A Study of Non-Gaussian Properties in Emotional EEG in Stroke Using Higher-Order Statistics

The stroke patients often suffered from emotional disturbances, and this leads to perceive emotions differently than normal control subjects; the emotional impairment of the stroke patients can be effectively analyzed using EEG signal. The EEG signal has been known to have non-Gaussian properties, and the non-Gaussianity characteristics of the EEG differ under different emotional states. The analysis of non-Gaussianity in EEG signal was performed by using higher-order statistics measures such as the skewness and kurtosis. In this study, the non-Gaussianity was examined in the emotional EEG signal of stroke patients and normal control subjects. The estimation of the emotional EEG distribution from the results was symmetrically non-Gaussian for both stroke and normal groups. Particularly, it was found that the normal subjects have more non-Gaussian EEG distribution than the stroke patients.