Revolutionizing Heart Failure Prediction with Artificial Intelligence

Abstract

Heart disease remains a leading cause of death worldwide, exacerbated by contemporary lifestyle factors and the ongoing global coronavirus pandemic. This reality underscores the urgent need for accurate and early prediction systems to mitigate the risk of heart failure. Coronary artery disease (CAD) is among the most prevalent forms of heart disease, making its prediction a critical challenge in the medical field. Existing prediction models, while numerous, often fall short due to various limitations, necessitating the development of more robust solutions. This paper aims to address these limitations by proposing an enhanced heart failure prediction system using a dataset from Kaggle with 11 key attributes. Several machine learning algorithms were employed, including K-Nearest Neighbors (KNN), Random Forest, Decision Tree, and Support Vector Machine (SVM). Among these, Random Forest demonstrated the highest accuracy at 88.04%. The paper also presents a comparative analysis of these algorithms and discusses model validation techniques to identify the most suitable approach for this scenario.