Review of Computational Techniques for the Analysis of Abnormal Patterns of ECG Signal Provoked by Cardiac Disease

Abstract

The 12-lead ECG aids in the diagnosis of myocardial infarction and is helpful in the prediction of cardiovascular disease complications. It does, though, have certain drawbacks. For other electrocardiographic anomalies such as Left Bundle Branch Block and Left Ventricular Hypertrophy syndrome, the ECG signal with Myocardial Infarction is difficult to interpret. These diseases cause variations in the ST portion of the ECG signal. It reduces the clarity of ECG signals, making it more difficult to diagnose these diseases. As a result, the specialist is misled into making an erroneous diagnosis by using the incorrect therapeutic technique. Based on these concepts, this article reviews the different procedures involved in ECG signal pre-processing, feature extraction, feature selection, and classification techniques to diagnose heart disorders such as Left Ventricular Hypertrophy, Bundle Branch Block, and Myocardial Infarction. It reveals the flaws and benefits in each segment, as well as recommendations for developing more advanced and robust methods for diagnosing these diseases, which will increase the system’s accuracy. The current issues and prospective research directions are also addressed.