Leap motion controlled videogame-based

therapy for rehabilitation of elderly patients

with subacute stroke: a feasibility pilot study
Marco Iosa1, Giovanni Morone1, Augusto Fusco1, Marcello Castagnoli1,2,
Francesca Romana Fusco3, Luca Pratesi1, Stefano Paolucci1
1
Clinical Laboratory of Experimental Neurorehabilitation, Fondazione Santa Lucia, I.R.C.C.S., Rome, Italy,
2
Scuola di Fisioterapia, Università di Roma, Tor Vergata, sede Fondazione Santa Lucia, Rome, Italy, 3Laboratory
of Neuroanatomy, Fondazione Santa Lucia, I.R.C.C.S., Rome, Italy

Background: The leap motion controller (LMC) is a new optoelectronic system for capturing motion of both
hands and controlling a virtual environment. Differently from previous devices, it optoelectronically tracks
the fine movements of fingers neither using glows nor markers.
Objective: This pilot study explored the feasibility of adapting the LMC, developed for videogames, to
neurorehabilitation of elderly with subacute stroke.
Methods: Four elderly patients (71.50+ 4.51 years old) affected by stroke in subacute phase were enrolled
and tested in a cross-over pilot trial in which six sessions of 30 minutes of LMC videogame-based therapy
were added on conventional therapy. Measurements involved participation to the sessions, evaluated by
means of the Pittsburgh Rehabilitation Participation Scale, hand ability and grasp force evaluated respect-
ively by means of the Abilhand Scale and by means of the dynamometer.
Results: Neither adverse effects nor spasticity increments were observed during LMC training.
Participation to the sessions was excellent in three patients and very good in one patient during the
LMC trial. In this period, patients showed a significantly higher improvement in hand abilities (P50.028)
and grasp force (P50.006).
Conclusions: This feasibility pilot study was the first one using leap motion controller for conducting a
videogame-based therapy. This study provided a proof of concept that LMC can be a suitable tool
even for elderly patients with subacute stroke. LMC training was in fact performed with a high level of
active participation, without adverse effects, and contributed to increase the recovery of hand abilities.
Keywords: Stroke rehabilitation, Virtual reality, Hand dexterity, Videogame-based therapy

Introduction and computer interface-based systems in rehabilita-

In a famous scene of science fiction movie ‘‘Minority
Report,’’ the main character moves virtual objects
just by moving his hands in midair, without touching
a keyboard or a controller. This technology recently
became available in reality. In fact, some devices for
markerless movement capture have recently been
developed, mainly for game playing, with the aim of
facilitating immersion, imagination, and interaction
of the player with the virtual environment.1
The last few years have witnessed a significant
increase in the applications of virtual reality
Correspondence to: G. Morone, Clinical Laboratory of Experimental
Neurorehabilitation, Santa Lucia Foundation, I.R.C.C.S., Via Ardeatina 306,
00179 Rome, Italy. Email: g.morone@hsantalucia.it
tion, for a variety of impairments resulting from
lesions of the nervous system.2,3 One of the possible
targets is the population of patients with stroke, in
particular with respect to the rehabilitation of func-
tions of the upper extremities.4
Stroke is the leading cause of major adult disability
in Western Society.5 After stroke, a proper walking
recovery typically occurs only in about half of reha-
bilitated patients,6 and the recovery of upper limb
functions is even poorer. It has been suggested that
this is due to inappropriate or inadequate therapeutic
interventions.7 In light of this, many new therapeutic
approaches in addition to conventional therapy, such
as robotic interventions,8–11 peripheral or brain
electrical stimulations,12 virtual reality,13 and other

