Survival Analysis of Heart Failure Patients Using Machine Learning

Abstract

The global burden of death from heart attacks has increased dramatically in the modern era. South Asians are more likely than those in other parts of the world to get a heart attack at a young age. Being able to accurately and rapidly forecast the stage of a heart attack patient requires extensive expertise as well as a deep level of understanding. The medical industry has access to a great quantity of data that may be utilized to make informed judgments thanks to all the concealed information. We will be able to predict heart attack patients' states or stages rapidly with good judgment and a few excellent data mining methods like logistic regression and decision trees. Support vector machine (SVM), random forest classifier, decision tree, logistic regression, KNN, and Gaussian Naive Bayes are the six algorithms we employed in our system (GaussianNB). The accuracy of our final model, which applies the SVM method, is 92%.