A virtual reality system integrated with robot-assisted haptics to simulate pinch-grip task: Motor ingredients for the assessment in chronic stroke.

BACKGROUND Among numerous studies on virtual reality (VR) technology for motor rehabilitation, the majorities have only used it as a motor training system and did not use the kinematic or kinetic information obtained as the basis for therapeutic effect assessment. OBJECTIVE Incorporating the kinematic and kinetic information acquired through a robot-assisted VR system on pinch-grip training, this study intends to propose novel motor ingredients represented as indices for the interpretation of motor behaviors and verify the efficacy of them for the assessment in chronic stroke. METHODS Clinical trials were conducted on eight stroke patients and clinical assessments were made as the objective standard. Non-parametric test and correlation analysis were performed on results of motor indices and clinical assessments. RESULTS Non-parametric test indicated that 50% of the motor indices significantly improved while the others showed a trend of improvement, consisting with the findings in clinical assessments. Correlation analysis showed that the three clinical assessment items were correlated to items in the motor ingredients. CONCLUSIONS The proposed motor ingredients gathered from the robot-assisted VR training system were able to contribute to the assessment of therapeutic effect in chronic stroke and were capable for the interpretation of behavior phases.