Lupus & antiphospholipid syndrome
Development of a machine learning-based nomogram for predicting atherosclerosis risk in patients with systemic lupus erythematosus
Authors' conclusion
This exploratory, hypothesis-generating single-centre study provides an internally validated, interpretable prediction approach for atherosclerosis in SLE.
Study context
- Population
- Systemic Lupus Erythematosus (SLE) inpatients, Affiliated Hospital of North Sichuan Medical College, 2013–2023
- Comparison
- Training dataset vs Testing dataset
- Outcome
- Discrimination (AUC) of a nomogram predicting atherosclerosis risk in SLE
Methodological detailssingle-centre, retrospective electronic medical records, LASSO regression, machine learning (Adaboost), SHAP, logistic regression nomogram, internal validation
- Single-centre
- Retrospective electronic medical records
- LASSO regression
- Machine learning (Adaboost)
- SHAP
- Logistic regression nomogram
- Internal validation
Performance
As reported in the abstract
AUC for atherosclerosis prediction
| Test or model | AUC (95% CI) |
|---|---|
| Training dataset | 0.905 (0.880-0.930) |
| Testing dataset | 0.890 (0.841-0.939) |
AUC (95% CI)
Training dataset0.905 (0.880-0.930)
Testing dataset0.890 (0.841-0.939)
Further findings
Calibration and decision-curve analysis
- Model showed strong applicability and significant clinical benefits
Original publication
Development of a machine learning-based nomogram for predicting atherosclerosis risk in patients with systemic lupus erythematosus. Clin Exp Rheumatol. Published 28 October 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →