Skip to main content
Fig. 3 | Journal of NeuroEngineering and Rehabilitation

Fig. 3

From: Competing at the Cybathlon championship for people with disabilities: long-term motor imagery brain–computer interface training of a cybathlete who has tetraplegia

Fig. 3

Time-varying decoding accuracy (DA) and TSD results for the 2-class classifiers. AD Comparison of the time-varying DA obtained for the four binary classifiers (LR, FR, LF, and TX). A Time-varying DA obtained from cross-validation (CV) during BCI calibration using dataset recorded in 2019 at sessions 2, 4, 5. B Cross-session average of time-varying DA obtained in 2019 at sessions 6–10. C Time-varying DA obtained from cross-validation (CV) during BCI calibration using dataset recorded in 2020 at sessions 3, 4, 5, 6. D Cross-session average of time-varying DA obtained in 2020 sessions 7–13. E Comparison of time-varying decoding accuracy obtained in 2019 vs. 2020 for each classifier (LR, FR, LF, and TX), separately. F and G Time-varying TSD values obtained for each binary classifier in 2019 during sessions 6–10 (B) and 2020 during sessions 7–13 (E). The onset of the task performance is indicated in each plot with a vertical dotted line (at 2 s). The solid lines in shaded areas of E–G panels indicate the mean value from analyzed sessions, and the shaded area is the standard deviation

Back to article page