[Brain-computer interface-based motor imagery training for patients with neurological movement disorders].

AIM

To study the electric brain activity during motor imagery task in the brain-computer interface (BCI) in motor-disabled patients to determine the optimal ways for using BCI-based ideomotor training in medical rehabilitation.

MATERIAL AND METHODS

The study included 26 patients with arm motor dysfunction caused by a stroke or a spinal cord injury. They were involved in motor imagery training in the BCI. The power and localization of electroencephalographic (EEG) event-related desynchronization during imagery of different arm movements were measured. The accuracy in the two-command BCI was assessed.

RESULTS AND CONCLUSION

The pattern of imagery-related EEG desynchronization showed the typical localization for such tasks. Despite the fact that the power of EEG reactions during motor imagery in motor-disabled patients was on average lower than in healthy subjects during a similar task, all the patients were able to achieve high accuracy in the two-command BCI system after several (at least three) training sessions. Our results demonstrate the great potential for using BCI-based motor imagery training for neurorehabilitation of patients with motor dysfunctions.