A Comparative Analysis of Machine Learning and Deep Learning Approaches for Prediction of Chronic Kidney Disease Progression

Abstract

Chronic kidney disease is a significant health problem worldwide that affects millions of people, and early detection of this disease is crucial for successful treatment and improved patient outcomes. In this research paper, we conducted a comprehensive comparative analysis of several machine learning algorithms, including logistic regression, Gaussian Naive Bayes, Bernoulli Naive Bayes, Support Vector Machine, X Gradient Boosting, Decision Tree Classifier, Grid Search CV, Random Forest Classifier, AdaBoost Classifier, Gradient Boosting Classifier, XgBoost, Cat Boost Classifier, Extra Trees Classifier, KNN, MLP Classifier, Stochastic gradient descent, and Artificial Neural Network, for the prediction of kidney disease. In this study, a dataset of patient records was utilized, where each record consisted of twenty-five clinical features, including hypertension, blood pressure, diabetes mellitus, appetite and blood urea. The results of our analysis showed that Artificial Neural Network (ANN) outperformed other machine learning algorithms with a maximum accuracy of 100%, while Gaussian Naive Bayes had the lowest accuracy of 94.0%. This suggests that ANN can provide accurate and reliable predictions for kidney disease. The comparative analysis of these algorithms provides valuable insights into their strengths and weaknesses, which can help clinicians choose the most appropriate algorithm for their specific requirements.