Multiscale entropy: Recent advances

Abstract

Multiscale entropy is a widely used metric for characterizing the complexity of physiological time series. The fundamental difference to classical entropy measures is it enables quantification of nonlinear dynamics underlying physiological processes over multiple time scales. The basic idea of multiscale entropy was initially developed in 2002 and has since witnessed considerable progress in methodological expansions along with growing applications. Here, we provide an overview of some recent developments in the theory, identify some methodological constraints of the originally introduced multiscale entropy analysis, and discuss some improvements that we, and others, have made regarding the definition of the time scales, its multivariate extension and improved methods for estimating the basic technique. Finally, the application of multiscale entropy to the analysis of cardiovascular data is summarized.