Multi‐modal models using functional magnetic resonance imaging (fMRI) , urine and serum biomarkers for classification and risk prognosis in diabetic kidney disease
Authors' conclusion
In clinical practice, the multi-modal models presented in this study demonstrate potential to enhance clinical decision-making capabilities regarding patient classification and prognosis prediction.
Study context
- Population
- Patients with DKD
- Participants
- 172 patients
- Comparison
- Classification models 1-6 and risk-prognostic models A-E
- Outcome
- Classification and prognostic performance
Methodological detailsmachine learning, multi-modal
- Machine learning
- Multi-modal
Performance
As reported in the abstract
| Test or model | AUC at 2-year follow-up | AUC at 3-year follow-up | Accuracy |
|---|---|---|---|
| Model 5 | — not stated in the abstract | — not stated in the abstract | 0.833 (95% CI 0.653-0.944) |
| Model 6 | — not stated in the abstract | — not stated in the abstract | 0.923 (95% CI 0.749-0.991) |
| Model E | 0.975 | 0.932 | — not stated in the abstract |
AUC at 2-year follow-up
Model 5— not stated in the abstract
Model 6— not stated in the abstract
Model E0.975
AUC at 3-year follow-up
Model 5— not stated in the abstract
Model 6— not stated in the abstract
Model E0.932
Accuracy
Model 50.833 (95% CI 0.653-0.944)
Model 60.923 (95% CI 0.749-0.991)
Model E— not stated in the abstract
— not stated in the abstract
Original publication
Multi‐modal models using fMRI , urine and serum biomarkers for classification and risk prognosis in diabetic kidney disease. Diabetes Obes Metab. Published 2 July 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →