Multifaceted ECG Feature Extraction for AFIB Detection: Using Traditional Machine Learning Techniques

This paper presents a novel approach for the early diagnosis of Atrial Fibrillation (AFIB) using electrocardiogram (ECG) signals. Specifically, the 12-lead ECG, a standard in cardiac monitoring, are exploit to extract meaningful patterns indicative of AFIB. Unlike the prevalent deep learning methods which often require significant computational resources, this study leverages traditional machine learning techniques. This choice is motivated by the aim to reduce computational costs and resource requirements, making the solution more accessible and feasible for widespread clinical use. The feature extraction process from ECG signals is emphasizes the identification of key characteristics relevant to AFIB. The proposed methodology demonstrates not only the efficacy of machine learning in medical diagnostics but also its potential in providing cost-effective, scalable solutions for early disease detection.