The Diagnosis of Heart Attacks: Ensemble Models of Data and Accurate Risk Factor Analysis Based on Machine Learning
Shaymaa Hussein Nowfal, Sudhakar Sengan, J G, Serwes Bhatta, V. Saravanan, B Veeramallu
Abstract
Recent studies in clinical studies have observed a rampant increase in the rate of heart attacks, even among the newer population. Medical experts compute a multitude of factors as origins of a heart attack. But, the medical community is not able to explain the exact reasons for the prediction of heart attacks. ML algorithms are now evading the healthcare sector to assist healthcare providers in diverse ventures. This work analyses the potential causes of heart attacks among different age groups besides predicting attacks from biological conditions. The proposed ensemble model constellates the prowess of Support Vector Machine (SVM), K-Nearest Neighbors (K-NN), Random Forest (RF), and Extreme Gradient Boost (XGB) to predict heart attacks. The performance of this ML is tested on a heart attack prediction dataset, and the results promise the model's power over its peers. The proposed system achieved a classification accuracy of 92.8% for the test set in the ensemble model.