Deep learning-based surgical phase recognition in laparoscopic cholecystectomy.

Autor:	Yang HY; Department of Liver Transplantation and Hepatobiliary and Pancreatic Surgery, Ajou University School of Medicine, Suwon, Korea., Hong SS; Department of Hepatobiliary and Pancreatic Surgery, Yonsei University College of Medicine, Seoul, Korea., Yoon J; Hutom Corp., Seoul, Korea., Park B; Hutom Corp., Seoul, Korea., Yoon Y; Hutom Corp., Seoul, Korea., Han DH; Department of Hepatobiliary and Pancreatic Surgery, Yonsei University College of Medicine, Seoul, Korea., Choi GH; Department of Hepatobiliary and Pancreatic Surgery, Yonsei University College of Medicine, Seoul, Korea., Choi MK; Hutom Corp., Seoul, Korea., Kim SH; Department of Hepatobiliary and Pancreatic Surgery, Yonsei University College of Medicine, Seoul, Korea.
Jazyk:	angličtina
Zdroj:	Annals of hepato-biliary-pancreatic surgery [Ann Hepatobiliary Pancreat Surg] 2024 Nov 30; Vol. 28 (4), pp. 466-473. Date of Electronic Publication: 2024 Jul 29.
DOI:	10.14701/ahbps.24-091
Abstrakt:	Backgrounds/aims: Artificial intelligence (AI) technology has been used to assess surgery quality, educate, and evaluate surgical performance using video recordings in the minimally invasive surgery era. Much attention has been paid to automating surgical workflow analysis from surgical videos for an effective evaluation to achieve the assessment and evaluation. This study aimed to design a deep learning model to automatically identify surgical phases using laparoscopic cholecystectomy videos and automatically assess the accuracy of recognizing surgical phases. Methods: One hundred and twenty cholecystectomy videos from a public dataset (Cholec80) and 40 laparoscopic cholecystectomy videos recorded between July 2022 and December 2022 at a single institution were collected. These datasets were split into training and testing datasets for the AI model at a 2:1 ratio. Test scenarios were constructed according to structural characteristics of the trained model. No pre- or post-processing of input data or inference output was performed to accurately analyze the effect of the label on model training. Results: A total of 98,234 frames were extracted from 40 cases as test data. The overall accuracy of the model was 91.2%. The most accurate phase was Calot's triangle dissection (F1 score: 0.9421), whereas the least accurate phase was clipping and cutting (F1 score: 0.7761). Conclusions: Our AI model identified phases of laparoscopic cholecystectomy with a high accuracy.
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu