Detection of Acute Atrial-Ventricular Arrhythmias Based on ECG Delineator: Evaluation on MIT/BIH Standard Databases

In this paper we use an efficient arrhythmia algorithm based on wavelet transform. In first step, QRS complexes are detected. Then each QRS is delineated by detecting and identifying the peaks of the individual waves, complex onset and end. Then the determination of P and T wave peaks, onset and ends are performed. Finally the Ventricular flutter, ventricular tachycardia, supra ventricular tachycardia, ventricular fibrillation, atrial fibrillation and atrial flutter that are kinds of acute ventricular-atrial arrhythmias are detected. In the proposed algorithm, we used a second order spline as mother wavelet and improved the previous algorithms proposed by other investigators. We evaluated the algorithm on some manually annotated single ECG signals selected from MIT-BIH arrhythmia databases. This algorithm may achieve the mean detection accuracy of about 80% in these arrhythmias.