Explainable AI-based analysis of human pancreas sections identifies traits of type 2 diabetes
Authors:
Open Access
- 9 February 2026
- journal article
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Abstract
Type 2 diabetes (T2D) is a chronic disease currently affecting around 500 million people worldwide with often severe health consequences. Yet, histopathological analyses are still inadequate to infer the glycaemic state of a person based on morphological alterations linked to impaired insulin secretion and β -cell failure in T2D. Giga-pixel microscopy can capture subtle morphological changes, but data complexity exceeds human analysis capabilities. In response, we generate a dataset of pancreas whole-slide images from living donors with multiple chromogenic and multiplex immunofluorescence stainings and train deep learning models to predict the T2D status. Using explainable AI, we make the learned relationships interpretable, quantify them as biomarkers, and assess their association with T2D. Remarkably, the highest prediction performance is achieved by simultaneously focusing on islet α - and δ -cells and neuronal axons, alongside subtle pancreatic alterations in T2D donors such as larger adipocyte clusters, altered islet-adipocyte proximity and smaller islets. This data-driven approach provides a foundation for future research into relevant diagnostic and therapeutic targets, refining several hypotheses regarding tissue alterations associated with T2D.Keywords
Funding Information
- Helmholtz Association (Project DIADEM, ZT-1-PF-5 139)
This publication has 69 references indexed in Scilit:
- Intrapancreatic adipocytes and beta cell dedifferentiation in human type 2 diabetesDiabetologia, 2025
- The insulin resistant brain: impact on whole-body metabolism and body fat distributionDiabetologia, 2024
- Unravelling innervation of pancreatic isletsDiabetologia, 2022
- Longitudinal association of fatty pancreas with the incidence of type-2 diabetes in lean individuals: a 6-year computed tomography-based cohort studyJournal of Gastroenterology, 2020
- Human pancreatic neuro-insular network in health and fatty infiltrationDiabetologia, 2017
- Metabolic crosstalk between fatty pancreas and fatty liver: effects on local inflammation and insulin secretionDiabetologia, 2017
- Pancreatic biopsy by minimal tail resection in live adult patients at the onset of type 1 diabetes: experiences from the DiViD studyDiabetologia, 2014
- A local glucagon-like peptide 1 (GLP-1) system in human pancreatic isletsDiabetologia, 2012
- Pancreatic fat is negatively associated with insulin secretion in individuals with impaired fasting glucose and/or impaired glucose tolerance: a nuclear magnetic resonance studyDiabetes/Metabolism Research and Reviews, 2010
- Islet-associated macrophages in type 2 diabetesDiabetologia, 2009