Development of a predictive model for the risk of microalbuminuria: comparison of 2 machine learning algorithms.
Source check: Source checked. Not independent scientific confirmation.
Finding: A Random Forest model using nine clinical variables predicted MAU among 981 hospitalized T2DM patients (test AUC 0.87); SHAP associated HbA1c, NLR and 25(OH)D with risk. This is a single-center retrospective prediction, not evidence of clinical benefit.
Technology: logistic screening selects nine variables, then Random Forest and BP neural-network models predict MAU; SHAP interprets RF feature contributions.
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- logistic screening selects nine variables, then Random Forest and BP neural-network models predict MAU; SHAP interprets RF feature contributions.
- Test and result
- retrospective data from 981 hospitalized T2DM patients (333 MAU) produced an RF risk model with reported test AUC 0.87 and associations for HbA1c, NLR and 25(OH)D.
- Limitation / still unknown
- Methods: random 70/30 training/test split, SMOTE applied to training data, and five-fold cross-validation. Validation is internal and single-center; it does not establish clinical benefit or external performance.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://link.springer.com/article/10.1007/s40200-024-01440-4
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.