Hepatic and abdominal adiposity in type 2 diabetes as assessed with machine learning on computed tomography scans
Authors' conclusion
Hepatic steatosis, hepatomegaly and visceral adiposity on CT are associated with type 2 diabetes. Hepatic changes may influence spleen size effects on diabetes. VSR can serve as an alternative to traditional obesity metrics to accurately reflect diabetes risk.
Study context
- Population
- Patients with abdominal computed tomography (CT) scans in the Penn Medicine BioBank; 62 years, 49.4% male, 59.3% White
- Participants
- 1,594 patients
- Exposure
- T2D vs no diabetes
- Diabeticsn = 644
- Nondiabeticsn = 950reference
- Outcome
- CT image-derived phenotypes
Methodological detailsdeep learning on CT, medical biobank
- Deep learning on CT
- Medical biobank
Main finding
As reported in the abstract
- Diabetics vs Nondiabetics: P = 1.34 × 10-8
- Diabetics vs Nondiabetics: P = 4.85 × 10-9
- Diabetics vs Nondiabetics: P = 1.69 × 10-3
Observational data: associations, not effects.
Further findings
Multivariate association of diabetes with CT traits (adjusted for age, sex, race, body mass index (BMI))
- Spleen-hepatic attenuation difference (SHAD), OR
- 1.04 [1.02-1.05] (95% CI)
- Spleen volumes (SV), OR
- 4.53 [1.89-10.99]
- Visceral-to-subcutaneous ratio (VSR), OR
- 2.87, [1.96-4.20]
Combined regression analysis
- Splenomegaly controlling for hepatic factors, OR (no relationship)
- 1.08, [0.95-1.23]
- VSR, OR
- 1.40, [1.20-1.63]
- BMI, OR
- 1.14, [1.01-1.29]
Original publication
Hepatic and abdominal adiposity in type 2 diabetes as assessed with machine learning on computed tomography scans. Diabetes Obes Metab. Published 18 February 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →