Somatosensory & Motor Research

ISSN: 0899-0220 (Print) 1369-1651 (Online) Journal homepage: http://www.tandfonline.com/loi/ismr20

Effects of Kinect-based virtual reality game

training on upper extremity motor recovery in
chronic stroke

Ayhan Aşkın, Emel Atar, Hikmet Koçyiğit & Aliye Tosun

To cite this article: Ayhan Aşkın, Emel Atar, Hikmet Koçyiğit & Aliye Tosun (2018): Effects of
Kinect-based virtual reality game training on upper extremity motor recovery in chronic stroke,
Somatosensory & Motor Research, DOI: 10.1080/08990220.2018.1444599

To link to this article: https://doi.org/10.1080/08990220.2018.1444599

Published online: 13 Mar 2018.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=ismr20
SOMATOSENSORY & MOTOR RESEARCH, 2018
https://doi.org/10.1080/08990220.2018.1444599

ORIGINAL ARTICLE

Effects of Kinect-based virtual reality game training on upper extremity motor

recovery in chronic stroke
Ayhan Aşkına , Emel Atarb
itc and Aliye Tosuna
, Hikmet Koçyig
a
Physical Medicine and Rehabilitation, Izmir Katip Celebi University, Izmir, Turkey; bPhysical Medicine and Rehabilitation, Sultan Abdulhamid
Han Training and Research Hospital, Uskudar, Turkey; cPhysical Medicine and Rehabilitation, Izmir Katip Celebi University Atat€ urk Training
and Research Hospital, Izmir, Turkey

ABSTRACT ARTICLE HISTORY

Background: Therapeutic benefits of Kinect-based virtual reality (VR) game training in rehabilitation Received 12 February 2018
encourage its use to improve motor function. Accepted 19 February 2018
Objective: To assess the effects of Kinect-based VR training on motor recovery of the upper extremity
KEYWORDS
and functional outcomes in patients with chronic stroke.
Motor function;
Methods: In this randomized controlled trial, group A received 20 sessions of physical therapy (PT) þ 20 rehabilitation; stroke;
sessions of Kinect-based VR training and group B received only 20 sessions of PT. Clinical outcome meas- virtual reality
ures were assessed at baseline and at the end of the treatments. Primary outcome measures that assess
stroke patients’ motor function included upper extremity (UE) Fugl-Meyer Assessment (FMA). Secondary
outcome measures were Brunnstrom Recovery Stages (BRS), Modified Ashworth Scale (MAS), Box and
Block test (BBT), Motricity index (MI), and active range of motion (AROM) measurement.
Results: Statistically significant improvements in game scores (p < 0.05) were observed in group A. In
within-group analysis, there were statistically significant improvements in all clinical outcome measures
except for the BRS-hand, MAS-distal, and MAS-hand in group A; MAS-(proximal, distal, hand) and BRS-
(UE, hand) in group B compared with baseline values. Differences from baseline of FMA, MI, and AROM
(except adduction of shoulder and extension of elbow) were greater in group A (p < 0.05).
Conclusions: To conclude, our results suggest that the adjunct use of Kinect-based VR training may
contribute to the improvement of UE motor function and AROM in chronic stroke patients. Further
studies with a larger number of subjects with longer follow-up periods are needed to establish its
effectiveness in neurorehabilitation.

Introduction 2017). Numerous treatments have been used in stroke

As one of the leading causes of death and disability, stroke
has been described as a worldwide epidemic. It is the second
most common cause of death and the third most common
cause of disability from all causes (Feigin et al. 2017).
Recovery after stroke is frequently incomplete and a substan-
tial number of stroke survivors are left with motor, sensory,
and/or cognitive impairments. Upper extremity motor impair-
ments include muscle weakness, alterations in muscle tone,
impaired motor control, limitations in the range of motions
of joints, contractures, or laxity. The major goal of stroke
rehabilitation is improving functional levels of patients; how-
ever, optimal recovery of upper extremity functions cannot
be achieved in most of the survivors (Lang et al. 2013).
Therefore, patients usually have difficulties in performing
activities of daily living such as reaching, grasping, picking
up or holding objects, using tools like cell phones, and par-
ticipation in social activities (Meyer et al. 2015; Silva
et al. 2017).
Neurorehabilitation strategies aim to improve functional
recovery via enhancing neuronal plasticity, relearning proc-
esses, and functional reorganization (Raffin and Hummel
rehabilitation to accelerate the recovery. Beneficial effects of
constraint-induced movement therapy (CIMT), mental prac-
tice, mirror therapy, interventions for sensory impairment,
and a relatively high dose of repetitive task practice have
been revealed in studies (Pollock et al. 2014). Likewise, restor-
ation of motor function should focus on high-intensity,
repetitive task-specific practice with feedback on perform-
ance (Langhorne et al. 2009). On the other hand, because of
longer waiting times, increased medical costs, or inadequate
access to rehabilitation specialists, patients with stroke are in
need of low-cost motivational rehabilitation technologies
(Lange et al. 2011). The use of virtual reality (VR) games as
rehabilitation tools, such as the Xbox Kinect or Wii, has
recently gained much interest in the rehabilitation arena with
moderate clinical evidence (Sin and Lee 2013).
The Kinect-based VR system for the Xbox enables users to
control and interact with the game console without the need
to touch a game controller. It has an infra-red depth sensor
for the detection of limb and motion, and players use their
own body as the controls to play a game (Bao et al. 2013).
Through using the motivational characteristics of Kinect-
based non-immersive VR games, interventions can be

