Machine Learning-Based Approach for Early Detection of Heart Disease

ABSTRACT: Heart disease is responsible for a staggering number of deaths globally each year nearly 18 million placing it among the top leading cause of mortality around the world. Due to sedentary lifestyles and hereditary predispositions, cardiovascular problems are becoming more common, which makes early identification and diagnosis extremely difficult. Conventional diagnostic techniques are frequently expensive, time-consuming, and have poor prediction accuracy. In this this research, we propose a framework that utilizes machine learning techniques to the prediction of heart disease utilizing important patient health markers in order to overcome these drawbacks. Several supervised machine learning methods, such as logistic regression, SVM, KNN, decision tree, random forest, and gradient boosting, are used in the study, which makes use of a benchmark dataset on heart disease with 14 clinical parameters. Random Forest and Gradient Boosting models provide higher prediction accuracy, while preprocessing methods enhance the quality of the data. With an emphasis on its potential for early cardiac disease identification and intervention, the study investigates the application of machine learning (ML) in healthcare. It looks at several ML classifiers and concludes that ML algorithms can improve patient outcomes, timeliness, and diagnosis accuracy, possibly cutting down on pointless testing.