Using Computer Vision to Automate Hand Detection and Tracking of Surgeon Movements in Videos of Open Surgery
This work provides a new automated tool for objectively evaluating surgical techniques in open surgery, which could benefit surgeons and medical educators by offering insights into movement patterns and economy of motion.
This paper addresses the lack of objective evaluation tools for open surgery by developing an automated computer vision approach to detect and track surgeons' hands in surgical videos. The method utilizes a convolutional neural network for hand detection, which significantly outperforms existing hand-detection datasets, and combines it with an object tracker to analyze intra-operative movement patterns.
Open, or non-laparoscopic surgery, represents the vast majority of all operating room procedures, but few tools exist to objectively evaluate these techniques at scale. Current efforts involve human expert-based visual assessment. We leverage advances in computer vision to introduce an automated approach to video analysis of surgical execution. A state-of-the-art convolutional neural network architecture for object detection was used to detect operating hands in open surgery videos. Automated assessment was expanded by combining model predictions with a fast object tracker to enable surgeon-specific hand tracking. To train our model, we used publicly available videos of open surgery from YouTube and annotated these with spatial bounding boxes of operating hands. Our model's spatial detections of operating hands significantly outperforms the detections achieved using pre-existing hand-detection datasets, and allow for insights into intra-operative movement patterns and economy of motion.