Table 4 Comparison of different preprocessing and feature extraction methods using the proposed BPNN on EGGMMIDB (4-fold cross validation).

From: Towards decoding motor imagery from EEG signal using optimized back propagation neural network with honey badger algorithm

Pre-processing Method

Feature Extraction Method

Precision

Recall

Accuracy

F-Measure

Band pass filter41

FFT37

80.2 ± 1.0

81.0 ± 1.1

82.4 ± 1.1

80.5 ± 1.0

Statistical features38

82.5 ± 0.9

83.2 ± 0.9

83.8 ± 1.0

82.8 ± 0.9

ICA23

84.8 ± 0.8

85.0 ± 0.8

85.5 ± 0.9

84.9 ± 0.8

PCA + CSP39

86.8 ± 0.7

87.0 ± 0.8

87.4 ± 0.8

86.9 ± 0.7

FBCSP44

85.9 ± 0.9

86.5 ± 0.9

86.7 ± 0.8

86.2 ± 0.9

STFT40

85.8 ± 0.9

86.2 ± 0.8

86.1 ± 0.8

85.9 ± 0.8

PCMICSP10

91.0 ± 0.7

91.5 ± 0.6

91.8 ± 0.6

91.2 ± 0.6

Wavelet analysis42

FFT37

82.0 ± 1.0

82.3 ± 1.1

82.8 ± 1.1

82.1 ± 1.0

Statistical features38

83.8 ± 0.9

84.2 ± 0.8

84.5 ± 0.9

83.9 ± 0.9

ICA23

85.5 ± 0.9

86.0 ± 0.8

86.2 ± 0.8

85.7 ± 0.8

PCA + CSP39

88.0 ± 0.8

88.5 ± 0.7

88.6 ± 0.8

88.2 ± 0.7

FBCSP44

86.8 ± 0.8

87.2 ± 0.8

87.4 ± 0.8

86.9 ± 0.8

STFT40

86.5 ± 0.9

86.8 ± 0.8

87.0 ± 0.8

86.6 ± 0.8

PCMICSP10

92.0 ± 0.6

91.8 ± 0.6

91.8 ± 0.6

91.9 ± 0.6

Adaptive Noise Cancellation (ANC)43

FFT37

80.8 ± 1.0

81.2 ± 1.0

82.5 ± 1.0

80.9 ± 1.0

Statistical features38

82.7 ± 0.8

83.5 ± 0.9

83.9 ± 0.9

83.0 ± 0.8

ICA23

85.0 ± 0.8

85.7 ± 0.8

85.9 ± 0.8

85.3 ± 0.8

PCA + CSP39

87.2 ± 0.8

87.8 ± 0.7

88.0 ± 0.7

87.4 ± 0.7

FBCSP44

86.1 ± 0.8

86.6 ± 0.8

86.9 ± 0.8

86.3 ± 0.8

STFT40

86.0 ± 0.8

86.2 ± 0.8

86.5 ± 0.8

86.1 ± 0.8

PCMICSP10

91.2 ± 0.6

91.0 ± 0.7

91.0 ± 0.7

91.1 ± 0.6

HHT9

FFT37

82.0 ± 1.0

82.3 ± 1.1

82.8 ± 1.1

82.1 ± 1.0

Statistical features38

83.8 ± 0.9

84.2 ± 0.8

84.5 ± 0.9

83.9 ± 0.9

ICA23

85.5 ± 0.9

86.0 ± 0.8

86.2 ± 0.8

85.7 ± 0.8

PCA + CSP39

88.0 ± 0.8

88.5 ± 0.7

88.6 ± 0.8

88.2 ± 0.7

FBCSP44

86.8 ± 0.8

87.3 ± 0.8

87.5 ± 0.8

87.0 ± 0.8

STFT40

86.5 ± 0.9

86.8 ± 0.8

87.0 ± 0.8

86.6 ± 0.8

PCMICSP10

93.0 ± 0.6

93.5 ± 0.6

93.5 ± 0.6

93.4 ± 0.6