Lipids & metabolic liver disease
An interpretable machine learning model for predicting metabolic dysfunction‐associated steatotic liver disease in patients with type 2 diabetes
Authors' conclusion
The model developed enables reliable and interpretable MASLD risk prediction in T2DM patients based on selected commonly available clinical data, providing a practical tool for routine identification and stratified management.
Study context
- Population
- Hospitalised T2D patients, Nanjing Drum Tower Hospital, January 2018–May 202513 features selected
- Participants
- 3,836 patients
- Exposure
- Eight machine learning models for MASLD risk prediction
- Outcome
- Discrimination for MASLD (AUROC, AUPRC)
Methodological detailsprediction model development, single-centre, machine learning, Shapley additive explanations, unsupervised clustering
- Prediction model development
- Single-centre
- Machine learning
- Shapley additive explanations
- Unsupervised clustering
Main finding
As reported in the abstract
MASLD prediction
| Group | AUROC | Area under the precision-recall curve (AUPRC) |
|---|---|---|
| XGB | 0.873 | 0.904 |
XGB
AUROC0.873
Area under the precision-recall curve (AUPRC)0.904
Observational data: associations, not effects.
Further findings
MASLD incidence rate
- MASLD incidence rate
- 55.9%
Unsupervised clustering
- Several high-risk subgroups with distinct metabolic characteristics identified
Original publication
An interpretable machine learning model for predicting metabolic dysfunction‐associated steatotic liver disease in patients with type 2 diabetes. Diabetes Obes Metab. Published 29 September 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →