Predicting Heart Disease at Early Stages using Machine Learning: A Survey

TLDR

Heart disease remains a major health challenge, with costly treatments and risk factors such as alcohol, tobacco, and inactivity, and machine learning has proven effective for early detection and decision support. The study aims to predict heart disease at early stages to enable timely interventions worldwide. The authors review supervised ML methods—including ANN, DT, RF, SVM, NB, and k‑NN—for early heart disease prediction. The performance of these algorithms is summarized, highlighting their comparative effectiveness.

Abstract

Predicting and detection of heart disease has always been a critical and challenging task for healthcare practitioners. Hospitals and other clinics are offering expensive therapies and operations to treat heart diseases. So, predicting heart disease at the early stages will be useful to the people around the world so that they will take necessary actions before getting severe. Heart disease is a significant problem in recent times; the main reason for this disease is the intake of alcohol, tobacco, and lack of physical exercise. Over the years, machine learning shows effective results in making decisions and predictions from the broad set of data produced by the health care industry. Some of the supervised machine learning techniques used in this prediction of heart disease are artificial neural network (ANN), decision tree (DT), random forest (RF), support vector machine (SVM), naïve Bayes) (NB) and k-nearest neighbour algorithm. Furthermore, the performances of these algorithms are summarized.