CONTACT Ayhan Aşkın ayhanaskin@hotmail.com Physical Medicine and Rehabilitation, Izmir Katip Celebi University, Izmir 35620, Turkey
ß 2018 Informa UK Limited, trading as Taylor & Francis Group
2 A. AŞKIN ET AL.
developed that include a high number of repetitions, while
keeping the patients engaged and motivated (Chang et al.
2011; Johnson et al. 2018). Hence, the use of these systems
has led to practice-dependent enhancement of the function
of the affected limb through the facilitation of cortical
reorganization and provides rehabilitation training for stroke
patients (Saposnik et al. 2011; Laver et al. 2017). Functional
magnetic resonance imaging studies revealed that Kinect-
based VR training facilitates functional neuroplasticity in the
brain after stroke (You et al. 2005; Bao et al. 2013).
Therapeutic benefits of Kinect-based VR game training in
rehabilitation encourages its use to improve motor function
(Park et al. 2017). Besides, several clinical studies have sug-
gested that additional VR with commercially available com-
puter gaming improved outcomes compared with
conventional physical therapy (PT) in adults post-stroke (Sin
and Lee 2013; Thomson et al. 2014; Park et al. 2017).
However, relatively few controlled trials are available about
stroke-specific Kinect-based task-oriented software interven-
tions (Laver et al. 2017).
Therefore, the present study aimed to investigate the
effects of additional VR training on the motor recovery of the
upper extremity and functional outcomes in patients with
chronic stroke. We hypothesized that Kinect-based VR gam-
ing combined with PT would yield enhancement of upper
extremity motor function and thus functional outcomes.
Materials and methods
Participants
Forty chronic stroke patients (mean ages: 55.1 ± 10.2 years;
range 34–72 years; 28 males, 12 females) with upper limb
paresis who met the following inclusion criteria were
included in the study: (1) ischemic or hemorrhagic stroke; (2)
adults (>18 years old); (3) stroke onset >6 months; (4)
absence of cognitive, visual impairment or neglect that made
it impossible to follow the instructions during treatment ses-
sions; and (5) voluntary shoulder, elbow, and wrist movement
that can perform game commands. Two patients were lost to
follow-up and the study was completed with 38 patients
(mean ages: 55.0 ± 10.4 years; range 34–72 years; 27 males, 11
females). Patients were excluded if they were clinically
unstable or had concomitant neurodegenerative disorders,
other intracranial diseases, history of seizure, significant med-
ical or psychiatric illness, fixed contractures or bony deform-
ities of the affected extremity, taking any drug that could
affect balance, and severe cognitive problems (Mini-mental
test score of <20/30).
Demographic data of the patients (age, gender, educa-
tional status, body mass index), time since stroke, affected
extremity, types of stroke, and comorbidities were recorded.
Participants were informed about the study and provided
written informed consents. The protocol was performed in
accordance with the ethical standards laid down in the 1964
Declaration of Helsinki and approved by the local Research
Ethics Committee (Approval number: 03.06.2015/103).
Study design
This was an assessor-blinded, randomized controlled study.
Forty patients were randomized into two groups by generat-
ing a random allocation sequence with a computer software
program. After randomization, group A received 20 sessions
of PT (5 days/week, 4 weeks) and 20 sessions of Kinect-based
VR game training (1 hour a day, 5 days/week, 4 weeks). Group
B received only 20 sessions of PT.
Kinect-based VR game training
Xbox Kinect (Xbox 360, Microsoft, Redwood, WA, USA) was
used for training. The Xbox Kinect platform uses an infra-red
camera to capture users’ body movement in 3D space for
interaction within game activities and does not require the
user to hold an interface device. Instead, the user’s body is
the game controller operating in 3D space. For the training,
the screen and Kinect sensor were set up in a room that was
not influenced by external factors. The sensor was positioned
and the patients participated while either sitting or standing
1.5–2 m from the screen. Two games which required the use
of the upper extremity were selected. These games were a
part of the free software platform called “KineLabs”. KineLabs
has been developed by a research team of Hong Kong
Polytechnic University to facilitate stroke patients to have
rehabilitation training and aged people to do exercise (Tong
2014). The details of these two games are:
 Good View Hunting: The participant is asked to delete the
dirty spots in different regions of the screen or to clean a
water stain, fog, kid’s drawing or car dust using his or her
affected extremity. A high score can be obtained by mov-
ing fast. The patients usually performed the active move-
ments of shoulder and elbow on the affected side in this
game part.
 Hong Kong Chef: It consists of two sections. Hong Kong
Egg Tart and BBQ Pork with Rice. The aim of both sec-
tions is to achieve the stages of making the food (drag-
ging and dropping ingredients, mixing, cutting and
cooking). Patients are required to be as fast as possible. A
high score can be obtained by moving fast. Particularly
active shoulder, elbow, and wrist movements are per-
formed in this game part. The Kinect game system and
the position of the patient are shown in Figure 1.
Before the treatment sessions, demonstrations about the
games were made by an experienced doctor. In all sessions,
the patients played games under the supervision of the same
trained medical staff. Patients were allowed to play all three
game parts for 1 hour, respectively. The highest scores for
each of the three game parts were recorded before and
after treatment.
Physical therapy
All the participants received the same standardized PT proto-
cols in neurological rehabilitation. The PT program included