ß W. S. Maney & Son Ltd 2015

306 DOI 10.1179/1074935714Z.0000000036 Topics in Stroke Rehabilitation 2015 VOL . 22 NO . 4

emerging rehabilitative technologies have been
suggested.4–14 All these devices aim to enhance the
effectiveness of rehabilitation with restorative or
adaptive purposes, both as add-on or stand-alone
therapeutic interventions.15
There are many potential advantages in using
videogame-based therapy:1–4 the artificial environ-
ment can easily be modified, allowing for designing
an optimal individualized therapy, it can provide
functional, rich stimuli and a motivating context,
increasing the active participation of the subject in
his/her rehabilitation program. This, in particular,
is a fundamental aspect for the outcome of rehabili-
tation of patients with stroke in subacute phase.16
Virtual reality-based therapy could benefit from
further development, wider diffusion and the pro-
gressive cost reduction of 3D monitors and of optoe-
lectronic markerless motion capture systems
occurring in these last years.14–17
However, limited positive results were found with
virtual reality and videogame-based therapy for
patients compared to conventional rehabilitation tech-
niques. Previous reviews18,19 and also a recent
Cochrane13 described beneficial effects of virtual reality
and interactive video-gaming in terms of an improve-
ment in upper limb functions and a slightly better ability
to manage everyday activities, but not in terms of global
functions, grip strength or gait speed. Further, these
limited positive effects were found soon after the end
of the treatment and it is not clear whether the effects
are long lasting.13 Moreover, cognitive requirements
needed for interaction with virtual environment greatly
diminished the possibility of using these devices for
patients with severe central nervous system damage.
To date, many ongoing studies are being conducted at
the level of feasibility and proof-of-concept for using in
rehabilitation commercial devices not certified for
medical deployment, such as, by way of example,
Nintendo WiiH and Sony X-BoxH. Furthermore, most
of these studies were performed on young patients and
in chronic phase of stroke for the narrow requirements
related to the use of these interfaces,13 The few studies
enrolling patients with subacute stroke and with a wide
age range (between 18 and 85 years) reported a low rate
of subjects matching inclusion/exclusion criteria (such
as the ability of following the given instructions or the
ability of standing alone) and completing the sessions
of videogame-based therapy: 15% in a study on the use
of Nintendo Wii with hand controller20 and 29% in a
study on the use of Nintendo Wii-Fit with force plat-
form.21 Analogously to Nintendo Wii-Fit, also for
using Kinect of Sony XboxH, there is the need of enrolling
only patients able to stand alone.
Iosa et al. Leap motion controller for stroke rehabilitation
It has also been reported that psychological features,
such as anxiety and an external perceived recovery locus
of control, negatively influence the rehabilitation
outcome of robotic therapy, but not that of convention-
al physical therapy.22 These aspects should be con-
sidered in proposing the new emerging technologies as
possible tools for neurorehabilitation.
The Leap Motion Controller (LMCH; Leap Motion,
Inc., San Francisco, CA, USA) is a new device suitable
for hand gesture controlled user interfaces not
specifically developed for rehabilitation, but for
human–computer interaction especially related to
game playing. It can track both hands and all ten fin-
gers, modeling all physiological hand joints, and it
has to be positioned between the user and the computer
monitor as shown in Fig. 1. Because of its low cost, its
reduced dimensions compared to other devices, and
especially its easiness to use, absence of markers and
the captivating aspects of its technology, the number
of its potential applications is wide and includes video-
game-based application also in neurorehabilitation.
Similarly to other devices for videogame-based
therapy, the aim of its use could be increasing patients’
participation, both in terms of time spent in training
and in terms of an active cognitively involvement in
that training. But differently from the other devices,
LMC could allow for including in videogame-based
therapies also elderly patients in subacute phase of
stroke when most of them have cognitive impairments
(an exclusion criterion of studies using complex inter-
faces), are confined on wheelchair (an exclusion
criterion of studies using for example, Sony Kinect or
Nintendo Wii-Fit) and have difficulties in handling a
controller (an exclusion criterion of studies using for
example Nintendo Wii hand controller). Furthermore,
it could enhance the time in which patients are involved
in rehabilitation program. A previous study showed in
fact that inpatients spent only 13% of their daily time
engaged in activities with the potential to improve
recovery of mobility and prevent complications,
whereas they are inactive and alone for more than
60% of time.23 The simplicity of LMC could facilitate
patients’ approach to technology, increasing their
feeling of immersion into the virtual environment,
their imagination and the interaction with the virtual
environment (being immersion, imagination and inter-
action known as the three I’s of virtual reality1).
The aim of this pilot study was to test the feasi-
bility, the compliance and the potential efficacy of
using a LMC-based system for enhancing the
recovery of paretic hand functions in a small
group of elderly patients with ischemic stroke in
subacute phase.
Topics in Stroke Rehabilitation 2015 VOL . 22 NO . 4 307
Iosa et al. Leap motion controller for stroke rehabilitation

Figure 1 The use of Leap Motion Controller. Above, on the left the patients is experiencing an enhanced feedback looking
the virtual hands together with real hands (under the left hand is visible the Leap Motion Controller), under that photo the
details of the LMC (the three lights are related to the three infrared light emitters, visible in the photo but not during the
exercise); on the right a schematic superimposition of virtual and real hand. Below, two examples of the performed game
during the training (in the supplementary online material a video shows the us of LMC).

Material and methods subacute phase (less than 3 months from the acute
Participants and study design event) with upper limb motor impairment due to
According to the aim of this study, inclusion criteria hemiparesis with ischemic lesions confirmed by com-
were: patients with first-ever ischemic stroke in puted tomography or magnetic resonance imaging,

308 Topics in Stroke Rehabilitation 2015 VOL . 22 NO . 4


and age w65 years. Exclusion criteria included the
presence of subarachnoid hemorrhages, sequelae of
prior stroke or other chronic disabling pathologies,
Mini-Mental State Examination score v24, and
severe hemispatial neglect, as evaluated by a neurop-
sychologist. The main demographical and clinical
characteristics of the patients enrolled in this pilot
study are reported in Table 1 (the time from acute
event was lower than 2 months for all patients at
the beginning of this study).
A cross-over pilot trial was designed as shown in the
flowchart reported in Fig. 2. The four enrolled inpatients
were monitored for four weeks during their rehabilita-
tion program, and six sessions of LMC training were
added to conventional therapy three times at week for
two weeks, according to the doses reported in previous
studies.34,35 According to the cross-over design, two
patients performed LMC training in the first 2 weeks
and the other two in the second 2 weeks The study was
performed in accordance to the Declaration of Helsinki
about ethical principles for medical research involving
human subjects, and patients signed informed consent.
Conventional therapy
As usually done in our neurorehabilitation hospital,
patients were involved in a conventional therapy pro-
gram, formed by two daily sessions of physiotherapy,
each one lasting 40 minutes, 5 days per week.
One daily session was dedicated to arm and hand
training, focused on the facilitation of movements on
the paretic side, upper-limb exercises for reaching and
grasping and for improving proprioception. The
second daily physiotherapy aimed to improve
balance, trunk stabilization, standing, weight transfer,
sitting, transferring, and when possible walking.
Training protocol
A LMC videogame-based training, specifically
designed by a trained therapist, was added to six
sessions of conventional rehabilitation therapy. The
LMC-based training lasted 2 weeks, three times for
week. Each session lasted 30 minutes, during which
the exercises were performed by paretic limb, and,
in some exercises, also by non-paretic limb.
Table 1 Demographical and clinical characteristics of the
patients
Side Barthel
Age Typeof of Index at
Patients (years) Sex Stroke hemiparesis admission
P1 78 Female Ischemic Right 66
P2 69 Male Ischemic Right 47
P3 68 Female Ischemic Right 80
P4 71 Male Ischemic Left 17
Iosa et al. Leap motion controller for stroke rehabilitation
Because the aim of this study was to test the feasi-
bility of LMC training in elderly patients also for
enhancing their time spent in rehabilitation, LMC-
based therapy was performed in addition and not
in substitution of conventional therapy.
During each session, patient sat in a front of a
table on which there was a 27 inch monitor at a dis-
tance of about 0.80 m. The Leap Motion Controller,
was located between patient and monitor, as shown
in Fig. 1. The Leap Motion Controller consists of
three infrared light emitters and two infrared cam-
eras. Hence, the LMC can be categorized into optical
tracking systems based on Stereo Vision. The LMC,
in conjunction with its interface, delivers positions in
Cartesian space of predefined objects, such as hands
and fingers. The delivered positions are relative to the
Leap Motion Controller’s center point, which is
located at the position of the second infrared
emitter.24 The accuracy of LMC was 0.2 mm for
static objects and 1.2 mm for dynamic objects.24
While other devices, such as Microsoft KinectTM,
can track large movements of the entire body, the
LMC captures fine hand movements, hence, it is
potentially more suitable for training upper limb
fine movements and hand functions.
The initial standard calibration required by LMC
was performed before the beginning of videogame-
based therapy and it took about 2 minutes. This cali-
bration phase has been also used for allowing subject
to become familiar with the LMC technology.
As shown in a video added to this paper as
supplementary material, this technology has an
intrinsic easiness that facilitates the understanding
of its use by the patient. Different types of
videogames were used, progressively increasing in
complexity. Therapist explained the game to the
patients and provided them all the needed
instructions, also showing what to do if necessary.
For example, in the game ‘‘caterpillar count’’
(right picture below in Fig. 1) therapist explained to
the subject that he/she should move a caterpillar,
moving his/her second finger, towards number fol-
lowing numerical sequence and collecting them to
make the caterpillar bigger. In the game ‘‘dots’’
(left picture below in Fig. 1), therapist explained to
the subject that he/she should form a geometric
shape connecting different dots moving his hand
without passing more than one time for each dot.
The time to complete each session of the game was
recorded for both the hands.
A therapist supervised the patient, explained the
objective of the game, and helped him/her focusing
on the rehabilitative goals, correcting movements and
Topics in Stroke Rehabilitation 2015 VOL . 22 NO . 4 309
Iosa et al. Leap motion controller for stroke rehabilitation
Figure 2 Flowchart diagram.
trunk posture, helping him/her when needed. The
target of this protocol was mainly related to improve
upper limb and hand functional performances, to aug-
ment the interaction with the external environment
(formed by virtual static and dynamic objects), to
facilitate the movements related to the activities of
daily living. Hence, the training involved motor
training, as well as cognitive aspects related to
problem solving process involved in upper limb
motor control.
Assessment protocol and data analysis
The primary outcome measure of this study was the
compliance of patients assessed in terms of active
310 Topics in Stroke Rehabilitation 2015 VOL .
participation measured by means of Pittsburgh
Rehabilitation Participation Scale (PRPS).25 PRPS
is a six-point scale completed by the therapist, who
evaluates patient’s participation in the therapy
session scoring it as follows: 1, refusal or no partici-
pation in any session; 2, no participation in at least
half of the sessions; 3, participation in most or all
exercises, but without maximal effort or not finishing
most exercises; 4, good participation in all exercises
with good effort and finishing most but not all exer-
cises; 5, very good participation in all exercises with
maximal effort and finishing all exercises; 6, excellent
participation in all exercises with maximal effort,
finishing all exercises, and taking an active interest
22 NO . 4

in exercises and/or future therapy sessions (when in
doubt, the lower rating was assigned).
Then, similarly to a previous pilot clinical trial and
proof of principle study about the use of Nintendo
WiiTM gaming system in rehabilitation,20 we also
measured the total time spent by patients in the inter-
vention compared to the planned one, the occurrence
of any training-related adverse events during the
training sessions, any eventual increments of spasti-
city assessed by Modified Ashworth Scale at
shoulder, elbow and wrist level.26
Secondary outcome measures were related to the
efficacy of intervention. Motor ability was assessed
at T0 (before the beginning of the trial), T1 (between
3rd and 4th sessions) and T2 (after the end of the
trial). Hand dexterity was assessed using the nine-
hole peg test.27,28 Dynamometers were used for
measuring grasp and pinch forces.29 Abilhand Scale
was used for assessing manual ability. Abilhand
Scale consists of a structured interview in which
patients are asked to estimate the ease or difficulty
in performing daily activities that require the use of
the upper limbs, whatever the strategies involved.
Its scores were normalized converting raw ordinal
scores into linear measures of manual ability [this
normalization was performed by online rasch anal-
ysis on the official site of this scale, obtaining a
patient’s continuous measure in logits (PM)].30
Stroke Upper Limb Capacity Scale (SULCS) was
used for assessing upper limb capacity. SULCS con-
sists of ten items related to arm capacity without,
with basic and with complex hand capacities, each
item having a possible score of 0 or 1.31
Data analysis
Parametric statistics was used for describing (mean
and standard deviation) and analyzing continuous
measures, whereas non-parametric statistics was
used for describing (median and quartiles) and ana-
lyzing ordinal measures (such as the scores of clinical
scales). Hence, two-tail Student’s t-test was used to
compare the changes recorded in continuous
measures during the period in which LMC was
added to conventional therapy with those of the
period in which it was not added. Wilcoxon signed
ranks test was used to compare changes in ordinal
measures. Finally, data were used to assess the mini-
mum number of subjects to involve in a further study
having as primary outcome measure the efficacy of
LCM-training using a sample size calculator.
Because our study had a cross-over design and
hence, we compared with-in subjects the differences
between the period with and that without
Iosa et al. Leap motion controller for stroke rehabilitation
LMC training, we used a one-sample calculation
comparing the mean and standard deviation of the
with-in subject differences with the zero-value setting
alpha level at 0.05 and beta level at 0.20, for having a
power of 80%. For variables found already statisti-
cally significant, the power analysis was performed
to calculate the probability to correctly reject the
null hypothesis when it is false.
Results
None of the participants experienced any serious
adverse event during the experimental trial, neither
when LMC was added to conventional therapy nor
when it was not added.
The feasibility of using LMC in neurorehabilita-
tion was estimated measuring the level of subject’s
participation to the training, using the Pittsburgh
Scale. The median scores recorded among the six
sessions of LMC training was 6 (meaning excellent
participation) for three patients (P1, P3, P5) and 5
(‘‘very good’’) for one patient (P2). None of the
patients asked to withdraw the trial sessions.
On the contrary, three out of four patients (P1, P3,
P4) asked to prolong the LMC training also after
the planned 30 minutes in about half of the sessions.
During LMC training, spasticity was reduced of
one point in the score of Modified Ashworth Scale
at wrist and elbow level in one patient, and it was
unchanged in the other patients. A reduction of one
point in Modified Ashworth Scale score was
observed at shoulder level also in two patients
during the period in which they performed only
conventional therapy without the adjunction of
LMC training.
With respect to the proof of concept of LMC training
efficacy, the improvements in terms of hand (pinch and
grasp) strength, hand ability and hand dexterity are
shown in Fig. 3. In terms of hand strength, the main
changes occurred in grasp force: three patients
improved between T0 and T2, with bigger changes
when LMC training was added to conventional therapy,
while one patient did not show any changes during the
period with LMC training and worsened in the period
with conventional therapy alone. The changes were sig-
nificantly different between the period with and that
without LMC training (P50.006). Fewer, and not sig-
nificant, changes were observed in terms of pinch
strength (P50.540). Hand ability improved in all
patients, with more evident changes occurring during
trial period with LMC training than during that without
it. The variations observed in period with LMC training
compared to those observed in the period without LMC
training were statistically significant (P50.042).
Topics in Stroke Rehabilitation 2015 VOL . 22 NO . 4 311
Iosa et al. Leap motion controller for stroke rehabilitation

Figure 3 Results for each single patient about hand force (above, grip, and pinch strength assessed using dynamometers),
hand ability (in the middle, assessed by normalized Abilhand score), and hand dexterity (below, assessed by velocity in
completing the nine-hole peg test). Solid lines represent the periods in which LMC training has been added to conventional
therapy (between T0 and T1 for P1 and P2, and between T1 and T2 for P3 and P4). Dot-lines represent the periods without the
adjunction of LMC training.

312 Topics in Stroke Rehabilitation 2015 VOL . 22 NO . 4


Hand dexterity, assessed by velocity in completing
the 9HPT, improved in all patients from T0 to T2, and
for all of them, improvements were more evident
during LMC training (solid lines in Fig. 3) than during
conventional therapy alone (dotted lines); however,
such changes were not statistically different
(P50.181). The score of Stroke Upper Limb Capacity
Scale was 10 for all the patients at all the three assess-
ments and was hence not reported in Figure. Interest-
ingly, patient P2, who reported a median score at
Pittsburgh Scale lower than the other ones, was the
one showing less improvement in hand dexterity and
even a reduction in hand force, both with and without
LMC training.
Sample size calculation was performed for the data of
each one of the four variables reported showed in Fig. 3.
Obviously, for the grasp strength and hand ability
(assessed by Abilhand) the sample of four subjects
resulted already sufficient, because of P values already
under the threshold of 0.05. For these two parameters,
the power of the analyses resulted of 99% and 93%,
respectively. For hand dexterity, assessed in terms of
velocity in completing the nine hole peg test, the
needed sample size resulted of eight subjects. Finally,
for having a statistically significant difference in terms
of pinch force the sample size should be of 52 subjects.
Figure 4 shows the time spent to complete one of the
proposed game (caterpillar count, chosen as example
because it was the simplest game and it was used from
the first LMC training session) as recorded in the first
trial of each LMC session for the paretic and non-pare-
tic hand. Despite the difficulty of the game progressively
increased, the difference between the performances of
the two hands was progressively reduced from the
third to the last session.
Figure 4 The performance of patients with paretic hand (in blue), non-paretic hand (in grey), and their differences (in red) in
terms of mean and standard deviation of time spent to complete the trial.
Iosa et al. Leap motion controller for stroke rehabilitation
Discussion
Our results showed that patients completed all the ses-
sions of LMC training with a level of participation
scored between very good and excellent. The increment
of time spent by patients in daily rehabilitation probably
contributed for providing a significant increase in their
grasping force and hand ability. Despite the fact that
other kinds of intervention could be used for increasing
this time, the LMC resulted a suitable technological tool
for being integrated into the therapy program.
The motor re-learning trend emerged quite evidently
in Fig. 4, with the differences between paretic and non-
paretic hand progressively reduced until to reach the
same time to complete the game at the end of LMC
sessions. Even though further studies are needed to
demonstrate that effect, data sounds promising to sus-
tain the need for specific therapies to treat hand
disabilities.
We took into account also safety of using LMC
videogame-based therapy and monitored the possible
adverse effects during training, including mild effects
such as a possible increment of spasticity. However,
no patient experienced intervention-related adverse
events, neither during LMC videogame-based therapy
nor during the physical conventional therapy. Safety is
an important issue for using in rehabilitation commer-
cial devices not specifically developed for medical use;
however, it should be noted that, being LMC an optoe-
lectronic system, patients just moved their hands in
midair, without being in contact with the device. The
absence of adverse effects was in accordance with the
Cochrane revision on the use of virtual reality for reha-
bilitation of patients with stroke, which highlighted that
few adverse events were reported across studies and
those reported were relatively mild.13
Topics in Stroke Rehabilitation 2015 VOL . 22 NO . 4 313
Iosa et al. Leap motion controller for stroke rehabilitation
In that Cochrane review,13 which reported about
19 studies involving 565 patients, significant improve-
ments in the outcomes related to activity of daily
living, but not in grip strength, in respect of conven-
tional therapy, were observed. In our pilot study, we
also observed an increment in grasping force, prob-
ably due to the fact that the LMC training is a
motion tracking system that is specific for the hand
and it can capture and reproduce all the physiologi-
cal finger degrees of freedom. Laver and colleagues
highlighted in that Cochrane as the sample sizes of
the included studies were generally small and the
majority of participants were relatively young and
in chronic phase of stroke.13 The sample size of our
study was small as well, being a feasibility study,
but in contrast with other studies, we enrolled a
sample of elderly subjects in subacute phase of
stroke. This sample is more representative of the
population of patients with stroke that are hospital-
ized in a neurorehabilitation unit. Thus, it is concei-
vable that if LMC training was feasible for being
used with elderly patients and in subacute phase of
stroke, it could very well be feasible in younger
patients and in chronic phase.
Despite the limited evidence about the benefits of
using virtual reality and interactive videogaming for
improving upper limb recovery compared to the
same dose of conventional therapy alone,13 video-
game-based therapy could be feasibly used in combi-
nation or as a suitable add-on to conventional
therapy for increasing the time spent by patients in
activities potentially beneficial, especially if the
alternative is to remain inactive.23 Our results
showed that the LMC can be feasibly used in add-
on to conventional therapy, even in elderly patients
in subacute phase of stroke. In our pilot study, we
selected a low dose of videogame-based therapy
(six sessions of 30 minutes, three per weeks), but
according to our results it is conceivable that this
dose could be augmented without adverse effects.
The above Cochrane suggested the need of adminis-
tering at least 15 hours of videogame-based
treatments for achieving a good efficacy of the inter-
vention.13 However, further studies should investi-
gate the best dose of LMC training (in terms of
frequency, duration, and timing) when it has been
added to conventional therapy in elderly patients in
subacute phase, as in our case.
Differently from previous studies involving mild to
moderately affected patients (as young patients with
chronic stroke),13 or from those reporting a low com-
pliance to videogame-based therapy among elderly
patients in subacute phase,20,21 our results showed a
314 Topics in Stroke Rehabilitation 2015 VOL . 22
participation from very good to excellent,
accompanied by significant improvement in motor
outcomes even in elderly patients within 2 months
from stroke event. It confirms the feasibility of
using LMC for neurorehabilitation also of this kind
of patients for its intrinsic easiness. In fact, differ-
ently from other devices, LMC neither needs any
marker/glow/controller, nor requires a subject able
to stand alone.
It was important to test elderly subjects because
they are usually less accustomed to computer tech-
nology and could have aging-related subclinical defi-
cits in cognitive functions. Conversely, videogame
therapy usually requires cognitive abilities such as
those related to selective attention, task switching,
multi-tasking problem solving, visual short-term
memory, spatial, and temporal attention for alert.31
The progressive reduction of costs of motion capture
devices (such as Sony’s PlayStation EyeTM and PlaySta-
tion MoveTM, Nintendo’s Wii Remote PlusTM, and
Microsoft’s KinectTM) allows to use them in clinical
rehabilitation.4–33 Hence, they have the potentiality to
provide low-cost and motivating training. In fact,
despite these commercial systems were not specifically
developed for rehabilitation, they matched some of
the rehabilitation guidelines related to the use of mean-
ingful functional activities, to retrain skills, for cognitive
functions, and for the development of alternative (com-
pensatory) motor strategies.33 The simplicity of a LMC
system allows its use in addition to conventional therapy
in a population of subjects normally elderly with physi-
cal and cognitive impairments.
The main limit of our study was the relatively small
sample size. In light of this, the confidence in statistical
results, despite some of them significant, deserves
caution. However, our data were helpful for evaluating
the suitable sample size necessary for a study on the effi-
cacy of videogame-based therapy involving the use of
LMC. Our small sample size was sufficient to highlight
differences in terms of grasp force and hand ability,
whereas it should be probably doubled (eight subjects)
for highlighting if also hand dexterity improves.
Conversely, the size of sample found as needed for
pinch force seemed to suggest that LMC training can
not improve this function. Caution in relying on the
LCM training effects is needed also because in subacute
phase of stroke the effects of a treatment can be superim-
posed to those of spontaneous recovery. Another limit of
this report is the short time planned for LMC training: as
stated above, six sessions of 30 minutes was a dose similar
to that of previous studies,34,35 but lower than that
recommended by Cochrane review for using virtual rea-
lity in upper limb functions in stroke.13 However,
NO . 4

