Medical Data Analysis and Machine Learning Models for Kidney Disease Prediction: Exploratory Study

Numerous recent breakthroughs in the field of ML and AI have made significant progress in the prediction and identification of healthcare catastrophes, disease communities, disease states, and immune system reactions. One characteristic of chronic kidney disease (CKD), sometimes referred to as chronic kidney failure, is a gradual decrease of the functioning of the kidneys. The Kidneys remove contaminants and surplus fluid from the bloodstream to produce urine. In the event of severe chronic renal illness, the human body may amass dangerously high levels of liquids, electrolytes, and toxic substances. The illness is called "chronic" because kidney damage in humans develops subsequently over a long period. Wastes may build up throughout your system as a result of this disease. These days, one of the most serious illnesses is CKD, for which a prompt diagnosis is essential. We use a variety of machine learning techniques in our current research to predict CKD. This work implements seven different ML algorithms to examine the kidney disease data and also finds the most suitable model. The results show that the three models Decision Tree, Random Forest, and Gradient Boosting perform very well with an accuracy rate of 98.03%.