An Explainable AI for Risk Stratification and Severity Prediction of CKD in Type 2 Diabetes using BMI, Lifestyle and Clinical Factors

This study presents a machine-learning framework designed to stratify risk and predict the severity of CKD, leveraging BMI, family history of diabetes, and lifestyle factors. A novel Severity Index was developed by integrating demographic, lifestyle, and medical attributes, facilitating patient categoriza-tion into distinct risk levels. Employing advanced clustering techniques like K-Means and Hierarchical Clustering, along with a logistic regression model, the model achieved a 91% accuracy in predicting severity levels. Model ex-plain ability via SHAP were utilized to identify significant predictors, includ-ing BMI, smoking habits, and the Severity Index, ensuring model transparen-cy and clinical relevance. This framework highlights the potential for AI to transform CKD and diabetes management by offering insights into patient-specific risk factors, thereby supporting early intervention and personalized care. Future work will focus on expanding the dataset, exploring multimodal inputs, and validating the framework across diverse clinical environments to enhance its robustness and clinical acceptability.