ARTICLE IN PRESS

Effects of Virtual Reality Training using Xbox Kinect on Motor

Function in Stroke Survivors: A Preliminary Study

Dae-Sung Park, PT, PhD,* Do-Gyun Lee, PT, MSc,† Kyeongbong Lee, PT, PhD,‡ and
GyuChang Lee, PT, PhD§

Background: Although the Kinect gaming system (Microsoft Corp, Redmond, WA)
has been shown to be of therapeutic benefit in rehabilitation, the applicability of
Kinect-based virtual reality (VR) training to improve motor function following a
stroke has not been investigated. This study aimed to investigate the effects of VR
training, using the Xbox Kinect-based game system, on the motor recovery of pa-
tients with chronic hemiplegic stroke. Methods: This was a randomized controlled
trial. Twenty patients with hemiplegic stroke were randomly assigned to either the
intervention group or the control group. Participants in the intervention group (n = 10)
received 30 minutes of conventional physical therapy plus 30 minutes of VR train-
ing using Xbox Kinect-based games, and those in the control group (n = 10) received
30 minutes of conventional physical therapy only. All interventions consisted of daily
sessions for a 6-week period. All measurements using Fugl–Meyer Assessment (FMA-
LE), the Berg Balance Scale (BBS), the Timed Up and Go test (TUG), and the 10-
meter Walk Test (10mWT) were performed at baseline and at the end of the 6 weeks.
Results: The scores on the FMA-LE, BBS, TUG, and 10mWT improved signifi-
cantly from baseline to post intervention in both the intervention and the control
groups after training. The pre-to-post difference scores on BBS, TUG, and 10mWT
for the intervention group were significantly more improved than those for the
control group (P < .05). Conclusions: Evidence from the present study supports
the use of additional VR training with the Xbox Kinect gaming system as an ef-
fective therapeutic approach for improving motor function during stroke rehabilitation.
Key Words: Stroke—Motor activity—Virtual reality—Video game.
© 2017 National Stroke Association. Published by Elsevier Inc. All rights reserved.

From the *Department of Physical Therapy, Konyang University,
Daejeon, Republic of Korea; †Department of Physical Therapy, Misodle
Hospital, Seoul, Republic of Korea; ‡Physical Therapy Part, Physi-
cal and Rehabilitation Medicine, Samsung Medical Center, Republic
of Korea; and §Department of Physical Therapy, Kyungnam Uni-
versity, Changwon, Republic of Korea.
Received February 1, 2017; revision received April 30, 2017; accepted
May 14, 2017.
Conflict of interest: No financial or nonfinancial competing inter-
est exists for either of this paper’s authors.
Address correspondence to GyuChang Lee, PT, PhD, Department
of Physical Therapy, Kyungnam University, 7 Kyungnamdaehak-ro,
Masanhappo-gu, Changwon-si, Gyeongsangnam-do 51767, Republic
of Korea. E-mail: leegc76@kyungnam.ac.kr; leegc76@hanmail.net.
1052-3057/$ - see front matter
© 2017 National Stroke Association. Published by Elsevier Inc. All
rights reserved.
http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2017.05.019
Introduction
The therapeutic approach to the rehabilitation of pa-
tients with hemiplegic stroke requires repetitive training
in combination with continuous modification of the train-
ing program to maintain patient engagement. Recreational
programs often provide emotional support and practi-
cal environments for intervention as a major therapeutic
effect.1 In patients with hemiplegic stroke, continuous
virtual feedback facilitates corticospinal activation to a
greater extent than the continuous visual feedback pro-
vided by the performance of activities in front of a real
mirror.2 Virtual reality (VR)-based feedback has also been
shown to increase activation of the primary sensorimo-
tor cortex, the supplementary motor area, and the
cerebellum during hand-clenching tasks.3
VR-based systems that use sensor technology to monitor
whole-body movements can elicit patients to perform

