Optimized Ensemble Machine Learning Model for Chronic Kidney Disease Prediction

Abstract

The prevalence of chronic kidney illness has been increasing by a rate of 41.5 over the past few years, posing a significant challenge for healthcare systems worldwide. The disease is characterized by a gradual loss of kidney function over time, resulting in the accumulation of waste products and fluids in the body. Early detection and management of CKD are crucial for improving patient outcomes and reducing healthcare costs. Machine learning (ML) techniques have been increasingly employed for predicting the risk of CKD and developing effective screening tools. In this study, an ensemble ML model for CKD forecast is projected, which attained an accurateness of 96.5% by utilizing the Kaggle-based chronic kidney disease dataset. The model was constructed using two algorithms, random forest, and bagging, which were trained on a large dataset of CKD patients' clinical and demographic information. The ensemble model demonstrated superior performance compared to each algorithm separately, indicating the efficacy of the ensemble approach in enhancing predictive accuracy. The results of this research indicate that the suggested ensemble model could be a valuable tool for early identification and management of CKD, thus reducing the burden of CKD on individuals and healthcare systems.