STATEMENT OF PROBLEM

About half of the children diagnosed with cerebral palsy (CP) suffer from upper extremity (UE) dysfunction, which makes activities involving reaching, grasping, and manipulation a challenge (1).  Therapeutic treatment for children with CP includes improving control of their UEs by practicing reaching movements.  Practice, or repetition of movement, is a key feature of the motor skill learning therapeutic treatments given to children with spastic CP because it aids brain plasticity (2).  Therapeutic techniques must be engaging and interesting to capture the child’s interest, but at the same time, must be flexible enough to allow the child to make purposeful movements, and offer adequate feedback to the child on his or her performance.  One of the current technologies therapists use to achieve such goals is through the use of virtual reality (VR) therapy.  VR refers to the use of interactive simulations created with computer hardware and software to present users with opportunities to engage in environments that appear to look and feel similar to real world objects and events.  Users interact with displayed images, move and manipulate virtual objects, and perform other actions in ways that attempt to “immerse” them within a simulated environment, allowing them to feel that they are within the virtual world (3).  Therapeutic treatment can be supplemented with VR because it enables therapists to offer their patients individualized treatment by adjusting their practice intensity and positive visual and auditory feedback.  VR also provides three-dimensional correspondence between degree of movement in real environment and degree of movement observed on computer screen (1).  The literature has indicated that VR can enable children with CP to practice movements without feeling embarrassed or concerned about injuring themselves, which may lead to not only improved motor performance, but self-efficacy (4).  Recent studies with current VR gaming systems, such as Sony PlayStation II EyeToy and Vivid GX, have shown that these systems cannot be graded to match the skill level of children with CP who have severe UE impairment (1,5).  Further, these systems are often very expensive, which make them impractical for the clinical setting.

SOLUTIONS CONSIDERED

At the Rehabilitation Engineering Research Center (RERC) on Technology for Children with Orthopedic Disabilities at New Jersey Institute of Technology, “Hands-Up”, an adaptable open-source gaming environment, has been created for children with orthopedic disabilities.  The unique environment allows users to play video games for recreational and therapeutic purposes, all of which can be done in their home, by tracking speed, accuracy, level of difficulty, and duration of game play.  The core of the gaming platform is a playing area created by graphical axes and game pieces represented by graphical patches which are programmed to interact with each other and the boundaries of the gaming environment.  Behaviors include moving in a pattern, changing shape, color or size and even disappearing altogether, which are generally triggered by interaction with other pieces or the boundary of the environment (6).  

APPROACH

To evaluate Hands-Up’s usability, a study consisting of five intervention sessions was conducted with three children with CP.  During intervention sessions, the children played designated games and games of their choice; and engaged in simple functional tasks to evaluate the platform’s ability to aid them in meeting their therapeutic goals, as determined by their therapists.  Usability questionnaires completed by the therapists who treat them were also incorporated in the study.

Participants

The three participants were male children between the ages of 4-12 and each had a different form of CP, as depicted in Table 1.  The participants were identified and recruited by clinicians at the Pediatric Rehabilitation Department of JFK Johnson Rehabilitation Institute in Edison, NJ.  They had been receiving UE therapy at the site prior to participating in the study and continued to receive therapy throughout the course of the study.  The participants and their parents provided informed consent and assent forms that had been approved by the Institutional Review Board at New Jersey Institute of Technology.  All intervention sessions occurred in a conference room in JFK Johnson’s Pediatric Rehab Department, where all subjects attended therapy sessions.  The evaluation consisted of one intervention session per week for five weeks, with each intervention session lasting for about an hour.  The therapists of the participants were invited to oversee intervention sessions and were asked to complete usability surveys to document their experiences and provide feedback.