Journal of Stroke and Cerebrovascular Diseases, Vol. ■■, No. ■■ (■■), 2017: pp ■■–■■ 1
 ARTICLE IN PRESS
2 D.-S. PARK ET AL.
high-intensity and high-energy movements. Kinect- Participants
based systems (Microsoft Corp, Redmond, WA) reduce
The study participants were inpatients at the rehabil-
the demand on staff time for intervention and increase
itation center with a clinical diagnosis of hemiplegic stroke.
patients’ motivation toward rehabilitation. 4 Kinect-
To recruit participants, the purpose and procedures of
based VR training that includes auditory and visual
the study were described in an advertisement placed
stimulation, feedback information about “winning” or
throughout the hospital. Twenty-five volunteers were re-
“losing,” and repetitions of the same motion can provide
cruited and screened by a research assistant on the basis
a variable rehabilitation tool that reduces barriers to in-
of the following inclusion and exclusion criteria. The in-
dividuals performing rehabilitation exercises.3
clusion criteria were a period of more than 6 months
Commercial active gaming systems, such as those that
between stroke and randomization, hemiplegic stroke as
fall within the category of exergaming,5 those used by
diagnosed by a neurologist, a total score of 21 or greater
the Kinerehab system,4 and those for the Wii console
on the Mini–Mental State Examination (MMSE), no prob-
(Nintendo, Kyoto, Japan), can be effectively used in re-
lems with auditory or visual functioning, an ability to
habilitation programs to assist in patients’ recovery of motor
walk more than 10 m with or without assistive devices,
function.5,6 The most recently released Xbox Kinect system
not taking any medication that could influence balance,
has an RGB camera and a dual infrared depth sensor for
stable vital signs, and a capacity to provide informed
the automatic detection of limb and body position and
consent. The exclusion criteria were severe conditions that
motion. The system uses these elements to capture data
require medical care, such as uncontrolled blood pres-
to create a 3-dimensional human body model in real time,
sure or angina; musculoskeletal impairments of the lower
called an avatar, which allows for players to use their
extremity; psychological conditions; or the refusal to use
own body as the controls to play a game.7 The reliable
a video game. One volunteer refused to use a video game
visual feedback from the on-screen player’s avatar may
and was excluded.
provide accurate feedback on movement. The built-in com-
mercial games of the Kinect-based system can be used
for real-time rehabilitation training. An important ad- Ethical Consideration
vantage of the Kinect software development kit (SDK) All participants who fulfilled the inclusion criteria par-
or the OpenNI SDK for noncommercial use is the avail- ticipated in the study after the purpose and procedures
ability of open-source tools that allow for the adaptation of the study were fully explained to them. All proce-
or creation of new games.7 The OpenNI SDK enables de- dures were approved by the Sahmyook University
velopers to create or modify the software. Institutional Review Board, and all patients signed an in-
These Kinect-based games have been adapted and used formed consent before participating in the study.
to treat patients with neurological problems. The
Kinect-based choice reaction time task is 1 prototype game Randomization
that has been developed for step training in elderly pa-
tients in clinical practice.8 Kinect-based games have been The participants were allocated to the intervention group
effectively used in children with degenerative ataxia to and the control group using random number tables. The
improve whole-body coordination, dynamic balance ca- allocation of the groups was initially concealed. A re-
pacities, and step length over an 8-week training program.9 search assistant, with no involvement in any of the
Significant improvement has also been seen in the upper assessment or intervention components of the trial, opened
limb function of patients with chronic stroke after VR train- envelopes consecutively based on the patient enroll-
ing using the Kinect-based game “Fruit” during a 3-week ment sequence and forwarded the random number to the
training program.3 researcher in charge of enrollment to arrange interven-
Although the Kinect gaming system has been shown tion and outcome measurements for each participant.
to be of therapeutic benefit in rehabilitation, the appli-
cability of Kinect-based VR training to improve motor Interventions
function in patients following the occurrence of a stroke
The intervention group participated in a 30-minute VR
has not been evaluated. Therefore, the present study aimed
training session using Xbox Kinect, followed by a 30-
to evaluate the effects of additional VR training with Xbox
minute session of conventional physical therapy. The control
Kinect on the motor recovery of the lower extremity in
group participated in a 30-minute conventional physi-
patients with a chronic hemiplegic stroke.
cal therapy session only. All interventions were performed
