Comparison of nonlinear indices in analyses of heart rate variability.

Recent advances in study of heart rate variability (HRV) call for a paradigm shift from linear to nonlinear analysis. However, whether HRV per se is chaotic or merely random has remained an open problem. Myriad idealized measures of nonlinear dynamics or fractals which have been proposed in the literatures to characterize HRV are limited by their sensitivity to discernable noise. In this work, we compare the robustness of some nonlinear indices and evaluate their applicability to HRV analysis in the presence of noise.