Skip to main content

Table 5 Performance of transformer model for EEG signals with eyes close on different methods for three class (ADHD, MDD, and SMC) classification

From: Transformers for autonomous recognition of psychiatric dysfunction via raw and imbalanced EEG signals

Decision

Method

Accuracy

F1-score

Precision

Recall

Window-level

Trans + CCE

67.97

64.86

66.25

63.92

Trans + Focal Loss (\(\gamma\) = 2)

69.87

60.55

66.21

57.48

Trans + Focal Loss (\(\gamma\) = 0.5)

68.64

59.13

63.61

63.54

Trans + WCCE

66.18

62.63

67.15

61.15

Patient-level

Trans + CCE

67.19

64.83

72.75

60.92

Trans + Focal Loss (\(\gamma\) = 2)

70.31

61.19

75.52

56.05

Trans + Focal Loss (\(\gamma\) = 0.5)

70.31

61.87

66.53

65.36

Trans + WCCE

65.62

60.91

72.62

57.43

  1. The best performance is in bold