Limited discriminative performance of endoscopic deep learning for Helicobacter pylori status assessment in gastric cancer patients: a retrospective study - Takeaways - MDSpire

Limited discriminative performance of endoscopic deep learning for Helicobacter pylori status assessment in gastric cancer patients: a retrospective study

  • By

  • Wenyi Zhou

  • Yixing Wang

  • Yuhong Wang

  • Yongluo Jiang

  • Wencan He

  • Binbin Xu

  • July 15, 2026

Share

  • 1

    The study evaluated endoscopic image-based classification models for Helicobacter pylori status in gastric cancer patients.

  • 2

    A total of 602 endoscopic images from 337 gastric cancer patients were initially collected, with 576 images retained for analysis.

  • 3

    The best-performing model achieved a median test AUC of 0.6255, indicating limited discrimination ability for H. pylori status.

  • 4

    Sensitivity was 0.7308, specificity was 0.3714, and accuracy was 0.6092, highlighting challenges in image-based assessment.

  • 5

    The findings suggest that assessing H. pylori status from endoscopic images in gastric cancer patients remains challenging.

Original Source(s)

Limited discriminative performance of endoscopic deep learning for Helicobacter pylori status assessment in gastric cancer patients: a retrospective study

  • by Wenyi Zhou, Yixing Wang, Yuhong Wang, Yongluo Jiang, Wencan He, Binbin Xu

  • July 15, 2026

Related Content

Efficacy and safety of bupleurum-containing formulas combined with chemotherapy for triple-negative breast cancer: a systematic review and meta-analysis

Age-stratified diagnostic performance of the ovarian-adnexal reporting and data system for adnexal masses: a focus on school-age children, early, and middle adolescents

Case Report: Preoperative embolization for a large intracranial pheochromocytoma metastasis: a novel approach to minimize intraoperative risk