Expert surgeons and deep learning models can predict the outcome of surgical hemorrhage from 1 min of video
Problem
At the onset of major vascular injury in minimally invasive surgery, surgeons cannot predict blood loss or whether hemorrhage can be controlled.
Why it matters
Uncontrolled surgical bleeding is catastrophic and often fatal; early prediction could guide response.
What the AI did
A deep neural network (SOCALNet, ResNet-based) trained on 147 SOCAL simulator videos predicted hemostasis control from the first minute of video, compared with four blinded expert surgeons.
Results
not stated in source
Impact
Title claims experts and the DNN can predict hemorrhage outcome, but the abstract is truncated before quantitative results.
Record
| Venue | Scientific Reports |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Dhiraj J. Pangal, Guillaume Kugener, Yichao Zhu, Aditya Sinha, Vyom Unadkat, David J. Coté, Ben Allen Strickland, Martin J. Rutkowski et al. |
| Citations | 26 (OpenAlex, at import time) |
| References | 69 |
| Model or method | SOCALNet (ResNet-based DNN) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1038/s41598-022-11549-2 |
| Open access | gold |
| Keywords | deep learning, surgical video, hemorrhage, hemostasis, prediction, SOCAL |
Importance rating
significant — thousands to millions affected
Abstract states major vascular injury is a catastrophic and often fatal surgical complication; significant patient-safety benefit but scale not quantified.