average tortuosity of the infiltration paths (a normalized ratio of the actual path lengths to the direct Euclidean distance between endpoints) with the number of branch points in the constructed minimum spanning tree, scaled by a conversion factor to produce a value in micrometers. This metric, derived from spatial metrics within uniformly sized patches, allows for direct comparisons across different patient cases. |
| asoy6n1j7u4c | M | AVG_TORTUOSITY | AVG_TORTUOSITY is a normalized parameter reflecting the mean ratio between the actual pathway lengths of macrophage branches (traced through the minimum spanning tree) and the direct Euclidean distances between their endpoints. This ratio effectively captures the degree of deviation from a straight path, making it a dimensionless indicator that can be reliably compared across patches from different patient samples. |
| b16szz5zpcvb | T | SHANNON_ENTROPY | SHANNON_ENTROPY: This parameter quantifies the heterogeneity of tumor subclones within a patch by calculating the Shannon diversity index. It uses the proportion of cells in each morphological cluster to capture the diversity, with higher values reflecting greater subclonal variation. |
| b16szz5zpcvb | T | MAX_SUBCLONE_PROPORTION | MAX_SUBCLONE_PROPORTION: This parameter represents the highest proportion of tumor cells within any identified subclone cluster in the patch. It indicates the dominance of a particular morphological group among the tumor cells, allowing for cross-patient comparisons due to its normalized nature. |
| b16szz5zpcvb | T | MIN_SUBCLONE_PROPORTION | MIN_SUBCLONE_PROPORTION: This parameter denotes the smallest proportion of tumor cells in any of the morphological subclone clusters. It highlights the presence of rare morphological variants within the tumor and is normalized for consistent comparison across different cases. |
| b16szz5zpcvb | T | MEAN_CELL_COMPACTNESS | MEAN_CELL_COMPACTNESS: This parameter measures the average compactness of tumor cells in the patch by evaluating the cell shape. The compactness metric is computed as a ratio that compares the cell's perimeter squared to its area, normalized against an ideal circular shape, thereby providing insight into morphological irregularities. |
| b1apmb2q2o2v | TL | BRIDGING_INDEX | BRIDGING_INDEX is a normalized ratio that measures the fraction of lymphocytes bridging between two distinct tumor subclones relative to the total number of lymphocytes within a patch. It provides an assessment of potential immune selection pressure and tumor heterogeneity by comparing bridging events across different patient cases. |
| b1apmb2q2o2v | TL | MEAN_BRIDGE_DISTANCE_UM | MEAN_BRIDGE_DISTANCE_UM represents the average maximum distance, in micrometers, from bridging lymphocytes to their nearest tumor cells. This parameter captures the spatial scale of the bridging event by averaging the distances where lymphocytes interact with tumor subclones, enabling meaningful comparisons across different patches and patients. |
| bgew6kslidda | L | FRAGMENTATION_INDEX | The FRAGMENTATION_INDEX parameter is a normalized measure representing the ratio of the number of distinct lymphocyte clusters (formed by spatially adjacent lymphocytes in the stromal region of lung cancer tissue) to the total number of lymphocytes in that region. This ratio circumvents the limitations of raw counts by providing a comparable metric across different patches and patient cases, thereby offering insight into the degree of lymphocyte dispersal and fragmentation within the tumor stroma. |
| buly5aezuq1y | T | TUMOR_CELL_IRREGULARITY_SCORE | TUMOR_CELL_IRREGULARITY_SCORE measures the average irregularity of tumor cell borders per patch by computing the ratio of the cell perimeter to the square root of its area. This normalization ensures that the measurement is independent of cell size, making it comparable across different patient cases. Higher values indicate more irregular and potentially more invasive tumor cell shapes. |
| buly5aezuq1y | T | TUMOR_CELL_IRREGULARITY_SD | TUMOR_CELL_IRREGULARITY_SD represents the standard deviation of the irregularity indices calculated for the tumor cells within a patch. This parameter quantifies the variability in cell border irregularity among the cells analyzed. A higher standard deviation suggests greater heterogeneity in cell shapes, while a lower value indicates more consistent morphology among tumor cells. |
| cfdv0gog3b7l | T | NUCLEAR_INTENSITY_GRADIENT | NUCLEAR_INTENSITY_GRADIENT measures the slope of the linear regression between the distance from the tumor boundary and the nuclear intensity of tumor cells. This parameter indicates the gradient of nuclear basophilia, where a negative slope suggests higher intensity at the tumor edge compared to the core, reflecting a potential aggressive phenotype in the invasive front. |
| cfdv0gog3b7l | T | GRADIENT_R2 | GRADIENT_R2 represents the R-squared value of the linear regression applied between the distance from the tumor edge and nuclear intensity. It quantifies the goodness-of-fit of the linear model, indicating how well the model explains the variability in nuclear intensity with respect to the distance, and is a normalized metric applicable across different cases. |
| cfdv0gog3b7l | T | MEAN_EDGE_INTENSITY | MEAN_EDGE_INTENSITY calculates the average nuclear intensity of tumor cells located in the outer 25% region of the tumor patch, which is defined as the area closest to the invasion front. This parameter provides insight into the staining characteristics and potential biological behavior at the tumor periphery. |
| cfdv0gog3b7l | T | MEAN_CORE_INTENSITY | MEAN_CORE_INTENSITY computes the average nuclear intensity of tumor cells found in the inner 75% region of the tumor patch. It serves as a reference measurement for the nuclear intensity in the central, less invasive part of the tumor, thereby aiding in comparative analyses with the edge region. |
| ci0k4ftnhnc1 | TLF | TRIAD_CONTACT_SCORE | TRIAD_CONTACT_SCORE represents the normalized ratio between the number of identified tumor-lymphocyte-fibroblast triads and the total number of relevant cells within a patch. It quantifies the relative frequency of close spatial interactions in the tumor microenvironment, allowing for meaningful comparisons across different patient cases. |
| cm62oy8x05s2 | TE | MEAN_EOS_TUMOR_DIST | MEAN_EOS_TUMOR_DIST measures the average of the minimum distances, in micrometers, from each eosinophil to its nearest tumor cell within a patch. This parameter reflects the overall spatial proximity between eosinophils and tumor cells, allowing for comparative analysis across different patient cases. |
| cm62oy8x05s2 | TE | MEDIAN_EOS_TUMOR_DIST | MEDIAN_EOS_TUMOR_DIST represents the median value of the minimum distances, in micrometers, from eosinophils to their closest tumor cell in a patch. It provides a robust measure of the central tendency of the spatial relationship, minimizing the effect of outliers. |
| cm62oy8x05s2 | TE | MIN_EOS_TUMOR_DIST | MIN_EOS_TUMOR_DIST captures the smallest recorded minimum distance, in micrometers, between any eosinophil and the nearest tumor cell in a patch. This metric highlights the closest point of interaction and can indicate localized areas of intense proximity. |
