Development and Validation of an Interpretable Machine Learning Model for Noninvasive Differentiation of Diabetic Kidney Disease and Non-Diabetic Kidney Disease in Type 2 Diabetes: A Multicenter Retrospective Cohort Study

Multicentre Study Validates Interpretable Machine Learning Model to Distinguish Diabetic from Non-Diabetic Kidney Disease Without Biopsy — Implications for Nephrologists

A completed multicentre Chinese cohort study involving 2,201 biopsy-confirmed patients has developed and externally validated an interpretable machine learning model to noninvasively distinguish diabetic kidney disease from non-diabetic kidney disease in type 2 diabetes. The web-based clinical decision support tool could help nephrologists stratify biopsy necessity and accelerate appropriate therapy — with significant implications for Australian practice where DKD drives the highest burden of end-stage kidney disease.