Wavelet Transform-Based Classification of Arrhythmia: Unraveling Cardiac Dynamics for Enhanced Diagnosis

Abstract

This report presents a comprehensive investigation into the analysis of electrocardiography (ECG) waveforms, inspired by the Moody Challenge hosted on PhysioNet. Leveraging a dataset comprising ECG recordings from over 21,000 patients sourced from the PTBXL database on PhysioNet, the study aims to shed light on the critical role of ECG waveform analysis in evaluating cardiac health. With a specific emphasis on the QRS complex, the report delves into the intricate details of this waveform component and its significance in diagnosing a spectrum of cardiac conditions, including arrhythmias. Furthermore, the study explores the application of advanced signal processing techniques such as the Maximal Overlap Discrete Wavelet Transform (MODWT) and Symlet4 wavelet to enhance the diagnostic accuracy and efficiency of ECG-based assessments. By synthesizing insights from contemporary research and clinical practice, the report underscores the importance of precise ECG interpretation in guiding clinical decision-making and optimizing patient care pathways. Additionally, a comparative analysis between MODWT and traditional bandpass filtering methods is conducted, demonstrating the superiority of MODWT in achieving more precise and reliable results in cardiac waveform classification.