daily for 6 weeks.
Materials and Methods
Design and Setting VR Training using Xbox Kinect
A randomized clinical trial was conducted in a reha- For the VR training, the Xbox Kinect system, which con-
bilitation hospital in Seoul, South Korea. sists of a Kinect sensor and console, was used. The Kinect
 ARTICLE IN PRESS
VIRTUAL REALITY TRAINING USING XBOX KINECT
Table 1. Description of each game in the Kinect Sports Pack and the Kinect Sports Pack 2
Games Contents
Boxing Players are encouraged to use their left and
right arms to punch and block both at the
head and the body levels.
Table tennis Players are encouraged to use their limbs and
trunk to hit a ball with a racket on a virtual
table.
Soccer Players are encouraged to use their lower
extremity, head and neck, and trunk while
standing kicking a ball in a virtual soccer
field.
Golf Players are encouraged to use a sensor to
substitute for the golf club in a virtual field.
After every swing, the golf ball is shown
flying down the fairway. Because full swing
and putt is needed, various types of swing
are available.
Ski Players are encouraged to shift their weight to
the right and left and up and down. The
virtual slope can be seen on the screen;
players have to avoid the barriers and
follow the slope.
Football Players are encouraged to move their lower
extremity to run with a ball. When they
catch the ball, they walk as fast as possible
to the end line while avoiding the opponent
players in the virtual field.
sensor is an infrared camera that can recognize the posi-
tions and motions of the player without the need for a
special controller. The console controls the various games.
For the VR training, the Xbox Kinect, console, and monitor
were set up in a dedicated space. The patient was placed
1.5-2 m away from the Kinect sensor. Before the start of
the training session, the research assistant adjusted the po-
sition of the sensor while the patient was sitting to ensure
optimal position and motion capture, and loaded games
into the system. After the setup was completed, the re-
search assistant demonstrated games included in the Kinect
Sports Pack and the Kinect Sports Pack 2. For the train-
ing, the following games were used: boxing, table tennis,
and soccer from the Kinect Sports Pack; and golf, ski, and
football from the Kinect Sports Pack 2. All games re-
quired the use of the upper and lower extremities while
standing. All games were provided to kindle the interest
of subjects with stroke. The descriptions of each game in
the Kinect Sports Pack and the Kinect Sports Pack 2 are
presented in Table 1. If a subject exhibited fatigue, abnor-
malities in breathing, or complained of pain, training was
stopped immediately. The games were adaptable to pa-
3
Performed movements
– Active movement of upper extremity
– Weight-shifting and weight-bearing training
– Balance training
– Active movement of upper extremity
– Trunk rotation
– Weight-shifting and weight-bearing training
– Balance training
– Active movement of lower extremity (hip flexion, abduction,
and external-internal rotation, knee flexion and extension,
ankle dorsiflexion and plantarflexion)
– Weight-shifting and weight-bearing training
– Balance training
– Active movement of upper extremity
– Trunk rotation
– Weight-shifting and weight-bearing training
– Balance training
– Active movement of lower extremity (hip flexion, abduction,
and external-internal rotation, knee flexion and extension,
ankle dorsiflexion and plantarflexion
– Trunk rotation
– Weight-shifting and weight-bearing training
– Balance training
– Active movement of lower extremity (hip flexion, abduction,
and external-internal rotation, knee flexion and extension,
ankle dorsiflexion and plantarflexion
– Trunk rotation
– Weight-shifting and weight-bearing training
– Balance training
tients with different levels of function after stroke. Verbal
encouragement was provided by a therapist to elicit maximal
effort. Following a demonstration of the games, partici-
pants stood up and practiced the games. To prevent fall
events during training, patients with stroke performed stand-
ing activities with a harness mounted to the ceiling. The
programs were conducted for a total of 30 minutes per
session. During the 6 weeks of the intervention, partici-
pants used all the provided games.
Conventional Physical Therapy
The conventional physical therapy regimen included
tasks from the traditional treatment program, which in-
cluded a range of motion exercises, muscle strengthening,
functional training, balance training, and gait training.
The specific tasks were selected by the therapist based
on the requirement of each patient. Usually, the tech-
niques of neurodevelopmental treatment and proprioceptive
neuromuscular facilitation were selected by the physi-
cal therapists in charge.10,11 The program was performed
for 30 minutes.
 ARTICLE IN PRESS
4 D.-S. PARK ET AL.
Outcome Measures and Follow-up
The clinical profile of each participant was recorded,
including affected side, type of stroke, duration of time
after stroke, MMSE score, and Brunnstrom stage of re-
covery. All participants were assessed at baseline and at
