Advancing Heart Failure Outcomes Using Machine Learning: A Data Science Breakthrough

Abstract

The scientific community and media are paying significant attention to machine learning. Such algorithms offer a lot of possibilities to improve patient care and personalize it in the medical field, including heart failure diagnosis and treatment. This study examined the risk variables in heart failure patients. In 2015, 299 heart failure patients were collected for our analysis. Using the Decision Tree, K-Nearest Neighbor, Support Vector, and Random Forest classifier, a heart failure outcome model is created with an accuracy of 80%, 77%, 82%, and 88% respectively. Current uses of machine learning techniques to diagnose, classify, and predict heart failure are evaluated in an overview of machine learning geared toward practicing clinicians.