i-CKD Score : A Smart Digital Calculator for Chronic Kidney Disease Risk Prediction in Type 2 Diabetes

About the Project

Chronic kidney disease (CKD) is a major complication of type 2 diabetes mellitus (T2DM) and may progress silently before becoming clinically apparent. In routine diabetes care, large amounts of clinical information are already collected, yet these data are not routinely translated into individualised CKD risk prediction. The i-CKD Score was developed as a smart digital risk calculator that transforms routinely available clinical information into an interpretable CKD risk category to support earlier identification of patients requiring further renal assessment and monitoring. Developed using the Johor National Diabetes Registry, the tool incorporates ten routinely collected predictors and was evaluated across development, internal validation and temporal validation cohorts. Six machine-learning models were benchmarked. Logistic regression demonstrated the best overall performance and was subsequently translated into a practical point-based scoring system. The model demonstrated stable discrimination over time, with AUROC values of 0.715, 0.717 and 0.723 across the three cohorts, respectively. The scoring system classifies patients into low, moderate and high CKD risk groups and has been deployed as a web-based digital calculator for rapid use in clinical settings. Because the tool uses routinely collected information and requires no additional diagnostic test, it is low-cost, scalable and suitable for integration into diabetes registries, electronic medical records and primary care workflows. The i-CKD Score converts existing clinical data into actionable risk information, with potential for wider implementation across diabetes services and adaptation to other healthcare settings.