Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Multicenter Study
. 2024 Oct;100(4):616-625.e8.
doi: 10.1016/j.gie.2024.04.013. Epub 2024 Apr 16.

Computer-aided diagnosis improves characterization of Barrett's neoplasia by general endoscopists (with video)

Jelmer B Jukema  1 , Carolus H J Kusters  2 , Martijn R Jong  1 , Kiki N Fockens  1 , Tim Boers  2 , Joost A van der Putten  2 , Roos E Pouw  1 , Lucas C Duits  1 , Bas L A M Weusten  3 , Lorenza Alvarez Herrero  4 , Martin H M G Houben  5 , Wouter B Nagengast  6 , Jessie Westerhof  6 , Alaa Alkhalaf  7 , Rosalie Mallant-Hent  8 , Pieter Scholten  9 , Krish Ragunath  10 , Stefan Seewald  11 , Peter Elbe  12 , Francisco Baldaque Silva  13 , Maximilien Barret  14 , Jacobo Ortiz Fernández-Sordo  15 , Guiomar Moral Villarejo  15 , Oliver Pech  16 , Torsten Beyna  17 , Nahid S M Montazeri  18 , Fons van der Sommen  2 , Peter H de With  2 , A Jeroen de Groof  1 , Jacques J Bergman  19 ; BONS-AI consortium
Collaborators, Affiliations
Free article
Multicenter Study

Computer-aided diagnosis improves characterization of Barrett's neoplasia by general endoscopists (with video)

Jelmer B Jukema et al. Gastrointest Endosc. 2024 Oct.
Free article

Abstract

Background and aims: Characterization of visible abnormalities in patients with Barrett's esophagus (BE) can be challenging, especially for inexperienced endoscopists. This results in suboptimal diagnostic accuracy and poor interobserver agreement. Computer-aided diagnosis (CADx) systems may assist endoscopists. We aimed to develop, validate, and benchmark a CADx system for BE neoplasia.

Methods: The CADx system received pretraining with ImageNet and then consecutive domain-specific pretraining with GastroNet, which includes 5 million endoscopic images. It was subsequently trained and internally validated using 1758 narrow-band imaging (NBI) images of early BE neoplasia (352 patients) and 1838 NBI images of nondysplastic BE (173 patients) from 8 international centers. CADx was tested prospectively on corresponding image and video test sets with 30 cases (20 patients) of BE neoplasia and 60 cases (31 patients) of nondysplastic BE. The test set was benchmarked by 44 general endoscopists in 2 phases (phase 1, no CADx assistance; phase 2, with CADx assistance). Ten international BE experts provided additional benchmark performance.

Results: Stand-alone sensitivity and specificity of the CADx system were 100% and 98% for images and 93% and 96% for videos, respectively. CADx outperformed general endoscopists without CADx assistance in terms of sensitivity (P = .04). Sensitivity and specificity of general endoscopists increased from 84% to 96% and 90% to 98% with CAD assistance (P < .001). CADx assistance increased endoscopists' confidence in characterization (P < .001). CADx performance was similar to that of the BE experts.

Conclusions: CADx assistance significantly increased characterization performance of BE neoplasia by general endoscopists to the level of expert endoscopists. The use of this CADx system may thereby improve daily Barrett surveillance.

PubMed Disclaimer

Conflict of interest statement

Disclosure The following authors disclosed financial relationships: M. Barret: research funding from Pentax Medical; consulting fees from Medtronic; medical training honorarium from Olympus; and board membership with Norgine and Ambu. J. J. Bergman: research support from Olympus Tokyo; and support for institutional review board (IRB)-approved research from Medtronic. T. Beyna: paid consultancy and lecture fees from Olympus and MicroTech; paid consultancy from Boston Scientific; and lecture fees from Fujifilm, Pentax, and Erbe. A. J. de Groof: research support from Olympus Tokyo. R. Mallant-Hent: consulting fees from Janssen. O. Pech: speaker honorarium from Boston Scientific, Medtronic, Fujifilm, Olympus, Aohua, Falk, and BMS. R. Pouw: consulting fees from Medtronic and MicroTech; and speaker fees from Pentax Medical. K. Ragunath: consulting fees from Olympus and Boston Scientific; research support from Olympus. F. van der Sommen: research support from Olympus Tokyo. P. de With: research support from Olympus Tokyo. B. Weusten: financial support for IRB-approved studies, speaker fees, and consulting fees from Pentax Medical; research support from Aqua Medical; financial support for IRB-approved studies from the St. Antonius research fund; and chair of the European Society of Gastrointestinal Endoscopy working group for the revision of the BE guideline. All other authors disclosed no financial relationships. This study received financial and logistical support from Olympus (Tokyo, Japan). Olympus did not have an active role in the design of the study, analysis of the results, or the writing of the manuscript.

LinkOut - more resources