Multilag extension of quadratic sample entropy for distress recognition with EEG recordings

Abstract

Distress has become one of the major issues in developed countries because of its negative effects in physical and mental health. In order to control its consequences, a number of researchers have studied distress from an electroencephalographic point of view by means of the use of different nonlinear metrics. However, those studies are only based on non-lag approaches, thus many nonlinear dynamics of brain signals could not be properly assessed. In this sense, this work applies a multilag extension of a nonlinear regularity-based metric called quadratic sample entropy, in order to check the influence of the selection of a time lag for the recognition of distress with electroencephalographic recordings.