1247
Virtual Reality and Haptics as a Training Device for
Movement Rehabilitation After Stroke: A Single-Case Study
Jurgen Broeren, MSc, OT, Martin Rydmark, PhD, MD, Katharina Stibrant Sunnerhagen, PhD, MD
ABSTRACT. Broeren J, Rydmark M, Sunnerhagen KS.
Virtual reality and haptics as a training device for movement
rehabilitation after stroke: a single-case study. Arch Phys Med
Rehabil 2004;85:1247-50.
Objective: To investigate whether training in a virtual en-
vironment with a haptic device will improve motor function in
the left hemiparetic arm of a stroke subject.
Design: Single case, A-B-A design.
Setting: University hospital research laboratory.
Participant: A man in his late fifties (right handed), with a
right-hemisphere lesion that caused a deficit in the left upper
extremity.
Intervention: The subject trained with a 3-dimensional
computer game during a 4-week period that consisted of twelve
90-minute sessions.
Main Outcome Measures: Three tests (Purdue pegboard
test, dynamometer hand-grip strength, upper-extremity test)
and a subjective interview were used to evaluate motor perfor-
mance.
Results: Improvements were found in fine manual dexterity,
grip force, and motor control of the affected upper extremity.
The subject reported that there was a change in his day-to-day
use of the upper extremity and that he was able to use it in
activities that were previously impossible for him.
Conclusions: Training with virtual reality and haptics can
promote motor rehabilitation.
Key Words: Hemiparesis; Rehabilitation; Stroke; Virtual
systems.
2004 by the American Congress of Rehabilitation Medi-
cine and the American Academy of Physical Medicine and
Rehabilitation
T HE IMPAIRMENT OF the upper extremity is a prime
obstacle in reacquiring competence in managing everyday
activities. Stroke survivors perceive the loss of upper-extremity
function as a major problem.1 Broeks et al2 assessed the long-
term motor and functional recovery of arm function after stroke
and concluded, as many have before, that there is a need to
develop active treatment methods for upper-extremity function.
The recovery of the affected upper extremity is, like every
other training situation, dependent on regularity and intensity
From Mednet (Broeren, Rydmark); and Department of Clinical Neuroscience and
Rehabilitation Medicine (Broeren, Sunnerhagen), Sahlgrenska Academy, Goteborg
University, Goteborg, Sweden.
Presented in part at the 4th International Conference on Disability, Virtual Reality
and Associated Technologies, September 18 20, 2002, Veszprem, Hungary.
Supported in part by the Swedish Stroke Victims Association, the Hjalmar Svens-
son Research Foundation, and the Federation of Swedish County Councils (VG-
region).
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit upon the authors(s) or upon any
organization with which the author(s) is/are associated.
Reprint requests to Jurgen Broeren, MSc, OT, Dept of Clinical Neuroscience and
Rehabilitation Medicine, Sahlgrenska Academy, Goteborgs University, Guldhedsga-
tan 19, SE 413 45 Goteborg, Sweden, e-mail: jurgen.broeren@mednet.gu.se.
0003-9993/04/8508-8397$30.00/0
doi:10.1016/j.apmr.2003.09.020
in training, as well as on specificity.3 To stimulate recovery, a
training program must allow the individual to develop and
reestablish the motor system. Training programs should be
made appealing, to motivate the patient. Target-oriented reha-
bilitation approaches and individually adapted training pro-
grams are essential for recovery. New models must be devel-
oped for patients and health professionals. Virtual reality
technology is presently being explored as an assessment and
training device for improving motor recovery after stroke.
Virtual reality or virtual environment can be defined as a
computer-generated environment, which imitates reality. Vir-
tual reality generates a convincing interface, so that users
believe they actually perceive sensory information that is sim-
ilar to that of the real world.
Virtual reality offers therapists new approaches for improv-
ing the effects of rehabilitation.4-7 Within a virtual environ-
ment, a patient can practice in a controlled environment
individuals can practice simple tasks or more complex tasks
and they can measure their success in real time. Although the
individual practices various tasks, it is possible for therapists to
track movements for analysis.7 However, when an individual
performs tasks with the affected upper extremity, sensory feed-
back is important for the capability to grasp. When manipulat-
ing objects in virtual environments, the users information must
closely approximate information he/she would use in a natural
setting. Consequently, the need for haptic force feedback be-
comes evident. The term haptic means interaction with a
3-dimensional environment created in a computer, which be-
sides the visual impressions gives the user physical interaction,
enabling him/her to touch the object in the virtual environment.
Our report concerns a man whose affected upper extremity was
trained with a haptic device in a virtual environment.
