RESEARCH REPORT POSTER DISPLAY

PERSONALIZED TASK PRACTICE: EFFECTIVENESS OF A VIRTUAL REALITY SYSTEM FOR UPPER EXTREMITY REHABILITATION IN PERSONS WITH CHRONIC HEMIPARESIS. Whitford M, Chen S, Jill S, Carolee W, Shih-Ching Y, Albert R, Hyunjin Y, Lei L, Youngbo J, Margaret M, Rebecca L; University of Southern California, Los Angeles, California, USA

PURPOSE: Our study examined the feasibility and effectiveness of a new virtual reality (VR) system for upper extremity (UE) rehabilitation in adults with chronic hemiparesis. The VR environment used a systematic, personalized progression of task-specific practice with goals to promote active engagement, motivation, and transfer to out of laboratory functional tasks. RELEVANCE: Evidence suggests that intense, focused task-specific training is efficacious for promoting recovery after centrally-induced paresis. Virtual reality systems simulate the real world using a human-machine interface and can provide a varied and enjoyable practice environment compared with traditional approaches. Virtual systems can be adapted to provide a systematic and personalized task practice program which allows for substantial time on task. PARTICIPANTS: Six adults>1 year post-cerebral injury (mean duration 29.6 mos), aged 39-88 yrs and with initial upper extremity Fugl-Meyer motor score ranging from 14-48 were selected from a sample of convenience. Three healthy adults were matched for age and treatment dosage. Institution-approved consent was obtained for all participants. METHODS: Participants attended 11-12 sessions of 1-2 hours over 3 weeks. During the intervention, participants practiced four VR games designed to promote specific UE movements (e.g. elevation, supination, prehension) with the hemiparetic arm. The level of game difficulty was chosen based on each participant’s ability and difficulty was progressed based on individual performance. Healthy controls practiced the same tasks. Standardized assessments and measures of activity, usability, and confidence in arm and hand tasks were administered at three time points. ANALYSIS: Group analyses were performed on the activity measures using a repeated measures (pre, mid, post) ANOVA. A paired sample t-test was used to measure mean group changes (pre/post) across intervention for the Confidence in Arm and Hand Movements (CAHM), a measure of confidence for real-world task performance, and the Stroke Impact Scale. Usability data were compiled and analyzed by task, visual display, and preference. Control subjects’ task performance values (time, success rate) were analyzed and compared to those with hemiparesis. RESULTS: For brevity, results are reported for only the activity measures. Participants with hemiparesis spent from 33% to 54% of total session time on task, and completed between 27-90 trials of each game. Despite heterogeneity of severity, as a group there were significant improvements in the Functional Test of the Hemiparetic UE (p = 0.010), bilateral component of the TEMPA (p = 0.034) and total TEMPA score (p = 0.056). The CAHM showed a significant decrease post-intervention (p = 0.041). CONCLUSIONS: Our findings demonstrate that a new VR system was both feasible and efficacious for use in upper extremity rehabilitation of adults with centrally-induced hemiparesis. Individualized systematic practice in a game-like environment optimized trial presentation for task progression that transferred to real world tasks. Future work will focus on the development of an algorithm for optimal task progression,development of a task library,optimization of system interface and implementation of a more cost-effective design. IMPLICATIONS: Virtual reality task practice is a promising tool for rehabilitation that allows customized and personalized progression. Additionally, the versatility allows promotion of skill development that is generalizable to real world functional activities for individuals with a wide range of severity levels. KEYWORDS: virtual reality, PHANToM. FUNDING ACKNOWLEDGEMENTS: Interdisciplinary Study of Neuroplasticity and Stroke Rehabilitation (ISNSR) an NIH Exploratory Center for Interdisciplinary Research (Grant # P20 RR20700-01) and the Integrated Media Systems Center, an NSF Engineering Research Center (Cooperative Agreement # EEC-9529152) at the University of Southern California. CONTACT: whitford@usc.edu

ETHICS COMMITTEE: Institutional Review Board of the University of Southern California