Endoscopist sentiments on artificial intelligence in endoscopy: A Large-scale intercontinental opinion survey
Endoscopy, vol. 56, pp. S353
Abstract
Aims In recent years, there has been rapid development of AI devices in endoscopy. Little is however known about endoscopist perception regarding AI in clinical practice. The aim of this study was to undertake an international endoscopist opinion survey on behalf of the World Endoscopy Organization AI committee about the use of artificial intelligence. Methods A prospective questionnaire study was conducted as part of the World Endoscopy Organization AI committee activities. The committee developed 23 statements on the use of AI in endoscopy which were distributed to endoscopists using an online survey platform. Endoscopists responded to each of the statements by using a 5-point Likert-scale, ranging from strongly disagree (1) to strongly agree (5). Results In total, 465 endoscopists completed the survey (from 83 countries). The endoscopists background was general (36%), specialist interventional (55%) and trainee (9%). 39% had experience of using AI in clinical practice. The majority believed that AI would improve endoscopy quality (90%), efficiency (87%) and patient outcomes (86%). Most thought that CADe would help detect clinically relevant polyps (86%) and upper GI neoplasia (87%). 68% felt more comfortable leaving in a diminutive rectosigmoid polyp with CADx assistance. Financial re-imbursement was considered important (72%) and most were concerned about medical liability (47%), with a mixed view on accountability (endoscopist 57%, manufacturer 22% and hospital 8%). Potential de-skilling was a concern (37%). A small majority felt that AI would lead to unnecessary resections and biopsies in the upper and lower GI tract, and also lengthen procedure times. Overall, the majority would use upper GI CADe (90%), lower GI CADe (83%) and CADx (87%) when available. Conclusions In this international endoscopist survey, participants strongly believed that AI would improve endoscopy quality and outcomes. Financial re-imbursement, concerns about medical liability and potential de-skilling are important issues. A small majority felt that AI would lead to unnecessary resections and increase procedure times. Overall, the vast majority would use AI when available. Publication History Article published online: 15 April 2024 © 2024. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Rüdigerstraße 14, 70469 Stuttgart, Germany
Authors 14
-
Amsterdam University Medical Centers · Amsterdam UMC Location Vrije Universiteit Amsterdam
Affiliation as printed
Amsterdam UMC, locatie VUmc, Amsterdam, Netherlands
-
Affiliation as printed
Division of Surgery and Interventional Sciences, University College London (UCL), London, United Kingdom
-
Megan M.L. Engels Aachen
Leiden University Medical Center
Affiliation as printed
Leiden University Medical Center (LUMC), Leiden, Netherlands
-
Amsterdam UMC Location Vrije Universiteit Amsterdam
Affiliation as printed
VU University Medical Center, Amsterdam, Netherlands
-
Affiliation as printed
Mayo Clinic, Rochester, United States of America
-
Renmin Hospital of Wuhan University
Affiliation as printed
Renmin Hospital of Wuhan University, Wuhan, China
-
Affiliation as printed
Tenwek Hospital, Bomet, Kenya
-
Affiliation as printed
Swedish Medical Center, Seattle, United States of America
-
Affiliation as printed
General Hospital Lahore, Lahore, Pakistan
-
University Medical Center of the Johannes Gutenberg University Mainz
Affiliation as printed
University Medical Center Mainz, Interventional Endoscopy Center, Mainz, Germany
-
Affiliation as printed
West Virginia University, Morgantown, United States of America
-
Affiliation as printed
Vancouver General Hospital, Vancouver, Canada
-
Amsterdam UMC Location Vrije Universiteit Amsterdam
Affiliation as printed
VU University Medical Center, Amsterdam, Netherlands
-
Affiliation as printed
Jichi Medical University, Shimotsuke, Japan
Cited by 0 stored of 0
No patents citing this paper on Lens.org (checked 2026-10-11).