CIRDataset: A large-scale Dataset for Clinically-Interpretable lung nodule Radiomics and malignancy prediction

We release 956 radiologist QA/QC’ed spiculation/lobulation annotations on segmented lung nodules for two public datasets, LIDC (with visual radiologist malignancy RM scores for the entire cohort and pathology-proven malignancy PM labels for a subset) and LUNGx (with pathology-proven size-matched benign/malignant nodules to remove the effect of size on malignancy prediction). We also release our multi-class Voxel2Mesh extension (available on our Clinically-Intrepretable Radiomics GitHub) to provide a good baseline for end-to-end deep learning lung nodule segmentation, peaks’ classification (lobulation/spiculation), and malignancy prediction; Voxel2Mesh is the only published method to our knowledge that preserves sharp peaks during segmentation and hence its use as our base model.

The primary motivation of this work comes from our collaborators in radiology inquiring about the importance of clinically-reported LUNG-RADS features such as spiculation/lobulation in state-of-the-art deep learning malignancy prediction methods. Previous methods have performed malignancy prediction for LIDC and LUNGx datasets but without robust attribution to any clinically reported/actionable features (see extensive literature on sensitivity of attribution methods to hyperparameters). This motivated us to annotate clinically-reported features at voxel/vertex-level on public lung nodule datasets (using our negative area distortion metric computed via spherical parameterization to annotate spiculations/lobulations on meshes followed by radiologist QA/QC) and relating these to malignancy prediction (bypassing the “flaky” attribution schemes). With the release of this comprehensively-annotated dataset, we hope that previous malignancy prediction methods can also validate their explanations and provide clinically-actionable insights. We also release our entire pipeline to generate the spiculation/lobulation annotations from scratch for LIDC/LUNGx as well as new datasets.