Advancing Chronic Kidney Disease Prediction: Comparative Analysis of Machine Learning Algorithms and a Hybrid Model

Abstract

Chronic kidney disease (CKD) presents a significant global health challenge, necessitating early detection and precise prediction for effective intervention. Recent advancements in machine learning have shown promise in enhancing CKD risk assessment by leveraging extensive datasets and complex pattern recognition. This study conducts a comparative analysis of machine learning algorithms, including XGBoost, Random Forest, Logistic Regression, AdaBoost, and a novel Hybrid Model, using real-world data from the UCI Chronic Kidney Failure dataset. The Hybrid Model emerges as the most accurate and robust approach, achieving superior performance metrics such as accuracy (94.99%), precision (95.21%), recall (95.11%), F-1 Score (95.32%), and AUROC (95.56%). This model not only surpasses individual algorithms but also integrates their strengths to provide reliable predictions, highlighting its potential to transform CKD diagnosis and management. Future research directions include validation across diverse datasets and populations, integration of advanced features, and longitudinal studies to assess long-term predictive efficacy.