Deep learning-powered quantification of endocrine cells and CD3+ T cells in the natural history of type 1 diabetes
Authors' conclusion
The deep learning-powered approach enabled our study to include clusters of endocrine cells scattered throughout WSIs, providing precise quantitative evidence of cluster-level infiltration.
Study context
- Population
- Pancreases from control, autoantibody-positive and T1D organ donors; head, body and tail sections stained for insulin, glucagon and CD3
- Participants
- 77 participants231 whole-slide images (WSIs)
- Exposure
- Control vs sAAb+ vs mAAb+ vs recent-onset T1D vs longstanding T1D donors
- Controln = 32reference
- sAAb+n = 12
- mAAb+n = 8
- Recent-onset T1Dn = 6
- Longstanding T1Dn = 19
- Outcome
- Disease stage-associated patterns of endocrine cells and CD3+ T cells
Methodological detailshistopathological analysis of organ donor pancreases, deep learning-assisted analysis workflow
- Histopathological analysis of organ donor pancreases
- Deep learning-assisted analysis workflow
Main finding
As reported in the abstract
Islets analysed
>82,000
Clusters analysed
>26,000
Observational data: associations, not effects.
Original publication
Deep learning-powered quantification of endocrine cells and CD3+ T cells in the natural history of type 1 diabetes. Diabetologia. Published 7 May 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →