Detection of Ventricular Fibrillation by Support Vector Machine Algorithm

Abstract

With the increasing of sudden cardiac death, the developing of a reliable and portable electrocardiograph (ECG) monitor is imminent, especially automated external defibrillators (AEDs). A pivotal component in AEDs is the detection of ventricular fibrillation (VF) by means of appropriate detection algorithms. Various algorithms were proposed, here we proposed a new algorithm, which is based on support vector machine (SVM), Hurst index, and the time-delay algorithm [phase space reconstruction (PSR)]. For the new VF detection algorithm we calculated the sensitivity, specificity, positive predictivity and accuracy, then we compared these values with the results from an earlier investigation of several VF detection algorithms under equal conditions, using same databases and all of data without any preselection. We used the BIH-MIT arrhythmia database and the CU database. The result shows that the proposed algorithm has a high detection quality and outperforms all other investigated algorithms.