Decision Support System Based on Machine Learning Techniques to Diagnosis Heart Disease Using Four-Lead ECG Recordings

Cardiovascular disease is one of the leading causes of death in the world. Accurately and rapidly diagnosis of this disease remains an important challenge in research. Different decision support systems (DSS) implementing Machine Learning (ML) techniques have been proposed in the literature. In this study, we attempt to build a DSS that implements ML and try to classify patients according to their clinical characteristics and four-lead Electrocardiogram (ECG) recorded by mean of a smart watch. The principal goal of this study is to check whether the proposed DSS can play the same role as a standard ECG. Indeed, the obtained results suggest that this DSS can accurately screen and diagnose heart conditions and may be used as an alternative to a standard ECG. Moreover, among the four optimized ML techniques by grid search optimization technique, the Support Vector Machines achieved the highest accuracy score.