| cm62oy8x05s2 | TE | MAX_EOS_TUMOR_DIST | MAX_EOS_TUMOR_DIST identifies the largest observed minimum distance, in micrometers, among all eosinophils to their nearest tumor cell within a patch. It offers insights into the spatial range and variability of cell proximities within the sample. |
| cm62oy8x05s2 | TE | STD_EOS_TUMOR_DIST | STD_EOS_TUMOR_DIST quantifies the standard deviation of the minimum distances from eosinophils to the nearest tumor cell in a patch, measured in micrometers. This parameter indicates the variability and dispersion in the distances, supporting comparisons across different patient samples. |
| d0p2056dg4vl | L | MITOTIC_FREQUENCY | MITOTIC_FREQUENCY is a normalized parameter representing the ratio of mitotically active lymphocytes to the total number of intratumoral lymphocytes within a given tissue patch. It is calculated by identifying lymphocytes within tumor and stroma regions of the patch, processing their high-resolution grayscale nucleus images to detect potential mitotic figures via image smoothing and edge enhancement techniques, and then computing the ratio of those lymphocytes that show mitotic activity to the total lymphocyte count. This normalized ratio facilitates meaningful comparisons across different patient cases. |
| d5or6c6xclzq | LPMNE | OVERALL_POLARITY_SCORE | OVERALL_POLARITY_SCORE measures the overall spatial polarity of all immune cells combined in a tumor patch. It is normalized by comparing the maximum and minimum quadrant counts to produce a value between 0 and 1, where 0 indicates a uniform distribution and 1 represents maximum polarization. This normalized metric enables comparison between different patient cases. |
| d5or6c6xclzq | LPMNE | LYMPHOCYTE_POLARITY_SCORE | LYMPHOCYTE_POLARITY_SCORE quantifies the asymmetry in the distribution of lymphocytes around the tumor cell cluster. It is derived by comparing the highest and lowest counts of lymphocytes among the four quadrants, yielding a normalized score between 0 and 1 that reflects how unevenly lymphocytes are distributed. |
| d5or6c6xclzq | LPMNE | PLASMA_POLARITY_SCORE | PLASMA_POLARITY_SCORE represents the degree of spatial polarization in the distribution of plasma cells relative to the tumor center. The score, ranging from 0 to 1, is calculated by evaluating the imbalance between quadrants with the highest and lowest plasma cell counts, providing a normalized metric for assessing distribution bias. |
| d5or6c6xclzq | LPMNE | MACROPHAGE_POLARITY_SCORE | MACROPHAGE_POLARITY_SCORE captures the polarization of macrophages in the tumor patch by analyzing their counts across the four quadrants. A score between 0 (uniform distribution) and 1 (maximally polarized) is computed from the relative differences, making it a normalized, numeric indicator that is comparable across different cases. |
| d5or6c6xclzq | LPMNE | NEUTROPHIL_POLARITY_SCORE | NEUTROPHIL_POLARITY_SCORE assesses the spatial distribution of neutrophils around the tumor center. By comparing the quadrant with the highest count to the one with the lowest count, the score is normalized between 0 and 1, allowing evaluation of the degree of polarization in neutrophil distribution among different patches. |
| d5or6c6xclzq | LPMNE | EOSINOPHIL_POLARITY_SCORE | EOSINOPHIL_POLARITY_SCORE measures the distribution imbalance of eosinophils in a tumor patch. It is computed similarly by comparing counts in the most and least populated quadrants, resulting in a normalized score between 0 and 1 that reflects the extent of spatial polarization. |
| e6irk8erlvyj | L | TUMOR_CLUSTER_COUNT | TUMOR_CLUSTER_COUNT: This parameter represents the number of distinct lymphocyte clusters identified in the tumor compartment within a fixed-size patch. The count is obtained by applying a clustering algorithm on the spatial coordinates of lymphocytes, excluding noise points. Since the patches are of consistent size, this count can be reliably compared across different patient cases. |
| e6irk8erlvyj | L | STROMA_CLUSTER_COUNT | STROMA_CLUSTER_COUNT: This parameter denotes the number of lymphocyte clusters detected in the stroma compartment in each patch. Like its tumor counterpart, it uses a clustering approach to group spatially proximate lymphocytes and omits noise, making it a standardized metric for comparison across patient samples. |
| e6irk8erlvyj | L | TUMOR_MEAN_CLUSTER_SIZE | TUMOR_MEAN_CLUSTER_SIZE: This parameter measures the average number of lymphocytes per cluster in the tumor compartment. It is computed by averaging the sizes of all clusters found in a patch. Because the calculation is performed on patches of constant area, the result is normalized and comparable across different patient cases. |
| e6irk8erlvyj | L | STROMA_MEAN_CLUSTER_SIZE | STROMA_MEAN_CLUSTER_SIZE: This parameter reflects the average number of lymphocytes per cluster in the stroma compartment. It is derived by calculating the mean cluster size from the detected clusters in each patch, ensuring that it serves as a normalized metric that can be used to compare different patients. |
| e6irk8erlvyj | L | TUMOR_MEAN_CLUSTER_AREA | TUMOR_MEAN_CLUSTER_AREA: This parameter represents the average spatial area of lymphocyte clusters in the tumor compartment, expressed in square microns. It is determined by computing the convex hull of each cluster’s centroids and then averaging these areas. The consistent patch size ensures that this metric is normalized for inter-patient comparisons. |
| e6irk8erlvyj | L | STROMA_MEAN_CLUSTER_AREA | STROMA_MEAN_CLUSTER_AREA: This parameter measures the average area of clusters within the stroma compartment, calculated in a similar manner by determining the convex hull area of cluster centroids and converting it to square microns. The normalization achieved by using fixed-size patches allows for valid comparisons across patient cases. |
| ekzviqtcboel | P | PERITUMORAL_PLASMA_DENSITY | PERITUMORAL_PLASMA_DENSITY quantifies the density of plasma cells within a specifically defined peritumoral ring surrounding the tumor. This parameter is calculated by taking the number of plasma cells found in the peritumoral ring, multiplying by 1000, and then dividing by the area of the ring expressed in square micrometers. The normalization to 1000 μm² allows for direct comparison between different patient cases and patches. Thus, it provides a standardized measure that reflects local immune response in the tumor microenvironment. |
| enbnaeoprjmf | T | MEAN_POLARITY_DISTORTION | MEAN_POLARITY_DISTORTION represents the average normalized apical-basal polarity distortion index calculated across tumor cells within a patch. This index is derived from quantifying the deviation in the angular difference between vectors computed from cell centroid to half-cell centroids, normalized by the ideal 180° separation. A value of 0 indicates perfect polarity alignment, while values closer to 1 indicate higher distortion. |
| enbnaeoprjmf | T | MEDIAN_POLARITY_DISTORTION | MEDIAN_POLARITY_DISTORTION captures the median value of the normalized polarity distortion indices for tumor cells in a given patch. This metric provides a robust measure of central tendency that reduces the influence of outlier distortion values, effectively representing the typical degree of polarity disruption in the tumor area. |
