Performance Analysis of Various Machine Learning Techniques for Classifying Cardiac Arrhythmia - Survey

Cardiac rhythm can be defined as the irregular electrical activity of the heart. It can be monitored using electrocardiogram (ECG). As ECG doesn't require any kind of surgery or need to put an instrument inside the body for reading the electrical activity of the heart, it is extensively used for identifying rhythms. The pulsation of the heart can be monitored through ECG which consists of P Wave, PQ Interval, QRS Complex, ST Segment, T Wave, and sometimes U Wave. The ECG can vary for the same individual depending on their lifestyle. Hence detecting arrhythmias is hard even for the most proficient cardiologist. Thus, system for autonomously identifying and classifying arrhythmia is in high demand. Feature extraction and segmentation are important parameters for autonomously identifying the majority of cardiovascular diseases. The primary goal of this paper is to examine the various ml techniques for detecting and distinguishing arrhythmias.