Publication Name	World Journal of Gastroenterology
Manuscript ID	121242
Country	Germany

Category	Gastroenterology & Hepatology
Manuscript Type	Retrospective Study
Article Title	Domain specific training with three-dimensional virtual colonoscopy images improves artificial intelligence based automated size prediction for large lesions
Manuscript Source	Unsolicited Manuscript
All Author List	Dominik Schulz, Alanna Ebigbo, Eva Holley, Sandra Nagl, Olga Nichiporyk, Andreas Probst, Markus W Scheppach, Christoph Römmele, David Roser, Stephan Zellmer and Helmut Messmann
Funding Agency and Grant Number
Corresponding Author	Dominik Schulz, Senior Scientist, III. Medizinische Klinik, Universitätsklinikum Augsburg, Stenglinstraße 2, Augsburg 86156, Germany. dominik.schulz@uk-augsburg.de
Key Words	Polyp size; Lesion size; Colonoscopy; Artificial intelligence; Monocular depth estimation
Core Tip	Accurate measurement of colorectal polyp size is mandatory influencing the choice of resection technique and surveillance intervals. However, conventional visual estimation by endoscopists is prone to significant error rates up to 66%. In this work, we compare three state-of-the art monocular depth estimation artificial intelligence models for automated size measurement. When applied natively to colonoscopy, they show poor accuracy in size prediction. By domain specific training using virtual colonoscopy images we could significantly improve size prediction. The fine-tuned model achieved a mean absolute percentage error of 33.9%-37.0%. This performance was comparable to trainee endoscopists but remained inferior to experts.
Citation	Schulz D, Ebigbo A, Holley E, Nagl S, Nichiporyk O, Probst A, Scheppach MW, Römmele C, Roser D, Zellmer S, Messmann H. Domain specific training with three-dimensional virtual colonoscopy images improves artificial intelligence based automated size prediction for large lesions. World J Gastroenterol 2026; In press

Received	2026-03-19 01:23
Peer-Review Started	2026-03-19 01:24
First Decision by Editorial Office Director	2026-04-13 09:34
Return for Revision	2026-04-13 09:34
Revised	2026-05-18 07:48
Publication Fee Transferred	2026-05-29 11:45
Second Decision by Editor	2026-07-03 02:40
Second Decision by Editor-in-Chief
Final Decision by Editorial Office Director	2026-07-03 06:42
Articles in Press	2026-07-03 06:42
Edit the Manuscript by Language Editor
Typeset the Manuscript

ISSN	1007-9327 (print) and 2219-2840 (online)
Open Access	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.
Copyright	©Author(s) (or their employer(s)) 2026. No commercial re-use. See Permissions. Published by Baishideng Publishing Group Inc.
Permissions	For details, please visit: http://www.wjgnet.com/bpg/gerinfo/207
Publisher	Baishideng Publishing Group Inc, 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA
Website	http://www.wjgnet.com