Table 3 Comparison of different preprocessing and feature extraction methods using the proposed BPNN on EGGMMIDB (2-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

79.5\(\:\pm\:\)1.2

78.8\(\:\pm\:\)1.3

81.5\(\:\pm\:\)1.4

79.1\(\:\pm\:\)1.4

Statistical features38

81.2\(\:\pm\:\)1.1

82.5\(\:\pm\:\)1.0

83.8\(\:\pm\:\)1.2

81.8\(\:\pm\:\)1.2

ICA23

83.0\(\:\pm\:\)1.0

84.2\(\:\pm\:\)1.1

84.6\(\:\pm\:\)1.1

83.6\(\:\pm\:\)1.0

PCA + CSP39

85.2\(\:\pm\:\)1.0

86.0\(\:\pm\:\)1.0

86.1\(\:\pm\:\)1.0

85.6\(\:\pm\:\)1.0

FBCSP44

86.1 ± 1.0

87.0\(\:\pm\:\)1.0

87.2 ± 1.0

86.5 ± 1.0

STFT40

84.0\(\:\pm\:\)1.1

85.0\(\:\pm\:\)1.1

84.9\(\:\pm\:\)1.3

84.4\(\:\pm\:\)1.2

PCMICSP10

88.8\(\:\pm\:\)0.9

89.5\(\:\pm\:\)0.9

89.3\(\:\pm\:\)0.9

89.1\(\:\pm\:\)0.9

Wavelet analysis42

FFT37

81.8\(\:\pm\:\)1.3

82.0\(\:\pm\:\)1.2

82.8\(\:\pm\:\)1.3

81.9\(\:\pm\:\)1.3

Statistical features38

83.5\(\:\pm\:\)1.2

84.0\(\:\pm\:\)1.1

84.2\(\:\pm\:\)1.2

83.7\(\:\pm\:\)1.1

ICA23

85.0\(\:\pm\:\)1.1

85.8\(\:\pm\:\)1.1

85.7\(\:\pm\:\)1.1

85.4\(\:\pm\:\)1.1

PCA + CSP39

87.2\(\:\pm\:\)1.0

87.8\(\:\pm\:\)1.0

87.4\(\:\pm\:\)1.0

87.5\(\:\pm\:\)1.0

FBCSP44

88.0 ± 1.0

88.5\(\:\pm\:\)1.0

88.6 ± 1.0

88.2 ± 1.0

STFT39

85.2\(\:\pm\:\)1.2

85.5\(\:\pm\:\)1.2

85.3\(\:\pm\:\)1.2

85.3\(\:\pm\:\)1.2

PCMICSP10

90.\(\:2\pm\:\)0.8

90.0\(\:\pm\:\)0.8

90.1\(\:\pm\:\)0.8

90.1\(\:\pm\:\)0.8

Adaptive Noise Cancellation (ANC)43

FFT37

79.8\(\:\pm\:\)1.3

80.2\(\:\pm\:\)1.2

81.5\(\:\pm\:\)1.4

80.0\(\:\pm\:\)1.2

Statistical features38

82.\(\:0\pm\:\)1.2

82.5\(\:\pm\:\)1.2

83.0\(\:\pm\:\)1.3

82.2\(\:\pm\:\)1.2

ICA23

83.5\(\:\pm\:\)1.2

84.5\(\:\pm\:\)1.2

84.8\(\:\pm\:\)1.2

83.9\(\:\pm\:\)1.1

PCA + CSP39

86.0\(\:\pm\:\)1.0

86.8\(\:\pm\:\)1.0

86.7\(\:\pm\:\)1.1

86.3\(\:\pm\:\)1.0

FBCSP44

86.8 ± 1.0

87.5\(\:\pm\:\)1.0

87.7 ± 1.1

87.0 ± 1.0

STFT40

85.0\(\:\pm\:\)1.0

85.5\(\:\pm\:\)1.0

85.2\(\:\pm\:\)1.1

85.2\(\:\pm\:\)1.1

PCMICSP10

89.5\(\:\pm\:\)0.9

89.3\(\:\pm\:\)0.9

89.4\(\:\pm\:\)0.9

89.4\(\:\pm\:\)0.9

HHT9

FFT37

80.5 ± 1.2

80.0 ± 1.2

81.0 ± 1.3

80.2 ± 1.2

Statistical features38

82.2 ± 1.1

82.8 ± 1.1

83.5 ± 1.2

82.5 ± 1.1

ICA23

84.0 ± 1.0

84.5 ± 1.0

84.8 ± 1.1

84.2 ± 1.0

PCA + CSP39

86.5 ± 1.0

87.0 ± 1.0

86.8 ± 1.0

86.6 ± 1.0

FBCSP44

87.2 ± 1.0

87.8 ± 1.0

87.9 ± 1.0

87.5 ± 1.0

STFT40

85.5 ± 1.1

86.0 ± 1.1

85.7 ± 1.1

85.6 ± 1.1

PCMICSP10

90.5 ± 0.8

90.5 ± 0.8

90.7 ± 0.8

90.6 ± 0.8