the end of the 6 weeks by a trained physical therapist
who was blinded to participants’ group allocation. In ad-
dition, a research assistant assessed the rate compliance
to participation in the planned sessions during the 6 weeks.
Primary Outcome Measure
Motor recovery was measured using the lower extrem-
ity subscale of the Fugl–Meyer Assessment (FMA). The
FMA provides a quantitative assessment of the recov-
ery of motor and sensory function in patients who have
sustained a stroke. The FMA has a 3-point scale: 0 (not
able to perform), 1 (partially able to perform), and 2 (per-
fectly able to perform). The total possible score for lower
extremity function is 34. It was reported that reliability
of the FMA for testing motor performance in patients fol-
lowing stroke is high.12
Secondary Outcome Measures
Secondary outcome measures included balance and gait.
Balance was measured using the Berg Balance Scale (BBS),
and gait was evaluated using the Timed Up and Go test
(TUG) and the 10-meter Walk Test (10mWT). The BBS
is a 14-item scale designed to measure balance function
and is a valid instrument to use to quantitatively de-
scribe balance function in a clinical population and to
evaluate the effectiveness of interventions in clinical prac-
tice and research.13,14 The TUG test assesses a person’s
mobility function using components that require both static
and dynamic balance. The outcome measure of the TUG
test is the time required for a person to rise from a chair,
walk 3 m, turn around, walk back to the chair, and sit
down. The TUG test has excellent reliability (Intraclass
Correlation Coefficient > .95).15 The 10mWT is recom-
mended to assess functional walking capacity; the test
has high inter- and intra-rater reliabilities.16 In the 10mWT,
participants are instructed to walk at a comfortable pace,
and the time taken to travel 10 m is recorded with a stan-
dard stopwatch. Participants are allowed to use their usual
walking devices during the test.
Sample Size Calculation
Using the G Power 3.1 statistical tool to achieve a sta-
tistical power of 80% with statistical significance at P < .05
(2-tailed test) and a large effect size of r = .8 (based on
FMA results and the minimal detectable difference [MDC]
of a previous study and 10% attrition rate),17 a total sample
size of 30 participants was required in the present trials.
Data Analysis
We used SPSS Statistics 18.0 (IBM Corporation, Armonk,
NY) for statistical analysis. The demographic variables
of participants are reported as means and standard de-
viations. Student’s t-tests and exact tests were used to
compare demographic data between the intervention group
and the control group. According to the Kolmogorov-
Smirnov test, the distribution of scores for the primary
and secondary outcome measures did not deviate sig-
nificantly from normality. Paired t-test analysis was used
to evaluate changes in outcomes scores from pre- to post
intervention. Differences between the 2 groups at the end
of the 6-week intervention period were assessed using
the independent t-test. Differences were considered sig-
nificant at P < .05.
Results
Over the course of the 6-week trial, 2 participants from
the intervention group dropped out because of early dis-
charge from the hospital, and 2 in the control group did
not conform to the required participation regimen in phys-
ical therapy sessions. Therefore, the final measurements
of 20 stroke survivors were considered; a flow chart of the
number of participants is shown in Figure 1. The rate of
compliance to the planned sessions during the 6-week trial
is shown in Table 2. In addition, the characteristics of the
participants who were included in the analyses are shown
in Table 2. We collected side effects data, but no partici-
pants reported any side effects like nausea, dizziness,
headache, or motion sickness while using the VR system.
All participants in the trial had sustained strokes more
than 6 months before the trial, had an MMSE score greater
than 21, and were classified within the following
Brunnstrom stages of motor recovery: 3, 4, 5, or 6. There
were no significant differences at baseline in the distri-
bution of characteristics among the participants in the
intervention group and the control group with regard to
age, gender, affected side, types of stroke, duration after
stroke, MMSE score, or Brunnstrom stage of recovery.
Pre- and post-intervention scores on the primary and
secondary outcomes are reported in Table 3. Scores on
the FMA-LE, BBS, TUG, and 10mWT improved from base-
line over the 6 weeks of training in both the intervention
group and the control group. Pre-to-post difference scores
on the FMA-LE for the intervention group were not sig-
nificantly greater than those for the control group. However,
for the BBS, TUG, and 10mWT, the intervention group
demonstrated significant improvement as compared with
the control group (P < .05).
Discussion
Main Findings
To our knowledge, this was the first study to evalu-
ate the effects of VR training using the Xbox Kinect system
 ARTICLE IN PRESS
VIRTUAL REALITY TRAINING USING XBOX KINECT 5