Figure 1. Kinect game system and patient position.
activities to improve the active range of motion, strength,
flexibility, transfers, posture, balance, coordination, and activ-
ities of daily living.
Clinical outcome measures
Clinical outcome measures were assessed at baseline and at
the end of the treatments by an experienced assessor who
was familiarized with the scales and tests used in this study
and who was unaware of the group assignment.
Primary outcome measures that assess stroke patients’
motor function included upper extremity Fugl-Meyer
Assessment (FMA). Secondary outcome measures were
Brunnstrom Recovery Stages (BRS), Modified Ashworth Scale
(MAS), Box and Block test (BBT), Motricity index (MI), and
active range of motion (AROM) measurement.
Primary outcome measure
The Fugl-Meyer Assessment (FMA) scale is a 226-point multi-
item Likert-type scale developed as an evaluative measure of
recovery from stroke. It is divided into five domains: motor
function, sensory function, balance, joint range of motion,
and joint pain. Each domain contains multiple items, each
scored on a 3-point ordinal scale (0 ¼ cannot perform,
1 ¼ performs partially, 2 ¼ performs fully). The motor domain
includes items measuring movement, coordination, and reflex
action about the shoulder, elbow, forearm, wrist, hand, hip,
SOMATOSENSORY & MOTOR RESEARCH 3
knee, and ankle. The motor score ranges from 0 to a max-
imum of 100 points (normal motor performance), divided
into 66 points for the upper extremity (UE) (Fugl-Meyer
et al. 1975).
Secondary outcome measures
Brunnstrom Recovery Stages (BRS) is a six-stage evaluation
tool with three different parts concerning the UE, hand (H),
and the lower extremity (Stages 1–6, 1: no activity of the
limb; 2: spasticity appears, and weak basic flexor and exten-
sor synergies are present; 3: spasticity is prominent; the
patient voluntarily moves the limb, but muscle activation is
all within the synergy patterns; 4: patient begins to activate
muscles selectively outside the flexor and extensor synergies;
5: spasticity decreases; most muscle activation is selective
and independent from the limb synergies; and 6: isolated
movements in a smooth, well-coordinated manner). UE and
hand BRS were used in this study (Smith and Sharpe 1994).
The Modified Ashworth Scale (MAS) is considered the pri-
mary clinical measure of muscle spasticity in patients with
neurological conditions. It is a 6-point scale. Scores range
from 0 to 4, where lower scores represent normal muscle
tone and higher scores represent spasticity or increased
resistance to passive movement. In this study, spasticity was
rated as follows: 0 (MAS 0) ¼ no increased resistance; 1 (MAS
1) ¼ minimal resistance at the end of elbow ROM; 2 (MAS
1þ) ¼ minimal resistance throughout less than half of the
elbow ROM; 3 (MAS 2) ¼ clear resistance throughout most of
the elbow ROM; 4 (MAS 3) ¼ strong resistance, passive move-
ment is difficult; and 5 (MAS 4) ¼ rigid elbow extension
(Bohannon and Smith 1987).
Box and Block test (BBT) was used to measure unilateral
gross manual dexterity. This test consists of moving, one by
one, the maximum number of blocks from one compartment
of a box to another of equal size within 60 seconds. The
patients had 15 seconds to practice and then the number of
blocks moved within 60 seconds was recorded (Desrosiers
et al. 1994).
Motricity index (MI) provides a brief assessment of the
motor function of the arm and total arm score is obtained by
evaluation of pinch grip (0: no movement; 11: beginnings of
prehension; 19: grips cube but unable to hold against gravity;
22: grips cube, held against gravity but not against weak
pull; 26: grip cube against pull but weaker than the other
side; 33: normal pinch grip), elbow flexion and shoulder
abduction (0: no movement; 9: palpable contraction in
muscle but no movement; 14: movement seen but not
range/not against gravity; 19: full range against gravity, not
against resistance; 25: movement against resistance but
weaker than the other side; 33: normal power) (Collin and
Wade 1990).
Active range of motion (AROM) was measured using a
handheld goniometer. AROM of flexion, extension, adduction,
abduction, and internal–external rotation of the shoulder;
flexion and extension of the elbow; and flexion and exten-
sion of the wrist were measured.
4 A. AŞKIN ET AL.
Statistical analysis
A priori sample size based on the work of Sin and Lee
(2013) was calculated on the basis of changes in the Fugl-
Meyer UE scores. For calculation, the GPower (GPower
version 3.1.9.2, Du€sseldorf, Germany) program was used. In
the analysis for the t-test in independent groups using 5%
significance level and 80% power it was found that at least
20 patients (totally 40) in each group should be included in
the study.
Database management and statistical analyses were per-
formed by an independent researcher who was blinded to
group allocations. The statistical analysis was performed by
SPSS 16.0 (IBM Corporation, Armonk, NY, USA) and
MedCalc 14 (MedCalc Software, Ostend, Belgium) software.
The normal distribution of measured data was examined
by the Kolmogorov–Smirnov test and the homogeneity of
variance tested using Levene’s test of equality of variances.
Pearson chi-square test, Fisher’s Exact test, and independ-
ent samples t-test were performed to determine differences
between the demographic and clinical characteristic of the
groups. Wilcoxon’s signed-rank test and Mann–Whitney
U-test were used to determine differences within and
between the groups’ baseline and post-intervention out-
come parameters. A Mann–Whitney U-test was performed
for comparison of differences from baseline between the
groups. Numeric data are expressed as mean ± SD and
median (min–max). The variables examined in the 95%
confidence level and p-values less than 0.05 were consid-
ered significant.
Figure 2. Flow diagram of the study.
Results
No adverse events were reported by any of the participants.
A flow chart of the study is shown in Figure 2. Demographic
and clinical characteristics of the patients are given in
Table 1. No statistically significant difference was found in
baseline demographical, clinical characteristics, and outcome
measures of the participants (p > 0.05).
Clinical outcome measures of baseline and post-interven-
tion are given in Table 2. In group A, all patients improved
their game scores significantly (p < 0.05). In within-group ana-
lysis, there were statistically significant improvements in all
clinical outcome measures except for the BRS-hand, MAS-dis-
tal, and MAS-hand in group A; MAS-(proximal, distal, hand)
and BRS-(UE, hand) in group B compared with baseline val-
ues. In group A, while there were significant improvements
in the AROM of all directions of shoulder, elbow, and wrist;
in group B, significant improvements were revealed only in
the AROM of flexion, abduction, external rotation of the
shoulder, and extension of the elbow (p < 0.05).
Differences from baseline of FMA, MI, and AROM (except
adduction of shoulder and extension of elbow) were greater
in group A (p < 0.05) (Table 3).
Discussion
The results of this study showed significant improvements in
all clinical outcome measures except for the BRS-hand,
MAS-distal, and MAS-hand in the Kinect-based VR group;
MAS-(proximal, distal, hand) and BRS-(UE, hand) in the
 SOMATOSENSORY & MOTOR RESEARCH 5

Table 1. Demographic and clinical characteristics of the groups.

Group A (n ¼ 18) Group B (n ¼ 20)
PT þ Kinect PT p
Age (years) 53.27 ± 11.19 56.55 ± 9.85 0.344
Gender (female/male), n (%) 5/13 (28/72) 6/14 (30/70) 0.880
Educational status, n (%) 0.423
Primary school 10 (56) 8 (40)
High school 3 (16) 7 (35)
University 5 (28) 5 (25)
Body mass index 24.72 ± 3.48 24.45 ± 2.40 0.780
Time since stroke (months) 20.27 ± 5.47 19.40 ± 4.48 0.591
Affected extremity, n (%)
Right/left 8/10 (44/56) 10/10 (50/50) 0.732
Dominant/non-dominant 9/9 (50/50) 9/11 (45/55) 0.758
Type of stroke (ischemic/hemorrhagic), n (%) 16/2 (89/11) 19/1 (95/5) 0.485
Comorbidities, n (%)
Transient ischemic attack 1 (6) 1 (5) 0.939
Diabetes mellitus 5 (28) 6 (30) 0.880
Hypertension 7 (39) 10 (50) 0.492
Ischemic heart disease 5 (28) 7 (35) 0.632
Arrhythmia 1 (6) 1 (5) 0.939
Smoke 8 (44) 9 (45) 0.973
Alcohol 2 (11) 4 (20) 0.453
Values are mean ± standard deviation or n (%).
PT: physical therapy.
Statistical analysis: Pearson chi-square test, Fisher’s Exact test, independent samples t-test.

