Towards reliable morphological subtyping and cancer detection on Barrett's esophagus pathology through ensembling foundation models
In brief
Pathology AI ensembles detect esophageal cancer with 96% accuracy
Across 3,528 tissue slides from 790 patients, combining pathology AI models detected esophageal adenocarcinoma with 95.9% accuracy and classified Barrett's tissue subtypes with an AUROC of 0.907. Gains varied across models and were limited for dysplasia, so the results support further testing rather than clinical use; external validation is still needed.
- Journal
- NPJ digital medicine (Q1)
- Published
- 21 October 2026
- Study design
- Unclassified
- Evidence level
- Level 5, Expert Opinion (CEBM 5)
- Authors
- Azar Kazemi, Camillo Saueressig, Jingsong Liu, Han Li, Julius Shakhtour, Bianca Grosser, et al.
- PMID
- 42919834
- DOI
- 10.1038/s41746-026-03329-3
Why clinicians should know about it
- Picked for Pathology and Forensic Medicine (top studies of the week, 25 October 2026): AI benchmarking for Barrett morphology and cancer detection
Abstract
The reliability of emergent histopathology foundation models (Histo-FMs) for real-world diagnostic tasks remains unproven. Given the complexity of diagnostic tasks and the challenges inherent in real-world data, we utilized Histo-FMs to investigate their utility for morphological subtyping of Barrett's esophagus (BE) and detecting esophageal adenocarcinoma (EAC) across the BE progression spectrum. We benchmarked Histo-FMs on the BarrettNet cohort, spanning diagnostic categories ranging from normal to EAC (N2EAC). The dataset comprised 3528 hematoxylin and eosin (H&E)-stained whole-slide images (WSIs) from 790 patients. A multi-magnification, multi-backbone aggregation of the five most expert-consistent single-scale models yielded incremental performance gains (AUROC of 0.907, F1-score of 0.696, accuracy of 0.795, and κ of 0.651 for morphological subtyping; AUROC of 0.909, F1-score of 0.836, accuracy of 0.959, and κ of 0.673 for EAC detection), with non-uniform gains across single-model comparisons and limited benefit for dysplasia. The findings support the capability of Histo-FMs for BE morphological subtyping and EAC detection in a multicenter clinical cohort, while highlighting persistent challenges in dysplasia assessment and the need for external validation. This study provides a foundation for future clinical validation.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.