Video and Accelerometer-Based Motion Analysis for Automated Surgical\n Skills Assessment

Résumé fourni par la source

Purpose: Basic surgical skills of suturing and knot tying are an essential\npart of medical training. Having an automated system for surgical skills\nassessment could help save experts time and improve training efficiency. There\nhave been some recent attempts at automated surgical skills assessment using\neither video analysis or acceleration data. In this paper, we present a novel\napproach for automated assessment of OSATS based surgical skills and provide an\nanalysis of different features on multi-modal data (video and accelerometer\ndata). Methods: We conduct the largest study, to the best of our knowledge, for\nbasic surgical skills assessment on a dataset that contained video and\naccelerometer data for suturing and knot-tying tasks. We introduce "entropy\nbased" features - Approximate Entropy (ApEn) and Cross-Approximate Entropy\n(XApEn), which quantify the amount of predictability and regularity of\nfluctuations in time-series data. The proposed features are compared to\nexisting methods of Sequential Motion Texture (SMT), Discrete Cosine Transform\n(DCT) and Discrete Fourier Transform (DFT), for surgical skills assessment.\nResults: We report average performance of different features across all\napplicable OSATS criteria for suturing and knot tying tasks. Our analysis shows\nthat the proposed entropy based features out-perform previous state-of-the-art\nmethods using video data. For accelerometer data, our method performs better\nfor suturing only. We also show that fusion of video and acceleration features\ncan improve overall performance with the proposed entropy features achieving\nhighest accuracy. Conclusions: Automated surgical skills assessment can be\nachieved with high accuracy using the proposed entropy features. Such a system\ncan significantly improve the efficiency of surgical training in medical\nschools and teaching hospitals.\n

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.