Case Description
METHODS
The patient was a man in his late fifties who was right
handed and who had a first-occurrence stroke. The time since
the stroke was 12 weeks. The cause was an infarction, and the
diagnosis was determined by a neurologist after clinical exam-
ination and was confirmed by computed tomography scan. The
lesion was situated in the right genus of the internal capsule.
On admission to the ward, the patient had symptoms in the
form of contralateral hemiparesis; this persisted until the con-
clusion of the study and when training began. The patient was
able to ambulate independently and could move his arm par-
tially, with some wrist extension. He had no limitations in
finger extension and flexion, but he had weak lateral grasp. His
body awareness and spatial competence were normal. There
were no limitations in understanding the information given.
Most of his activities of daily living (ADLs) were accom-
plished by compensating with the right arm.
Research Design
A single-subject design was used to evaluate pre- and post-
training changes and changes at 20-week follow-up (35wk
poststroke) posttraining.8 This included a baseline (A), an
Arch Phys Med Rehabil Vol 85, August 2004
1248 VIRTUAL REALITY AND HAPTICS FOR STROKE REHABILITATION, Broeren

right-hand dominant, with a mean age of 50 years (range,

Fig 1. Virtual reality setup. The Reachin display, a stereoscopic
display system with the PHANToM haptic device and Crystal Eyes
CE-2 setup.
intervention phase (B), and a follow-up phase (A), with with-
drawal of the direct intervention provided the occupational
therapist.
Procedure
The patient was selected because he had an obvious deficit in
his left arm that could be observed in ADLs. Three tests and a
subjective interview were used to evaluate motor performance
in the university hospitals laboratory. The first test was the
Purdue pegboard test, which measures unilateral and bilateral
dexterity for 2 types of activity: gross movements of hands,
fingers, and arms, and finger dexterity. Two subtests were used:
(1) the ability to put as many pegs as possible with the left hand
into holes on the left side of the board and (2) the ability, using
both hands, to insert as many pegs as possible into the board.
The score was the number of pins handled as quickly as
possible in a 30-second period.9 In the second test, the dyna-
mometer hand-grip strength by Grippita was recorded.10 Peak
maximum grip force and the mean value of the 10-second
sustained grip were measured. The third test was an upper-
extremity test, in which the subject moved the PHANToMb
haptic device to different targets (generated positions) on the
screen. The target placements (32) appeared randomly to the
patient, but were actually set according to a preset cinematic
scheme. The targets were numbered from 1 to 9, starting with
number 1 at the lower right-hand side of the computer screen,
continuing with 2 and 3 to the left. The fourth target was placed
above target 1, 5 over 2, and so on. Motions of the upper
extremity were divided in different directionsthat is, side-
ways (left to right, right to left), up and down, diagonal (left to
right), up and down. The x, y, and z coordinates for the haptic
device and the targets were continuously sampled. Time,
speed, trajectory distance (actual pathway of the upper extrem-
ity), and intertarget distance (geometric shortest distance be-
tween the targets) to complete the exercise were recorded to
evaluate the performance. For purposes of comparison, a ref-
erence group was added to this test. Nine healthy men, all
46 73y), completed the same test as did the subject. In the
fourth test, after the follow-up, the subject was interviewed
about use of the affected upper extremity in ADLs at home.
Apparatus
Haptics were added to a computer workstation by using
Reachin 3.0.c Reachin API is a programming environment that
allows interactive manipulation of a 3-dimensional model by
using a haptic interface. Haptic force feedback is provided by
using the PHANToM haptic device. The patient connects to the
mechanism by grasping the stylus, which provides a force-
reflecting interface between a human user and a computer and
creates a convincing illusion of interaction with solid physical
objects. This sense of touch includes not only translational
force feedback, but also rotational or torque feedback. Stereo-
scopic visualization is accomplished by using a standard Crys-
tal Eyesd CE-2 setup (fig 1). Crystal Eyes allows the user to
view 3-dimensional objects through a special display monitor
mode and special stereo viewing glasses. The left and right
images are alternated rapidly on the monitor screen and create
the illusion that on-screen objects have depth and presence
outside of the display.
Training
With his left upper extremity, the subject played a computer
game, 3-dimensional bricks (fig 2), developed by Reachin
Technologies. At the start of the game, the subject grasped the
haptic device with his left hand. He then viewed a ball in a
court filled with bricks. The game starts when the subject
strikes the ball. The subject receives credit for the bricks
knocked down. On the return, when the subject misses the ball,
he collects minus points. The game was customized to operate
at 4 speeds. Modifying the speed from 4 to 3 to 2 to 1 caused
the program to increase the velocity of the ball. The level of
difficulty was changed after the subject had reached a given
score in 3 consecutive games.
