Development of A New Recognition System Based on Support Vector Machines for Shockable ECGs and Its Performance Analysis.

The aim of this paper is development of a new recognition system based on a Support Vector Machine (SVM) for serious arrhythmias such as Ventricular Fibrillation (VF) and Ventricular Tachycardia (VT). In order to develop the proposed SVM-based detection system, feature parameters for shockable/non-shockable Electrocardiograms (ECGs) are extracted. Moreover by using the extracted feature parameters, the proposed SVM-based recognition system is constructed. In addition, the achievable performance of the proposed recognition system is evaluated.