Table 1: Participant Demographics

Participant	Gender	Age	Diagnosis
1	Male	4	Hemplegic CP (right side)
2	Male	12	Spastic Diplegic
3	Male	10	Diplegic

Equipment

Hands-Up was created with Matlab’s Simrobot toolbox and currently runs on the 2007b version of Matlab (Mathworks, Natick, MA).  The platform also consists of a 30 frames per second webcam, which allows for external user input by taking snapshots to capture real-time hand movements (6,7).  The webcam used was the Logitech Quickcam Pro 500.  A color detection scheme has been devised to use the photo information to locate the most probable position for one of three colored markers in the gaming environment: red, green, or blue (6).  This is accomplished by capturing the real-time images of the picture at a 240 x 320 resolution and calculating the median of the center point of the red-green-blue pixel of the tracker in the player’s hand.  The player views representations of his hands, allowing him to move and interact within the gaming environment (7).  Hands-Up was played on a Dell Latitude D620 laptop.  The trackers were selected based on the therapeutic objectives of each subject.  For example, Participant 3 need to improve bilateral control and played some games with a blue cup, as depicted in Photo 1.  A Flock of Birds (Ascension Technologies) electromagnetic position/rotation sensor was attached to the tracker to collect the position of the subject’s hand in three-dimensional space.  Position data were collected at 100 frames per second.

EVALUATION AND OUTCOMES

Functional Tasks

At the beginning of the sessions, participants completed two functional tasks related to their therapeutic goals.  One task was to pull 2 attached strips of Velcro apart and put them back together.  Participants were told to complete the task as many times as possible in one minute.  For the second task, the participants flipped the top card of a deck of cards with their most impaired hand.  The number of cards flipped over was recorded.  

Comparing the first session to the last session, Participants 2 and 3 both increased in their ability to complete the functional tasks, as seen in Table 2.  Participant 1 was not able to participate in two of the sessions due to illness; however, his results increased for the Velcro Task over the sessions.  Participant 1 was able to complete the task 6 times during the first session and 16 times his last session, for a mean of 9.8 (+/- 5.39) times for the three sessions in which he participated.

Hands-Up Games

The participants played two types of games on the Hands-Up: designated games and games of their choice.  The designated games were the “Number Find” games which were played each of the four levels for approximately 5 times for 4 sessions (the first session was used to test different games to determine the appropriateness, motivation, and cognition levels of all participants).  As the levels increased, the placement and patterns of the objects on the screen as well as the number of objects on the screen increased in complexity.  The games the participants selected to play upon completing the designated games were based on their interests.  Participant 1 liked cartoons, so he chose games that featured cartoon characters, such as Spongebob Squarepants and Diego.  Participants 2 and 3 were motivated to play more challenging games, so they played games that had continuously moving objects that had to be caught.  To make the game more challenging, the speed of the falling objects could be increased.  Participants 2 and 3 were very competitive, so they were interested in playing the games quicker to beat their previous game times.      

Usability Testing

A physical therapist, an occupational therapist, and an occupational therapy student completed the usability questionnaires.  The questionnaires included a background questionnaire, a pre-game questionnaire, and a post-game questionnaire.  All three believed that VR based gaming systems could help their patients achieve their therapeutic goals.  They even reported that noticing an improvement of their patient’s reaching ability.  Participant 1’s therapist even noted that Hands-Up helped him achieve the therapeutic goals she had set for them.  All were in agreement that their patients became more comfortable with the platform after participating in a few intervention sessions.  They all indicated that they would consider using Hands-Up in their clinical setting.

IMPLICATIONS

The feedback provided by the therapists reveals that they are accepting of Hands-Up and consider it useful in the clinical setting.  Their feedback also reveals that Hands-Up is adaptable and customizable for children with different forms of CP to play the games successfully, while achieving therapeutic goals and improving reaching ability.  The likability of Hands-Up by the participants reveals that the platform is well tolerated by children ranging in age.  Moreover, the intervention sessions encouraged the participants to play the designated games quicker, so that they could play the games that they found more fun.  The outcomes from this evaluation serve as a basis for designing a larger scale feasibility study to examine the effectiveness of Hands-Up in enhancing UE control of children with CP by computing kinematic variables that describe their reaching patterns.