Figure 1. Flow diagram of the present study.

Table 2. Characteristics of participants in the intervention group and the control group

Intervention group (n = 10) Control group (n = 10)

Age, y, mean (SD) 62.00 ± 17.14 65.30 ± 10.51

Gender (male/female) 5/5 5/5
Affected side (left/right) 4/6 7/3
Type of stroke (infarct/hemorrhage) 8/2 5/5
Duration after stroke 10.78 ± 7.06 14.10 ± 7.73
MMSE 23.40 ± 4.12 24.00 ± 2.94
Brunnstrom stage (stage 3/4/5/6) 2/2/3/3 3/3/2/2
Compliance rate (%) 92.30 ± 13.53 83.75 ± 15.64

Abbreviations: MMSE, Mini–Mental State Examination; SD, standard deviation.

Values are mean ± SD or mode.

on the motor recovery of patients with hemiplegic stroke. Previous studies have reported improved upper ex-
The most important finding of the present study was that tremity function in patients with stroke following use of
additional VR training using the Xbox Kinect can be a commercial VR gaming systems, much like the Kinect
positive intervention for improving motor function. system with its open-source–based VR game repertoire

Table 3. Baseline and post-intervention scores, within and between groups

Intervention group (n = 10) Control group (n = 10)

Between
Pretest Posttest △Values Pretest Posttest △Values groups i

FMA-LE 16.30 ± 10.52 26.10 ± 7.31 −9.80 ± 4.85* 21.30 ± 8.82 27.50 ± 5.19 −6.20 ± 5.22* −3.60 ± 2.25 −1.597
BBS 35.80 ± 8.61 50.00 ± 6.27 14.20 ± 4.26* 37.30 ± 11.98 44.70 ± 7.47 7.40 ± 5.78* 6.80 ± 2.27 2.995*
TUG 44.35 ± 33.49 44.93 ± 27.41 −7.79 ± 3.98* 44.93 ± 27.41 41.16 ± 27.68 −3.78 ± 4.51* −4.01 ± 1.90 −2.107*
10mWT 57.14 ± 22.61 44.73 ± 20.87 −12.41 ± 5.60* 53.90 ± 26.06 47.77 ± 22.98 −6.13 ± 4.68* −6.29 ± 2.31 −2.725*