| enbnaeoprjmf | T | STD_POLARITY_DISTORTION | STD_POLARITY_DISTORTION computes the standard deviation of the normalized polarity distortion indices across the tumor cells within a patch. It reflects the variability or heterogeneity in polarity distortion among the cells, indicating whether the cell population exhibits consistent polarity changes or a mix of normal and severely distorted patterns. |
| enbnaeoprjmf | T | MAX_POLARITY_DISTORTION | MAX_POLARITY_DISTORTION denotes the maximum normalized polarity distortion index observed among the tumor cells in the patch. This parameter highlights the cell with the most severe polarity distortion, serving as an indicator of the extreme disruption in cellular organization within the tumor region. |
| fgec4i1ljwnr | TL | ADHESION_PROBABILITY | ADHESION_PROBABILITY measures the fraction of possible tumor-lymphocyte pairs that exhibit close adhesion events. This metric is calculated as the number of adhesion events (where the distance between tumor and lymphocyte is less than or equal to a specified threshold) divided by the total number of all potential tumor-lymphocyte pairs. Being a normalized ratio, it allows for meaningful comparisons across different patient cases and tissue patches. |
| fgec4i1ljwnr | TL | MEAN_ADHESION_DISTANCE_UM | MEAN_ADHESION_DISTANCE_UM calculates the average distance, in micrometers, between tumor cells and lymphocytes that are sufficiently close to be considered in interaction. This parameter reflects the typical physical separation between tumor and immune cells within tissue patches and, by using a mean value, provides a normalized metric that enables cross-case comparisons. |
| hbx0izkb5u70 | M | RUFFLING_INDEX_MEAN | RUFFLING_INDEX_MEAN is the average ruffling index of macrophages within a patch. It integrates measurements of cell boundary irregularity by combining fractal dimension and deviation from ideal circularity. This normalized metric enables comparisons across different patient cases and patches. |
| hbx0izkb5u70 | M | RUFFLING_INDEX_STD | RUFFLING_INDEX_STD represents the standard deviation of the ruffling index values among macrophages in the patch. It quantifies the heterogeneity of membrane ruffling, offering insight into the variability of macrophage activation across tissue regions. This parameter is normalized and useful for cross-sample comparisons. |
| hbx0izkb5u70 | M | RUFFLING_INDEX_MIN | RUFFLING_INDEX_MIN is the minimum ruffling index recorded in the patch, capturing the lowest level of membrane ruffling. This value provides a reference for the least activated macrophage phenotype and is normalized to allow comparison across different studies. |
| hbx0izkb5u70 | M | RUFFLING_INDEX_MAX | RUFFLING_INDEX_MAX is the maximum ruffling index in the patch, reflecting the highest degree of membrane ruffling present among the macrophages. It captures the most activated state and, being normalized, facilitates comparisons across different patient cases. |
| huamv7hzyd8u | L | LYMPHOCYTE_SPATIAL_UNIFORMITY_SCORE | LYMPHOCYTE_SPATIAL_UNIFORMITY_SCORE quantifies the deviation of the spatial distribution of intratumoral lymphocytes from what would be expected under complete spatial randomness. The parameter is calculated by first extracting the coordinates of lymphocytes located in tumor regions of each image patch, then computing Ripley’s K function over a range of radii. For each radius, the procedure calculates the squared difference between the observed clustering and the expected value (based on a uniform random distribution), and these deviations are averaged to produce a final normalized score. This normalization ensures that the metric is suitable for comparing different patient cases, and as a numeric value, it allows an assessment of whether the intra-tumoral lymphocyte spread tends more towards a uniform dispersion or a clustered pattern. |
| i9jt1x7fp6ni | LPMNE | LYMPHO_DEPL_RATIO | LYMPHO_DEPL_RATIO is a normalized and numeric parameter that measures the ratio of lymphocytes located in the tumor core to the total number of lymphocytes in the entire tumor area. The tumor core is defined by eroding the convex hull of tumor cell centroids, which distinguishes the inner region from the periphery, allowing for the assessment of local immune cell depletion. |
| i9jt1x7fp6ni | LPMNE | PLASMA_DEPL_RATIO | PLASMA_DEPL_RATIO is a normalized and numeric parameter that quantifies the ratio of plasma cells located in the eroded tumor core compared to the overall plasma cell count in the tumor area. This metric allows for consistent comparisons across different patient samples. |
| i9jt1x7fp6ni | LPMNE | MACRO_DEPL_RATIO | MACRO_DEPL_RATIO is a normalized and numeric parameter that represents the ratio of macrophages present within the tumor core, identified by an eroded convex hull method, to the total macrophage count across the full tumor region. It is used to assess immune cell depletion in a standardized manner. |
| i9jt1x7fp6ni | LPMNE | NEUTRO_DEPL_RATIO | NEUTRO_DEPL_RATIO is a normalized and numeric parameter that indicates the ratio of neutrophils found in the tumor core relative to the total number of neutrophils in the entire tumor area. This parameter is calculated by dividing the count within the core by the overall count, ensuring comparability between patient cases. |
| i9jt1x7fp6ni | LPMNE | EOSINO_DEPL_RATIO | EOSINO_DEPL_RATIO is a normalized and numeric parameter that calculates the ratio of eosinophils in the tumor core to the total number of eosinophils across the whole tumor area. Like the other ratios, it is designed to facilitate comparisons by standardizing the count as a value between 0 and 1. |
| ipnoosl8ivky | EF | HELICAL_CONVERGENCE_FACTOR | HELICAL_CONVERGENCE_FACTOR quantifies the degree of spiral-like clustering between eosinophils and fibroblasts within localized tumor patches. It is derived by computing the angle between each eosinophil and its nearby fibroblast partners, calculating the circular variance of these angles, and then normalizing this value to lie between 0 and 1. A higher value indicates strong directional clustering, making this metric comparable across different patient cases. |
| ipnoosl8ivky | EF | MEAN_INTERACTION_DISTANCE_UM | MEAN_INTERACTION_DISTANCE_UM represents the average physical distance, in micrometers, between interacting eosinophil-fibroblast pairs within each patch. This distance is calculated from the positional differences between cell centroids and converted from pixels to micrometers using a fixed conversion factor, ensuring it is normalized and comparable across different patient cases. |
| j7jpij09oitl | LF | LYMPH_FIB_RATIO | LYMPH_FIB_RATIO calculates the ratio of lymphocytes that have fibroblast as their nearest neighbor compared to those with another lymphocyte as their nearest neighbor in a given patch of the tumor stroma. This metric is normalized, allowing for reliable comparisons between different patient cases by standardizing the measure; it is derived from computing Euclidean distances in local cell neighborhoods, classifying the nearest neighbor cell type, and then forming the ratio. It reflects potential shifts in cell-to-cell interactions that might signal alterations in the tumor microenvironment impacting metastasis. |
