Measure the structure in heart rate variability

It has been proposed that the heartbeat may display complex nonlinear dynamics, and that such behaviors may be useful for the diagnosis of risk of sudden death. Many efforts to apply such measures as the metric entropy, Lyapunov exponents, and fractal dimensions directly to physiological systems have met with considerable difficulties. According to the empirical explanation that the statistical complexity is the level of “randomness” caused by nonlinear dynamics, the authors presented a simple statistical complexity measure that is composed of Shannon entropy and Kullback Leibler information distance derived from time irreversibility. As a general indicator of correlative structure, it allowed reliable detection of periodic, quasi-periodic, linear stochastic and chaotic dynamics. The authors analyzed electrocardiograms from four groups of subjects. The RR intervals exhibited correlative structures in all subjects, more important, the estimated statistical complexities were correlated to different cardiac dynamics. These findings suggest that cardiac nonlinear dynamics is prevalent in heart, and the proposed statistical complexity measure may be a diagnostic tool for increased risk of sudden cardiac death