Clinical Decision Support System for All Stages of Gastric Carcinogenesis in Real-Time Endoscopy: Model Establishment and Validation Study
Open accessPublished version of a preprint
Abstract
Background: Our research group previously established a deep-learning–based clinical decision support system (CDSS) for real-time endoscopy-based detection and classification of gastric neoplasms. However, preneoplastic conditions, such as atrophy and intestinal metaplasia (IM) were not taken into account, and there is no established model that classifies all stages of gastric carcinogenesis. Objective: This study aims to build and validate a CDSS for real-time endoscopy for all stages of gastric carcinogenesis, including atrophy and IM. Methods: A total of 11,868 endoscopic images were used for training and internal testing. The primary outcomes were lesion classification accuracy (6 classes: advanced gastric cancer, early gastric cancer, dysplasia, atrophy, IM, and normal) and atrophy and IM lesion segmentation rates for the segmentation model. The following tests were carried out to validate the performance of lesion classification accuracy: (1) external testing using 1282 images from another institution and (2) evaluation of the classification accuracy of atrophy and IM in real-world procedures in a prospective manner. To estimate the clinical utility, 2 experienced endoscopists were invited to perform a blind test with the same data set. A CDSS was constructed by combining the established 6-class lesion classification model and the preneoplastic lesion segmentation model with the previously established lesion detection model. Results: The overall lesion classification accuracy (95% CI) was 90.3% (89%-91.6%) in the internal test. For the performance validation, the CDSS achieved 85.3% (83.4%-97.2%) overall accuracy. The per-class external test accuracies for atrophy and IM were 95.3% (92.6%-98%) and 89.3% (85.4%-93.2%), respectively. CDSS-assisted endoscopy showed an accuracy of 92.1% (88.8%-95.4%) for atrophy and 95.5% (92%-99%) for IM in the real-world application of 522 consecutive screening endoscopies. There was no significant difference in the overall accuracy between the invited endoscopists and established CDSS in the prospective real-clinic evaluation (P=.23). The CDSS demonstrated a segmentation rate of 93.4% (95% CI 92.4%-94.4%) for atrophy or IM lesion segmentation in the internal testing. Conclusions: The CDSS achieved high performance in terms of computer-aided diagnosis of all stages of gastric carcinogenesis and demonstrated real-world application potential.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Some CDSS can generate ranked lists of potential diagnoses for disease differentiation, including disease subtype classification or stages, to guide subsequent treatment planning.
“Some CDSS can further generate ranked lists of potential diagnoses for disease differentiation [219,220,221], including disease subtype classification or stages, to guide subsequent treatment planning [222].”
From Toward Safer Diagnoses: A SEIPS-Based Narrative Review of Diagnostic Errors (2026) · cites this paper as [222]
The Discussion (Principal Findings; page not provided) says the CDSS classifies stages of gastric carcinogenesis, and the Methods (Development of the CDSS; page not provided) describes its six diagnostic classes. Those passages support the stage-classification part attributed to this source, but do not say that its classifications guide subsequent treatment planning. The ranked-list portion of the sentence is attached to other citations. (LLM, conf=0.84)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 27 · Cited by 14
