Tomer Golany1, Amit Aides2, Daniel Freedman1, Nadav Rabani2, Yun Liu2, Ehud Rivlin1, Greg S Corrado2, Yossi Matias3, Wisam Khoury4, Hanoch Kashtan5, Petachia Reissman6,7. 1. Verily Life Sciences, Tel Aviv, Israel. 2. Google Health, Tel Aviv, Israel. 3. Google Research, Tel Aviv, Israel. 4. Department of Surgery, Rappaport Faculty of Medicine, Carmel Medical Center, Technion, Haifa, Israel. 5. Department of Surgery, Rabin Medical Center, The Sackler School of Medicine, Tel-Aviv University, Petah Tikva, Israel. 6. Department of Surgery, The Hebrew University School of Medicine, Sharee Zedek Medical Center, Jerusalem, Israel. reissman@szmc.org.il. 7. Digestive Disease Institute, Shaare-Zedek Medical Center, The Hebrew University School of Medicine, P.O. Box 3235, 91031, Jerusalem, Israel. reissman@szmc.org.il.
Abstract
BACKGROUND: The potential role and benefits of AI in surgery has yet to be determined. This study is a first step in developing an AI system for minimizing adverse events and improving patient's safety. We developed an Artificial Intelligence (AI) algorithm and evaluated its performance in recognizing surgical phases of laparoscopic cholecystectomy (LC) videos spanning a range of complexities. METHODS: A set of 371 LC videos with various complexity levels and containing adverse events was collected from five hospitals. Two expert surgeons segmented each video into 10 phases including Calot's triangle dissection and clipping and cutting. For each video, adverse events were also annotated when present (major bleeding; gallbladder perforation; major bile leakage; and incidental finding) and complexity level (on a scale of 1-5) was also recorded. The dataset was then split in an 80:20 ratio (294 and 77 videos), stratified by complexity, hospital, and adverse events to train and test the AI model, respectively. The AI-surgeon agreement was then compared to the agreement between surgeons. RESULTS: The mean accuracy of the AI model for surgical phase recognition was 89% [95% CI 87.1%, 90.6%], comparable to the mean inter-annotator agreement of 90% [95% CI 89.4%, 90.5%]. The model's accuracy was inversely associated with procedure complexity, decreasing from 92% (complexity level 1) to 88% (complexity level 3) to 81% (complexity level 5). CONCLUSION: The AI model successfully identified surgical phases in both simple and complex LC procedures. Further validation and system training is warranted to evaluate its potential applications such as to increase patient safety during surgery.
BACKGROUND: The potential role and benefits of AI in surgery has yet to be determined. This study is a first step in developing an AI system for minimizing adverse events and improving patient's safety. We developed an Artificial Intelligence (AI) algorithm and evaluated its performance in recognizing surgical phases of laparoscopic cholecystectomy (LC) videos spanning a range of complexities. METHODS: A set of 371 LC videos with various complexity levels and containing adverse events was collected from five hospitals. Two expert surgeons segmented each video into 10 phases including Calot's triangle dissection and clipping and cutting. For each video, adverse events were also annotated when present (major bleeding; gallbladder perforation; major bile leakage; and incidental finding) and complexity level (on a scale of 1-5) was also recorded. The dataset was then split in an 80:20 ratio (294 and 77 videos), stratified by complexity, hospital, and adverse events to train and test the AI model, respectively. The AI-surgeon agreement was then compared to the agreement between surgeons. RESULTS: The mean accuracy of the AI model for surgical phase recognition was 89% [95% CI 87.1%, 90.6%], comparable to the mean inter-annotator agreement of 90% [95% CI 89.4%, 90.5%]. The model's accuracy was inversely associated with procedure complexity, decreasing from 92% (complexity level 1) to 88% (complexity level 3) to 81% (complexity level 5). CONCLUSION: The AI model successfully identified surgical phases in both simple and complex LC procedures. Further validation and system training is warranted to evaluate its potential applications such as to increase patient safety during surgery.
Authors: Lawrence W Way; Lygia Stewart; Walter Gantert; Kingsway Liu; Crystine M Lee; Karen Whang; John G Hunter Journal: Ann Surg Date: 2003-04 Impact factor: 12.969
Authors: Pietro Mascagni; Armine Vardazaryan; Deepak Alapatt; Takeshi Urade; Taha Emre; Claudio Fiorillo; Patrick Pessaux; Didier Mutter; Jacques Marescaux; Guido Costamagna; Bernard Dallemagne; Nicolas Padoy Journal: Ann Surg Date: 2020-11-16 Impact factor: 13.787