GastritisMIL: An interpretable deep learning model for the comprehensive histological assessment of chronic gastritis
Patterns · 2025-08 · 6 (8) · 101286
DOI: 10.1016/j.patter.2025.101286 ↗Kurz gefasst
Fachartikel in der Zeitschrift Patterns, 2025-08. Verfasst von Kun Xia und 22 weiteren Autor:innen. Beteiligte Einrichtung: Department of Gastroenterology, The National Key Clinical Specialty, Zhongshan Hospital of Xiamen University, School of Medicine, Xiamen University, …. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Fujian Province, Natural Science Foundation of Xiamen Municipality.
Abstract (Originalsprache)
The comprehensive histological assessment of chronic gastritis is imperative for guiding endoscopic follow-up strategies and surveillance of early-stage gastric cancer, yet rapid and objective assessment remains challenging in clinical workflows. We propose a powerful deep learning model (GastritisMIL) to effectively identify pathological alterations on H&E-stained biopsy slides, thereby expediting pathologists' evaluation and improving decision-making regarding follow-up intervals. We have trained and tested GastritisMIL by using retrospective data from 2,744 patients and evaluated discriminative performance across three medical centers (467 patients). GastritisMIL attained areas under the receiver operating curve greater than 0.971 in four tasks (inflammation, activity, atrophy, and intestinal metaplasia) and superior performance comparable to that of two senior pathologists. Specifically, interpretable attention heatmaps generated by GastritisMIL effectively assist junior pathologists in locating suspicious lesion regions across the entire field and minimizing missed diagnosis risk. Moreover, the high generalizability of this developed model across multiple external cohorts demonstrates its potential translational value.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Gastroenterology, The National Key Clinical Specialty, Zhongshan Hospital of Xiamen University, School of Medicine, Xiamen University, Xiamen 361004, Fujian Province, P.R. China.
- Department of Computer Science, School of Informatics, Xiamen University, Xiamen 361005, Fujian Province, P.R. China.
- Clinical Research Center for Gut Microbiota and Digestive Diseases of Fujian Province, Xiamen Key Laboratory of Intestinal Microbiome and Human Health, Xiamen 361004, Fujian Province, P.R. China.
- Department of Digestive Disease, Institute for Microbial Ecology, Xiamen University, Xiamen 361004, Fujian Province, P.R. China.
- Xiamen Gastroenterology Quality Control Center, Xiamen 361004, Fujian Province, P.R. China.
- Department of Pathology, Zhongshan Hospital of Xiamen University, School of Medicine, Xiamen University, Xiamen 361004, Fujian Province, P.R. China.
Förderung
- Natural Science Foundation of Fujian Province
- Natural Science Foundation of Xiamen Municipality
Zitieren
APA
Xia, K., Hu, Y., Cai, S., Lin, M., Lu, M., Lu, H., Ye, Y., Lin, F., Gao, L., Xia, Q., Tian, R., Lin, W., Xie, L., Tan, D., Lu, Y., Lin, X., Yang, X., Zhong, L., Xu, L., …, & Xu, H. (2025). GastritisMIL: An interpretable deep learning model for the comprehensive histological assessment of chronic gastritis. Patterns, 6(8), 101286. https://doi.org/10.1016/j.patter.2025.101286
Vancouver
Xia K, Hu Y, Cai S, Lin M, Lu M, Lu H, et al. GastritisMIL: An interpretable deep learning model for the comprehensive histological assessment of chronic gastritis. Patterns. 2025;6(8):101286. doi:10.1016/j.patter.2025.101286
BibTeX
@article{Xia2025,
author = {Xia, Kun and Hu, Yihuang and Cai, Shuntian and Lin, Mengjie and Lu, Mingzhi and Lu, Huadong and Ye, Yuhan and Lin, Fenglian and Gao, Liang and Xia, Qingan and Tian, Ruihua and Lin, Weiping and Xie, Lei and Tan, Decheng and Lu, Yapi and Lin, Xunting and Yang, Xiaoning and Zhong, Lingfeng and Xu, Lei and Zhang, Zhixin and Wang, Liansheng and Ren, Jianlin and Xu, Hongzhi},
title = {GastritisMIL: An interpretable deep learning model for the comprehensive histological assessment of chronic gastritis},
journal = {Patterns},
year = {2025},
volume = {6},
number = {8},
pages = {101286},
doi = {10.1016/j.patter.2025.101286}
}
EUVIMED