Machine Learning-based Prognosis for Early Detection of Heart Disease – A Comparative Study

Abstract

Medical analysis is occasionally cited as a valuable source of insightful data. Coronary heart disease (CHD), a common and serious complication of diabetes mellitus type 2 (T2DM), is one of the most common chronic conditions characterized by an imbalance in insulin secretion. T2DM commonly has poor outcomes and even fatalities as a result of these complications. Identification of those who have an a higher risk of CHD problems is becoming more and more important due to the enormous number of people with T2DM, but a predictive method is still lacking. Early detection of CHD can help reduce mortality rates because it is one of the main causes of death in the globe. The challenge arises from the complexity of the data and relationship prediction using conventional methods. The purpose of this project is to use machine learning (ML) technologies and historical medical data to predict CHD. This study's primary objective is to identify correlations in CHD data using supervised learning methods such as Support Vector Machines (SVM), Decision Trees and Ensemble Classifiers. The researched ML techniques produce intelligent models. Empirical results demonstrate that probabilistic approaches are promising in diagnosing CHD using a variety of performance assessment parameters. Key Words: heart, health, machine learning, ensemble