A Multisite Fusion‐Based Deep Convolutional Neural Network for Classification of Helicobacter pylori Infection Status Using Endoscopic Images: A Multicenter Study
Journal of Gastroenterology and Hepatology · 2025-09 · 40 (9) · 2240-2247
DOI: 10.1111/jgh.70004 ↗Kurz gefasst
Multizentrische Studie in der Zeitschrift Journal of Gastroenterology and Hepatology, online erschienen am 19.07.2025. Verfasst von Duwei Dai und 17 weiteren Autor:innen. Beteiligte Einrichtung: Department of Gastroenterology, The Second Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China. Hauptthemen laut MEDLINE: Helicobacter Infections, Helicobacter pylori, Neural Networks, Computer, Deep Learning. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 21 Literaturangaben. Gefördert von: Key Research and Development Projects of Shaanxi Province, IIT Clinical Research Fund of the Second Affiliated Hospital of Xi'an Jiaotong University.
Abstract (Originalsprache)
Background and Aim We aimed to develop a deep convolutional neural network (DCNN) that integrates features from multiple sites of the stomach to classify Hp infection status, distinguishing between uninfected, previously infected, and currently infected.
Methods
Ten deep learning architectures were employed to develop DCNN models using a training dataset comprising 3380 white‐light images collected from 676 subjects across eight centers. External validation was conducted with a separate dataset consisting of images from 126 individuals. External testing was subsequently performed to assess and compare the diagnostic efficacy between single‐site and multisite fusion DCNN models.
Results
Among these models, the DCNN model using Wide‐ResNet emerged as the top performer, achieving a high accuracy of 68.11% (95% confidence interval [CI]: 63.36%–73.09%) with an area under the curve (AUC) of 75.06% (95% CI: 70.22%–80.24%) for noninfection, 69.18% (95% CI: 64.51%–74.03%) for past infection, and 77.04% (95% CI: 72.12%–82.39%) for current infection using images from a single site on the lesser gastric curvature. In comparison, the voting‐based multisite fusion DCNN model demonstrated superior accuracy (73.83%, 95% CI: 69.12%–78.65%) and AUC (77.51%, 95% CI: 72.89%–82.59%), particularly notable for noninfection and current infection. Additionally, the DCNN model exhibited heightened sensitivity, specificity, and precision compared to experienced endoscopists.
Conclusions
The DCNN model, crafted through a voting‐based multisite fusion, displayed stellar performance, excelling in the classification of Hp infection status into uninfected and currently infected.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Gastroenterology, The Second Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China.
- Institute of Medical Artificial Intelligence, The Second Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China.
- Department of Gastroenterology, Xi'an Central Hospital, College of Medicine, Xi'an Jiaotong University, Xi'an, China.
- Department of Gastroenterology, Xi'an Gem Flower Changqing Hospital, Xi'an, China.
- Department of Gastroenterology, Weinan Central Hospital, Weinan, China.
- Department of Gastroenterology, Ankang Central Hospital, Ankang, China.
ORCID-Profile: Yueqin Zheng
Förderung
- Key Research and Development Projects of Shaanxi Province
- IIT Clinical Research Fund of the Second Affiliated Hospital of Xi'an Jiaotong University
Zitieren
APA
Dai, D., Quan, X., Zheng, Y., Wang, Y., Hao, Q., Lei, Y., Gao, Y., Liang, J., Zhang, H., Huang, Y., Chen, C., Wang, J., Zhang, J., Wu, J., Zou, B., Li, L., Shi, H., & Qin, B. (2025). A Multisite Fusion‐Based Deep Convolutional Neural Network for Classification of Helicobacter pylori Infection Status Using Endoscopic Images: A Multicenter Study. Journal of Gastroenterology and Hepatology, 40(9), 2240-2247. https://doi.org/10.1111/jgh.70004
Vancouver
Dai D, Quan X, Zheng Y, Wang Y, Hao Q, Lei Y, et al. A Multisite Fusion‐Based Deep Convolutional Neural Network for Classification of Helicobacter pylori Infection Status Using Endoscopic Images: A Multicenter Study. Journal of Gastroenterology and Hepatology. 2025;40(9):2240-2247. doi:10.1111/jgh.70004
BibTeX
@article{Dai2025,
author = {Dai, Duwei and Quan, Xiaojing and Zheng, Yueqin and Wang, Yu and Hao, Quan and Lei, Yali and Gao, Yuan and Liang, Jianqun and Zhang, Hanhua and Huang, Yun and Chen, Changxin and Wang, Jia and Zhang, Jiantao and Wu, Jie and Zou, Baicang and Li, Lu and Shi, Haitao and Qin, Bin},
title = {A Multisite Fusion‐Based Deep Convolutional Neural Network for Classification of Helicobacter pylori Infection Status Using Endoscopic Images: A Multicenter Study},
journal = {Journal of Gastroenterology and Hepatology},
year = {2025},
volume = {40},
number = {9},
pages = {2240--2247},
doi = {10.1111/jgh.70004}
}
EUVIMED