Machine Learning Models To Predict The Response To Different Kidney Disease Treatments

Abstract

“If kidney disease is not identified and treated promptly, it might eventually result in kidney failure. Chronic kidney disease (CKD) is a degenerative illness marked by a progressive decrease of kidney function. Improving patient outcomes and stopping the disease's development depend on early CKD diagnosis and prognosis. This work investigates the use of different machine learning methods to clinical data-driven CKD prediction. Support Vector Machine (SVM), Random Forest, K-Nearest Neighbors (KNN), and Logistic Regression are among the methods that were assessed. A variety of performance criteria, such as accuracy, precision, recall, and F1-score, were used to evaluate each model.Both the SVM and KNN models attained 99% flawless accuracy rates, proving their remarkable ability to accurately categorize situations that are either CKD or not. The Random Forest model demonstrated great precision and was especially useful in recognizing nonCKD occurrences, although being marginally less accurate with an accuracy rate of 89%. With an accuracy rate of 99%, logistic regression demonstrated its dependability and strength as a prognostic tool for chronic kidney disease.The study underscores the noteworthy possibilities of machine learning algorithms in the prognosis of chronic kidney disease (CKD). It also shows the necessity of interdisciplinary cooperation among data scientists, physicians, and regulatory agencies to promote progress in this area and enhance patient care.