Table 2. Baseline/post-intervention comparisons in outcome measures within and between the groups.
Group A (n ¼ 18) Group B (n ¼ 20)
PT þ Kinect PT
Baseline Post-intervention Baseline Post-intervention p1 p2 p3
MAS
Proximal 1.0 (0.0–3.0) 1.0 (0.0–2.0) 1.5 (0.0–3.0) 1.0 (0.0–3.0) 0.063 0.025* 0.157
Distal 1.0 (0.0–3.0) 1.0 (0.0–3.0) 1.0 (0.0–4.0) 1.0 (0.0–4.0) 0.059 0.157 0.317
Hand 1.0 (0.0–3.0) 1.0 (0.0–3.0) 2.0 (0.0–4.0) 1.0 (0.0–4.0) 0.076 0.157 0.564
Brunnstrom Stages
UE 3.0 (2.0–4.0) 4.0 (2.0–4.0) 3.0 (1.0–5.0) 3.0 (1.0–5.0) 0.900 0.046* 0.317
Hand 3.0 (2.0–4.0) 3.0 (2.0–4.0) 3.0 (1.0–5.0) 3.0 (1.0–5.0) 0.242 0.157 0.999
Fugl-Meyer Assessment 39.0 (22.0–56.0) 42.0 (23.0–58.0) 30.5 (19.0–58.0) 31.5 (19.0–58.0) 0.319 <0.001* 0.034*
Box and Block test 4.0 (0.0–64.0) 5.0 (0.0–69.0) 0.5 (0.0–39.0) 1.0 (0.0–38.0) 0.668 0.044* 0.014*
Motricity index 40.0 (20.0–77.0) 55.5 (23.0–93.0) 37.5 (23.0–30.0) 40.0 (23.0–73.0) 0.369 0.001* 0.006*
Kinect VR game scores
Good View Hunting 1270.0 (760.0–1610.0) 1565.0 (910.0–1980.0) – – – <0.001* –
Hong Kong Egg Tart 46.0 (19.0–64.0) 64.0 (41.0–88.0) – – – <0.001* –
BBQ Pork with Rice 35.0 (9.0–54.0) 55.5 (17.0–89.0) – – – <0.001* –
Values are median (minimum–maximum). Statistically significant (p < 0.05).
MAS: Modified Ashworth Scale; PT: physical therapy; UE: upper extremity; VR: virtual reality.
Statistical analysis: p1, baseline comparison between the groups (Mann–Whitney U-test); p2, baseline and post-intervention comparison within group A (Wilcoxon
test); p3, baseline and post-intervention comparison within group B (Wilcoxon test).

control group. In addition, FMA, BBT, MI scores, and AROM
(shoulder flexion–abduction–external rotation and elbow
extension) were significantly increased in both groups, how-
ever, these changes were greater in the Kinect-based VR
group except for BBT score and AROM (adduction of shoul-
der and extension of elbow). Although both groups bene-
fited from rehabilitative procedures to some extent, Kinect-
based VR training seemed to contribute more to the
improvement of motor function and AROM of affected upper
limbs in chronic stroke patients.
Neurological reorganization plays an important role in the
restoration of UE functions. It can extend for a much longer
period of time and is of particular interest because it can be
influenced by rehabilitation training (Hebert et al. 2016).
There is strong evidence suggesting that a high dose of
repetitive task practice can improve recovery. Thus, increas-
ing therapy dose, also in the chronic phase of the disease,
might be a critical factor to achieve a better outcome
(Pollock et al. 2014). Kinect-based VR training is a novel tech-
nology which is increasingly being used in neurorehabilita-
tion to modulate rehabilitation sessions (Laver et al. 2017).
Many VR systems, often generic (i.e., not developed for
rehabilitation purposes) commercial computer games, are
used to perform a series of exercises (Sin and Lee 2013;
Saposnik et al. 2016). These systems provide a non-immersive
environment that is fun and extrinsically motivating for
patients (Park et al. 2017). And also, VR-based training is able
to provide high rehabilitation doses, both in terms of active
training time and repetitions per session. In our study,
patients received a total of 1 hour/day of active VR-based
training of the UE, with a total duration of the training ses-
sions of 20 hours (including breaks and time between
games). We think that this training duration—compatible
with other studies in the literature—is enough to accept the
treatment as high-dose repetitive intensive rehabilitation
(Perez-Marcos et al. 2017).
6 A. AŞKIN ET AL.

