Skip to main content
. 2020 Aug 14;20(16):4572. doi: 10.3390/s20164572

Table 1.

List of electroencephalogram (EEG) features evaluated in the present study.

Feature Mathematical Expression
Power Spectral Density (PSD) 1 1Nn=1Nxnei2πfn/N 
Differential Asymmetry (DASM) Difference between PSDs of
interhemispheric electrode pairs
Rational Asymmetry (RASM) Ratio between PSDs of
interhemispheric electrode pairs
Hjorth Parameters 2 [30] Activityx=1Ni=1Nxiμx2
Mobilityx=σ(x)σx
Complexityx=MobilityxMobilityx
Shannon Entropy 3 [31] i=1Np(xi)lnpxi , where i=1Np(xi)=1
Hurst Exponent 4 [32] logR/S/logN
Kolmogorov Complexity 5 [33] c(n)/b(n)
Higher-order Cumulants 6 [34] EXkXk+mXk+n
Common Spatial Pattern 7 [35] fp=logvarZpi=12mvarZi, where p = 1 to 2m

1x represents the EEG time-series data and N indicates the length of data. 2 x represents the EEG-time series data, μx represents the mean of x, x’ represents the derivative of x, and σ(x) represents the standard deviation of x. 3 x represents the EEG time-series data. 4 N is the length of the sample of data, R is the difference between the maximum deviation from the mean and the minimum deviation from the mean, and S is the standard deviation. 5 n is the length of the time series data and b(n) is the ratio between n and log(n). 6 X(n) is the Fourier transform of the signal x(n) and E[·] stands for the expectation operation. 7 Zp indicate the spatially filtered EEG signals associated with the largest p eigenvalues of sum of covariance matrix. More details can be found in [35]. Note that the value of p was set to 2 in this study.