RESULTS
Fine manual dexterity with the Purdue pegboard test im-
proved on both subtests for the subject: for the left hand
(impaired limb), it improved by 11% units after the interven-
tion period and continued to improve from baseline to 17%
Fig 2. Screenshot of the computer game, 3-dimensional bricks.

Arch Phys Med Rehabil Vol 85, August 2004

 VIRTUAL REALITY AND HAPTICS FOR STROKE REHABILITATION, Broeren
Fig 3. Fine manual-dexterity values with the Purdue pegboard test
for the left hand and both hands, measured as a percentage of a
normal population.
units at follow-up. The results of the bilateral test improved
with 22% and persisted during the follow-up (fig 3). The mean
grip force at baseline was 13% of grip force in the normal age-
and sex-matched population and increased to 57% at follow-up
(table 1). The median intertarget time to complete the virtual
reality test changed from 1.27 to 0.96 seconds posttraining and
further decreased to .86 seconds during the follow-up. The
reference group median intertarget time was .73 seconds, with
some variation. Figure 4 shows the subjects time results before
and after training and at 20-week follow-up, compared with a
reference group of 9 healthy men. The quotation of the trajec-
tory distance (actual pathway of the upper extremity) and
intertarget distance (geometric shortest distance between the
targets), which reflects the degree of superfluous movements of
the hand along the trajectory to the targets, did not indicate any
major changes after the intervention period and during the
follow-up. The values were close to the values of the reference
group after the training period and during the follow-up. As a
consequence, the velocity of movement was calculated and
expressed basically as the inverse of the time results. The
velocity to complete the assessment trial increased from pre-
training values of .18m/s to normal values of .31m/s (reference
group, .34m/s) after training, a result that persisted during the
follow-up. In the follow-up assessment, the subject reported
that a couple of weeks after the training he was able to use his
left arm in several ADLs that were previously impossible for
him (eg, buttoning his shirt, emptying the dishwasher with 2
hands). He felt that he could rely more on his left upper
extremity.
DISCUSSION
Virtual reality technology was found to have potential as an
assessment and training device in stroke rehabilitation. We
observed that finger dexterity, grip force, and endurance in-
creased after the treatment, which started 12 weeks after the
stroke. Time to complete the exercise decreased considerably
and approached normal values. The results imply that intensive
training with virtual reality can reduce motor impairment and
that it could optimize the stroke survivors motor skills. This is
Table 1: Grip Force Peak and Average Values During 10 Seconds
(newtons) for the Left Hand
Follow-Up
Pretraining Posttraining (20wk) Range
Peak 68 264 316 327549
Average 10s 52 176 228 324484
1249
Fig 4. The time to perform the test is shown as a boxplot (25th,
75th percentiles) at 3 different time settings, as well as the result
from the reference population (n9). Median value (thick black line)
with 10th and 90th percentiles are shown at the end of the lines.
in line with the results of the Copenhagen Stroke Study, which
showed that the function of upper-extremity recovery after a
stroke with initial paresis occurred predominantly within the
first 3 weeks and no further recovery of function should be
expected after 11 weeks.11 We used a single-subject design.
With a single-subject design, conclusions can be drawn about
treatment effectiveness.8 Grip-force improvement increased
considerably; this change could account for the improvement in
endurance while playing the 3-dimensional computer game. By
the end of the intervention period, the subject progressed to
game level 1 (most difficult).
The time it took the subject to complete the assessment trial
decreased from the pretraining period to the posttraining period
and approached normal values, a result that persisted during the
follow-up (fig 4). The trajectory/intertarget distance quotient
measure, indicating smoothness (precision, accuracy) of mo-
tion, did not change before or after training. These measures
were already normal at the initial preassessment, compared
with the reference group.
The basis for training routines should be based on the pa-
tients specific neurologic demands as detected (ie, during the
virtual reality haptic assessment), the neurologic status evalu-
ation, and the results of the occupational therapy (OT) assess-
ment. Our patient had an obvious deficit in his left arm, which
we observed during ADLs, and poor performance in time to
complete the virtual reality assessment, but he showed no lack
of accuracy and precision. Before playing the 3-dimensional
computer game, the level of difficulty was adjusted to his
needsthat is, we modified the speed of the ball for each
training session to maximize the effectiveness of the treatment.
