Development and application of methods for real-time fMRI neurofeedback.

摘要

Improving stroke recovery is a topic of considerable interest in the developed world. Motor deficits following stroke can significantly impact quality of life, and motor rehabilitation strategies are urgently needed to promote brain recovery. One technique that has had therapeutic success in other domains is neurofeedback (NF), a strategy whereby a person is trained to gain volitional control of their neurological activity. The NF technique has proven efficacious in disorders ranging from epilepsy to attention deficit disorders based on feedback of electroencephalography (EEG) signals, and the potential exists for NF applications in the recovery of motor function following stroke. Using functional magnetic resonance imaging (fMRI), NF can be applied to precisely defined regions of interest (ROIs), facilitating the targeted treatment of affected functional areas. The combination of fMRI and NF is still relatively new, and much work remains in the characterization and optimization of fMRI NF strategies.;In a two-pronged approach, this thesis focuses on the application of ROI-based NF to the motor system using traditional fMRI measurements and also the development and analysis of acquisition strategies intended for use in ROI-based NF. First, a study of the mechanisms governing the successful application of NF in primary motor cortex ROIs is presented, using kinaesthetic motor imagery (imagining the sensation and execution of movement) to engage the motor system. Second, an investigation of fMRI signal contrast enhancement properties using multi-echo fMRI acquisition methods in dense sampling regimes is considered. Third, a novel acquisition method is introduced, designed using parallel MRI principles to provide fast and detailed sampling of fMRI signals in selected ROIs, called Constrained Source Space Imaging (CSSI). To conclude, the potential future directions for fMRI NF research and the CSSI technique are discussed, including thoughts toward the continued development of NF as a potential motor therapy for stroke patients.