Arrhythmia Classification: A pipeline based Comprehensive Survey

Abstract

Nowadays, Artificial Intelligence (AI) plays an indispensable role in advancing healthcare data systems, particularly in the realm of intricate medical data analysis. Its efficacy in unveiling meaningful relationships has proven pivotal for diagnosis, treatment, and prediction across a spectrum of clinical scenarios. One such critical area is arrhythmia, a condition marked by deviations in the heart's electrical system, posing a substantial risk of sudden cardiac arrest and potential fatality. Electrocardiograph (ECG) signals serve as the primary medium for capturing and documenting the heart's electrical activity. This paper provides a comprehensive overview of the application of AI techniques at various stages of the arrhythmia classification process. A distinctive presentation approach was used as the survey is made in the form of pipeline. Encompassing the preprocessing of ECG data, extraction and selection of pertinent features, classifier training, and performance evaluation, the swift and accurate analysis of ECG signals is imperative for monitoring and treating individuals with heart conditions. The key goal is the deployment of these AI-driven solutions in clinical settings, ensuring enhanced patient care and outcomes.