AI-Driven segmentation and morphogeometric profiling of epicardial adipose tissue in type 2 diabetes
Authors' conclusion
This AI-based framework enables accurate segmentation for structurally complex EAT and reveals key morphogeometric differences associated with T2D, supporting its potential as a biomarker for cardiometabolic risk assessment.
Study context
- Population
- Participants with T2D and matched controls undergoing cardiac 3D Dixon MRI, enrolled 2014–2018, SCAPIS sub-study
- Participants
- 90 participants
- Comparison
- T2D vs age-, sex-matched controls
- T2Dn = 45
- Controlsn = 45reference
- Outcome
- EAT-Seg segmentation performance and EAT morphogeometric differences
Methodological detailsage-, sex-matched controls, sub-study of the Swedish SCAPIS cohort, deep learning segmentation (EAT-Seg)
- Age-, sex-matched controls
- Sub-study of the Swedish SCAPIS cohort
- Deep learning segmentation (EAT-Seg)
Main finding
As reported in the abstract
- Dice similarity coefficient (DSC)
- 0.881
- Hausdorff distance (HD95)
- 3.213 mm
- Average symmetric surface distance (ASSD)
- 0.602 mm
- Volume and thickness gradient-related features identified as key discriminators
- r > 0.8, P < 0.05
- Random Forest classification AUC
- 0.703
Original publication
AI-Driven segmentation and morphogeometric profiling of epicardial adipose tissue in type 2 diabetes. Cardiovasc Diabetol. Published 18 July 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →