Classification of Cardiac Arrhythmia using Kernelized SVM

Cardiovascular diseases are one of the major causes of death in the world. These diseases include abnormalities in the smooth functioning of the heart causing cardiac arrest, blockages, and other related problems. One such ailment is the irregularities in the heartbeat of the person. Due to this, the movements of the heart are not operating at the normal pace causing palpitations and cardiac arrest. Though Electrocardiogram (ECG) is one of the most popular and widely used methods for monitoring the heart's electrical activity, it becomes quite strenuous for understanding the ECG reports which is a manual approach. So, there is a need to develop a system that could determine the condition a prior and classify them according to its severity. This paper focuses on the ECG deflections, cardiac arrhythmia, and its types. The paper further dwells into the development of an automated system to detect and classify arrhythmia. Various Machine Learning algorithms like Support Vector Machine (SVM), Random Forest Classifier (RF) are analyzed that lead to the identification of the optimized machine learning algorithm for classification of cardiac arrhythmia to distinguish the patient with arrhythmia. Kernelized SVM has been identified as the most accurate model.