ECG Signal Classification Using Deep Learning

Abstract

Heart arrhythmia is an irregular state of heart. This problem occurs when the electrical signals that coordinate the heart's beats don't work properly. Electrocardiography (ECG) is used to check the electrical activity and your heart rhythms. It records the electrical activities of the heart of a patient using electrodes attached to the skin. Electrodes are taped to the chest to record the heart's electrical signals, which cause the heart to beat. The signals are shown as waves on an attached computer monitor or printer. An electrocardiogram records the electrical signals in your heart. ECG signals reflect the physiological conditions of the heart, medical doctors use ECG signals to diagnose heart’s normal and abnormal condition i.e., heart arrhythmia. Medical professionals use their most important skill of identifying the dangerous types of heart arrhythmia from ECG signals. However, ECG waveforms performed by professional medical doctor manually is tedious and time- consuming. As a result, the development of automatic techniques for identifying abnormal conditions from daily recorded ECG data is of fundamental importance. Moreover, timely first-aid measures can be effectively applied if such abnormal heart conditions can be detected automatically using health monitoring equipment which internally uses machine learning algorithms. Thus, machine learning will play an important role in this regard.