Proceedings of the 17th Iranian Conference of Biomedical E ngineering (ICBME 2010), 3-4 November 2010 978-1-4244-7484-4/10/$26.00 ©2010 IEEE Figure 1. Poincare plot of RR intervals of a healthy person with its standard descriptors SD1 and SD2 Utilizing Occurrence Sequence of Heart Rate’s Phase Space Points in order to Discriminate Heart Arrhythmia S Moharreri, S Parvaneh, N Jafarnia Dabanloo Islamic Azad University, Science and Research Branch Tehran, Iran sadaf.moharreri@gmail.com; parvaneh@ieee.org n_jafarnia@yahoo.com A M Nasrabadi Shahed University Tehran, Iran Abstract— Poincare plot analysis of RR time series allows a beat- to-beat approach to Heart Rate Variability (HRV), detecting patterns associated with nonlinear processes. Since the measurement of standards descriptors of Poincare plot is based on the point's distribution in relation to the line of identity (y=x), we have concentrated on it and evaluated the points behavior related to this line. For this purpose, we test two global and local analyses of points against the identity line. For evaluating these two novel features of Poincare plot, we try to use them for distinguishing four groups of subjects (Arrhythmia, Congestive Heart Failure (CHF), Atrial Fibrillation (AF) and Normal Sinus Rhythm (NSR)). Kruskal-Wallis test was used to define the level of significance of features. The results show that global feature discriminate different groups by p<6E-7, and local feature discriminate them by p<2E-7. Keywords- poincare plot; heart rate variability; temporal variations; occurrence I. INTRODUCTION Heart rate is an indicator of heart's condition [1]. Assessment of heart rate has been shown to aid clinical diagnosis and intervention strategies. It has been proved that nonlinear analysis of it might provide more valuable information for the physiological interpretation of heart rate fluctuations [2]. However, the variety of contradictory reports in this field indicates that there is a need for a more rigorous investigation of methods as aids to clinical evaluation [2]. The nonlinear analysis of Heart Rate Variability (HRV) is a valuable tool in both clinical practice and physiological research reflecting the ability of the cardiovascular system. The Poincare plot is a tool developed by Henry Poincare for analyzing complex systems [1]. It has found its use in such diverse fields as physics and astronomy, geophysics, meteorology, mathematical biology and medical sciences. In the context of medical sciences it is mainly used for quantifying HRV and proves to be quite an effective measure of this marker [3]. Poincare plot is a geometrical representation of RR time series to demonstrate patterns of heart rate dynamics resulting from nonlinear processes. Poincare plot analysis of RR time series allows a beat-to-beat approach to HRV, detecting patterns associated with nonlinear processes. It is a familiar method for the analysis of two-dimensional nonlinear dynamic systems[4]. Tulppo et. al. [5] fitted an ellipse to the distribution of the poincare plot and defined two standard descriptors SD1 and SD2 for quantification of the poincare plot geometry. These standard descriptors represent the minor axis and the major axis of the ellipse (Fig. 1) and guide the visual inspection of the distribution. In case of HRV, it reveals a useful visual pattern of the RR interval data by representing both short and long term variations of the signal [6]. But standard descriptors SD1 and SD2 are linear statistics and hence the measures do not directly quantify the nonlinear temporal variations in the time series contained in the poincare plot. Moreover, the limitations of the SD1/SD2 analysis are important to understand when attempting to investigate the physiological mechanisms in a time series, or when analyzing data where the occurrence of nonlinear behavior may be a distinguishing feature between health and disease[6]. The identity line (y = x) in the poincare plot has a simple physiological interpretation: the points on this line correspond to equal consecutive RR intervals, the points above it correspond to increasing heart rate and the points below this line to decreasing heart rate[7].Since the measurement of standards descriptors of poincare plot is based on the point's distribution in relation to the line of identity (y=x), we have concentrated on it and evaluated the points behavior related to