Predicting Chronic Kidney Disease using Machine Learning in the Early Stages

chronic kidney disease (CKD) is an illness that causes a decline in kidney function. The CKD early diagnosis is tough since it does not have initial symptoms. The early disease can stop the infection advancement with a well-timed diagnosis. The research aim is to implement and compare the various algorithm’s performance and identify the best combinations that can provide better detection accuracy. We have found the solution using Machine Learning (ML) approaches that can achieve risk minimization. Seven ML algorithms have been utilized in this work with Accuracy, Precision, Recall, F1-Score, and Support as model evaluation parameters. In this comparison, RF and AdaBoost algorithms provided the uppermost accuracy worth 99%. Finally, data analysis is performed such as frequency distribution, Correlation between the features, Feature distribution plots, classification, and prediction plots. The domain of predictive analysis of CKD can be significantly enhanced by ML algorithms. By collaborating with them, healthcare professionals will be able to design a computer-aided diagnosis system that can enhance their ability to treat patients effectively.