fMRI-EEG correlations in an interactive study of the cerebral motor network in ischemic stroke

Neurofeedback (NF) technologies provide an interactive mode of self-regulation of the brain motor system in stroke rehabilitation. Simultaneous presentation of signals of two modalities, hemodynamic (functional MRI) and electrophysiological (EEG), in the feedback loop allows to compensate for the limitations of each neuroimaging technology and helps to reveal the cerebral mechanisms of motor recovery. Aim of the study was to investigate coordinated changes in fMRI signal and EEG sensorimotor rhythm power during leg movement imagination in fMRIEEG-NF sessions in a patient with stroke. Materials and methods . A patient with right-sided hemiparesis after a 6-week stroke was trained to imagine movements of the paretic foot in five fMRI-EEG-NF sessions. The feedback scale was 2/3 determined by the level of fMRI signal amplification from the supplementary motor area (SMA) and foot representation in the motor cortex (M1F) of the left hemisphere and 1/3 by desynchronization of EEG rhythms in the mu (8–13 Hz) and beta2 (18–26 Hz) frequency bands over the central midline region (electrode Cz according to the 10–20 system). Results and discussion . In most of the NF runs, the patient achieved desynchronization of the mu and beta2 rhythms of the EEG and an increase in the fMRI signal of the regions of interest. Activation of the left M1F was associated with desynchronization of mu rhythms, which is consistent with the literature. Suppression of both mu and beta2 bands correlated (p<0.05) with activation of the premotor cortex bilaterally, the right primary motor cortex, and the anterior thalamus and anterior insula of the right hemisphere, indicating the involvement of the intact hemisphere in motor planning and control. The predominance of activation of homologous regions of the undamaged hemisphere, as well as the involvement of the nodes of the salience network and the dorsal attention network are consistent with the hypothesis of global functional reorganization of the brain after stroke; the same is evidenced by the strengthening of the intrahemispheric functional connectivity SMA–M1F bilaterally by the end of the course. Conclusions. For the first time, an analysis of the relationships between fMRI and EEG signals in sessions on the bimodal fMRI-EEG NF platform was performed in a patient with post-stroke foot paresis. It was shown that volitional control of the activity of M1F and SMA of the affected hemisphere also activates homologous regions of the opposite hemisphere and recruits nodes of cognitive networks, demonstrating associations with the power of the muand beta2band of EEG in the central leads.