Predictive Modeling of Cardiovascular Disease using Machine Learning Techniques

Abstract

Coronary artery disease (CAD), is consistently ranked among the leading causes of death around the globe. Over several decades, many non-invasive approaches for predicting and detecting coronary artery disease have been proposed. Despite the extensive study that has been conducted, the death rate due to CAD continues to be at an all-time high. It is possible that predictive models constructed with machine learning (ML) algorithms can help doctors discover CAD earlier, which in turn may improve patient outcomes. This study focuses on applying several machine learning algorithms to make predictions about coronary vascular disease. We rely on the Coronary Artery Disease Data Collection for our analysis. Python and the jupyter notebook environment are used to realize this project. Many machine learning techniques are utilized in this research to predict CAD results, including a random forest, a decision tree, a gradient-boosted tree, and a logistic regression. These algorithms are compared to each other in this paper, and the gradient-boosted tree algorithm obtained more accurate results than the other existing machine-learning methods.