Based on information theory, a number of entropy measures have been proposed since the 1990sto assess systems’ irregularity, such as approximate entropy, sample entropy, permutation entropy,intrinsic mode entropy, and dispersion entropy to cite only a few. Among them, sample entropy hasbeen used in a very large variety of disciplines for both univariate and multivariate data. However,improvements to the sample entropy algorithm are still being proposed because sample entropy isunstable for short time series, may be sensitive to parameter values, and can be too time-consumingfor long data.At the same time, it is worth noting that sample entropy does not take into account the multipletemporal scales inherent in complex systems. It is maximized for completely random processesand is used only to quantify the irregularity of signals on a single scale. This is why analyses ofirregularity—with sample entropy or its derivatives—at multiple time scales have been proposed toassess systems’ complexity.This Special Issue invited contributions related to new and original research based on the use ofsample entropy or its derivatives. The papers published in this Special Issue can be divided into twocategories. First, some papers present new applications of sample entropy or its derivatives. Second,some papers propose improvements to sample entropy or its derivatives.