Utilizing Gaming Mechanics to Optimize Telerehabilitation Adherence in Persons With Stroke

• Purpose/Specific Aims The overarching aim of this study is to provide a mechanism for patients to engage in progressive motor practice for a meaningful time period. The investigator's aim to improve on the positive outcomes demonstrated in patients in the chronic phase and the pilot work being done on patients in the acute in-patient phase post stroke to determine whether functional recovery can be further improved using a home based system.

Aim 1: Evaluate compliance with Home-Telerehabilitation simulated hand/arm gaming activities and two computer game groups, one with motivation enhanced: Home Training Motivation Enhanced (HTme) simulations and one with non-enhanced simulations: Home Training Unenhanced (HTu) versions. Hypothesis: Participants in the HTme group will show significant compliance as compared to the control group (HTu).

Aim 2: Evaluate the effectiveness of motivation enhanced HTme home-based virtually simulated hand/arm gaming activities for individuals with stroke as compared to a program unenhanced HTu versions of the same simulations. Hypothesis: Participants completing HTme training will exhibit significantly improved clinical, kinematic and neurophysiological outcomes as compared to the control group (HTu).

Aim 3: Evaluate the impact of the motivation enhancements designed into computer games to provide a more enjoyable training experience. Hypothesis: Enjoyment of the games will be a more valid predictor of compliance than personal factors.

• Background and Significance Studies have shown that sustained hand rehabilitation training is important for continuous improvement and maintenance of function following a stroke. It is unimaginably difficult to pursue education, employment and community participation without being able to independently use one's hands. The primary goal of this study is to test an exciting new technology that can be easily used in the home for long-term hand and upper extremity training. Recovery of hand function post brain injury is particularly recalcitrant to currently available interventions. To date, the best efforts of groups studying traditionally presented as well as technology-based therapeutic interventions for the hemiplegic hand and arm have produced measurable changes in motor function and motor control but fall far short of major reductions in disability.

If the amount of therapy is critical to rehabilitation, our current institutional limitations undermine the probabilities for successful outcomes. After discharge from the inpatient stay, access to rehabilitation therapy can be difficult for some patients. This is due in part to inadequate insurance, lack of transportation, and the patient's dependence on their caregiver. Having access to long-term rehabilitation training anywhere and at any time is necessary for sub-acute and chronic patients to continuously improve their functional abilities.

• Research Design and Methods This study will be a single blind randomized controlled trial. Subjects will be blinded to the purpose of the study. All outcome measures will be performed by a therapist blinded to group assignment. A controlled trial will be utilized to determine the additive effect of presenting rehabilitation activities in a virtual environment as compared to standard upper extremity exercise. The investigators will randomize subjects to treatment and control groups using a computerized random number generator.

3.1. Duration of Study

Each subject will perform a pre-study evaluation, train using one of the protocols for three months, perform a post study evaluation as well as one and six month retention evaluations.

3.2 Study Sites Testing and initial training will take place in the Bergen Building of the Rutgers Biomedical and Health Sciences Campus in Newark. Home training will take place in subjects' homes.

3.3 Sample Size Justification The investigators will seek sufficient power to detect a clinically significant difference in the Wolf score changes in these two pre-planned, primary comparisons. To evaluate these effects of training, we will assume a power level of .8 and a significance level of 0.05. With presumed correlation among repeated measures of 0.1 and effect size of 0.3, a sample size of 25 subjects in each of the two groups (HTme and HTu) to observe a significant effect for the first comparison (G*Power, version 3.1.5) is necessary. Although the investigators will screen for patients with homogeneous impairments, by its nature stroke is an extremely variable condition. Due to possible subject attrition, the investigators will use a total of 30 subjects in each of the two groups.

3.4 Subject Recruitment Subjects will be recruited through flyers, stroke support groups, and clinician referrals. The investigators will assume that approximately 15-20% of the population will satisfy our inclusion criteria based on our previous experience with upper extremity rehabilitation in this population. Hence the investigators will approach 300 persons.

3.5 Consent Procedures Example: The study will be explained to the potential subject by the study staff, the consent will be read, and their questions will be answered. If participants wish to enroll, the subject will sign the consent form. The study staff obtaining consent will also sign and date the consent form, and a copy will be given to the subject sought from each prospective subject or the subject's legally authorized representative, in accordance with federal & state law and institutional policy. If the study staff member performing the consent process identifies issues suggesting that the prospective subject may not be capable of participating in the consent process due to dementia, a Folstein Mini Mental Status will be performed. Prospective subjects screening positive for dementia will not be included in the study.

3.5.1 Subject Costs and Compensation There are no costs for the subjects. The subjects will be paid 100$ at each of the retention tests.

