Deep learning-based quantitative histopathology of endoscopic biopsies in Crohn’s disease: a retrospective cross-sectional validation study - Takeaways - MDSpire

Deep learning-based quantitative histopathology of endoscopic biopsies in Crohn’s disease: a retrospective cross-sectional validation study

  • By

  • Xin Jiang

  • Pan Li

  • Zhaojing Chen

  • Hui Yao

  • Hao Jia

  • Taiping Wang

  • Xuefeng Tang

  • June 5, 2026

  • 0 min

Share

  • 1

    A deep learning framework was developed for quantitative assessment of endoscopic biopsies in Crohn's disease, addressing histopathologic heterogeneity.

  • 2

    The study analyzed 3,641 biopsy slides from 687 patients, achieving high performance metrics with Dice coefficients and AUC values exceeding 0.95.

  • 3

    AI-derived crypt abscess and cryptitis ratios showed good concordance with pathologist assessments, though some cell-level outputs had weaker agreement.

  • 4

    Associations between AI-derived features and clinical indicators were generally weak to moderate, suggesting exploratory findings rather than definitive conclusions.

  • 5

    The AI framework offers objective quantification of CD histopathology, potentially complementing routine pathologic assessment and standardizing evaluations.

Original Source(s)

Deep learning-based quantitative histopathology of endoscopic biopsies in Crohn’s disease: a retrospective cross-sectional validation study

  • by Xin Jiang, Pan Li, Zhaojing Chen, Hui Yao, Hao Jia, Taiping Wang, Xuefeng Tang

  • June 5, 2026

Related Content

The role of the complement system in gastrointestinal-related diseases

Case Report: Bilateral panuveitis with serous ciliary body and choroidal detachment associated with ulcerative colitis

Editorial: Community series in towards precision medicine for immune-mediated disorders: advances in using big data and artificial intelligence to understand heterogeneity in inflammatory responses, volume III