CLASSIFICATION OF ARRHYTHMIAS ON THE BASIS OF HEART RATE VARIABILITY

Abstract

Heart rate variability (HRV) is concerned with the analysis of intervals between heartbeats. It indicates balance between parasympathetic and sympathetic nervous system. There is significant relationship between autonomic nervous system (ANS) and cardiovascular mortality and ANS is strongly related with HRV. In this paper HRV is analyzed by time domain measure, frequency domain measure and Poincare plot analysis. Initially R peak detection is performed by R wave detection algorithm using wavelet transform. Also we can detect all five peaks P, Q, R, S, T using the same. After detecting R peaks RR interval is calculated from which further parameters are calculated like SDNN, SDANN, RMSSD, pNN50, Heart rate (Time Domain parameters) total power, low frequency component, High frequency component and their ratio (frequency domain parameters). Unlike these methods Poincare plot analysis is non-linear visual technique for assessment of HRV. Poincare plot comments about short term and long term variability. Ratio of these two parameters is also an important factor for arrhythmia classification. Also position and orientation of RR interval in Poincare plot is helpful in visual identification of arrhythmias. ECG signals used are Normal Sinus Rhythm, Atrial Fibrillation, Supraventricular Arrhythmia, Long term ST change, Malignant Ventricular Ectopy, Arrhythmia. Signals are obtained from MIT-BIH (Massachusetts Institute of technology Beth Israel Hospital) database.