COMPARATIVE ANALYSIS OF KIDNEY FAILURE PREDICTION AT AN EARLY STAGE USING MACHINE LEARNING ALGORITHMS

Abstract

Chronic kidney disease (CKD) is a medical state of a person in which the kidney can’t filter the blood due to which the body fills with extra water and waste products. It can lead to stroke, heart attack, heart failure, swelling of the feet and kidney failure, which can lead to death. The global health problem is growing rapidly as more and more people are being diagnosed with CKD. With advancing technology, as well as ongoing medical research, machine learning is being used in the healthcare sector to diagnose many diseases early. ML algorithms and decoding methods have been very useful in extracting, analyzing data and making predictions when a person is positive or negative about a disease based on the given data sets. ML algorithms and in-depth reading have been proven to be very true in detecting CKD early. Machine learning algorithms Cat boost classifier, Support Vector Machine (SVM), Decision Trees (DT), Random Forests, KNN, Artificial Neural Networks have been studied and applied in this work to perform comparative analysis to create a ML model that can accurately predict if a person is positive to CKD or not. This paper uses pre-data processing, including background and above-mentioned machine learning algorithms to build the most accurate model to accurately detect this disease CKD and perform a comparative analysis of various Machine learning models for prognosis of CKD.