Abstract 18886: A Thought-Controlled Immersive Virtual Reality Platform for Motor Learning Applied With Cortical and Basal Ganglia Stroke Survivors
An immersive, virtual reality platform using a sensorimotor rhythm- brain-computer interface was used in a cohort of 6 hemiparetic stroke survivors. The subject’s controlled the system with accuracies of up to 81% in a binary classification task. The 3D virtual reality platform displayed photorealistic human hands moving in coordination with the patient’s thoughts, overlaying the patient’s own hands.
Background: The National Stroke Association estimated that stroke management cost the United States 73.7 billion dollars in 2010. Evidence based management of the patient’s rehabilitation relies heavily on constraint induced movement therapy, and close work with a physical therapist. Approaches that reduce the burden of retraining the motor system from both a patient and provider standpoint could be of substantial impact.
Methods: The system was presented to 6 subjects with histories of basal ganglia or cortical stroke leading to unilateral hemiparesis. The patient’s viewed the stimulus using 3D anaglyph glasses to produce the illusion that they were seeing their own arms through the lid of the stimulus box. EEG data were acquired at 1000Hz during all experimental trials.
Results: Using the system described, the subject’s were able to achieve control accuracies as high as 87.4% (Cursor) or 81% (VR) in a binary classification task (reach left or right) and showed progression of skill in as little as three, two-hour experimental sessions. A comparison of this system to a standard cursor feedback control task suggests that use of the immersive virtual reality system allowed subjects to produce more sustained and homogenous motor imagery as assessed by spectral analysis of subject EEG.
Discussion: Pairing the motor imagination task with the virtual reality photorealistic motor feedback promotes a consistent and sustained motor imagery signal as compared to a standard cursor control task. Recent controlled trials on the efficacy of brain-computer interface show statistically significant changes in Fugl-Meyer motor scores when using a sensorimotor rhythm based BCI with physical therapy. By creating a training platform that requires low caregiver overhead, the system could serve as an affordable rehabilitation option in the management of stroke.
- © 2013 by American Heart Association, Inc.