Table 1 Averaged classification accuracies (%) and ITRs (bits/min) in brackets of seven compared models across all subjects in the Benchmark datasets at the data lengths of 0.2 s, 0.4 s, 0.6 s, 0.8 s and 1.0 s.

From: Enhancing the performance of SSVEP-based BCIs by combining task-related component analysis and deep neural network

Methods

Data length (s)

0.2

0.4

0.6

0.8

1.0

CCA

3.14 (0.39)

8.92 (6.06)

19.88 (23.50)

34.77 (48.55)

47.60 (67.88)

FBCCA

3.51 (0.58)

12.94 (13.17)

31.17 (48.94)

49.77 (83.69)

64.17 (106.24)

TRCA

31.50 (80.72)

55.27 (144.23)

70.37 (169.13)

83.04 (181.52)

88.43 (174.00)

Compact-CNN

7.75 (5.59)

21.57 (31.07)

33.81 (52.50)

45.58 (70.58)

54.62 (81.12)

Conv-CA

48.65 (137.86)

77.42 (223.86)

83.91 (209.19)

88.58 (194.12)

92.99 (183.40)

bi-SiamCA

51.59 (163.81)

76.50 (229.45)

86.10 (226.16)

91.38 (211.17)

94.07 (91.73)

eTRCA + sbCNN

58.88 (197.76)

81.18 (249.47)

89.55 (239.08)

94.45 (220.94)

96.19 (197.48)

Back to article page