Entropy Index as a Measure of Heartbeat Irregularity

A method is proposed to analyze the heartbeat waveform that can yield a reliable characterization of the structure after only a few pulses. The measure suggested is entropy index that is related to the one found effective in describing chaotic behaviors in a wide variety of physical systems. When applied to the ECG data that include ventricular fibrillation, the index is shown to change drastically within a few pulses. Wavelet analysis is used to exhibit different scaling behaviors in different phases. Concepts from the theory of nonlinear dynamics and statistical physics have been applied to the study of nonstationary time series, such as human heartbeats [1, 2, 3, 4] and brain electrical activities[5, 6]. Those analyses generally involve the use of data recorded over a long period of time. In this paper we propose a method of analysis that needs only a short duration of the time series data, as short as, say, ten heartbeats, for example. Conceptually,