Skip to main content
. 2021 May 14;7:e523. doi: 10.7717/peerj-cs.523

Table 2. The extracted 25-time domain features from each band of EEG signals in our study.

Number of the feature
in EEG signals dataset
Name of the feature Its equation
1 Kurtosis xkur=i=1n(x(i)x¯)4(n1)S4
2 Skewness xske=i=1n(xix¯)3(n1)S3
3 *IQR IQR=iqr(x)
4 DK DK=(S/x¯)100
5 Geometric Mean G=x1++xnn
6 Harmonic Mean H=n/(1x1++1xn)
7 Activity-Hjort Parameters A=S2
8 Mobility-Hjort Parameters M=S12/S2
9 Complexity-Hjort Parameters C=(S22/S12)2(S12/S2)2
10 *Maximum xmax=max(xi)
11 Median x~={xn+12:xodd12(xn2+xn2+1):xeven
12 *Mean Absolute Deviation MAD=mad(x)
13 *Minimum xmin=min(xi)
14 *Central Moments CM=moment(x,10)
15 Mean x¯=1ni=1n=1n(x1++xn)
16 Average Curve Length CL=1ni=2n|xixi1|
17 Average Energy E=1ni=1nxi2
18 Root Mean Squared Xrms=1ni=1n|xi|2
19 Standard Error Sx¯=S/n
20 Standard Deviation S=1ni=1n(xix¯)
21 Shape Factor SF=Xrms/(1ni=1n|xi|)
22 *Singular Value Decomposition SVD=svd(x)
23 *25% Trimmed Mean T25=trimmean(x,25)
24 *50% Trimmed Mean T50=trimmean(x,50)
25 Average Teager Energy TE=1ni=3n(xi12xixi2)

Note:

The feature was computed using MATLAB, IQR Interquartile Range, DK Coefficient of Variation. In formulas, x it represents the signal. The properties with “*” were calculated with the help of the MATLAB library.