Research works printed but not linked: 21
- [1]Correa P Piazuelo MB The gastric precancerous cascade J Dig Dis 2012 01 13 1 2 9 10.1111/j.1751-2980.2011.00550.x 22188910 22188910 PMC3404600 · DOI
- [2]Bang CS Baik GH Attempts to enhance the eradication rate of Helicobacter pylori infection World J Gastroenterol 2014 05 14 20 18 5252 5262 10.3748/wjg.v20.i18.5252 24833855 24833855 PMC4017040 · DOI
- [3]Malfertheiner P Megraud F Rokkas T Gisbert JP Liou J Schulz C Gasbarrini A Hunt RH Leja M O'Morain C Rugge M Suerbaum S Tilg H Sugano K El-Omar EM Management of infection: the Maastricht VI/Florence consensus report Gut 2022 08 08 1724 1762 10.1136/gutjnl-2022-327745 35944925 gutjnl-2022-327745 35944925 · DOI
- [4]Bang CS Baik GH Shin IS Kim JB Suk KT Yoon JH Kim YS Kim DJ Helicobacter pylori eradication for prevention of metachronous recurrence after endoscopic resection of early gastric cancer J Korean Med Sci 2015 06 30 6 749 756 10.3346/jkms.2015.30.6.749 26028928 26028928 PMC4444476 · DOI
- [7]Kang SJ Kim JG Moon HS Kook M Lee JY Bang CS Tae CH Gong EJ Nam SY Kim HJ Clinical practice guideline for gastritis in Korea J Korean Med Sci 2023 04 03 38 13 e115 10.3346/jkms.2023.38.e115 37012690 38.e115 37012690 PMC10070048 · DOI
- [8]Cho B Bang CS Park SW Yang YJ Seo SI Lim H Shin WG Hong JT Yoo YT Hong SH Choi JH Lee JJ Baik GH Automated classification of gastric neoplasms in endoscopic images using a convolutional neural network Endoscopy 2019 12 51 12 1121 1129 10.1055/a-0981-6133 31443108 31443108 · DOI
- [9]Cho B Bang CS Lee JJ Seo CW Kim JH Prediction of submucosal invasion for gastric neoplasms in endoscopic images using deep-L=learning J Clin Med 2020 06 15 9 6 1858 10.3390/jcm9061858 32549190 jcm9061858 32549190 PMC7356204 · DOI
- [10]Bang CS Lim H Jeong HM Hwang SH Use of endoscopic images in the prediction of submucosal invasion of gastric neoplasms: automated deep learning model development and usability study J Med Internet Res 2021 04 15 23 4 e25167 10.2196/25167 33856356 v23i4e25167 33856356 PMC8085753 · DOI
- [11]Gong EJ Bang CS Lee JJ Baik GH Lim H Jeong JH Choi SW Cho J Kim DY Lee KB Shin S Sigmund D Moon BI Park SC Lee SH Bang KB Son D Deep learning-based clinical decision support system for gastric neoplasms in real-time endoscopy: development and validation study Endoscopy 2023 08 55 8 701 708 10.1055/a-2031-0691 36754065 36754065 · DOI · Paper page
- [12]Cho B Bang CS Artificial intelligence for the determination of a management strategy for diminutive colorectal polyps: hype, hope, or help Am J Gastroenterol 2020 01 115 1 70 72 10.14309/ajg.0000000000000476 31770118 00000434-202001000-00013 31770118 · DOI
- [13]Berzin TM Parasa S Wallace MB Gross SA Repici A Sharma P Position statement on priorities for artificial intelligence in GI endoscopy: a report by the ASGE Task Force Gastrointest Endosc 2020 10 92 4 951 959 10.1016/j.gie.2020.06.035 32565188 S0016-5107(20)34460-6 32565188 · DOI · Paper page
- [14]Bang CS Deep learning in upper gastrointestinal disorders: status and future perspectives Korean J Gastroenterol 2020 03 25 75 3 120 131 10.4166/kjg.2020.75.3.120 32209800 kjg.2020.75.3.120 32209800 · DOI
- [15]Bang CS Artificial intelligence in the analysis of upper gastrointestinal disorders Korean J Helicobacter Up Gastrointest Res 2021 12 21 4 300 310 10.7704/kjhugr.2021.0030 · DOI
