Chapter 7 - Automated arrhythmia classification for monitoring cardiac patients using machine learning techniques

Abstract

People with chronic illnesses, such as cardiovascular disease, stroke, and diabetes, can be continuously monitored with the help of remote patient monitoring (RPM) systems. The major concern in RPM systems, which were developed to monitor cardiac patients, is identification of the correct arrhythmia class from the electrocardiogram signal. Existing arrhythmia classification techniques report low accuracy for a few arrhythmia classes due to class overlap and class imbalance problems. Arrhythmias are classified into five classes: nonectopic beat (N), supraventricular ectopic beat (S), ventricular ectopic beat (V), fusion beat (F), and unknown beat (Q) by ANSI/AAMI EC57: 2012 standard. The algorithms used for arrhythmia classification incorporate preprocessing, feature extraction, and classification. Classification becomes complicated when class overlap and class imbalance problems occur together. The classification results obtained after addressing class imbalance and class overlap issues show an improvement over the previously reported results for automated arrhythmia classification systems.