Surgical video understanding
Collection
LoRA adapters, supervised baselines, and LemonFM linear probes trained for the SDSC x Chicago Booth surgical video understanding leaderboard. • 27 items • Updated
ResNet-50 — CholecT50 verb recognition
Supervised 10-way multi-label ResNet-50 baseline for current-action (verb) recognition on CholecT50 frames.
Trained as a baseline for the SDSC × Chicago Booth surgical video understanding leaderboard (Recommendations tab).
Prompt example
This closed-set example mirrors the leaderboard format, not a text-input API for this checkpoint.
[surgical frame]
Which surgical actions are being performed in this cholecystectomy frame?
Select every matching label.
- grasp
- retract
- dissect
- coagulate
- clip
- cut
- aspirate
- irrigate
- pack
- idle
Evaluation
Full 19,923-frame validation split (95% bootstrap CI):
| Metric | Value |
|---|---|
| Exact match | 52.9% (52.3–53.6) |
| Micro-averaged F1 | 75.2% (74.8–75.6) |
References
- Skobelev, K., Fithian, E., Baranovski, Y., et al. A Comparative Study in Surgical AI: Potential and Limitations of Data, Compute, and Scaling. arXiv:2603.27341, 2026.
- Dataset: Nwoye, C. I., Yu, T., Gonzalez, C., et al. Rendezvous: Attention Mechanisms for the Recognition of Surgical Action Triplets in Endoscopic Videos. Medical Image Analysis, 78, 102433 (2022).
Limitations
Research baseline only. Not a medical device.