Using a functional near-infrared spectroscopy-guided brain-computer interface to facilitate observational imitation after stroke

Background: A brain-computer interface (BCI) shows promise for facilitating motor imagery (MI) during observational imitation motor relearning of the upper extremity in people after stroke. Objective: To investigate the efficacy and mechanisms of a functional near-infrared spectroscopy (fNIRS)-based BCI in augmenting MI during observational imitation for poststroke upper extremity rehabilitation. Methods: A randomized trial was conducted among participants after stroke. In the real fNIRS-BCI group, participants engaged in kinesthetic MI. When the activation level over the corticomotor areas recorded by fNIRS surpassed a predefined threshold, an instructional video showing the target movement was triggered, and participants were instructed to observe and imitate the movement. The sham group received feedback at constant intervals without being contingent on individual brain signals. Upper extremity motor tests and mirror visual feedback (MVF)-induced sensorimotor event-related desynchronization (ERD) were assessed before and after intervention. MI-induced oxygenated hemoglobin (HbO) concentrations were extracted from participants receiving fNIRS-BCI. Results: Forty-four participants were enrolled. Observational imitation with or without BCI was effective in enhancing upper extremity function. However, there were no between-group differences in upper extremity motor improvement. fNIRS-BCI-driven observational imitation significantly enhanced MVF-induced beta sensorimotor ERD bilaterally more than sham BCI did. In participants receiving fNIRS-BCI, the capacity to upregulate MI-induced HbO over the ipsilesional sensorimotor cortex and supplementary motor area was significantly enhanced post-intervention. Conclusion: fNIRS-BCI shows promise for monitoring real-time brain activity during rehabilitation and enhancing the participants' ability to upregulate corticomotor activity through neurofeedback; however, it did not yield superior benefits in upper extremity measures. fNIRS-BCI may improve brain responsiveness to visual feedback after stroke. Future research should determine how these neurophysiological effects can be translated into better clinical outcomes.