Robots integrated with virtual reality simulations for customized motor training in a person with upper extremity hemiparesis: a case study

Abstract

Background and purpose: A majority of studies examining repetitive task practice facilitated by robots for the treatment of upper extremity paresis utilize standardized protocols applied to large groups. Others utilize interventions tailored to patients but do not describe the clinical decision-making process utilized to develop and modify interventions. This case study describes a robot-based intervention customized to match the goals and clinical presentation of person with upper extremity hemiparesis secondary to stroke.

Methods: The patient, P.M., was an 85-year-old man with left hemiparesis secondary to an intracerebral hemorrhage 5 years prior to examination. Outcomes were measured before and after a 1-month period of home therapy and after a 1-month robotic intervention. The intervention was designed to address specific impairments identified during his physical therapy examination. When necessary, activities were modified on the basis of response to the first week of treatment.

Outcomes: P.M. trained in 12 sessions, using six virtually simulated activities. Modifications to original configurations of these activities resulted in performance improvements in five of these activities. P.M. demonstrated a 35-second improvement in Jebsen Test of Hand Function time and a 44-second improvement in Wolf Motor Function Test time subsequent to the robotic training intervention. Reaching kinematics, 24-hour activity measurement, and scores on the Hand and Activities of Daily Living scales of the Stroke Impact Scale all improved as well.

Discussion: A customized program of robotically facilitated rehabilitation was associated with short-term improvements in several measurements of upper extremity function in a patient with chronic hemiparesis.

Figures

Figure 1

1a. Patient performing Cup Reach simulation using NJIT-RAVR system. 1b. Patient performing Piano Trainer simulation utilizing NJIT-TrackGlove system. 1c. The CyberGrasp, a cable actuated exoskeleton robot, worn over the TrackGlove system to assist in the extension of non-targeted fingers during Virtual Piano Trainer performance.

Figure 2

2a. Schematic of RTG test. 2b. Pre Test and Post-Test values for Time to Peak Velocity of Movement 1 and Time from Peak Velocity of Movement 1 to Initiation of Movement 2. 2c. Trajectories during ten repetitions of reach to grasp, transport, release and return of a 30mm diameter disc prior to training. 2d. Trajectories of same activity immediately after training.