| kdnu6a8nnoh4 | ME | ME_COOP_RATIO | ME_COOP_RATIO represents the ratio of cooperative clusters to the total number of stromal immune cell clusters within a tumor patch. A cooperative cluster is defined as a group of immune cells identified by a spatial clustering algorithm that contains both macrophages and eosinophils. This parameter is normalized, meaning it provides a comparative metric across different patient cases by expressing the number of cooperative clusters as a fraction of the total clusters, rather than a raw count. |
| kekfwzgc1vzf | TL | IS_TUMOR_CORE | IS_TUMOR_CORE is a binary normalized indicator that designates whether a patch qualifies as a high-density tumor core region based on a fixed threshold (at least 10 tumor cells). A value of 1 means the patch meets the threshold criteria and is considered a tumor core, while a value of 0 means it does not. This normalization enables direct and fair comparison across different patient cases. |
| kekns0djxlp0 | F | ORIENTATION_STD_DEG | ORIENTATION_STD_DEG represents the standard deviation of the orientation angles (in degrees) of fibroblasts located at the tumor-stroma interface. This parameter is computed by first identifying fibroblasts that are in close proximity (within 12.5 micrometers) to tumor cells, then extracting the orientation of each fibroblast based on the geometry of its nucleus. The resulting standard deviation quantifies the dispersion of these orientation angles, providing insight into the heterogeneity of fibroblast alignment, which may be indicative of extracellular matrix remodeling and related metastatic processes. Being a derived statistical measure rather than a raw count, it is normalized and suitable for comparing different patient cases. |
| kvqkdy3yp0h1 | T | FILOPODIAL_INDEX | The FILOPODIAL_INDEX quantifies the proportion of tumor cells exhibiting filopodial protrusions relative to the total number of tumor cells in a given image patch. It is calculated by identifying cells with irregular shapes—specifically those with a perimeter ratio greater than a set threshold (1.1), indicating the presence of thin, elongated extensions compared to their convex hull. Because it is represented as a ratio ranging from 0 to 1, this parameter is normalized and can be used to reliably compare results across different patient cases and patches. |
| l46qetz8g9dm | NE | STROMAL_NE_RATIO | STROMAL_NE_RATIO measures the relative abundance of neutrophils compared to eosinophils within the stromal compartment of each tumor patch. It is calculated as a ratio where the count of neutrophils is divided by the count of eosinophils, ensuring normalization across different patches and patient cases. This normalized metric supports comparative analysis even when absolute cell counts can vary widely, and it is designed to handle situations with zero eosinophils by returning NaN in such cases. |
| m7fhggpq6h5a | L | SUBREGION_COUNT | SUBREGION_COUNT represents the number of lymphocyte subregions (clusters) identified within a standardized patch of the tumor. These clusters are determined using a density-based spatial clustering method, ensuring that the count is comparable across different patient cases because the patch size is fixed. |
| m7fhggpq6h5a | L | MEAN_LYMPHO_AREA_UM2 | MEAN_LYMPHO_AREA_UM2 measures the average area of lymphocytes within each patch, calculated by extracting the area from each cell's polygon representation and converting it to square micrometers. By averaging these values over the identified clusters, this parameter provides a normalized metric of cell size across patient cases. |
| m7fhggpq6h5a | L | MEAN_CIRCULARITY | MEAN_CIRCULARITY captures the average circularity of lymphocytes in a patch. Circularity is determined by comparing the cell’s area to its perimeter, yielding a value between 0 and 1 that indicates how round the cells are. This normalized metric is useful for comparing morphological features across patches and patient samples. |
| m7fhggpq6h5a | L | MEAN_DENSITY | MEAN_DENSITY quantifies the average local density of lymphocytes within a patch. It is computed by evaluating the distances between each cell and its nearby neighbors, and then taking the inverse of the average distance. This results in a standardized density measure that is comparable across different patient cases. |
| m7fhggpq6h5a | L | TEMPORAL_SURROGATE_INDEX | TEMPORAL_SURROGATE_INDEX is a composite score that integrates multiple morphological and spatial density features of lymphocytes to infer a pseudo-temporal progression of lymphocyte exhaustion. This index is calculated using a weighted sum of the average lymphocyte area, circularity, and density, ensuring that the metric is normalized and can be compared across different patient cases. |
| m7wcf5yk16d6 | E | GRANULE_SPATIAL_DISPERSION_UM | GRANULE_SPATIAL_DISPERSION_UM represents the normalized spatial dispersion of eosinophil granules within a patch. It is computed as the standard deviation of the pairwise distances between granule centroids, and then converted to micrometers. This normalization allows for accurate comparison of granule dispersion across different patient cases. |
| m7wcf5yk16d6 | E | GRANULE_INTENSITY_VARIATION | GRANULE_INTENSITY_VARIATION measures the normalized variation in intensity of the granule regions within eosinophils. Calculated as the coefficient of variation of the mean intensities of detected granules, it provides a numeric indicator of the heterogeneity in granule staining, enabling consistent comparisons between samples. |
| m7wcf5yk16d6 | E | COMBINED_DISPERSION_FACTOR | COMBINED_DISPERSION_FACTOR is a composite metric that combines the normalized mean spatial dispersion of granules and their intensity variation. By normalizing each individual metric and applying a specific weighting scheme, this parameter yields a single numeric value that effectively captures both spatial and intensity-based variation, making it suitable for cross-patient analyses. |
| nry627bgn553 | LP | MIXED_CLUSTER_RATIO | MIXED_CLUSTER_RATIO measures the proportion of cell clusters in the tumor stroma that contain both lymphocytes and plasma cells compared to those composed of only one cell type. This ratio is normalized by design, allowing for meaningful comparisons across different patient cases and patches, as it reflects the balance between coordinated immune responses and more homogeneous cell groupings. |
| nry627bgn553 | LP | AVG_CLUSTER_SIZE | AVG_CLUSTER_SIZE represents the average number of cells contained within each detected cluster in the tumor stroma. By computing the mean cell count per cluster, this parameter provides an insight into the typical cluster composition and is comparable across different patches, avoiding biases associated with simple raw cell counts. |
| ogf2k1929q38 | MN | COLLOCATION_INDEX | COLLOCATION_INDEX measures the proportion of a tumor patch’s stroma area that exhibits spatially collocated clusters containing both macrophages and neutrophils. It is computed by dividing the total area of such collocated clusters by the total stroma area in the patch, yielding a normalized, dimensionless ratio that allows comparison across different patient cases. |
| p9iwk18gx1j8 | PL | MEAN_DISTANCE_UM | MEAN_DISTANCE_UM represents the average Euclidean distance between plasma cells and lymphocytes within the tumor region. This metric, measured in micrometers, provides an overall indication of how far apart these two cell types are spaced on average in each patch. |
