Copyright: ©Author(s) 2026. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial (CC BY-NC 4.0) license. No commercial re-use. See permissions. Published by Baishideng Publishing Group Inc.
World J Gastroenterol. Jul 21, 2026; 32(27): 119276
Published online Jul 21, 2026. doi: 10.3748/wjg.119276
Published online Jul 21, 2026. doi: 10.3748/wjg.119276
Artificial intelligence-based mucosa touch rate: A novel real-time quality control indicator for colonoscopy
Wen Chen, Meng Wu, Heng-Yu Wang, Hong-Bo Wu, Zhi-Hang Zhong, Lei Chen, Department of Gastroenterology, Southwest Hospital of Army Medical University, Chongqing 400038, China
Jie Li, Yu-Hao Sun, Fang Huang, Min Gao, Department of Technology Platform, Jinshan Science and Technology (Group) Co., Ltd., Chongqing 401120, China
Yan-Min Wu, Department of Internal Medicine, the 956th Hospital of the Chinese People’s Libe
Co-first authors: Wen Chen and Meng Wu.
Author contributions: Chen L conceived and designed this study; Chen W and Wu M designed this study, as they are co-first authors; Chen W, Wang HY and Wu HB collected and analyzed patient data; Chen W drafted and completed the manuscript; Li J, Sun YH, Huang F and Gao M constructed the model; Chen W, Zhong ZH and Wu YM prepared the tables and figures; all the authors have read and approved the final manuscript.
AI contribution statement: As far as we know, none of the authors directly relied on generative AI tools to create content. Before the initial submission, we had the manuscript reviewed by a professional language editing service — one that was either recommended by the journal or that follows the journal’s language polishing standards. We later became aware that this editing process might have included AI-assisted rewriting or polishing, something we did not supervise or intend to happen. Other than that, only the basic spelling checker in Microsoft Word was used. All the research ideas, data analysis, arguments, and conclusions were entirely written and developed by us. The Abstract, Introduction, Methods, Results, Discussion, and Conclusion all come from our own draft. No tool generated the scientific content for us. The authors themselves did not use AI for these tasks. A third-party language polishing service (as suggested by journal guidelines) was employed to improve readability and correct grammatical errors. We now suspect that part of its workflow might have incorporated AI-based language enhancement, which likely triggered the detection alarm. Data analysis was performed exclusively with standard statistical software (e.g., SPSS, R), without any AI involvement. The study design, the interpretation of results, and all scientific conclusions were purely our own intellectual work. No AI tool had any role in those parts. Every figure was produced from our own experimental or clinical data using ordinary scientific software. No AI-generated images were used.
Supported by the Chongqing Science and Health Joint Medical Research Project, No. 2023ZDXM007.
Institutional review board statement: This study was approved by the Committee of the First Affiliated Hospital of Army Medical University [No. (A)KY2023112].
Clinical trial registration statement: This study is registered at https://www.chictr.org.cn. The registration identification number is ChiCTR2400081562.
Informed consent statement: All study participants, or their legal guardian, provided informed written consent prior to study enrollment.
Conflict-of-interest statement: The authors declare that they have no conflict of interest.
CONSORT 2010 statement: The authors have read the CONSORT 2010 Statement, and the manuscript was prepared and revised according to the CONSORT 2010 Statement.
Data sharing statement: Technical appendix, statistical code, and dataset available from the corresponding author at xhl13228683896@tmmu.edu.cn.
Corresponding author: Lei Chen, MD, Full Professor, Department of Gastroenterology, South
Received: January 26, 2026
Revised: February 15, 2026
Accepted: March 25, 2026
Published online: July 21, 2026
Processing time: 165 Days and 13.5 Hours
Revised: February 15, 2026
Accepted: March 25, 2026
Published online: July 21, 2026
Processing time: 165 Days and 13.5 Hours
Core Tip
Core Tip: This study introduces an artificial intelligence-based mucosa touch rate (MTR) as a novel real-time quality indicator for colonoscopy. Using a deep learning model, MTR objectively quantifies mucosal contact during withdrawal. The results demonstrate a strong negative correlation between MTR and polyp detection rate (PDR). Prospective validation shows that real-time MTR feedback significantly improves PDR, particularly among less experienced endoscopists, highlighting its potential for real-time skill assessment and standardized training in colonoscopy quality control.