Heart Disease Prediction Using Machine Learning Methods

Abstract

A disease is an unusual occurrence that usually impacts one or more body parts of an individual. Diseases of all kinds are becoming more common as a result of lifestyle and environment. The most prominent of any of these diseases are heart disease, which also has the most catastrophic impacts of any condition. The fatality rate can be reduced by early recognition of cardiac diseases and active clinical supervision by experts. Unfortunately, since it requires additional intelligence, effort, and expertise, a reliable diagnosis of cardiac problems in all circumstances and 24-hour patient consultations by a physician are still not possible. In this research, we evaluated several machine-learning techniques. A comparative study was conducted using five widely used machine-learning strategies such as Logistic Regression, Random Forest, Decision Tree, Ada-Boost, and XG-Boost to predict cardiac illness in this work. Using heart disease datasets that were collected from the UCI machine learning repository database and subjected to a variety of assessment procedures, the performance of each technique was evaluated. With a 99-percent accuracy rate, Random Forest and XG-Boost classifiers proved to be the most effective approaches for heart disease detection.