A Proposed Model for Machine Learning-Based Early Chronic Kidney Disease Detection

Chronic kidney disease (CKD) is a sluggish and cumulative decline of kidney characteristics over the years. Early detection of CKD is essential when you consider that delayed detection and hospital treatment of the situation might eventually bring about the final stage of CKD and the patient’s death. The prediction of diseases has grown to be appreciably impacted by means of the utilization of machine learning (ML) in healthcare. Thus, ML methods are the simplest in CKD prediction. The major goal is to use distinct supervised ML techniques, such as Support Vector Machine, Random Forest, Artificial Neural Network, and Logistic Regression, to examine and verify how correct they may be at detecting CKD. The proposed model provides various strategies for detecting CKD using medical facts which incorporates data preprocessing, strategies for handling missing values, and data reduction. It is noted that the clinical data has been classified with an overall accuracy of $100 \%$ by the Random Forest algorithm. The suggested ML model will help doctors identify chronic kidney ailments early on, which will be an effective manner of preventing the illness.