Table 3. Comparison of differences from baseline between the groups. results also revealed significant improvements in FMA, MI,
Differences from baseline and AROM (except adduction of shoulder and extension of
Group A (n ¼ 18) Group B (n ¼ 20) elbow) in the Kinect-based training group when compared
PT þ Kinect PT p with the control group. However, we failed to find any sig-
Modified Ashworth Scale nificant difference in BBT scores in-between. Since the VR
Proximal 0.0 (1.0–0.0) 0.0 (2.0–1.0) 0.494 game selected in our study did not include training specific-
Distal 0.0 (1.0–0.0) 0.0 (1.0–0.0) 0.491
Hand 0.0 (1.0–0.0) 0.0 (1.0–1.0) 0.618 ally for manual dexterity, other games specifically designed
Brunnstrom Stages for manual dexterity can be added to the VR rehabilitation
UE 0.0 (0.0–1.0) 0.0 (0.0–1.0) 0.122 program for this purpose.
Hand 0.0 (0.0–1.0) 0.0 (0.0–0.0) 0.131
Fugl-Meyer Assessment 2.0 (1.0–10.0) 0.0 (0.0–2.0) <0.001* On the contrary, Adie et al. (2017) recently investigated
Box and Block test 0.5 (3.0–7.0) 0.0 (1.0–3.0) 0.744 the efficacy of using Nintendo Wii Sports to improve affected
Motricity index 6.0 (0.0–17.0) 0.0 (0.0–6.0) 0.019* arm function after stroke and concluded that the Wii was not
Active range of motion
Shoulder superior to arm exercises in home-based rehabilitation
Flexion 20.0 (0.0–40.0) 0.0 (0.0–10.0) <0.001* although it was more expensive than arm exercises. Similarly,
Extension 5.0 (0.0–30.0) 0.0 (0.0–10.0) 0.035* Kong et al. (2016) revealed that 12 sessions of augmented
Abduction 20.0 (10.0–30.0) 0.0 (0.0–10.0) <0.001*
Adduction 0.0 (0.0–10.0) 0.0 (0.0–10.0) 0.131 upper limb exercises via Wii gaming or conventional therapy
Internal rotation 5.0 (0.0–25.0) 0.0 (0.0–10.0) 0.005* over a 3-week period was not effective in enhancing upper
External rotation 10.0 (0.0–20.0) 0.0 (0.0–10.0) 0.002* limb motor recovery compared to controls. We assume that
Elbow
Flexion 0.0 (0.0–15.0) 0.0 (0.0–10.0) 0.040* differences in the design of the studies, sample sizes, and
Extension 0.0 (0.0–20.0) 0.0 (0.0–10.0) 0.637 gaming systems might have possibly caused these contra-
Wrist dictory results. The game used in this study is not commer-
Flexion 5.0 (0.0–20.0) 0.0 (0.0–10.0) 0.009*
Extension 0.0 (0.0–30.0) 0.0 (0.0–10.0) 0.022* cial, but has been developed by a research team of Hong
Values are median (minimum–maximum). Statistically significant (p < 0.05). Kong Polytechnic University specifically for stroke patients.
PT: physical therapy; UE: upper extremity. Therefore, significant improvements in most of the clinical
Statistical analysis: Mann–Whitney U-test. outcome measures superior to conventional rehabilitation
could be achieved. There is no other study in the literature
Recent studies revealed that the Kinect-based VR gaming that uses our Kinect-based VR gaming system, but our results
system is an effective therapeutic approach for improving are similar to the current data (Laver et al. 2017).
motor function during stroke rehabilitation. Sin and Lee Patients sometimes report low levels of fatigue and stress
(2013) evaluated the effects of additional VR training using generated during the rehabilitation training sessions. In a
Xbox Kinect on UE function (including AROM, FMA, and BBT) recent study that used Wii games for UE training in subacute
in stroke patients and found that there were significant dif- stroke, participants reported adverse events such as numb-
ferences between the two groups at follow-up for AROM ness, dizziness, pins and needles, headaches, or nausea
(flexion, extension, abduction of the shoulder and flexion of (Saposnik et al. 2016). However, no adverse event was
the elbow but not for AROM of the wrist), FMA score, and reported in our study. This, also, is very likely because our
BBT score. They finally concluded that VR training using Xbox Kinect-based VR exercises were specifically designed for
Kinect can improve the functioning of the UE. Park et al. stroke rehabilitation purposes by clinicians, and were vali-
(2017) investigated the effects of VR training, using the Xbox dated with stroke patients before (Tong 2014).
Kinect-based game system, on the motor recovery of patients There are several limitations of our study. First is the rela-
with chronic hemiplegic stroke and found significant tively small sample size and the absence of a sham group to
improvements for the Berg Balance Scale, the Timed Up and estimate the natural recovery of stroke. Second, the improve-
Go test, and the 10-meter Walk Test (but not for FMA scores) ments observed in the Kinect-based VR group may have
for the intervention group. It was reported that the use of been affected by the total intervention time (1 hour more
additional VR training with the Xbox Kinect gaming system is than the PT group). Third, due to our strict inclusion criteria
an effective therapeutic approach for improving motor func- and diversity of cerebrovascular diseases, the results of this
tion during stroke rehabilitation. Similarly, Fan et al. (2014) study may not be generalized to patients with stroke. Fourth,
revealed that VR gaming has immediate effects on motor due to the short duration of the study, we are unable to con-
recovery and provides motivation for treatment compliance clude on the long-term effect of VR-based training therapy
in stroke patients. Choi et al. (2014) suggested that the com- on UE motor function. One of the strengths of our study is
mercial gaming-based VR therapy was as effective as conven- that our VR-based gaming programs were designed for
tional occupational therapy on the recovery of UE motor and stroke patients. Therefore, we did not need intense supervi-
daily living function in subacute stroke patients. Manlapaz sion of medical staff to optimize VR training. Further studies
et al. (2010) determined that the use of Nintendo Wii pro- including a larger number of subjects with long-term follow-
vided marked improvement in the UE function of chronic up assessments are needed to interpret the results
stroke patients demonstrated within 6 weeks. Both Saposnik more accurately.
et al. (2010) and Lee et al. (2016) reported that VR technol- To conclude, our results suggest that the adjunct use of
ogy represents a safe, feasible, and potentially effective alter- Kinect-based VR training may contribute to the improvement
native to facilitate rehabilitation therapy and promote motor of UE motor function and AROM in chronic stroke patients.
recovery after stroke. Consistent with these findings, our Further studies with a larger number of subjects with longer

