Hybrid Brain-Computer Interface and Functional Electrical Stimulation for Sensorimotor Training in Participants With Tetraplegia. Issue 1 (January 2015)

• Record Type: Journal Article
• Authors: Vučković, Aleksandra; Wallace, Leslie; Allan, David B.
• Is Part Of: Journal of neurologic physical therapy. Volume 39:Issue 1(2015:Jan.)
• Publication Date: 2015-01
• Subjects: Nervous system -- Diseases -- Physical therapy -- Periodicals; Nervous system -- Diseases -- Treatment -- Periodicals; 616.80462
• DOI: 10.1097/NPT.0000000000000063
• Languages: English
• ISSNs: 1557-0576

Abstract

<abstract> <title> <x xml:space="preserve">Abstract</x> </title> <sec> <title>Background and Purpose:</title> <p>Impaired hand function decreases quality of life in persons with tetraplegia. We tested functional electrical stimulation (FES) controlled by a hybrid brain-computer interface (BCI) for improving hand function in participants with tetraplegia.</p> </sec> <sec> <title>Methods:</title> <p>Two participants with subacute tetraplegia (participant 1: C5 Brown-Sequard syndrome, participant 2: complete C5 lesion) took part in this proof-of-concept study. The goal was to determine whether the BCI system could drive the FES device by accurately classifying participants' intent (open or close the hand). Participants 1 and 2 received 10 sessions and 4 sessions of BCI-FES, respectively. A novel time-switch BCI strategy based on motor imagery was used to activate the FES. In one session, we tested a hybrid BCI-FES based on 2 spontaneously generated brain rhythms: a sensory-motor rhythm during motor imagery to activate a stimulator and occipital alpha rhythms to deactivate the stimulator. Participants received BCI-FES therapy 2 to 3 times a week in addition to conventional therapy. Imagery ability and muscle strength were measured before and after treatment.</p> </sec> <sec> <title>Results:</title> <p>Visual feedback was associated with a 4-fold increase of brain response during motor imagery in both participants. For participant 1, classification accuracy (open/closed) for motor imagery-based BCI was 83.5% (left hand) and 83.8% (right hand); participant 2 had a classification accuracy of 83.8% for the right hand. Participant 1 had moderate improvement in muscle strength, while there was no change for participant 2.</p> </sec> <sec> <title>Discussion and Conclusion:</title> <p>We demonstrated feasibility of BCI-FES, using 2 naturally generated brain rhythms. Studies on a larger number of participants are needed to separate the effects of BCI training from effects of conventional therapy.</p> <p> <bold>Video Abstract available.</bold> (see Video, Supplemental Digital Content 1, <ext-link ext-link-type="uri" xlink:href="http://links.lww.com/JNPT/A84" xlink:type="simple" xmlns:xlink="http://www.w3.org/1999/xlink">http://links.lww.com/JNPT/A84</ext-link>) for more insights from the authors.</p> </sec> </abstract>