Table 3 Performance of TL, TL-BLSTM, and TL-BLSTM-attention models based on EfficientNetB0 for different brain regions on unseen test data. The mean and standard deviation of each performance measure is presented.

From: A convolutional recurrent neural network with attention for response prediction to repetitive transcranial magnetic stimulation in major depressive disorder

  

Frontal

Central

Temporal

Parietal

Occipital

VGG16

ACC (%)

89.2 ± 4.6

86.5 ± 3.3

86.9 ± 4.1

87.1 ± 3.8

88.9 ± 4.5

SEN (%)

90.2 ± 3.8

86.4 ± 3.0

88.3 ± 2.5

89.4 ± 2.8

90.3 ± 3.5

SPE (%)

89.5 ± 3.4

86.0 ± 3.5

85.2 ± 3.1

85.5 ± 3.5

86.4 ± 3.7

AUC

0.88 ± 0.03

0.85 ± 0.04

0.84 ± 0.05

0.86 ± 0.06

0.86 ± 0.05

VGG16-BLSTM

ACC (%)

95.2 ± 2.9

93.6 ± 1.9

92.2 ± 2.5

93.9 ± 2.4

94.4 ± 2.9

SEN (%)

95.5 ± 1.5

95.4 ± 2.7

93.4 ± 2.9

94.5 ± 2.6

95.3 ± 3.0

SPE (%)

94.9 ± 3.1

93.2 ± 2.8

91.8 ± 2.9

92.5 ± 2.1

93.9 ± 2.3

AUC

0.95 ± 0.03

0.91 ± 0.03

0.93 ± 0.05

0.92 ± 0.03

0.94 ± 0.04

VGG16-BLSTM-Attention

ACC (%)

96.8 ± 2.4

95.2 ± 0.9

94.5 ± 1.0

94.7 ± 2.0

96.1 ± 1.7

SEN (%)

97.7 ± 1.2

95.4 ± 1.0

95.1 ± 1.5

95.5 ± 1.6

96.6 ± 0.9

SPE (%)

96.4 ± 1.4

94.8 ± 1.6

94.1 ± 1.5

94.5 ± 1.6

95.3 ± 0.8

AUC

0.96 ± 0.03

0.94 ± 0.02

0.93 ± 0.04

0.94 ± 0.04

0.95 ± 0.03

  1. Significant values are in bold.