Repetitive Bimanual Integrative Therapy (RABIT) Virtual Rehabilitation System

Grigore C. Burdea* and Colin Defais Rutgers Tele-Rehabilitation Institute, USA.

ABSTRACT

Virtual rehabilitation games have gained increased clinical acceptance in the therapy of patients post stroke or traumatic brain injury. This is due to their ability to motivate, as well as intrinsic large number of movement repetitions that are keys to inducing brain plasticity and facilitating recovery. Unfortunately, bi-manual therapy, while potentially more efficacious, is less developed at this time. The novel RABIT system combines physical and cognitive training in integrative game-play sessions. Using the Razer Hydra game interface, the patients play a series of custom games designed to improve focusing, decision making (executive function), short-term and long-term memory, progressing in difficulty over 6 weeks of therapy.

Keywords: Bi-manual therapy, VR games, stroke.

Index Terms: K.6.1 [Human-Centered Computing]: Human computer interaction—Virtual Reality.

1 INTRODUCTION

Stroke is the leading cause of disability in the US, with 795,000 Americans suffering one each year [1]. Traditional physical rehabilitation of the paretic arm involves passive movement, compensatory training, electrical stimulation, and more recently constraint induced therapy [2]. These are uni-manual training approaches which do not take into account the prevalence of daily activities which involve both arms. Another drawback of uni-manual training is diminished neural cross talk to mirror motor areas associated with bimanual activities. Recent works have proved the efficiency of bilateral training [3, 4, 5]. Researchers found, that training the healthy arm resulted in a 23% functional improvement in the non-trained paretic arm, and in better bilateral tasks performance.

There are indications that bimanual training induces higher functional improvements compared to uni-manual training. A randomised controlled study [6] was performed on patients’ chronic post-stroke, half doing bimanual training and half doing uni-manual training of the affected arm, showed that those doing bimanual training had a 9 points larger improvement in motor function (as measured by their Fugl Meyer Assessment [7] scores). More recently VR was used in a randomised study of 36 nursing home residents to try to lessen cognitive decline and improve memory function [8]. The experimental group showed significant improvements in long-term recall and in several other aspects of cognition, while controls showed progressive decline.

Stroke survivors, as well as other patient populations, present with both motor and cognitive deficits [9]. Typically their short term and long term memory are affected, as are decision making and the ability to focus. Most stroke patients also get depressed. Under the current fractionated care system, such patients are attended by therapists, psychologists or psychiatrists, in separate sessions. This method is costly and does not exploit fully the body-mind continuum. The elderly form the majority of stroke survivors. For them, the situation worsens due to age-related cognitive decline [10]. One such deficit is diminished ability for split attention (or dual-tasking). These patients need a system designed from the start for integrative cognitive and motor therapy that would use therapeutic games that pose both cognitive and whole arm motor demands, and trains dual tasks. The system should automatically adapt to the patient’s functioning level, thus making games winnable, so to reduce depression.

2 THE RABIT BIMANUAL THERAPY SYSTEM

The RABIT system (patent pending) consists of off-the shelf gaming hardware and custom therapeutic games written in Unity 3D Pro [11]. The games are rendered on an HP Z600 graphics workstation with an nVidia Quadro 2000 graphics accelerator. To facilitate immersion and help the patients in their manual tasks, the workstation is connected to an Asus VG236H 3D monitor, and the patients wear nVidia “3D Vision” glasses. The interaction with the games is through the Razer Hydra interface (Figure 1) [12] consisting of two hand-held pendants, each with a number of buttons and a trigger, and a stationary source connected to the workstation over an USB port. The system uses an analog trigger so to detect the degree of flexion/extension of the patient’s index. The pressing of these analog triggers controls the closing/opening of hand avatars, while the position/orientation of the hand avatars is determined by the position/orientation of the corresponding pendants relatively to the electro-magnetic source.

* burdea@jove.rutgers.edu, colindefais@gmail.com

Figure 1: The experimental system and a patient during therapy. Copyright Rutgers Tele-Rehabilitation Institute. Reprinted by permission.

99

IEEE Virtual Reality 201316 - 20 March, Orlando, FL, USA978-1-4673-4796-9/13/$31.00 ©2013 IEEE

2.1 Therapeutic Games Several games were developed to be played either uni-manually

or bimanually. This gives flexibility when the therapy focus is motor re-training (using uni-manual mode) or integrative motor and cognitive retraining (using bimanual mode). The requirement for developing a multi-game therapy stems from the need to address several cognitive areas (by targeted games), as well as to minimize boredom by alternating games during a session. Some of these games are based on a library developed earlier by our Institute [13].

2.1.1 Baseline The system uses three baselines, two for arm range, and one for

the analog triggers. A vertical baseline asks the patient to draw a circle on a virtual blackboard, to determine the full vertical space on the game scene. A horizontal baseline is similar, except now the patient is asked to draw a circle on a horizontal table. During bimanual play sessions each arm performs the baseline in sequence. Each arm has different gains to the virtual scene so the movement of their respective hand avatars appears equal in the virtual world. A third baseline measures the range of movement of the index of each hand.

