Table 3 Three-class accuracy of evaluated CNN training strategies using the CWT-extracted vectors for the considered strategies: without spatial dropping (CWT), spatial dropping by removing all sensorimotor electrodes out (CWT*), spatial dropping by excluding nonrelevant electrodes (CWT**)
Subjects | [43] | [44] | CWT | \(\kappa \) | CWT* | \(\kappa \) | CWT** | \(\kappa \) |
|---|---|---|---|---|---|---|---|---|
A03T | 86.1 | 76.7 | 83.6 ± 6.6 | 0.75 | 84.6 ± 8.3 | 0.76 | 82.0 ± 5.6 | 0.73 |
A07T | 83.6 | 82.8 | 73.5 ± 9.5 | 0.59 | 73.9 ± 7.5 | 0.60 | 75.0 ± 7.4 | 0.63 |
A08T | 63.1 | 78.2 | 73.0 ± 9.0 | 0.59 | 76.0 ± 9.5 | 0.64 | 75.1 ± 10.4 | 0.63 |
A01T | 80.8 | 84.1 | 73.0 ± 6.6 | 0.59 | 73.4 ± 6.1 | 0.60 | 75.7 ± 5.5 | 0.63 |
A09T | 54.3 | 75.2 | 72.2 ± 5.6 | 0.58 | 69.4 ± 5.7 | 0.58 | 73.0 ± 7.6 | 0.60 |
A04T | 54.0 | 62.4 | 69.7 ± 5.4 | 0.47 | 69.7 ± 6.9 | 0.47 | 69.9 ± 5.6 | 0.47 |
A06T | 51.3 | 51.8 | 68.3 ± 6.7 | 0.49 | 68.8 ± 8.5 | 0.52 | 68.4 ± 6.8 | 0.52 |
A02T | 71.2 | 57.7 | 65.0 ± 5.5 | 0.44 | 58.9 ± 6.5 | 0.46 | 62.4 ± 8.8 | 0.46 |
A05T | 52.2 | 48.1 | 59.1 ± 4.2 | 0.37 | 58.5 ± 5.6 | 0.36 | 57.5 ± 6.0 | 0.34 |
Average | 66.2 | 68.5 | 70.8 ± 6.5 | 0.54 | 70.3 ± 7.0 | 0.55 | 71.2 ± 7.0 | 0.56 |