A new approach for ICD rhythm classification based on support vector machines

Inappropriate shocks due to misclassification of supraventricular and ventricular arrhythmias remain a major problem in the care of patients with implantable cardioverter defibrillators (ICDs). The purpose of this study was to investigate the ability of a new covariance-based support vector machine classifier, to distinguish ventricular tachycardia from other rhythms such as supraventricular tachycardia. The proposed algorithm is applicable on both single and dual chamber ICDs and has a low computational demand. The results demonstrate that suggested algorithm has considerable promise and merits further investigation.