- [17]Luo W Phung D Tran T Gupta S Rana S Karmakar C Shilton A Yearwood J Dimitrova N Ho TB Venkatesh S Berk M Guidelines for developing and reporting machine learning predictive models in biomedical research: a multidisciplinary view J Med Internet Res 2016 12 16 18 12 e323 10.2196/jmir.5870 27986644 v18i12e323 27986644 PMC5238707 · DOI · Paper page
- [18]Zhou Z Siddiquee MR Tajbakhsh N Liang J UNet++: redesigning skip connections to exploit multiscale features in image segmentation IEEE Trans Med Imaging 2020 06 39 6 1856 1867 10.1109/TMI.2019.2959609 31841402 31841402 PMC7357299 · DOI
- [21]Redmon J Farhadi A YOLOv3: an incremental improvement arXiv. Preprint posted online on April 8, 2018 2021
- [23]Guimarães P Keller A Fehlmann T Lammert F Casper M Deep-learning based detection of gastric precancerous conditions Gut 2020 01 69 1 4 6 10.1136/gutjnl-2019-319347 31375599 gutjnl-2019-319347 31375599 · DOI
- [24]Zhang Y Li F Yuan F Zhang K Huo L Dong Z Lang Y Zhang Y Wang M Gao Z Qin Z Shen L Diagnosing chronic atrophic gastritis by gastroscopy using artificial intelligence Dig Liver Dis 2020 05 52 5 566 572 10.1016/j.dld.2019.12.146 32061504 S1590-8658(20)30003-7 32061504 · DOI
- [25]Zhao Q Chi T Deep learning model can improve the diagnosis rate of endoscopic chronic atrophic gastritis: a prospective cohort study BMC Gastroenterol 2022 03 23 22 1 133 10.1186/s12876-022-02212-1 35321641 10.1186/s12876-022-02212-1 35321641 PMC8941797 · DOI
- [26]Luo J Cao S Ding N Liao X Peng L Xu C A deep learning method to assist with chronic atrophic gastritis diagnosis using white light images Dig Liver Dis 2022 11 54 11 1513 1519 10.1016/j.dld.2022.04.025 35610166 S1590-8658(22)00474-1 35610166 · DOI
- [27]Siripoppohn V Pittayanon R Tiankanon K Faknak N Sanpavat A Klaikaew N Vateekul P Rerknimitr R Real-time semantic segmentation of gastric intestinal metaplasia using a deep learning approach Clin Endosc 2022 05 55 3 390 400 10.5946/ce.2022.005 35534933 ce.2022.005 35534933 PMC9178134 · DOI
- Authors
- Gong, E. J., Bang, C. S., Lee, J. J., Jeong, H. M., Baik, G. H., Jeong, J. H., Dick, S., & Lee, G. H.
- Affiliations
- Hallym University (KR)
- Corresponding author
- Bang, Chang Seok
- ORCID
- Venue
- Journal of Medical Internet Research
- Publisher
- JMIR Publications Inc.
- Published
- 2023-10-30
- Volume
- 25
- Pages
- e50448
- ISSN
- 1438-8871 (ISSN-L)
- PMC
- PMC10644184
- OpenAlex
- W4388013968
- Cited by
- 14
- References
- 27
- Citations by year
- 2026: 7 · 2025: 3 · 2024: 4
- FWCI
- 1.32
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- Journal of Medical Internet Research (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- National Research Foundation (RS-2023-00223501)
- Ministry of Science and ICT, South Korea (RS-2023 00223501, RS-2023)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
- Health Sciences › Medicine › Pulmonary and Respiratory Medicine › Gastric Cancer Management and Outcomes
- Health Sciences › Medicine › Oncology › Colorectal Cancer Screening and Detection
- Keywords
- clinical decision support systems, gastric carcinogenesis, atrophic gastritis, intestinal metaplasia, lesion segmentation, deep learning, early gastric cancer, dysplasia, endoscopic screening, precancerous lesions, model validation, external validation, clinical utility
- MeSH terms
- Atrophy, Humans, Metaplasia, Prospective Studies, Stomach Neoplasms, Endoscopy, Gastrointestinal, Decision Support Systems, Clinical
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
