An Improved Deep Learning Approach for Prediction of The Chronic Kidney Disease

Abstract

Kidney function is harmed by chronic kidney disease, leading to renal failure. Machine learning and data mining come in handy to detect kidney disease. Machine learning employs a variety of algorithms to make predictions and classify data. CT scans have been used to detect chronic renal disease. When CT scans are used to diagnose disease in the kidney, cross-infection occurs, and the results are delayed. The authors of the prior study developed a model for categorizing chronic renal illness utilizing multiple classification methods. A unique deep learning model is presented in this study for the early identification and prognosis of Chronic Kidney Disease (CKD). This study aims to build a neural network and evaluate its performance compared to other cutting-edge machine learning methods. Compared to the four different classifiers (K-Nearest Neighbor (KNN), Random Forest, Naive Bayes classifier, and probabilistic neural network), the suggested Deep neural model fared better by reaching higher accuracy. Nephrologists may find the proposed method helpful in the early detection of CKD.