| p9iwk18gx1j8 | PL | MEDIAN_DISTANCE_UM | MEDIAN_DISTANCE_UM is the median value of the Euclidean distances computed between plasma cells and lymphocytes within the tumor area. It offers a robust measure of central tendency that is less influenced by outliers compared to the mean. |
| p9iwk18gx1j8 | PL | MIN_DISTANCE_UM | MIN_DISTANCE_UM denotes the smallest distance recorded between any plasma cell and any lymphocyte pair in the tumor region. This metric reflects the closest proximity between the two cell types in a given patch. |
| p9iwk18gx1j8 | PL | MAX_DISTANCE_UM | MAX_DISTANCE_UM captures the largest Euclidean distance observed between any plasma cell and lymphocyte pair in the tumor region. It indicates the upper bound of spatial separation within the analyzed patch. |
| p9iwk18gx1j8 | PL | STD_DISTANCE_UM | STD_DISTANCE_UM is the standard deviation of the distances between plasma cells and lymphocytes in the tumor region. This parameter quantifies the variability in cell-cell distances, providing insight into the heterogeneity of spatial distribution. |
| pk97k70nnq5q | P | PLASMA_CELL_ANISOTROPY_MEAN | PLASMA_CELL_ANISOTROPY_MEAN represents the average elliptical ratio computed for plasma cells within a patch. This ratio, derived from the cell's polygon shape relative to its centroid, quantifies the degree of elongation, where higher values indicate more elongated plasma cells. |
| pk97k70nnq5q | P | PLASMA_CELL_ANISOTROPY_SD | PLASMA_CELL_ANISOTROPY_SD indicates the standard deviation of the elliptical ratios among plasma cells in a patch. This measure reflects the variability in plasma cell shape, capturing the differences in elongation across individual cells. |
| pk97k70nnq5q | P | PLASMA_CELL_ANISOTROPY_MIN | PLASMA_CELL_ANISOTROPY_MIN records the minimum elliptical ratio observed among plasma cells within a patch, identifying the most circular (least elongated) plasma cells in that region. |
| pk97k70nnq5q | P | PLASMA_CELL_ANISOTROPY_MAX | PLASMA_CELL_ANISOTROPY_MAX captures the maximum elliptical ratio among plasma cells in a patch, highlighting the most elongated (highest anisotropy) plasma cells. |
| pttgx6d8zkju | T | TUMOR_BUDDING_INDEX | TUMOR_BUDDING_INDEX measures the number of tumor buds normalized by the overall tumor area within a patch. This metric is calculated by dividing the number of small, detached tumor cell clusters (buds) that meet the detachment criterion by the total tumor area extracted from the tissue mask. Its normalized nature (buds per pixel) enables effective comparison between different patient cases, accounting for variations in tumor size. |
| pttgx6d8zkju | T | MEAN_BUD_DISTANCE_UM | MEAN_BUD_DISTANCE_UM quantifies the average distance, expressed in micrometers, between each detected tumor bud and the main tumor mass. It is computed by first determining the geometric distance between the convex hull of the main tumor mass and that of each budding cluster, then averaging these distances after a conversion factor is applied. This parameter provides an indication of the spatial separation of budding clusters and is a numeric metric useful for comparative analysis across cases. |
| qik3pcc9rwk9 | T | HOTSPOT_PROPORTION | The 'HOTSPOT_PROPORTION' parameter represents the fraction of tumor cells within a patch that are identified as part of proliferation hotspots. It is normalized because it is calculated by dividing the number of hotspot tumor cells by the total number of tumor cells in the patch, allowing meaningful comparisons across different patient cases. |
| qik3pcc9rwk9 | T | MEAN_NUCLEAR_AREA_UM2 | The 'MEAN_NUCLEAR_AREA_UM2' parameter denotes the average nuclear area of tumor cells in a patch, measured in square micrometers. Being an averaged metric, it is normalized and enables comparison of cellular characteristics across different tumors and patient cases. |
| r65d7exty0n5 | T | GAP_HETEROGENEITY_SCORE | GAP_HETEROGENEITY_SCORE is a normalized metric that represents the coefficient of variation of intercellular gap distances. It quantifies the variability in spatial distances between tumor cell centroids within a patch, facilitating direct comparison across different patient cases. |
| r65d7exty0n5 | T | MEAN_GAP_DISTANCE_UM | MEAN_GAP_DISTANCE_UM is the average distance between connected tumor cells, measured in micrometers. This parameter provides a measure of the typical gap size calculated from the spatial relationships derived through triangulation. |
| r65d7exty0n5 | T | STD_GAP_DISTANCE_UM | STD_GAP_DISTANCE_UM represents the standard deviation of the intercellular gap distances, measured in micrometers. It reflects the dispersion or variability in the spacing between connected tumor cells within a patch. |
| r65d7exty0n5 | T | MIN_GAP_DISTANCE_UM | MIN_GAP_DISTANCE_UM indicates the minimum distance observed between tumor cells within the patch, measured in micrometers. This value captures the smallest gap, highlighting the closest proximity among tumor cell pairs. |
| r65d7exty0n5 | T | MAX_GAP_DISTANCE_UM | MAX_GAP_DISTANCE_UM indicates the maximum distance observed between tumor cells within the patch, measured in micrometers. It represents the largest gap and aids in understanding the range of spacing within the tumor microenvironment. |
| rcil37000fw8 | T | BLEB_FREQUENCY | BLEB_FREQUENCY quantifies the proportion of tumor cells exhibiting membrane blebbing. This is determined by identifying tumor cells with a convexity ratio below a preset threshold (0.95) and calculating the ratio of these bleb-positive cells to the total number of tumor cells in each patch. The resulting value, normalized between 0 and 1, allows for comparisons across different patient cases. |
| rcil37000fw8 | T | MEAN_CONVEXITY_RATIO | MEAN_CONVEXITY_RATIO represents the average convexity ratio across all tumor cells in a patch. The convexity ratio is a measure derived from dividing the area of a cell's polygon by the area of its convex hull, indicating how irregular a cell shape is. As a normalized metric (with values typically ranging from 0 to 1), it provides insight into the overall morphology of tumor cells in a standardized form. |
| rcil37000fw8 | T | SD_CONVEXITY_RATIO | SD_CONVEXITY_RATIO indicates the standard deviation of the convexity ratios for tumor cells within a patch. It measures the variability or spread in cell shape irregularity among tumor cells, offering a numeric reflection of heterogeneity in cell morphology that is normalized for valid cross-patient comparisons. |
| rjni1pb4ja4a | L | MEAN_BLEBBING_INDEX | MEAN_BLEBBING_INDEX measures the average degree of membrane distortion among lymphocytes within a patch. Each lymphocyte’s blebbing index is computed as the relative difference between its actual boundary length and the length of its convex hull. This average provides a normalized metric that facilitates direct comparison between different patient cases by offsetting variations in the number of cells. |
| rjni1pb4ja4a | L | SD_BLEBBING_INDEX | SD_BLEBBING_INDEX represents the standard deviation of the blebbing indices for lymphocytes in a patch. It quantifies the variability in membrane irregularity across the cells. Since the underlying blebbing index is normalized, this standard deviation is also comparable across different patches and patient cases, highlighting the heterogeneity of lymphocyte morphology. |
