Machine Learning-Based Comparative Analysis for Chronic Kidney Disease Prediction Using Clinical Data

Chronic Kidney Disease (CKD) is a progressive and often asymptomatic condition that poses significant health risks if not detected early. With recent advances in machine learning (ML), predictive models have been developed to enable early CKD diagnosis using clinical data. This study presents a comparative review of existing ML-based approaches applied to datasets such as the UCI CKD dataset and electronic health records (EHRs). It examines algorithms like Logistic Regression, Support Vector Machine (SVM), Random Forest, K-Nearest Neighbors (KNN), and XGBoost. The models are compared using metrics such as accuracy, precision, recall, F1-score, and AUC-ROC. Results from reviewed studies show that ensemble methods, particularly Random Forest and XGBoost, offer superior performance and robustness in managing missing or imbalanced data. Additionally, the study highlights variations in data preprocessing, feature selection, and validation strategies.