Pietro Mascagni1,2, Deepak Alapatt3, Giovanni Guglielmo Laracca4, Ludovica Guerriero5, Andrea Spota6, Claudio Fiorillo7, Armine Vardazaryan3, Giuseppe Quero7, Sergio Alfieri7, Ludovica Baldari8, Elisa Cassinotti8, Luigi Boni8, Diego Cuccurullo5, Guido Costamagna7, Bernard Dallemagne9,10, Nicolas Padoy3,10. 1. ICube, University of Strasbourg, CNRS, c/o IHU-Strasbourg, 1, place de l'hôpital, 67000, Strasbourg, France. p.mascagni@unistra.fr. 2. Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy. p.mascagni@unistra.fr. 3. ICube, University of Strasbourg, CNRS, c/o IHU-Strasbourg, 1, place de l'hôpital, 67000, Strasbourg, France. 4. Department of Medical Surgical Science and Translational Medicine, Sant'Andrea Hospital, Sapienza University of Rome, Rome, Italy. 5. Department of Laparoscopic and Robotic General Surgery, Monaldi Hospital, AORN dei Colli, Naples, Italy. 6. Scuola di Specializzazione in Chirurgia Generale, University of Milan, Milan, Italy. 7. Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy. 8. Department of Surgery, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico di Milano, University of Milan, Milan, Italy. 9. Institute for Research Against Digestive Cancer (IRCAD), Strasbourg, France. 10. IHU-Strasbourg, Institute of Image-Guided Surgery, Strasbourg, France.
Abstract
BACKGROUND: A computer vision (CV) platform named EndoDigest was recently developed to facilitate the use of surgical videos. Specifically, EndoDigest automatically provides short video clips to effectively document the critical view of safety (CVS) in laparoscopic cholecystectomy (LC). The aim of the present study is to validate EndoDigest on a multicentric dataset of LC videos. METHODS: LC videos from 4 centers were manually annotated with the time of the cystic duct division and an assessment of CVS criteria. Incomplete recordings, bailout procedures and procedures with an intraoperative cholangiogram were excluded. EndoDigest leveraged predictions of deep learning models for workflow analysis in a rule-based inference system designed to estimate the time of the cystic duct division. Performance was assessed by computing the error in estimating the manually annotated time of the cystic duct division. To provide concise video documentation of CVS, EndoDigest extracted video clips showing the 2 min preceding and the 30 s following the predicted cystic duct division. The relevance of the documentation was evaluated by assessing CVS in automatically extracted 2.5-min-long video clips. RESULTS: 144 of the 174 LC videos from 4 centers were analyzed. EndoDigest located the time of the cystic duct division with a mean error of 124.0 ± 270.6 s despite the use of fluorescent cholangiography in 27 procedures and great variations in surgical workflows across centers. The surgical evaluation found that 108 (75.0%) of the automatically extracted short video clips documented CVS effectively. CONCLUSIONS: EndoDigest was robust enough to reliably locate the time of the cystic duct division and efficiently video document CVS despite the highly variable workflows. Training specifically on data from each center could improve results; however, this multicentric validation shows the potential for clinical translation of this surgical data science tool to efficiently document surgical safety.
BACKGROUND: A computer vision (CV) platform named EndoDigest was recently developed to facilitate the use of surgical videos. Specifically, EndoDigest automatically provides short video clips to effectively document the critical view of safety (CVS) in laparoscopic cholecystectomy (LC). The aim of the present study is to validate EndoDigest on a multicentric dataset of LC videos. METHODS: LC videos from 4 centers were manually annotated with the time of the cystic duct division and an assessment of CVS criteria. Incomplete recordings, bailout procedures and procedures with an intraoperative cholangiogram were excluded. EndoDigest leveraged predictions of deep learning models for workflow analysis in a rule-based inference system designed to estimate the time of the cystic duct division. Performance was assessed by computing the error in estimating the manually annotated time of the cystic duct division. To provide concise video documentation of CVS, EndoDigest extracted video clips showing the 2 min preceding and the 30 s following the predicted cystic duct division. The relevance of the documentation was evaluated by assessing CVS in automatically extracted 2.5-min-long video clips. RESULTS: 144 of the 174 LC videos from 4 centers were analyzed. EndoDigest located the time of the cystic duct division with a mean error of 124.0 ± 270.6 s despite the use of fluorescent cholangiography in 27 procedures and great variations in surgical workflows across centers. The surgical evaluation found that 108 (75.0%) of the automatically extracted short video clips documented CVS effectively. CONCLUSIONS: EndoDigest was robust enough to reliably locate the time of the cystic duct division and efficiently video document CVS despite the highly variable workflows. Training specifically on data from each center could improve results; however, this multicentric validation shows the potential for clinical translation of this surgical data science tool to efficiently document surgical safety.
Authors: John D Birkmeyer; Jonathan F Finks; Amanda O'Reilly; Mary Oerline; Arthur M Carlin; Andre R Nunn; Justin Dimick; Mousumi Banerjee; Nancy J O Birkmeyer Journal: N Engl J Med Date: 2013-10-10 Impact factor: 91.245
Authors: Thomas M Ward; Pietro Mascagni; Yutong Ban; Guy Rosman; Nicolas Padoy; Ozanan Meireles; Daniel A Hashimoto Journal: Surgery Date: 2020-12-01 Impact factor: 3.982
Authors: Yue-Yung Hu; Laura M Mazer; Steven J Yule; Alexander F Arriaga; Caprice C Greenberg; Stuart R Lipsitz; Atul A Gawande; Douglas S Smink Journal: JAMA Surg Date: 2017-04-01 Impact factor: 14.766
Authors: Andru P Twinanda; Sherif Shehata; Didier Mutter; Jacques Marescaux; Michel de Mathelin; Nicolas Padoy Journal: IEEE Trans Med Imaging Date: 2016-07-22 Impact factor: 10.048
Authors: Thomas M Ward; Pietro Mascagni; Amin Madani; Nicolas Padoy; Silvana Perretta; Daniel A Hashimoto Journal: J Surg Oncol Date: 2021-08 Impact factor: 3.454
Authors: Rebecca E Scully; Shanley B Deal; Michael J Clark; Katherine Yang; Greg Wnuk; Douglas S Smink; Jonathan P Fryer; Jordan D Bohnen; Ezra N Teitelbaum; Shari L Meyerson; Andreas H Meier; Paul G Gauger; Rishindra M Reddy; Daniel E Kendrick; Michael Stern; David T Hughes; Jeffrey G Chipman; Jitesh A Patel; Adnan Alseidi; Brian C George Journal: J Surg Educ Date: 2020-03-20 Impact factor: 2.891
Authors: Daniel A Hashimoto; Guy Rosman; Elan R Witkowski; Caitlin Stafford; Allison J Navarette-Welton; David W Rattner; Keith D Lillemoe; Daniela L Rus; Ozanan R Meireles Journal: Ann Surg Date: 2019-09 Impact factor: 12.969