Explainable AI-based analysis of human pancreas sections identifies traits of type 2 diabetes
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.
Authors
- Alfred Königsrainer (ORCID: https://orcid.org/0000-0002-2301-4080)
- Paul F. Jäger (ORCID: https://orcid.org/0000-0002-6243-2568)
- Andreas L. Birkenfeld (ORCID: https://orcid.org/0000-0003-1407-9023)
- Silvia Wagner (ORCID: https://orcid.org/0009-0002-5756-8046)
- Nicole Kipke
- Yanni Morgenroth (ORCID: https://orcid.org/0000-0002-3555-7429)
- Daniela Friedland
- Lukas Klein (ORCID: https://orcid.org/0000-0003-3725-068X)
- Felicia Gerst
- Karol Gotkowski (ORCID: https://orcid.org/0000-0003-0453-2816)
- Eyke Schöniger (ORCID: https://orcid.org/0000-0002-3692-1747)
- Ellen Geibelt
- Silvio Nadalin
- F. Fend
- Annika Seiler
- Marius Distler
- Andre L. Mihaljevic
- Daniela Aust
- Sebastian Ziegler
- Robert K. Wagner
- Martin Heni
- Daniel Hartmann
- Hans-Ulrich Häring
- Fabian Isensee
- Reiner Jumpertz-von Schwartzenberg
- Hajime Yamazaki
- Susanne Ullrich
- Klaus Maier-Hein
- Michele Solimena
- Jürgen Weitz
Institutions
- Fukushima Medical University (JP)
- German Cancer Research Center (DE)
- Kyoto University (JP)
- Heidelberg University (DE)
- University Hospital Heidelberg (DE)
- Düsseldorf University Hospital (DE)
- Institute for Biomedical Engineering (CH)
- Google (United Kingdom) (GB)
- University Children's Hospital Tübingen (DE)
- Center for Systems Biology Dresden (DE)
- DKFZ-ZMBH Alliance (DE)
- Deutsches Diabetes-Zentrum e.V. (DE)
- University Hospital Ulm (DE)
- German Center for Diabetes Research (DE)
- University Hospital Carl Gustav Carus (DE)
- Paul Langerhans Institute Dresden (DE)
- Heinrich Heine University Düsseldorf (DE)
- École Polytechnique Fédérale de Lausanne (CH)
- Technische Universität Dresden (DE)
- University of Tübingen (DE)
Publication Details
- Journal
- Nature Communications
- Published
- 2026-02-09
- DOI
- https://doi.org/10.1038/s41467-026-69295-2
- Primary Topic
- Pancreatic function and diabetes
- Type
- article
- Field-Weighted Citation Impact
- 0.00