Investigating the use of neurofeedback to promote self-regulation of cortical motor activities as a potential rehabilitation tool post-stroke

Thesis

Stroke-induced damage often leads to a more bilateral motor activation pattern, with under-activation of affected ipsilesional motor areas and over-activation in undamaged contralesional (i.e., ipsilateral) motor areas during movement of the stroke-impaired limb. This change in the functional organization of the brain, affecting cortical laterality can significantly impact individuals’ ability to perform activities of daily living. Therefore, a potential rehabilitative strategy for post-stroke motor recovery is promoting self-regulation of cortical motor activity to rebalance this abnormal brain activation pattern.

This thesis explored the use of two different neuroimaging modalities as the mode of neurofeedback to promote self-modulation of cortical motor brain activities. fMRI offers a higher spatial resolution of activation over other neurofeedback technologies. On the other hand, EEG offers higher temporal resolution, lower cost and increased portability over other neurofeedback techniques, making it more amenable for widespread clinical use. By examining multiple neuroimaging modalities, this thesis hopes to explore the potential of various neurofeedback setups for motor rehabilitation after stroke.

The first experiment examined the effect of one fMRI neurofeedback training session on increasing the laterality of cortical motor activity during the movement execution in stroke patients using the “gold standard” of clinical trials. Using a double-blind, sham-controlled, between-subject design, 24 stroke patients underwent a single session of real or sham real-time fMRI neurofeedback training. Results suggest that in the presence of real neurofeedback, lateralisation was higher during the movement of the affected hand in stroke participants.

The second experiment investigated the potential of using a low-cost, mobile, wireless EEG system for neurofeedback training. Using a within-subject, sham-controlled design, fourteen healthy volunteers underwent four EEG neurofeedback training sessions to modulate event-related desynchronization(ERD) activity using a simple motor execution task. Functional MRI before and after the training were also collected to assess training effects and possible individual differences more comprehensively. Results suggest that neurofeedback training with a low-density EEG setup can be effective for increasing lateralised ERD. BOLD activity in the non-dominant hemisphere was shown to be a predictor for an individual’s ability to control neurofeedback self-regulated ERD activities using motor execution.

Taken together, this thesis provides a preliminary investigation of using fMRI and low-density EEG as modes of neurofeedback in the context of motor rehabilitation for stroke. Nevertheless, more studies are needed before translation to clinical practice can be achieved.

Authors: Huang, Y

• DOI: 10.5287/ora-nbvek0q88
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: sham; fMRI; neurofeedback; stroke; EEG; lateralisation; double-blind; rehabilitation; vent-related desynchronization(ERD)
• Subjects: Neurofeedback; fMRI; Rehabilitation; EEG; vent-related desynchronization(ERD); Stroke