Maximizing the rehabilitative effectiveness of training pro-
grams as a means of reinforcing the patients motor perfor-
mance requires both commitment and engagement in the pro-
cess. The subject reported an increased awareness in using his
affected left upper extremity; this is consistent with methods
Arch Phys Med Rehabil Vol 85, August 2004
1250 VIRTUAL REALITY AND HAPTICS FOR STROKE REHABILITATION, Broeren
used in OT, which are intended to encourage the patient to use
the impaired upper extremity in ADLs.
After discharge from hospital units, many patients who have
had a stroke still need qualified rehabilitation. Yet, the alter-
natives offered to patients after hospital discharge are limited in
availability and vary locally and regionally. Home rehabilita-
tion might be an important complement that can produce re-
sults comparable to institutional rehabilitation.12 The virtual
reality setup presently used is easy to distribute but still ex-
pensive ($20,000). We hope that a refined commercial alter-
native eventually will be affordable for health care providers.
This could be installed, for example, in patients homes, and
the Internet could be used for data transfer to the clinic,
allowing the occupational therapist to monitor a patients
progress remotely. In this way, one could reduce treatment
costs in the long-term training process and improve patient
quality of life.
CONCLUSIONS
The virtual environments potential to support tasks that
address the specific needs of motor rehabilitation is promising.
The use of meaningful and rewarding activities has been shown
to improve a patients motivation to practice.13 Our results
confirm the efficacy of the treatment and justify further clinical
trials. A larger trial is required to determine the differences in
improvement in motor ability of the upper extremities in stroke
survivors using virtual reality with haptics.
References
1. Feys HM, De Weerdt WJ, Selz BE, et al. Effect of a therapeutic
intervention for the hemiplegic upper limb in the acute phase after
stroke: a single-blind, randomized, controlled multicenter trial.
Stroke 1998;29:785-92.
2. Broeks JG, Lankhorst GJ, Rumping K, Prevo AJ. The long-term
outcome of arm function after stroke: results of a follow-up study.
Disabil Rehabil 1999;21:357-64.
Arch Phys Med Rehabil Vol 85, August 2004
3. Kwakkel G, Wagenaar RC, Twisk JW, Lankhorst GJ, Koetsier JC.
Intensity of leg and arm training after primary middle-cerebral-
artery stroke: a randomised trial. Lancet 1999;354:191-6.
4. Piron L, Cenni F, Tonin P, Dam M. Virtual reality as an assess-
ment tool for arm motor deficits after brain lesions. Stud Health
Technol Inform 2001;81:386-92.
5. Merians AS, Jack D, Boian R, et al. Virtual reality-augmented
rehabilitation for patients following stroke. Phys Ther 2002;82:
898-915.
6. Broeren J, Bjorkdahl A, Pascher R, Rydmark M. Virtual reality
and haptics as an assessment device in the postacute phase after
stroke. Cyberpsychol Behav 2002;5:207-11.
7. Jack D, Boian R, Merians AS, et al. Virtual reality-enhanced
stroke rehabilitation. IEEE Trans Neural Syst Rehabil Eng 2001;
9:308-18.
8. Zhan S, Ottenbacher KJ. Single subject research designs for
disability research. Disabil Rehabil 2001;23:1-8.
9. Hamm N, Curtis D. Normative data for the Purdue Pegboard on a
sample of adult candidates for vocational rehabilitation. Percept
Mot Skills 1980;50:309-10.
10. Nordenskiold UM, Grimby G. Grip force in patients with rheu-
matoid arthritis and fibromyalgia and in healthy subjects. A study
with the Grippit instrument. Scand J Rheumatol 1993;22:14-9.
11. Nakayama H, Jorgensen HS, Raaschou HO, Olsen TS. Recovery
of upper extremity function in stroke patients: the Copenhagen
Stroke Study. Arch Phys Med Rehabil 1994;75:394-8.
12. Early Supported Discharge Trialists. Services for reducing dura-
tion of hospital care for acute stroke patients. Cochrane Database
Syst Rev 2002;(1):CD000443.
13. Nelson DL, Konosky K, Fleharty K, et al. The effects of an occupa-
tionally embedded exercise on bilaterally assisted supination in per-
sons with hemiplegia. Am J Occup Ther 1996;50:639-46.
Suppliers
a. AB Detector, Box 17124, 402 61 Gothenburg, Sweden.
b. SensAble Technologies Inc, 15 Constitution Way, Woburn, MA
01801.
c. Reachin Technologies AB, rstaangsvagen 24, 117 43 Stockholm,
Sweden.
d. StereoGraphics Corp, 2171 E Francisco Blvd, San Rafael, CA,
94901.