ReXGroundingCT is a large-scale 3D chest CT dataset linking free-text radiology findings to pixel-level segmentations in volumetric imaging. It comprises 3,142 non-contrast chest CT scans with 8,028 annotated findings (16,301 entities) from the CT-RATE dataset. ReXGroundingCT enables sentence-level grounding for both focal and non-focal lung and pleural abnormalities across 14 categories. On ReXrank, we are hosting ReXGroundingCT's testset, which contains 100 CT scans with exhaustive radiologist annotations for all visible findings.
Global Dice: Average Dice per finding per case
Global HIT Rate: Proportion of findings that have Dice >= 0.1
Instance Precision: TP / (TP + FP), where True Positives are instances that have a Dice >= 0.2
Instance Recall: TP / (TP + FN)
Instance F1: Harmonic mean of Instance Precision and Instance Recall
| Model | Global Dice | Global HIT Rate | Instance Precision | Instance Recall | Instance F1 |
|---|---|---|---|---|---|
| SAT-FT | 0.209 | 0.473 | 0.074 | 0.369 | 0.123 |
| BiomedParseV2 | 0.025 | 0.066 | 0.007 | 0.072 | 0.012 |
| SegVol | 0 | 0 | 0 | 0 | 0 |
| SAT | 0 | 0 | 0 | 0 | 0 |