Predicting Early-Stage Heart Failure Using Artificial Intelligence Techniques

Heart failure is an incurable condition in which the heart gradually loses its ability to pump blood effectively. It is a growing global health concern affecting millions of people worldwide. The risk of heart failure increases with age, highlighting the need for machine learning models capable of predicting heart failure at an early stage. Early predictions can help reduce disease progression, lower hospitalization rates, and improve patients’ quality of life. The primary objective of this study is to predict patients in the early stages of heart failure using machine learning techniques based on health-related attributes. By leveraging the Cleveland dataset, which includes 13 key health features, our system predicts heart failure with high precision, enabling early intervention and more effective treatment planning. These models were tested and evaluated using standard performance metrics. Among them, the Random Forest classifier, implemented using RapidMiner, achieved the highest accuracy of 92.16%, outperforming other models in predictive capability.