An affordable real-time assessment system for surgical skill training

Abstract

This research proposes a novel computer-vision-based approach for skill assessment by observing a surgeon's hand and surgical tool movements in minimally invasive surgical training, which can be extended to the evaluation in real surgeries. Videos capturing the surgical field are analyzed using a system composed of a series of computer vision algorithms. The system automatically detects major skill measuring features from surgical task videos and provides real-time performance feedback on objective and quantitative measurement of surgical skills.

A Survey of Video-based Action Quality Assessment

“…Traditional medical skill evaluation methods usually need contact sensors which will affect the operation of doctors. To overcome this disadvantage, [43], [44] proposed a surgical skill method based on visual information. Based on Islam et al [43], [44] improved the algorithm to realtime, added feedback function and tested the model in the real surgical video.…”

“…To overcome this disadvantage, [43], [44] proposed a surgical skill method based on visual information. Based on Islam et al [43], [44] improved the algorithm to realtime, added feedback function and tested the model in the real surgical video. In order to deal with the semantic gap between low-level visual features and high-level surgical skills, Chen et al [45] fused multi-modal information and completed meaningful surgical skill analysis by looking for hidden space.…”