240701_4DHSI_Bovine_Nerve
Authors/Creators
Contributors
Other (2):
Description
A 4D hyperspectral imaging (HSI) dataset capturing autofluorescence of bovine nerve and ligament tissues. Imaging was performed under multiple excitation conditions, representing the fourth dimension. Ground truth masks are included for tissue segmentation analysis.
Materials & Methods
Sample Preparation
Nerve and ligament were isolated from a bovine trotter acquired at the local market. The trotter was kept in water before acquisition by us, but was stored in the freezer before the experiment date.
For imaging, the nerve and ligament were put in a 90 mm plastic petri dish, attached to the bottom using a tape, and covered with 30 mL medical-grade normal saline (0.9%) to avoid drying of the sample (see in Photos)
Tissues in HSI images*, from top to bottom:
-
Nerve
-
Ligament
* The original dataset included three pieces of tissues, one of which identity was uncertain, so we cropped it and included only two tissues.
Sample Excitation/Illumination for HSI
For illuminating the sample:
-
Instrument: Tunable Light Source (model: TLS130B-300X, Newport Corporation)
-
Fiber Bundle Focusing Assembly was used to obtain higher intensities (model: 77776, Newport Corporation)
-
Simple concave mirror was used to direct light from the instrument to the sample
Light source optical power was measured at each wavelength using Optical Power Meter (model: 1936-R, Newport Corporation) with a Photodiode Sensor (model: 918D-UV-OD3R, Newport Corporation). Measurement was repeated 5 times and averaged. Respective data are available in metadata file(s).
4D HSI
Nuance FX Imaging System (Model: N-MSI-FX, CRi, Woburn, MA, USA) was used to acquire HSI data with parameters described in:
-
Metadata.csv
Metadata Variables
|
Variable |
Description |
|
cube_name |
Identifier or filename of the hyperspectral image cube, often including sample or acquisition details. |
|
excitation_nm |
Central wavelength (in nm) of the excitation/illumination light used for fluorescence or reflectance imaging. |
|
emission_start_nm |
Shortest wavelength (in nm) captured in the emission spectrum, defining the lower detection limit. |
|
emission_end_nm |
Longest wavelength (in nm) captured in the emission spectrum, defining the upper detection limit. |
|
step_nm |
Spectral resolution or step size (in nm) between consecutive bands in the hyperspectral image. |
|
exposure_time_ms |
Duration (in milliseconds) for which the sensor is exposed to light per spectral band. |
|
acquisition_timestamp |
Timestamp of the imaging process. |
|
notes |
Free-text field for additional comments, observations, or metadata. |
|
light_source |
Type and/or model of the illumination source (e.g., laser, LED, halogen lamp). |
|
light_power_uW |
Power of the illumination/excitation light in microwatts (µW), measured approximately at the sample position. |
|
light_fwhm_nm |
Spectral width (FWHM in nm) of the excitation light source. |
|
emission_wavelengths_nm |
List or array of specific wavelengths (in nm) at which emission signals are recorded. |
|
emission_fwhm_nm |
List or array of full width at half maxima (FWHM in nm) for each emission wavelength, indicating spectral resolution. |
|
sensor_gain |
Amplification factor applied to the sensor signal. |
|
custom_field |
User-defined or placeholder column for additional metadata. |
Ground Truth Masks
Masks generated manually and semi-manually are available to serve as a ground truth reference for tissues in the HSI images.
-
Mask_Manual.png
-
Mask_SemiManual.png
Pixel labels in masks:
-
Background: 0
-
Ligament: 100
-
Nerve: 200
Background has some fluorescence - it represents essentially a plastic petri dish
Files
Cubes.zip
Files
(219.7 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:dd6bddbd1099ada806c3cd5927fd17b7
|
213.3 MB | Preview Download |
|
md5:2d9b7312a77420c97475584bd272cf02
|
1.3 kB | Preview Download |
|
md5:0a1a02877c9a3c1a923e9f40a2c79a5e
|
1.5 kB | Preview Download |
|
md5:e5480400ea2ab9a8c695ba5cec14ecbf
|
2.6 kB | Preview Download |
|
md5:52449fda443c08af32bfacb347973783
|
6.1 MB | Preview Download |
|
md5:81e96cddf9b0dc7c42e09fee662d90cf
|
246.5 kB | Preview Download |
Additional details
Funding
Dates
- Created
-
2024-07-01