Abnormal Entropy Modulation of the EEG Signal in Patients With Schizophrenia During the Auditory Paired-Stimulus Paradigm

Abstract

The complexity change in brain activity in schizophrenia is an interesting topic clinically. Schizophrenia patients exhibit abnormal task-related modulation of complexity, following entropy of electroencephalogram (EEG) analysis. However, complexity modulation in schizophrenia patients during the sensory gating (SG) task, remains unknown. In this study, the classical auditory paired-stimulus paradigm was introduced to investigate SG, and EEG data were recorded from 55 normal controls and 61 schizophrenia patients. Fuzzy entropy (FuzzyEn) was used to explore the complexity of brain activity under the conditions of baseline (BL) and the auditory paired-stimulus paradigm (S1 and S2). Generally, schizophrenia patients showed significantly higher FuzzyEn values in the frontal and occipital regions of interest (ROIs). Relative to the BL condition, the normalized values of FuzzyEn of normal controls were decreased greatly in condition S1 and showed less variance in condition S2. Schizophrenia patients showed a smaller decrease in the normalized values in condition S1. Moreover, schizophrenia patients showed significant diminution in the suppression ratios of FuzzyEn, attributed to the higher FuzzyEn values in condition S1. These results suggested that entropy modulation during the process of sensory information and SG was obvious in normal controls and significantly deficient in schizophrenia patients. Additionally, the FuzzyEn values measured in the frontal ROI were positively correlated with positive scores of Positive and Negative Syndrome Scale (PANSS), indicating that frontal entropy was a potential indicator in evaluating the clinical symptoms. However, negative associations were found between the FuzzyEn values of occipital ROIs and general and total scores of PANSS, likely reflecting the compensation effect in visual processing. Thus, our findings provided a deeper understanding of the deficits in sensory information processing and SG, which contribute to cognitive deficits and symptoms in patients with schizophrenia.

Alteration of Power Law Scaling of Spontaneous Brain Activity in Schizophrenia

“…Yang et al (2015) applied entropy analysis to resting-state fMRI data from patients with schizophrenia and documented that the decreased resting-state brain activity complexity can be characterized by increased regularity in the temporal lobe and increased randomness pattern in the frontal cortex, which explains the patients' withdrawal behavior and auditory hallucinations. Nonlinear characteristics of brain signals were also reported in studies with various biomedical signal recording techniques, such as electroencephalogram (EEG) (Akar et al, 2016;Carlino et al, 2012;Cerquera et al, 2017;Di Lorenzo et al, 2015;Fernandez et al, 2013;Ibanez-Molina et al, 2018;Keshavan et al, 2004;Kim et al, 2000;Lee et al, 2008;Li et al, 2008;Raghavendra et al, 2009;Takahashi et al, 2010;Xiang et al, 2019;Yu et al, 2016) or magnetoencephalography (MEG) (Brookes et al, 2015;Fernandez et al, 2013;Fernandez et al, 2011;Kotini and Anninos, 2002;Robinson and Mandell, 2015).…”

Altered Complexity of Spontaneous Brain Activity in Schizophrenia and Bipolar Disorder Patients

“…3 Similar results were obtained in a study by Hager et al 10 However, in Yang's study, decreased entropies in SC patients were shown in the occipital, cingulate, frontal, temporal regions and cerebellum, while Hager found differences only in the frontal regions and thalamus, which are regions that also detected more brain activity than normal controls in other studies. [11][12][13] Regarding bipolar disorder, according to Hager's results, the occipital, precentral, frontal, and cingulate regions were found to have decreased entropies, although other studies have shown that these brain regions have increased entropies. 14 Several other regions, including the thalamus, temporal region, and cerebellum, have also shown abnormal complexity.…”

A hybrid deep neural network for classification of schizophrenia using EEG Data

“…The function of the two layers is formalized by Eqs. (19) and (20). www.nature.com/scientificreports/ Therefore, the cell state C t of the current chain is a combination of the reserved historical information of C t−1 , and the updating information selected from C t (Eq.…”

“…The entropy value measured by fuzzy entropy is continuously stable and less sensitive to the noise of EEG data, which renders it more suitable for analyzing chaotic signals. Previous studies have proven that the ability of FuzzyEn to detect and recognize signals is superior to the ability of other entropies for both epilepsy 18 and schizophrenia 19 .…”