Surgical Phase Recognition in Laparoscopic Cholecystectomy
This work addresses surgical workflow analysis for medical professionals, but it is incremental as it builds on existing action segmentation methods.
The authors tackled surgical phase recognition in laparoscopic cholecystectomy by proposing a Transformer-based method with calibrated confidence scores for a 2-stage inference pipeline, achieving improved performance over the baseline on the Cholec80 dataset.
Automatic recognition of surgical phases in surgical videos is a fundamental task in surgical workflow analysis. In this report, we propose a Transformer-based method that utilizes calibrated confidence scores for a 2-stage inference pipeline, which dynamically switches between a baseline model and a separately trained transition model depending on the calibrated confidence level. Our method outperforms the baseline model on the Cholec80 dataset, and can be applied to a variety of action segmentation methods.