| rjni1pb4ja4a | L | MAX_BLEBBING_INDEX | MAX_BLEBBING_INDEX indicates the highest blebbing index observed among lymphocytes in a patch. By identifying the most extreme instance of membrane irregularity, this normalized metric can reveal localized regions with significant morphological disturbances and supports cross-patient comparisons. |
| s0gc176ao7c4 | TN | AVG_NEUTROPHIL_DENSITY | AVG_NEUTROPHIL_DENSITY is a normalized numerical measure representing the average density of neutrophils found within a fixed radius (50μm) of tumor margin cells. It is calculated by counting the neutrophils near each tumor cell at the margin, normalizing the count over the area of a circle (using π*(50μm)^2), and then averaging these values across the patch. This metric allows comparison between different patches and patient cases. |
| s0gc176ao7c4 | TN | MARGIN_IRREGULARITY_INDEX | MARGIN_IRREGULARITY_INDEX is a dimensionless numeric indicator that quantifies the morphological irregularity of the tumor margin. It is determined by first identifying tumor cells at the margin (i.e., those in proximity to stroma cells), calculating the centroid for these cells, and then computing the spread (standard deviation) of their distances from the center of mass relative to the mean distance. This provides a normalized measure of how irregular the tumor boundary is. |
| s0gc176ao7c4 | TN | EROSION_SCORE | EROSION_SCORE is the composite numeric parameter that integrates both neutrophil density and margin irregularity. It is computed by multiplying the average neutrophil density (AVG_NEUTROPHIL_DENSITY) with the morphological irregularity index (MARGIN_IRREGULARITY_INDEX). This score reflects the combined effect of the local immune cell activity on the erosion of the tumor margin, allowing standardized comparisons across different cases. |
| setv5642bhib | P | MEAN_PLASMA_CELLS_PER_ZONE | MEAN_PLASMA_CELLS_PER_ZONE is the average number of plasma cells per standardized grid zone (100x100 pixels) within a patch. Since the zones are of constant size, this metric is normalized across different patches and patient cases, enabling comparability despite being derived from cell counts. |
| setv5642bhib | P | MEAN_PLASMA_DENSITY_PER_ZONE | MEAN_PLASMA_DENSITY_PER_ZONE quantifies the average plasma cell density by dividing the number of plasma cells in each zone by the zone area (converted to micrometers squared). This normalized density measurement allows for fair comparison across tumors and patient cases. |
| setv5642bhib | P | MEAN_PLASMA_NN_DIST | MEAN_PLASMA_NN_DIST represents the average nearest neighbor distance among plasma cells within each zone (calculated only when more than one plasma cell is present). This distance, expressed in micrometers and normalized by zone dimensions, provides insight into spatial aggregation patterns that are directly comparable across different patient cases. |
| sr9tmf7aldah | E | VACUOLE_INDEX | VACUOLE_INDEX measures the average number of vacuoles per eosinophil in tumor regions of a tissue patch. It is computed by detecting vacuole-like structures within each eosinophil and then averaging these counts over the total number of eosinophils, thereby providing a normalized metric that can be compared across different patient cases. |
| sr9tmf7aldah | E | MEAN_VACUOLE_AREA_UM2 | MEAN_VACUOLE_AREA_UM2 measures the mean area of the identified vacuoles, expressed in square micrometers (μm²), per eosinophil in tumor regions. The area of each detected vacuole is summed for all eosinophils and then divided by the number of eosinophils to produce a normalized parameter suitable for inter-patient comparisons. |
| t5j4xihitr5y | P | CHROMATIN_UNIFORMITY_MEAN | CHROMATIN_UNIFORMITY_MEAN represents the average score derived from the texture analysis of plasma cell nuclei within a patch. This metric is obtained by comparing the symmetry of chromatin distribution across opposite angular sectors of the polar-transformed nucleus image. Lower values of this mean indicate a more uniform and symmetric chromatin pattern, reflective of the typical 'cartwheel' appearance of plasma cell nuclei. |
| t5j4xihitr5y | P | CHROMATIN_UNIFORMITY_SD | CHROMATIN_UNIFORMITY_SD quantifies the variability of the uniformity scores within a patch by calculating the standard deviation of these scores across the analyzed plasma cells. A lower standard deviation suggests that the chromatin patterns are consistently uniform across the cells, while higher values indicate a greater variability in the chromatin texture. |
| t5j4xihitr5y | P | CHROMATIN_SYMMETRY_SCORE | CHROMATIN_SYMMETRY_SCORE represents the median symmetry score computed from the texture analysis of plasma cell nuclei in a patch. It is calculated by taking the median of absolute differences between texture descriptors of opposite angular sectors obtained from a polar-transformed nucleus image. This metric provides a robust measure of the overall chromatin symmetry, with lower values indicating a more balanced and uniform chromatin arrangement. |
| uasgmrcmc4in | P | MOTT_LIKE_PLASMA_FREQ | MOTT_LIKE_PLASMA_FREQ measures the normalized frequency of Mott-like plasma cells in the peritumoral region of lung cancer tissue patches. It is calculated as the ratio of plasma cells, classified as Mott-like based on having a dark-stained area (indicative of Russell bodies) constituting more than 30% of the total cell area, to the total number of plasma cells within the peritumoral region. This parameter is numeric and ranges from 0 to 1, allowing comparison across different patient cases. The analysis is conducted on patches of tissue by first identifying plasma cells within the stromal region, determining their proximity to tumor cells using a set distance threshold, and then applying an intensity threshold to detect relevant intracellular features. |
| umzmyafid66j | L | BARRIER_DENSITY_MEAN | BARRIER_DENSITY_MEAN represents the average density of lymphocytes around tumor boundary cells in a defined circular region. It is computed by counting lymphocytes within a fixed radius around each tumor boundary cell, normalizing by the area of the circle, and then taking the mean of these densities across all boundary cells in a tumor patch. |
| umzmyafid66j | L | BARRIER_DENSITY_MAX | BARRIER_DENSITY_MAX indicates the highest lymphocyte density observed among all tumor boundary cells in the patch. This parameter highlights the tumor-stroma region that has the most intense lymphocyte aggregation, providing insight into peak protective barrier formation. |
| umzmyafid66j | L | BARRIER_DENSITY_MIN | BARRIER_DENSITY_MIN reflects the lowest lymphocyte density measured in any tumor boundary area within a patch. It helps identify regions where the lymphocyte barrier is weakest, informing on areas of potential vulnerability in the tissue microenvironment. |
| umzmyafid66j | L | BARRIER_DENSITY_STD | BARRIER_DENSITY_STD is the standard deviation of the lymphocyte densities calculated for the tumor boundary cells. This metric quantifies the variability in lymphocyte barrier formation within the patch, indicating uniformity or heterogeneity in immune response at the tumor margin. |
