fMRI-EEG-neurofeedback in stroke rehabilitation: theory and practice (review)

Neurofeedback (NF) is a noninvasive neuromodulation technology achieved by teaching the subject the skill of selfregulation of certain parameters of their own brain activity in a feedback loop. It is believed that such mental training leads to changes in the functional architecture of global cerebral networks, involving neuroplasticity mechanisms, and therefore may have potential in stroke rehabilitation. EEG-NF has a long history and traditionally uses frequency bands of EEG rhythms associated with known behavioral functions as an adjustable parameter. The development of MRI technology has made it possible to obtain topographically accurate functional images of the brain (fMRI) in real time with the prospect of creating fMRI-NF-platforms. Online fusion of signals from two modalities (EEG and fMRI) in a feedback loop for self-regulation training (fMRI-EEG-NF) is attractive due to its therapeutic and research potential, promising a more detailed understanding of spatio-temporal dynamics in the brain after stroke, which is impossible to obtain using each modality separately. However, such a multimodal functional neuroimaging requires non-trivial hardware and computational solutions. The objective of the review was to trace the vector of development of the NF technology in relation to stroke rehabilitation in the historical and technological aspect. For this purpose, the theoretical and practical prerequisites for fusion of fMRI and EEG signals in a feedback loop are summarized and data on the effectiveness of NF methods as a scientifically based method of recovery after stroke are presented.