2.1.2 Games to train focusing Two games were developed to train patient’s ability to focus.

The Kites game presents two kites flying over water. One kite is green, one red, and they have to be piloted through like-colored target circles. The circles alternate randomly in their color and their position on the screen, and the difficulty of the game is modulated by the speed of the circles, the duration of the game, the visibility (a foggy sky gives less time to react) and the presence of air turbulence (acting as a disturbance). The game calculates the percentage of targets entered vs. those available, and displays it at the end of the game as summative feedback on performance.

The Breakout 3D game is a bimanual adaptation of the game developed earlier for uni-manual training on the Rutgers Arm system [13]. The scene depicts an island with an array of crates placed in a forest clearing. Two paddle avatars of different color, each controlled by one of the patient’s hands are located on each side of the crates. The patient needs to bounce a ball with either paddle, so to keep it in play, and attempt to destroy all the crates. The sound of waves is added to help the patient relax. The difficulty of the game is modulated by the speed of the ball, the size of the paddles, and the number of crates to be destroyed.

2.1.3 Game to train decision making (executive function) Towers of Hanoi 3D is similar to the version being played with

a mouse online. The patient has to restack a pile of disks of different diameters, from one pole to another pole, using a third pole as way-point. The game trains decision making by setting the condition that no disk can be placed on top of a smaller one. The scenes shows two hand avatars, one green and one red and similarly colored red and green disks. Each hand avatar is allowed to manipulate only a disk of its color. The difficulty of the game depends on the number of disks (2- easy, 3- medium, 4- difficult). The number of moves in the game is counted and compared to the ideal number of moves to complete the task.

2.1.4 Games to train memory In Card Island patients see an island and an array of cards

placed face down. The array is divided symmetrically by a central barrier, such that each hand avatar has to stay on its half of the island. When a hand avatar overlaps a card, the patient can turn it face up by squeezing the pendant trigger. The task is to turn cards

face up so to find matching pairs. Since non-matching cards turn face down again, the patient has to remember where a given card was seen before (short term visual memory).

Remember this card is a game that trains long-term visual and auditory memory. In the first part the patient is presented with a number of cards placed face down. Each card needs to be turned face up, at which time a sound in played associated with the image on the card. The patient selects one, by flexing the hand avatar over the card, and is prompted with “Remember this card” text. After other games are played the patients is presented with the same cards and has to remember which one was selected (long-term visual and auditory memory).

2.1.5 Dual tasking and therapy gradation As stated before, dual tasking is typically problematic with

older populations. Breakout 3D asks patients to bounce a ball with either hand paddle avatar to destroy an array of cubes. Dual tasking requires a momentary squeeze at the moment of bounce, lest the ball passes through the paddle. When a sustained grasp is required, the movement of the paddle is decoupled from that of the hand when the trigger is not squeezed. Naturally, the introduction of the squeezing requirement further increases game difficulty. Thus the weeks of therapy are gradated in terms of session duration and game difficulty, with dual tasking introduced in the last weeks of therapy.

REFERENCES

[1] C.D.C. Heart Disease and Stroke Prevention at a Glance. 2011 http://www.cdc.gov/chronicdisease/resources/publications/aag/pdf/2011/Heart-Disease-and-Stroke-AAG-2011.pdf

[2] C.Y. Wu, Chuang LL, Lin K.C. Randomized trial of distributed constraint-induced therapy versus bilateral arm training for the rehabilitation of upper-limb motor control and function after stroke. Neurorehabilitation and Neural Repair, 25(2):130-9, 2011.

[3] J.H. Cauraugh, H. Lodha, S.K. Naik et al. Bilateral Movement Training and Stroke Motor Recovery Progress: A structured Review and Meta-Analysis. Human Movement Science, 29(5):853-870, 2010.

[4] C. Ausenda and M. Carnovali. Transfer of motor skill learning from the healthy hand to the paretic hand in stroke patients: a randomized controlled trial. Eur J Phys Rehabil Med, 47(3):417-425, Sep 2011.

[5] G. Burdea. Virtual Rehabilitation-Benefits and Challenges. JMethods Information Med, Schattauer, Germany, 519-523, 2003.

[6] K. Lin, Y. Chen, C. Chen, et al. The Effects of Bilateral Arm Training on Motor Control and Functional Performance in Chronic Stroke: A Randomized Controlled Study. Neurorehabil Neural Repair, 24; 42, 2010.

[7] P.W. Duncan, Probst M., Nelson S.G. Reliability of the Fugl-Meyer assessment of sensorimotor recovery following cerebrovascular accident. Physical Therapy, 63:1606–1610, 1983.

[8] G. Optale G, C. Urgesi, V Busato, et al. Controlling memory impairment in elderly adults using virtual reality memory training: a randomized controlled pilot study. Neurorehabil Neural Repair,24(4):348-57, May 2010.

[9] C. Brooks, Gabella B., Hoffman R., et al. Traumatic brain injury: designing and implementing a population-based follow-up system. Arch Phys Med Rehabil, 78,S26-30, 1997.

[10] M. Wang, Gamo N.J., Yang Y, Neuronal basis of age-related working memory decline. Nature. 2011 Jul 27

[11] Unity Technologies. Reference Manual. San Francisco, CA. 2010. http://unity3d.com/support/documentation/Components/index.html

[12] Sixense Entertainment. Razer Hydra Product page. 2012. [13] G. Burdea, D. Cioi, J. Martin, et al. The Rutgers Arm II

Rehabilitation System – A feasibility study,” IEEE Trans Neural Sys Rehab Eng, 18(5):505-514, 2010.

100