| uqi9h9gb75u2 | L | ASYMMETRY_INDEX | ASYMMETRY_INDEX is a normalized metric that quantifies the degree of directional lymphocyte infiltration skew within a tumor patch. It is computed by subtracting the minimum lymphocyte count from the maximum lymphocyte count across eight angular sectors and then dividing the result by the average count across all sectors. This parameter captures relative differences in infiltration directionality, allowing comparison between different tumor cases without being affected by patch size or total cell counts. |
| uqi9h9gb75u2 | L | MAX_MIN_RATIO | MAX_MIN_RATIO is a normalized measure that offers an alternative view of the directional bias in lymphocyte distribution. It is calculated as the ratio of the maximum lymphocyte count to the minimum lymphocyte count (with a slight smoothing correction to avoid division by zero). This metric reflects the disparity between the most and least infiltrated sectors and enables consistent comparisons across different patient samples. |
| urig21w4ab21 | TE | MEAN_DEGRANULATION_TUMOR_OVERLAP | MEAN_DEGRANULATION_TUMOR_OVERLAP measures the average ratio of overlap between the eosinophil degranulation clouds and tumor cell regions within a 1x1 mm patch. This ratio is computed by dividing the area of intersection between the buffered eosinophil (representing its degranulation cloud) and the union of tumor cell polygons by the total area of the degranulation cloud. The normalized ratio facilitates comparison across different patient cases and patches. |
| urig21w4ab21 | TE | MAX_DEGRANULATION_TUMOR_OVERLAP | MAX_DEGRANULATION_TUMOR_OVERLAP captures the maximum overlap ratio observed among all eosinophils in a given patch. This metric identifies the eosinophil with the greatest proportional interaction with tumor regions, offering a normalized value that supports direct comparisons among patches from multiple patients. |
| urig21w4ab21 | TE | STD_DEGRANULATION_TUMOR_OVERLAP | STD_DEGRANULATION_TUMOR_OVERLAP reflects the standard deviation of the overlap ratios of eosinophil degranulation clouds with tumor regions within a patch. It quantifies the variability of these normalized ratios, thereby highlighting the consistency or disparity in the spatial interactions across different eosinophils in the same patch. |
| uycmsvecwb4g | LM | LM_ADJACENCY_RATIO | LM_ADJACENCY_RATIO is a normalized metric that quantifies the proportion of lymphocytes in a tumor stroma patch which are in close proximity to at least one macrophage. It is computed as the ratio of lymphocytes with an adjacent macrophage within a preset threshold to the total lymphocytes observed in the patch, enabling comparison across different patient cases. |
| uz1pgqa1u9k1 | T | MEAN_MITO_SCORE | MEAN_MITO_SCORE represents the average normalized mitochondrial surrogate score per patch. It is calculated by first quantifying texture features from the eosin stained channel of tumor cell cytoplasmic regions, which serve as a surrogate indicator of mitochondrial abundance. This value is then normalized by the nuclear area of each cell, ensuring that the metric can be compared across different patches and patient cases by adjusting for cell size. |
| uz1pgqa1u9k1 | T | MEDIAN_MITO_SCORE | MEDIAN_MITO_SCORE represents the median normalized mitochondrial surrogate score per patch. This metric captures the central tendency of per-cell normalized scores within each patch, providing a robust measure that is less influenced by extreme values in the distribution of mitochondrial content. |
| uz1pgqa1u9k1 | T | SD_MITO_SCORE | SD_MITO_SCORE is the standard deviation of the normalized mitochondrial surrogate scores within each patch. This parameter quantifies the heterogeneity of mitochondrial content among tumor cells in a patch, allowing for an assessment of the variability in mitochondrial surrogate measurements after normalization by nuclear area. |
| vm0hzhvi7g62 | N | MEAN_LOBE_COUNT | MEAN_LOBE_COUNT represents the average number of nuclear lobes detected in neutrophils per patch. It is derived by identifying and counting the concave regions in the neutrophil nucleus polygon for each cell and then computing the average across all neutrophils in a given patch. This average value normalizes the measurement across different patient cases. |
| vm0hzhvi7g62 | N | MEAN_IRREGULARITY_INDEX | MEAN_IRREGULARITY_INDEX represents the average measure of nuclear contour irregularity in neutrophils per patch. The index is computed by comparing the actual nucleus perimeter with the perimeter of its convex hull, resulting in a value between 0 and 1, where higher values indicate more irregular nuclear shapes. The averaging process over the patch enables normalized comparison between patient cases. |
| vm0hzhvi7g62 | N | MEAN_COMPLEXITY_SCORE | MEAN_COMPLEXITY_SCORE represents the average nuclear complexity of neutrophils per patch. It is calculated as the product of the lobe count and the irregularity index for each neutrophil, combining both morphological complexity and shape irregularity into one metric. The score is averaged across the cells in each patch, making it a normalized parameter suitable for comparing different patient cases. |
| vsee52flg0b4 | P | BRIDGING_RATIO | BRIDGING_RATIO is a normalized measure that represents the proportion of plasma cell pairs showing physical bridging contact relative to the total number of evaluated pairs. This ratio facilitates comparison across different patches and patient cases by accounting for variations in the total number of cell pairs. |
| vsee52flg0b4 | P | MEAN_BRIDGE_DISTANCE_UM | MEAN_BRIDGE_DISTANCE_UM is a normalized, numerically computed parameter that indicates the average distance (in micrometers) between plasma cell pairs that are identified as having a bridging contact. By averaging the distances exclusively from the bridged pairs, this parameter provides a consistent measure of how closely these cellular interactions occur across different tumor regions. |
| woxlvu8ngaa7 | MF | COLLABORATION_INDEX | COLLABORATION_INDEX represents the normalized measure of the interaction between macrophages and fibroblasts within a tissue patch. It is computed as the percentage of macrophage-fibroblast pairs that are in close proximity (within a distance threshold equivalent to 50 micrometers) compared to the total number of potential pairs. This percentage value, ranging from 0 to 100, allows for direct comparison across different patient cases, making it a robust metric for assessing cellular spatial interactions in the tumor microenvironment. |
| y8dhnwb45gnj | N | MEAN_GRANULARITY | MEAN_GRANULARITY represents the average intensity variation score across all neutrophils in a patch. It is obtained by first measuring the standard deviation of pixel intensities within each neutrophil cell – a proxy for granularity – and then computing the mean of these scores over the patch. This metric allows for a normalized comparison of neutrophil granularity across different patient cases because each patch is of a standardized size. |
| y8dhnwb45gnj | N | SD_GRANULARITY | SD_GRANULARITY quantifies the variability in the granularity scores of neutrophils within a patch. By calculating the standard deviation of the granularity scores across individual cells, this parameter provides insights into the heterogeneity of neutrophil granule intensity variations. This numeric measure is comparable between patient cases due to the standardized patch sampling. |
