|About GSBS | FAQ | Job Opportunities | Search UMDNJ|
Microsoft Kinect® Based Mini-Games to Investigate Stepping in Healthy Adults, Clinicians and Persons Post-Stroke"
Robbie R. Gosine
Biomedical Engineering Program
M.S.E.E 2007, Florida Atlantic University, FL
B.S.E.E 2004, Florida Atlantic University, FL
Thesis Advisor: Judith E. Deutsch, PT, PhD, FAPTA
Department of Rehabilitation & Movement Sciences, SHRP
Friday, January 15, 2016
10:30 A.M., SSB 921
Off-the-shelf video games have been embraced in rehabilitation settings. They are appealing tools for clinicians because they provide serious games, or virtual environments as a movement rehabilitation tool at a low price point. Their limitation is that they are not designed for rehabilitation. This research bridged this gap by developing custom software in the form of serious games also called virtual environments for rehabilitation using the Microsoft Kinect® (Kinect). The research was organized with four main aims and three groups of participants: a group of healthy clinicians (Clinicians), a group of healthy younger adults (Healthy) and a group of adults post-stroke (Stroke). The first aim investigated validating the measurement capability of the Kinect by comparing it to a commonly used motion capture system; the Vicon-Motus (Vicon). The second aim investigated (using validated questionnaires) the usability of the system and the level of flow state it produced in the participants. The third aim characterized the stepping performance of the participants and investigated the step response to manipulating the step distances. The final aim investigated the effect of music on stepping during game play.
A high level of agreement and consistency (ICC of: Stroke – 0.99, Clinician – 0.98 and Healthy – 0.99) was found between the kinematics measured by the Kinect and Vicon. Usability was deemed acceptable by the three groups: Stroke participants rated the games 81.3/100, Clinicians – 82.8/100, and Healthy – 77.8/100. Stroke participants had the highest Flow rating a 4.9/5 on Flow State Scale (FSS), followed by the Healthy Adults (3.94/5) and the Clinician (3.72/5) groups in that order. All participants increased their stepping displacement with manipulation of the step target distance; eighty percent of the healthy subjects, and sixty-seven percent of the Stroke group were able to symmetrically accommodate their step displacements. While there was evidence of changes in stepping frequency and stepping displacement in all three groups with the addition of music, the results indicated that subjects were not able to increase step frequency and step displacement simultaneously. The VSTEP was successfully developed and validated. While it produced mixed results, it also provided substantial evidence in support of the primary objectives.