Abbreviations: 10mWT, 10-meter Walk Test; BBS, Berg Balance Scale; FMA-LE, Fugl–Meyer Assessment of Lower Extremity; TUG,
Timed Up and Go.
Values are mean ± SD.
*P < .05.
 ARTICLE IN PRESS
6 D.-S. PARK ET AL.
18
that was used in our study. Additionally, commercial
Microsoft Xbox Kinect games (table tennis, Light Race,
20,000 Leaks) have been shown to improve dynamic
balance training in children with degenerative ataxia.9 The
evaluation of VR training for lower extremity function
has not been reported. The reason for this may be that,
because of impairments in balance, it is difficult to leave
patients with hemiplegic stroke alone when they are using
Kinect-based games. As a therapist is usually required
to assist the patient, the Kinect sensor can make an error
in distinguishing the patient from the assistant. In ad-
dition, the Xbox Kinect game requires a fast response time
that might be difficult for some patients to achieve during
rehabilitation. Also, the therapist’s help is needed to start
and restart the commercial Kinect-based game.
In spite of these limitations, the completeness and good
graphic quality of the Kinect commercial gaming system
provides an immersive environment that is fun and ex-
trinsically motivating for participants, offering an alternative
to programs of repetitious exercise that are often per-
formed in front of a mirror, and can provide low intrinsic
motivation. In addition, VR systems used in rehabilita-
tion provide feedback to assist patients in performing a
task and can be used with little or no risk or burden of
high cost. The feedback on performance provided by the
Xbox Kinect systems is based on the movement of the
player; this is an advantage, as it replicates the real move-
ment of the patient’s impaired extremity rather than
replacing it with visually normal movement. Various ver-
sions of Kinect have been reported to have rehabilitative
functions, and the Kinect-based game console can be used
for stroke patients. In the present study, a combination
of 2 games (the Kinect Sports Pack and the Kinect Sports
Pack 2) was used for stroke patients. The programs re-
quired active movements, and the participants usually
performed the active movements of the hip, knee, or ankle,
as well as the upper extremity, on the affected side. VR
training may have contributed to the improvements in
balance and gait function observed in the participants in
the intervention group by encouraging more intense con-
centration on the task, which could increase motor
learning19 and enhance the physical performance of in-
dividuals with stroke during activities of daily living.20
Previous research has shown that repetitive movement
of the hemiplegic lower extremity increases neural
plasticity.20 Wüest et al21 used a 12-week VR training
program consisting of 5 rehabilitation games for balance
training in untrained elderly persons. The VR had a pos-
itive effect on improving participants’ balance and TUG
scores. This is in agreement with the results of our study.
In a previous study,17 the MDC and the 95% confidence
interval of BBS was 4.66 and 10%, respectively, in pa-
tients with stroke. In our study, there was a significant
improvement of 14.20 in BBS score for participants in the
VR training group, as compared with a 7.40 score im-
provement for the control group. Additionally, the TUG
score correlated with the strength of the affected ankle
plantar flexors (r = −.860), gait parameters (.620 < r < −.900),
and walking endurance (r = −.960).14 VR training can
improve walking endurance and walking function through
increased activities and participation in games. The MDC
of TUG was 7.84 in a previous study.17 In our study, there
were significant improvements in TUG scores in both
groups after training; however, the change in TUG scores
of the patients in the control group did not exceed the
MDC threshold. The minimal detectable change on the
FMA-LE score in patients with chronic stroke is 3.57, or
a 14% change from baseline.17 The increased FMA score
demonstrated was due to increased isolated movements.
Evidence from our trial supports the use of addition-
al VR training with the Xbox Kinect gaming system as
an effective therapeutic approach for improving motor
function during stroke rehabilitation.
Limitations
The limitations of our study need to be considered. First,
the dose of intervention was not the same between the
2 groups. As such, we could not conclude whether the
VR training method is a more effective approach than
conventional therapy. Second, we included only pa-
tients with a first instance of stroke who had not suffered
any cognitive impairment. Therefore, the results of this
study may not be generalizable to patients with stroke,
such as those who have undergone multiple strokes. Third,
