ECG Beat Classification Based on Mirrored Gauss Model and Cluster Template

摘要

Accurate electrocardiogram (ECG) beat classification is essential for automated detection of arrhythmias. A practical classification algorithm of the ECG beats, combining Mirrored Gauss Model (MGM) and cluster template had been proposed in this paper. High classification accuracy of normal beats and premature ventricular contraction (PVC) beats could be reached using MGM with an expense of huge time cost. Cluster template based on correlation matching and template queue could reducefitting times of MGM a lot effectively. It was proved by experiment carrying out using all of ECG records in MIT-BIH Arrhythmia Database that this algorithm combining cluster template with Mirrored Gauss Model could guarantee both accuracy and speed well. Hence it is an effective and practical classification algorithm of ECG beat.