Improving functional ability after a stroke requires task-oriented physical rehabilitation with the supervision of a therapist. However, with increasing medical costs and a shortage of rehabilitation specialists, post-stroke survivors sometimes receive a limited amount of individual treatment.\\ This paper proposes a low-cost computer-assisted rehabilitation system, called Virtual Coach that evaluates and guides a post-stroke survivor to engage in rehabilitation correctly at home with minimal supervision of a therapist. This system includes two major components: 1) motion analysis modules that evaluate exercise performance and guide a desirable joint trajectory and 2) a dialogue interface to provide feedback. The evaluation function of motion analysis modules computed exercise performance scores of 15 post-stroke survivors and achieved a 78% agreement with a practicing clinician. After developing a guidance function, the usability of the system will be evaluated with post-stroke survivors and therapists and iteratively improved.
[1]
P. Langhorne,et al.
Barriers to achieving evidence-based stroke rehabilitation
,
2000,
Clinical rehabilitation.
[2]
Peter Sandercock,et al.
Evidence-based practice for stroke
,
2009,
The Lancet Neurology.
[3]
Stefan Rennick Egglestone,et al.
Motivating mobility: designing for lived motivation in stroke rehabilitation
,
2011,
CHI.
[4]
Yao-Jen Chang,et al.
A Kinect-based upper limb rehabilitation system to assist people with cerebral palsy.
,
2013,
Research in developmental disabilities.
[5]
Xin Feng,et al.
Potential of a suite of robot/computer-assisted motivating systems for personalized, home-based, stroke rehabilitation
,
2007,
Journal of NeuroEngineering and Rehabilitation.
[6]
Bodo Rosenhahn,et al.
Human Motion: Understanding, Modelling, Capture, and Animation (Computational Imaging and Vision)
,
2007
.