Table 3.
The mean precision, recall and F-score of the different two classes of four sessions using S3 classification strategy in the decoder-training phase across all subjects (in %).
| Session 1 (MI) | Session 2 (ME) | Session 3 (MI) | Session 4 (ME) | |||||
|---|---|---|---|---|---|---|---|---|
| Left Hand | Right Hand | Left Hand | Right Hand | Left Hand | Both Feet | Left Hand | Both Feet | |
| Precision | 76.86 ± 2.38 | 75.91 ± 0.92 | 78.71 ± 1.27 | 79.69 ± 2.36 | 87.55 ± 3.24 | 88.30 ± 2.03 | 90.70 ± 1.37 | 91.96 ± 2.07 |
| Recall | 75.48 ± 1.09 | 77.27 ± 1.33 | 80.03 ± 0.74 | 78.45 ± 1.04 | 88.42 ± 3.03 | 87.43 ± 1.92 | 92.08 ± 0.87 | 90.56 ± 1.90 |
| F-score | 76.16 ± 0.71 | 76.58 ± 0.31 | 79.36 ± 0.92 | 79.07 ± 0.48 | 87.99 ± 1.06 | 87.86 ± 0.77 | 91.38 ± 0.92 | 91.25 ± 0.88 |
MI: motor imagery, ME: motor execution.