Using time causal quantifiers to characterize sleep stages

Abstract

Resent studies have found that sleep is closely connected with mood, playing substantial role in daily cognitive performance and memory. Sleep is a dynamic activity presenting different stages which progresses cyclically throughout night. It is very important to maintain these cycles for healthy body function when the person is awake. The construction of mathematical models of sleep that help us characterize the different sleep stages with an optimal use of sleep recording time is of paramount importance. For this purpose, in this work, we analyzed single-channel EEG signals from 17 healthy subjects. The signals were quantified using the permutation vector approach and applied five different information measures: i) Shannons entropy, ii) MPR statistical complexity, iii) Fisher information, iv) Renyì Min-entropy and v) Lempel-Ziv complexity over the signals. The results show that it is possible to quantify and classify the different sleep stages. In addition, we combine these measures in complexity/information-entropy space, extracting relevant information from the signals impossible to obtain using the measures separately. Information theory-based measures allow us to characterize the underlying dynamics of EEG time series, identifying the different sleep stages and defining electrophysiological biomarkers able to distinguish between pathological and normal conditions.

Gamma-Divergence. An Introduction to New Divergence Family

“…The presence of 11-16 Hz activity (sleep spindles) in N1, and more abundant alpha activity (8-13Hz) in REM sleep means that these two stages present activity at an overlapping frequency range, which explains the proximity of the divergence values obtained. Difficulty in detecting N1 and REM sleep has also been found using other measures [28,29].…”

Low Frequency Oscillations drive EEG’s complexity changes during wakefulness and sleep

“…Complex neural dynamics are thought to be necessary for consciousness (Tononi and Edelman, 1998;Oizumi et al, 2014). Different reports show that cortical activity exhibits complex patterns during Wakefulness, that are reduced during deep NREM sleep (Ouyang et al, 2010;Nicolaou and Georgiou, 2011;Abásolo et al, 2015;Schartner et al, 2017;Bandt, 2017;González et al, 2019González et al, , 2020Hou et al, 2021;Mondino et al, 2021;Mateos et al, 2021) or anesthesia (Jordan et al, 2008;Sitt et al, 2014;Sarasso et al, 2015;Fagerholm et al, 2016;Thul et al, 2016;Varley et al, 2020Varley et al, , 2021. However, it was unclear how the different frequency bands contribute to the observed complexity changes in EEG analyses.…”

“…Under this framework, it has been shown that EEG's complexity changes according to the behavioural state, but irrespective of the animal species (including mice, rats, cats, monkeys, and humans). In particular, it has been consistently reported (Nicolaou and Georgiou, 2011;Abásolo et al, 2015;Bandt, 2017;González et al, 2019González et al, , 2020Mateos et al, 2021;Varley et al, 2021;González et al, 2021;Pascovich et al, 2021) that Wake is a highly complex state, that complexity decreases during NREM when consciousness is lost, and that it increases during REM sleep when a state of altered consciousness emerges, i.e., dreams. However, it is still unclear how these complexity results are related to the classical frequency bands during the sleep-wake states.…”

“…Recently, the classical analysis of EEG per frequency bands has been complemented by the study of EEG's complexity (Jordan et al, 2008;Ouyang et al, 2010;Nicolaou and Georgiou, 2011;Sitt et al, 2014;Abásolo et al, 2015;Sarasso et al, 2015;Bandt, 2017;González et al, 2019González et al, , 2020Varley et al, 2020;Hou et al, 2021;Mateos et al, 2021;Varley et al, 2021;González et al, 2021); Sarasso et al (2021) provides a up-to-date literature review. Complexity analyses usually focus on the EEG signal as a whole, instead of its frequency components (or bands).…”

Complexity of Brain Dynamics as a Correlate of Consciousness in Anaesthetized Monkeys

“…In the literature there are different types of complexity-entropy planes [31][32][33], in this work we focus on the Permutation Lempel-Ziv complexity vs. permutation entropy plane (LZ × P E ) [34]. This plane has been used to distinguish between chaotic and random signals [34], to analyse electrophysiological signals in altered states of consciousness [35] and to characterise sleep states [36].…”