Novel framework for Efficient Detection of QRS Morphology for The Cardiac Arrhythmia Classification

Abstract

The abnormal conduction or disturbance in the cardiac activity is called cardiac arrhythmia except for sinus rhythm. Cardiac arrhythmias are placing a significant strain on the healthcare system as a result of the rising mortality rate in the world. According to the American Heart Association’s (AHA) updated health data records, heart disease is the leading cause of mortality, with 17.3 million stated in the recent annual report. A cardiac specialist frequently uses an electrocardiograph (ECG), a non-invasive instrument, to identify heart arrhythmia. Currently, studies have been directed at employing computer-aided techniques to diagnose cardiac arrhythmia. However, due to the interpatient variability issues in ECG signal, QRS morphology is difficult to analyze as it is regarded as the primary characteristic because of its wide range of variances. In literature analysis, we have found that accurate detection of the QRS morphology using computer-assisted methods still is quite a challenging task due to the different variations. In the field of medicine, the biased results may cause ineffective detection of cardiac arrhythmias and can lead to the serious lives threat of patients. Moreover, human error and time constraints are two additional concerns associated with manual cardiac arrhythmia analysis. This research seeks to offer a novel methodology for the extraction of the QRS morphological feature (E-QRSM) to classify Premature Ventricular Contraction (PVC) arrhythmia from ECG signals. This would save the patient time and medical professional effort. The exact morphological features that are pertinent to the arrhythmia must be extracted, which is the most important and difficult part of the ECG signal analysis. In this study, a novel E-QRSM algorithm for categorizing PVC arrhythmias is presented. Since QRS segments are thought to be the primary component of PVC arrhythmia, these components are fed to the classifier. The studies were carried out utilizing the MIT-BIH arrhythmia benchmark dataset as a public benchmark to assess the effectiveness of our suggested E-QRSM approach. E-QRSM found that the proposed methodology’s experimental analysis revealed that the novel algorithm delivers accurate and effective real-time analysis of QRS-related aspects with the conduction of aberrant rhythm in ECG data.