Multi-channel analysis of the EEG in Toutette’s syndrome during tic suppression

Abstract

Tourette\`s syndrome (TS) characterized by sudden, repetitive, and stereotyped motor and vocal tics has been known an involuntary. But most TS patients can suppress their tics consciously for some time period. We can infer from it that some cortical parts are involved in the inhibition of tic occurrence. The goal of this study is to find where brain regions concern in tic suppression in TS patients by several linear and nonlinear methods underlying the electroencephalogram (EEG). In this paper we calculated Power spectrum (PS) and Approximate entropy (ApEn) as a single channel analysis to find out how different regularity and complexity exists in each EEG channels, and also estimated Coherence and Transfer entropy (TE) as a multi channel analysis to measure the degree of functional connectivity among EEG channels obtained from 1) TS patients and normal controls (NC) in resting state, 2) tic suppression and resting state for TS patients. Especially TE is, a new nonlinear method which can quantify the direction and amount of information transition between two channels, estimated to find out the cortical areas on the scalp relating to tic suppression of TS patients. We used parzen window estimation method to get the probability density function (PDF) which is less dependent on the bin size for TE analysis. The comparison results between TS patients and normal controls in resting state showed that TS patients had higher power at frontal and temporal regions (Fz, T3, T5, and T6) for high frequency bands (slow-beta, fast-beta, and gamma) and more complex at frontal regions (F3, F4), and increased functional connectivity between frontal, central, temporal, and parietal regions than NC. Similarly, in case of comparison results between tic suppression and resting state for TS patients, TS patients in tic suppression showed higher power at frontal, central, and temporal regions (Fz, C4, T3, and T4) over all frequency bands except alpha band. The functional connectiv...