Machine Learning based Prediction of Chronic Kidney Disease Using PCA and Ensemble Techniques

Early diagnosis of chronic kidney disease (CKD) is crucial treatment and improved patient outcomes.This study investigates the performance of various machine learning models including logistic regression, random forest, adaboost, and gradient boosting for CKD classification using a publicly available dataset including 400 patient records with 25 clinical and physiological attributes.Pre-processing has been utilized to remove missing values and confirm uniformity.Principal component analysis (PCA) is used to reduce dimensionality and enhance model interpretability.The classification models performance are evaluated using standard metrics such as accuracy, precision, recall and F1-score.The experimental results showed that ensemble-based methods, particularly random forest and adaboost, achieved superior accuracy (97.50%) and F1-score (0.9750) which outperformed both logistic regression and gradient boosting.These findings demonstrate the robustness and reliability of ensemble approaches for medical diagnosis and applications, highlighting their potential for clinical decision support in CKD detection.