Heart Disease Detection using Hybrid Machine Learning and IoT (Software Based)

Abstract

heart disease is a leading cause of mortality worldwide, presenting a critical challenge for clinical data analysis in predicting cardiovascular disease. Machine learning (ML) has shown promise in assisting decision-making and predictions based on the large amounts of data generated by the healthcare industry. Advancements in the Internet of Things (IoT) have opened new avenues for the application of ML techniques in diverse domains. However, the current literature provides only a limited perspective on predicting heart disease with ML techniques. To address this gap, we propose a novel approach that leverages ML techniques to identify significant features that can improve the accuracy of heart disease prediction. By utilizing a variety of feature combinations and established classification techniques, our prediction model achieves a superior level of performance with an accuracy rate of 88.7% for predicting heart disease. The hybrid random forest with a linear model (HRFLM) was found to be particularly effective in achieving these results.