Enhancing Surgical Performance Through Automated Video Analysis Utilizing Computer Vision and Machine Learning

Abstract

Aim: Recent advancements in computer vision have enabled the development of automated systems that can assess surgeons’ expertise with high accuracy using automated performance metrics (APMs). This study aims to evaluate and enhance surgical performance through the use of APMs. Method: This is a prospective, quality-control, multicenter international cohort study. The primary outcome is the improvement of APMs extracted from two-dimensional laparoscopic or robotic colorectal procedure video films after feedback to the surgeons. The secondary outcome is the development of new metrics to measure the model’s performance beyond simple accuracy. The collaborators will send 2-3 real-world video films of colorectal procedures they have performed. They will then receive feedback on their films, including an APM data analytics report. After the feedback, the collaborators will send 2-3 videos of the same colorectal procedures. Data analysis of APMs comparing pre- and post-feedback operations will follow. Conclusion: The study will enable efficient training programs within constrained working hours and address heightened ethical considerations regarding patient safety. Moreover, the training of surgeons in low- and middle-income countries will benefit from the results of this study, as they can improve their skills without the need to spend months to years training in developed countries.