Wavelet-based Multifractal Analysis of Heartbeat Dynamics: Non-Gaussian Expansion vs. Cumulants

Nonlinear and non-Gaussian analyses contribute to a comprehensive characterization of autonomic nervous system control on heartbeat dynamics. Nevertheless, a statistical comparison between non-Gaussian features and cumulants in the frame of a wavelet-based multifractal analysis of heartbeat dynamics has not been performed yet. Here we exploit a multifractal features formulation based on wavelet p-leaders spectrum applied to instantaneous heartbeat estimates from inhomogeneous point processes. We then perform a non-Gaussian multiscale expansion and analyze physiologically-meaningful differences between resting state and cold-pressure test in 30 healthy subjects. Results show that nonlinear and non-Gaussian features are associated with statistical differences between physiological states, whereas cumulants from the multifractal spectrum up to the third order were not statistically different.