Method for Detecting Ventricular Activity of ECG Using Adaptive Threshold

The QRS complex in an electrocardiogram reflects the activity of the cardiac ventricles. Cardiac ventricle activity can provide information about ventricular arrhythmia. This study investigated whether the dominant activity of the ventricles can be used to analyze ventricular arrhythmia characteristics. To assess ventricular activity, we modified the adaptive threshold method proposed by Shin et al. and developed a ventricular activity segmentation approach. The proposed method was tested using five cardiac episodes, namely normal sinus rhythm, supraventricular tachycardia, ventricular tachycardia, ventricular flutter, and ventricular fibrillation, obtained from the MIT-BIH and Creighton University Ventricular Tachyarrhythmia databases. The average sensitivity was 95.77 %, the average positive predictivity was 98.09 %, and the average failed detection rate was 6.08 %.