Arrhythmia Detection From 12-Lead ECG with 2-Phase Feature Extraction: By Presenting the Evaluation of Atrial Fibrillation

Abstract

12-Lead Electrocardiography (ECG) is an essential diagnostic tool for detecting Cardiac Arrhythmias (CAs). In this study, Arrhythmia Detection (AD) was conducted using a 12-Lead ECG dataset. The dataset underwent specific preprocessing, and a hybrid 2-Phase Feature Extraction (2-PFE) method was proposed: (1) QRS detection using the Pan-Tompkins algorithm, and (2) P-Peak detection using windowing. The study specifically focused on analyzing Atrial Fibrillation (AFIB) separately from other arrhythmia types. This approach was evaluated through three classification models: SR and NON-SR, SR and NON-SR without AFIB, SR and AFIB.