What question did this study set out to answer?

Determine whether a deep learning-based image analysis model can accurately diagnose celiac disease and classify related histopathologic findings from whole-slide duodenal biopsy images.

August 1, 2026Clinical Gastroenterology and HepatologyOpen Access

Use of Artificial Intelligence for Duodenal Biopsy for Celiac Disease: Developing a Prediction Model for Histopathology Diagnosis

View Full Paper

Ask AI

Bookmark

Authors

CJClaire Jansson‐KnodellCleveland Clinic ESErica SavageCleveland Clinic KPKatherine PoissantCleveland Clinic

Discussion

Member takes

Implication

Retrospective study demonstrates high diagnostic accuracy of deep learning for celiac disease in adults, highlighting potential to reduce manual pathology workloads.

Key Points

Determine whether a deep learning-based image analysis model can accurately diagnose celiac disease and classify related histopathologic findings from whole-slide duodenal biopsy images.
Analyzed 2,406 whole-slide duodenal biopsy images from >2,100 adults (≥18 years) evaluated between 2008 and 2021, excluding those on a gluten-free diet.
Categorized cases across five histological groups: normal, celiac disease, intraepithelial lymphocytes (IELs) only, IELs with positive tTG-IgA, and seronegative villous atrophy.
Trained and evaluated a clustering-constrained attention multiple-instance learning algorithm using 10-fold cross-validation.
Differentiated celiac disease from normal histology with an average area under the ROC curve (AUC) of 0.992 and an accuracy of 97%.
Achieved an overall average AUC of 0.872 and an accuracy of 74% across all five diagnostic categories.
Distinguished celiac disease from seronegative villous atrophy with an average AUC of 0.810 and an accuracy of 76%.

Cite This Study

Jansson‐Knodell et al. (2026) studied this question.

synapsesocial.com/papers/6a9e9b17ef338a591ef9a230 https://doi.org/10.1016/j.cgh.2026.07.021

View Full Paper

Ask AI

Bookmark

Implication

Retrospective study demonstrates high diagnostic accuracy of deep learning for celiac disease in adults, highlighting potential to reduce manual pathology workloads.

Use of Artificial Intelligence for Duodenal Biopsy for Celiac Disease: Developing a Prediction Model for Histopathology Diagnosis

Discussion

Member takes

Implication

Key Points

Cite This Study

Use of Artificial Intelligence for Duodenal Biopsy for Celiac Disease: Developing a Prediction Model for Histopathology Diagnosis

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider