A Gated Recurrent Deep Model for Supraventricular Premature Beat Detection in Electrocardiography Signal

Heart diseases are mainly diagnosed by the electrocardiogram (ECG) or (EKG), which a mechanism of recording the electrical signals of the heart. Correct detection of ECG signals behavior helps in diagnosing heart diseases and assists the cardiologists in determining the correct treatment for the patient. In this paper, we proposed a novel Deep Gated Recurrent (DGRU) based model for heart supraventricular premature beat detection in a multivariate ECG signal. The proposed model achieved a high detection accuracy using a small implementation budget. Furthermore, it could be easily implemented on portable hardware devices.