• Study Variables

4.1 Independent Variables or Interventions

The two computer game groups, Motivation Enhanced (HTme) and Motivation Non-Enhanced (HTu) will use the NJIT- Home Virtual Rehabilitation System (HoVRS) to play a series of computer games developed to practice movement of the hand and fingers. Subjects will first come into our lab, perform pre-tests as well as a pre-intervention training session. Then a physical therapist and engineer will set up the apparatus in subject's home and will train them on how to use the system and play the games in their home during the first week. The physical therapist and engineer will be in contact with subjects throughout the training and will visit subjects' homes as needed if problems are encountered. Additionally, the system allows the therapist to remotely monitor each day's activity.

4.1.1 Device Description NJIT HoVRS has two sub-systems to deliver home-based training: 1) a patient based platform to provide the training and 2) a server based online data logging and reporting system. In the patient's home, a cross platform virtual reality training application runs video games (developed in the Unity 3D game engine using the language C#) on their home computer.

4.1.11 Hardware The Leap Motion Controller (LMC) a commercially developed infrared tracking device developed for home video game control is used to capture motion of the hand and arm movement without requiring wearable sensors. The device's USB controller reads the sensor data into its own local memory and performs any necessary resolution adjustments. This data is then streamed via USB to the Leap Motion image Application Programming Interface (API). From there, we programmed the system to feed tracking data into virtual reality activities by calling the Leap Motion API.

If the patient's arm is weak and cannot support the hand against gravity above the Leap Motion Controller, a commercially available, spring-based arm support, will be provided to the subject (Figure 1). The arm support provides 12 different levels of passive support allowing it to accommodate a wide range of patient sizes and strength levels. It requires a single setting that can be provided during the patient's initial evaluation

4.1.1.2 Software Patients will either use their own home computer or will be provided with a computer if needed. A user-friendly Graphic User Interface (GUI) lists all of the training activities allowing patients to choose which activity they want to begin with using just one mouse click. Currently twelve games have been developed, each one designed to focus on training a specific hand or arm movement such as wrist rotation or finger individuation. All games are downloadable via HoVRS website.

4.2 Dependent Variables: See Outcomes Measures

4.3 Risk of Harm

There is less than minimal risk involved. The virtual reality (VR) experiments are non-invasive and pose no obvious risk. Transient fatigue of the hand and arm are possible, but this risk is not greater than that posed by normal daily activities following a stroke.

4.4 Potential for Benefit The benefits of taking part in this study may be: Patient may regain better use of their hand and arm. However, it is possible that patients might receive no direct personal benefit from taking part in this study.

• Data Handling and Statistical Analysis All efforts will be made to keep subjects' personal information confidential. All subject names will be removed from the data and the data will be tagged using a coded identification (ID) number. Demographic, clinical outcome and survey data will first be recorded on paper. All kinematic and computerized performance data will be collected on computer. These computer files will be identified by the coded subject ID number. All data will be transferred to an Excel spreadsheet with subjects identified by this same ID number. Spreadsheets will be stored on a drive that is password protected. Data will only be accessible to study staff and will be retained for seven years. The link between subject identity and subject ID number will be destroyed when data collection is completed.

The primary outcome measures and all secondary outcome measures described above will be subjected to a repeated measured analysis of variance, with between-group factors Therapy Type (HTme, HTu) and within-group factor Test (Before, Post, One Month retention, Six Months Retention). Post-hoc analyses of the Therapy Type by Test interaction effects will focus on the Month 1 versus Month 6 comparison. The investigators will be quantifying training effects by comparing group means as well as by percent change in performance, and by comparing the recovery curves obtained from Tests 1-4. All clinical outcomes used are well established measures of upper extremity functional recovery with published minimum clinically important differences which will be used to evaluate the significance of our findings.

• Reporting Results

7.1 Individual Results No disease screening data will be collected. Patient's changes on clinical tests will be shared with them during testing sessions. These sessions are conducted by licensed Physical Therapists who have training to help persons with stroke interpret clinical examination findings.

7.2 Aggregate Results Subjects will not be informed of aggregate findings.

7.3 Professional Reporting De-identified, aggregate findings will be published in professional journals and presented at scientific meetings.

Inclusion criteria

• unilateral stroke
• score of 22 or greater on the Montreal Cognitive Assesment
• Score of 1 or better on extinction and inattention portion of NIH Stroke Scale
• Fugl-Meyer (FM) between 36-58/66 (
• Score of 1 or better on language portion of NIHSS
• intact cutaneous sensation (ability to detect <4.17 Newton stimulation using Semmes-Weinstein nylon filaments)

Exclusion criteria

Orthopedic issues that would limit the ability to perform regular upper extremity activity