the benefit observed in the VR training group may have
been influenced by the total intervention time for them
being more than that for the control group. Finally, there
were some errors in feedback introduced by the limita-
tions of the Kinect sensor to differentiate the participant
from the therapist providing assistance. Future studies
evaluating the therapeutic effectiveness of the Xbox Kinect
system in stroke rehabilitation will need to correct for
these limitations. We also recommend that the provi-
sion of a harness to assist shoulder movement and gait
aids to assist standing balance be considered for pa-
tients with severe motor impairment. This would allow
these patients to safely play the game. The harness will
need to be developed to limit recognition error by the
Kinect sensor.
References
1. Braddom RL. Physical medicine & rehabilitation. 3rd ed.
Philadelphia: Saunders, 2007.
2. Kang YJ, Park HK, Kim HJ, et al. Upper extremity
rehabilitation of stroke: facilitation of corticospinal
excitability using virtual mirror paradigm. J
Neuroengineering Rehabil 2012;9:71.
3. Bao X, Mao Y, Lin Q, et al. Mechanism of Kinect-based
virtual reality training for motor functional recovery of
upper limbs after subacute stroke. Neural Regen Res
2013;8:2904-2913.
 ARTICLE IN PRESS
VIRTUAL REALITY TRAINING USING XBOX KINECT
4. Chang YJ, Chen SF, Huang JD. A Kinect-based system
for physical rehabilitation: a pilot study for young adults
with motor disabilities. Res Dev Disabil 2011;32:2566-2570.
5. Mellecker RR, McManus AM. Active video games and
physical activity recommendations: a comparison of the
Gamercize Stepper, XBOX Kinect and XaviX J-Mat. J Sci
Med Sport 2014;17:288-292.
6. Saposnik G, Teasell R, Mamdani M, et al. Effectiveness
of virtual reality using Wii gaming technology in stroke
rehabilitation: a pilot randomized clinical trial and proof
of principle. Stroke 2010;41:1477-1484.
7. Davaasambuu E, Chiang CC, Chiang JY, et al. A Microsoft
Kinect based virtual rehabilitation system. In: The 5th
International Conference FITAT. Ulaanbaatar, Mongolia:
2012:44-50.
8. Garcia JA, Felix Navarro K, Schoene D, et al. Exergames
for the elderly: towards an embedded Kinect-based clinical
test of falls risk. Stud Health Technol Inform 2012;178:
51-57.
9. Ilg W, Schatton C, Schicks J, et al. Video game-based
coordinative training improves ataxia in children with
degenerative ataxia. Neurology 2012;79:2056-2060.
10. Thrasher TA, Zivanovic V, McIlroy W, et al. Rehabilitation
of reaching and grasping function in severe hemiplegic
patients using functional electrical stimulation therapy.
Neurorehabil Neural Repair 2008;22:706-714.
11. Mesci N, Ozdemir F, Kabayel DD, et al. The effects of
neuromuscular electrical stimulation on clinical
improvement in hemiplegic lower extremity rehabilitation
in chronic stroke: a single-blind, randomised, controlled
trial. Disabil Rehabil 2009;31:2047-2054.
12. Sanford J, Moreland J, Swanson LR, et al. Reliability of
the Fugl-Meyer assessment for testing motor performance
in patients following stroke. Phys Ther 1993;73:447-454.
7
13. Berg KO, Wood-Dauphinee SL, Williams JI, et al.
Measuring balance in the elderly: validation of an
instrument. Can J Public Health 1992;83:S7-11.
14. Blum L, Korner-Bitensky N. Usefulness of the Berg
Balance Scale in stroke rehabilitation: a systematic review.
Phys Ther 2008;88:559-566.
15. Ng SS, Hui-Chan CW. The timed up & go test: its
reliability and association with lower-limb impairments
and locomotor capacities in people with chronic stroke.
Arch Phys Med Rehabil 2005;86:1641-1647.
16. Scivoletto G, Tamburella F, Laurenza L, et al. Validity
and reliability of the 10-m walk test and the 6-min walk
test in spinal cord injury patients. Spinal Cord 2011;49:
736-740.
17. Hiengkaew V, Jitaree K, Chaiyawat P. Minimal detectable
changes of the Berg Balance Scale, Fugl-Meyer Assessment
Scale, Timed “Up & Go” Test, gait speeds, and 2-minute
walk test in individuals with chronic stroke with different
degrees of ankle plantarflexor tone. Arch Phys Med
Rehabil 2012;93:1201-1208.
18. Sin H, Lee G. Additional virtual reality training using
Xbox Kinect in stroke survivors with hemiplegia. Am J
Phys Med Rehabil 2013;92:871-880.
19. Walker ML, Ringleb SI, Maihafer GC, et al. Virtual
reality-enhanced partial body weight-supported treadmill
training poststroke: feasibility and effectiveness in 6
subjects. Arch Phys Med Rehabil 2010;91:115-122.
20. Yang YR, Tsai MP, Chuang TY, et al. Virtual reality-based
training improves community ambulation in individuals
with stroke: a randomized controlled trial. Gait Posture
2008;28:201-206.
21. Wüest S, Borghese NA, Pirovano M, et al. Usability and
effects of an exergame-based balance training program.
Games Health J 2014;3:106-114.