follow-up periods are needed to establish its effectiveness in
neurorehabilitation.
Ethics committee approval
The authors declared that the research was conducted according to the
principles of the World Medical Association Declaration of Helsinki
“Ethical Principles for Medical Research Involving Human Subjects”
(amended in October 2013).
University local Research Ethics Committee Approval number:
03.06.2015/103.
Informed consent
Written informed consent was obtained from the patients who partici-
pated in this study.
Acknowledgements
We acknowledge the research team of Hong Kong Polytechnic University
for their permission to use KineLabs’ software platform in our study.
Disclosure statement
The authors declared no conflicts of interest with respect to the author-
ship and/or publication of this article.
Funding
The authors received no financial support for the research and/or author-
ship of this article.
ORCID
Ayhan Aşkın http://orcid.org/0000-0001-9445-4430
Emel Atar http://orcid.org/0000-0002-8373-1196
Aliye Tosun http://orcid.org/0000-0003-2827-6255
References
Adie K, Schofield C, Berrow M, Wingham J, Humfryes J, Pritchard C,
James M, Allison R. 2017. Does the use of Nintendo Wii
Sports(TM) improve arm function? Trial of Wii(TM) in Stroke: a
randomized controlled trial and economics analysis. Clin Rehabil
31:173–185.
Bao X, Mao Y, Lin Q, Qiu Y, Chen S, Li L, Cates RS, Zhou S, Huang D.
2013. Mechanism of Kinect-based virtual reality training for motor
functional recovery of upper limbs after subacute stroke. Neural
Regen Res 8:2904–2913.
Bohannon RW, Smith MB. 1987. Interrater reliability of a modified
Ashworth scale of muscle spasticity. Phys Ther 67:206–207.
Chang YJ, Chen SF, Huang JD. 2011. A Kinect-based system for physical
rehabilitation: a pilot study for young adults with motor disabilities.
Res Dev Disabil 32:2566–2570.
Choi JH, Han EY, Kim BR, Kim SM, Im SH, Lee SY, Hyun CW. 2014.
Effectiveness of commercial gaming-based virtual reality movement
therapy on functional recovery of upper extremity in subacute stroke
patients. Ann Rehabil Med 38:485–493.
Collin C, Wade D. 1990. Assessing motor impairment after stroke: a pilot
reliability study. J Neurol Neurosurg Psychiatry 53:576–579.
Desrosiers J, Bravo G, Hebert R, Dutil E, Mercier L. 1994. Validation of the
box and block test as a measure of dexterity of elderly people:
SOMATOSENSORY & MOTOR RESEARCH 7
reliability, validity, and norms studies. Arch Phys Med Rehabil
75:751–755.
Fan SC, Su FC, Chen SS, Hou WH, Sun JS, Chen KH, Lin WH, Hsu
SH. 2014. Improved intrinsic motivation and muscle activation
patterns in reaching task using virtual reality training for stroke
rehabilitation: a pilot randomized control trial. J Med Biol Eng
34:399–407.
Feigin VL, Norrving B, Mensah GA. 2017. Global burden of stroke. Circ
Res 120:439–448.
Fugl-Meyer AR, J€a€asko € L, Leyman I, Olsson S, Steglind S. 1975. The post-
stroke hemiplegic patient. 1. A method for evaluation of physical per-
formance. Scand J Rehabil Med 7:13–31.
Hebert D, Lindsay MP, McIntyre A, Kirton A, Rumney PG, Bagg S, Bayley
M, Dowlatshahi D, Dukelow S, Garnhum M, et al. 2016. Canadian
stroke best practice recommendations: stroke rehabilitation practice
guidelines, update 2015. Int J Stroke 11:459–484.
Johnson L, Bird ML, Muthalib M, Teo WP. 2018. Innovative STRoke
Interactive Virtual thErapy (STRIVE) online platform for community-
dwelling stroke survivors: a randomised controlled trial protocol. BMJ
Open 8:e018388.
Kong KH, Loh YJ, Thia E, Chai A, Ng CY, Soh YM, Toh S, Tjan SY. 2016.
Efficacy of a virtual reality commercial gaming device in upper limb
recovery after stroke: a randomized, controlled study. Top Stroke
Rehabil 23:333–340.
Lang CE, Bland MD, Bailey RR, Schaefer SY, Birkenmeier RL. 2013.
Assessment of upper extremity impairment, function, and activity after
stroke: foundations for clinical decision making. J Hand Ther