| y8dhnwb45gnj | N | VARIATION_COEFFICIENT | VARIATION_COEFFICIENT is a normalized metric defined as the ratio of the standard deviation (SD_GRANULARITY) to the mean (MEAN_GRANULARITY) of the granularity scores within a patch. This parameter offers a relative measure of dispersion in granularity scores, effectively facilitating comparisons across different tumor regions and patient cases. |
| y9vgcyprsj9r | T | MICRO_LUMEN_INDEX_PER_MM2 | MICRO_LUMEN_INDEX_PER_MM2 quantifies the density of tumor cell clusters that contain micro-lumen structures normalized per square millimeter of tumor area. This parameter is derived by identifying clusters with a centrally located hollow region, calculating the number of such micro-lumen clusters, and then normalizing this count by the tumor area within the patch. The normalization allows for robust comparisons across different patient cases, reflecting the underlying biological hypothesis that micro-lumen formation is associated with an invasive phenotype in tumors. |
| yedb9ik4nimn | LPMNE | STROMAL_IMMUNE_DIVERSITY | The parameter 'STROMAL_IMMUNE_DIVERSITY' quantifies the diversity of immune cells in the stromal compartment of tumor patches using the Shannon diversity index. It is calculated by first filtering immune cells within the stroma, then computing the relative proportions of different immune cell subtypes, and finally applying the Shannon formula to derive a normalized, numeric diversity measure. This index allows for comparisons across different patient cases since it is based on cell proportions rather than raw counts and provides insights into the heterogeneity of the immune microenvironment, where higher values suggest greater diversity that may be associated with dynamic tumor behavior. |
| zdqpczi83kiv | T | EMT_SCORE_MEAN | EMT_SCORE_MEAN represents the average epithelial-to-mesenchymal transition-like morphology score computed for tumor cells within each patch. This score is derived from the cell shape measurements, particularly the cell's elongation (aspect ratio) and its roundness. The higher the EMT_SCORE_MEAN, the more pronounced the EMT-like features are, making it a normalized metric to compare different patient cases. |
| zdqpczi83kiv | T | EMT_SCORE_STD | EMT_SCORE_STD is the standard deviation of the EMT scores calculated for tumor cells in a patch. This metric indicates the variability or heterogeneity of the EMT-like morphological features among the cells in a patch. Since it is computed as a statistical measure from normalized cell scores, it enables comparative analysis across different patient samples. |
| zdqpczi83kiv | T | ASPECT_RATIO_MEAN | ASPECT_RATIO_MEAN captures the average value of the cell aspect ratios within a patch. The aspect ratio is determined by the ratio of the longer side to the shorter side of the minimum rotated rectangle that encloses the cell. This measure quantifies cell elongation in a normalized manner, allowing comparison between different regions or patients. |
| zdqpczi83kiv | T | ROUNDNESS_MEAN | ROUNDNESS_MEAN is the average roundness value of tumor cells in a patch. Roundness is computed based on the cell area and perimeter, reflecting how closely a cell's shape approximates a circle. This normalized metric supports cross-patient comparisons by summarizing the cell shape characteristics within each analyzed patch. |
| ze5y8vt7fsxr | TL | MEDIAN_TL_DISTANCE | MEDIAN_TL_DISTANCE represents the median of the Euclidean distances computed between the centroids of tumor cells and lymphocyte cells within a patch. The distances are converted to micrometers from pixels and this metric provides a robust measure of central tendency for tumor-to-lymphocyte proximity, enabling normalized comparisons across different patient cases. |
| ze5y8vt7fsxr | TL | MEAN_TL_DISTANCE | MEAN_TL_DISTANCE is the arithmetic average of the Euclidean distances between tumor cell centroids and lymphocyte cell centroids in a patch, with distances expressed in micrometers. This normalized metric summarizes the overall spatial separation between these cell types and facilitates comparison across various patches and patient samples. |
| ze5y8vt7fsxr | TL | STD_TL_DISTANCE | STD_TL_DISTANCE quantifies the variability in the distances calculated between tumor and lymphocyte cell centroids. By expressing the variability (standard deviation) in micrometers, it provides insight into the dispersion of spatial relationships, enabling normalized assessments of heterogeneity across patient cases. |
| ze5y8vt7fsxr | TL | HIGH_DEVIATION | HIGH_DEVIATION is a binary flag that indicates whether the patch exhibits abnormal spatial patterns based on preset thresholds for median and standard deviation distances. A flag value of 1 indicates that either the median distance exceeds 100 micrometers or the standard deviation exceeds 50 micrometers, thus enabling quick normalized identification of atypical cell clustering patterns. |
| zindts95vrfa | TP | PLASMA_CELL_VARIATION_INDEX | PLASMA_CELL_VARIATION_INDEX is a normalized metric that quantifies the variability in plasma cell infiltration across tumor subclones within a patch. It is computed as the coefficient of variation, which is the ratio of the standard deviation to the mean of the plasma cell counts from different subclones. This dimensionless value allows for the comparison of heterogeneity in plasma cell distribution across different patients and tumor regions. |
| zlzt4yp6vaia | M | POLARIZATION_INDEX | POLARIZATION_INDEX: A normalized metric computed as (2 times the minimum count of macrophages between two morphological clusters divided by the total number of macrophages) that yields values between 0 and 1. This measure reflects the balance between the two distinct macrophage subpopulations within a tumor stroma patch, with a value of 1 indicating a perfectly balanced distribution and 0 indicating a complete skew towards one cluster. |
| zlzt4yp6vaia | M | MEAN_AREA_UM2 | MEAN_AREA_UM2: An averaged measurement of the macrophage cell areas in square micrometers within each patch. The individual cell areas are calculated from the polygon properties after an appropriate conversion, and the mean is used to represent the typical cell size, facilitating comparisons across different patient cases. |
| zlzt4yp6vaia | M | MEAN_CIRCULARITY | MEAN_CIRCULARITY: An aggregated measure of the average circularity of macrophages within a patch. Circularity, derived from the relationship between cell area and perimeter, ranges from 0 to 1, where a value of 1 represents a perfectly circular cell. This parameter provides insights into cell shape uniformity and is normalized through averaging, allowing for consistent comparisons. |
| zz1vs1fmv9xw | P | MEAN_NUCLEUS_CYTOPLASM_RATIO | MEAN_NUCLEUS_CYTOPLASM_RATIO represents the average of the computed ratios of nucleus area to approximated cytoplasm area for plasma cells within each tissue patch. This normalized metric is derived from calculating the nucleus area and estimating the cytoplasm area by dilating the nucleus boundary, ensuring comparability across different patient cases. |
| zz1vs1fmv9xw | P | SD_NUCLEUS_CYTOPLASM_RATIO | SD_NUCLEUS_CYTOPLASM_RATIO quantifies the variability or dispersion of the nucleus-to-cytoplasm ratios computed for plasma cells in a tissue patch. This standard deviation measure provides insight into the heterogeneity of nuclear features relative to cytoplasmic space, and its normalized value allows for comparison across samples. |