Virtual Booth

Chronic kidney disease is a major complication of type 2 diabetes and may remain clinically silent during its early stages. This innovation developed and validated a machine learning driven risk stratification tool using Johor National Diabetes Registry data from 30,093 patients across development, internal validation and temporal validation cohorts. Six machine learning models were compared, with Logistic Regression demonstrating the best overall performance. AUROC remained stable at 0.715, 0.717 and 0.723 across the three phases, respectively, with consistently high negative predictive values. Using ten routinely available clinical predictors, the model stratifies patients into low, moderate and high CKD risk groups.