26:104–114.
Lange B, Chang CY, Suma E, Newman B, Rizzo AS, Bolas M. 2011.
Development and evaluation of low cost game-based balance
rehabilitation tool using the Microsoft Kinect sensor. Conf Proc IEEE
Eng Med Biol Soc 2011:1831–1834.
Langhorne P, Coupar F, Pollock A. 2009. Motor recovery after stroke: a
systematic review. Lancet Neurol 8:741–754.
Laver KE, Lange B, George S, Deutsch JE, Saposnik G, Crotty M. 2017.
Virtual reality for stroke rehabilitation. Cochrane Database Syst Rev
11:CD008349.
Lee S, Kim Y, Lee BH. 2016. Effect of virtual reality-based bilateral upper
extremity training on upper extremity function after stroke: a random-
ized controlled clinical trial. Occup Ther Int 23:357–368.
Manlapaz DG, Silverio LA, Navarro JA, Ang MF, Regacho M, Canaberal
KA, Dela Cruz RB. 2010. Effectiveness of using Nintendo Wii in
rehabilitation of chronic stroke patients with upper limb hemiparesis.
Hong Kong Physiother J 28:25.
Meyer S, Verheyden G, Brinkmann N, Dejaeger E, De Weerdt W, Feys H,
Gantenbein AR, Jenni W, Laenen A, Lincoln N, et al. 2015. Functional
and motor outcome 5 years after stroke is equivalent to outcome at 2
months: follow-up of the collaborative evaluation of rehabilitation in
stroke across Europe. Stroke 46:1613–1619.
Park DS, Lee DG, Lee K, Lee G. 2017. Effects of virtual reality training
using Xbox Kinect on motor function in stroke survivors: a preliminary
study. J Stroke Cerebrovasc Dis 26:2313–2319.
Perez-Marcos D, Chevalley O, Schmidlin T, Garipelli G, Serino A, Vuadens
P, Tadi T, Blanke O, Millan JDR. 2017. Increasing upper limb training
intensity in chronic stroke using embodied virtual reality: a pilot
study. J Neuroeng Rehabil 14:119.
Pollock A, Farmer SE, Brady MC, Langhorne P, Mead GE, Mehrholz J, van
Wijck F. 2014. Interventions for improving upper limb function after
stroke. Cochrane Database Syst Rev 11:CD010820.
Raffin E, Hummel FC. 2017. Restoring motor functions after stroke: mul-
tiple approaches and opportunities. Neuroscientist [Nov 1]; [1–17]. doi:
10.1177/1073858417737486.
Saposnik G, Cohen LG, Mamdani M, Pooyania S, Ploughman M, Cheung
D, Shaw J, Hall J, Nord P, Dukelow S, et al. 2016. Efficacy and safety of
nonimmersive virtual reality exercising in stroke rehabilitation
(EVREST): a randomised, multicentre, single-blind, controlled trial.
Lancet Neurol 15:1019–1027.
Saposnik G, Levin M, Outcome Research Canada (SORCan) Working
Group. 2011. Virtual reality in stroke rehabilitation: a meta-analysis
and implications for clinicians. Stroke 42:1380–1386.
8 A. AŞKIN ET AL.
Saposnik G, Teasell R, Mamdani M, Hall J, McIlroy W, Cheung D, Thorpe
KE, Cohen LG, Bayley M, Stroke Outcome Research Canada (SORCan)
Working Group. 2010. Effectiveness of virtual reality using Wii gaming
technology in stroke rehabilitation: a pilot randomized clinical trial
and proof of principle. Stroke 41:1477–1484.
Silva SM, Corre^a JCF, Pereira GS, Corr^ea FI. 2017. Social participation fol-
lowing a stroke: an assessment in accordance with the international
classification of functioning, disability and health. Disabil Rehabil [Dec
13]; [1–8]. doi: 10.1080/09638288.2017.1413428.
Sin H, Lee G. 2013. Additional virtual reality training using Xbox Kinect in
stroke survivors with hemiplegia. Am J Phys Med Rehabil 92:871–880.
Smith RH, Sharpe M. 1994. Brunnstrom therapy: is it still relevant to
stroke rehabilitation? Physiother Theory Pract 10:87–94.
Thomson K, Pollock A, Bugge C, Brady M. 2014. Commercial gaming
devices for stroke upper limb rehabilitation: a systematic review. Int J
Stroke 9:479–488.
Tong KY. 2014. Assisted technology for daily living. Springerplus
3:K2.
You SH, Jang SH, Kim YH, Hallett M, Ahn SH, Kwon YH, Kim JH, Lee MY.
2005. Virtual reality-induced cortical reorganization and associated
locomotor recovery in chronic stroke: an experimenter-blind random-
ized study. Stroke 36:1166–1171.