Sjögren's disease & overlap diseases
An accurate predictive model for depressive symptoms in patients with primary Sjögren's disease based on machine learning algorithms
Authors' conclusion
This study presents the first ML-based model for predicting depressive symptoms in pSjD patients, highlighting the significance of immuno-inflammatory and clinical factors in depression pathogenesis.
Study context
- Population
- Primary Sjögren's disease (pSjD) patients, Nanjing First Hospital, 2019–2022
- Participants
- 147 patients
- Outcome
- Depressive symptoms, PHQ-9 ≥5
Methodological detailsmachine-learning prediction model development and validation, single-centre, 70/30 training/testing split, stratified 10-fold cross-validation, LASSO selection, multiple imputation
- Machine-learning prediction model development and validation
- Single-centre
- 70/30 training/testing split
- Stratified 10-fold cross-validation
- LASSO selection
- Multiple imputation
Main finding
As reported in the abstract
Prediction of depressive symptoms (PHQ-9 ≥5), held-out test set
| Group | AUROC | AUPRC | Brier score |
|---|---|---|---|
| Support vector machine (SVM) | 0.929 | 0.959 | 0.106 |
Support vector machine (SVM)
AUROC0.929
AUPRC0.959
Brier score0.106
Observational data: associations, not effects.
Further findings
| Outcome | Values |
|---|---|
| Discrimination across repeated cross-validation | SVM mean AUROC ≈0.90 |
Discrimination across repeated cross-validation
- SVM showed the best overall discrimination of eight machine-learning (ML) algorithms, with strong precision and accuracy
Original publication
An accurate predictive model for depressive symptoms in patients with primary Sjögren's disease based on machine learning algorithms. Clin Exp Rheumatol. Published 14 August 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →