Time series Clustering and Analysis of ECG heart-beats using Dynamic Time Warping

A novel Time series Clustering and Analysis Method for ECG (Electro Cardiogram) heart-beat Analysis is proposed using K-medoids Clustering with Dynamic Time Warping (DTW) distance. The main objective of this paper is to identify the abnormalities in ECG heart beats through Clustering and Validation by using QRS complexes of ECG heart-beats. The ECG data obtained from MIT-BIH Arrhythmia Database, is used for experimentation. The 5 types of classes in ECG heart beats, used in this study are Normal (N), Left bundle branch blocks (LBBB), Right bundle branch blocks (RBBB), Premature ventricular contraction (PVC), Atrial premature contraction (APC).