it should be kept in mind that this was a feasibility study,
primarily aimed at investigating a proof of concept for
using a LMC system without severe adverse effects. Fur-
thermore, the limitations of using LMC system with
patients deserve attention. Despite a tested precision of
0.2 mm for static objects and 1.2 mm for dynamic
objects,24 it is still not possible to use LMC in kinematic
analysis. Further, tracking of finger movements can be
lost during activities, for example when hands are super-
imposed or curled because of severe spasticity being the
optical sensors just placed under the field of motion of
hands. Finally, it should be noted that all the open
issues related to the use of virtual reality for sensorimotor
rehabilitation post-stroke remain unsolved also for LMC
system: is it necessary to use virtual instead of other real
feedback? Is a 2D stimulus immersive? Does score pro-
gression at a videogame represent patient improvements?
What are the minimal cognitive requirements for video-
game-based rehabilitation? What patients can benefit
more from this virtual approach?19,36
In conclusion, our results showed as the LMC is a
new feasible potential tool in the hands of physiothera-
pists and rehabilitative staff for videogame-based
therapy aiming into training hand functions with a
high level of active participation of patients during ses-
sions. According to the positive results found in our
feasibility pilot study, further randomized controlled
studies should test the efficacy of LMC training on
wider samples of patients with stroke.
Disclaimer Statements
Contributors
MI designed the study, performed the analysis and
drafted the manuscript, GM and AF collected the
data and clinically interpreted the results, MC pro-
posed the study and administered treatment, FRF
and LP revised the manuscript for important intellec-
tual contents and provided English Editing, SP
supervised the study and obtained funding.
Funding
The study has been supported by a grant of Italian Min-
istry of Health titled ‘‘Assessment of altered motor scheme
in patologies of central nervous system’’ RC-G2014.
Conflicts of interest
The authors have declared that no conflict of inter-
ests exist.
Ethics approval
The study was performed in accordance to the
Declaration of Helsinki about ethical principles for
medical research involving human subjects, and
patients signed informed consent.
Iosa et al. Leap motion controller for stroke rehabilitation
References
1 Burdea GC, Coiffet P. Virtual Reality Technology. 2nd ed.
Hoboken, NJ: John Wiley & Sons, Inc. 2003.
2 Holden MK. Virtual environments for motor rehabilitation:
review. Cyberpsychol Behav. 2005;8(3):187–211.
3 Rose FD, Brooks BM, Rizzo AA. Virtual reality in brain
damage rehabilitation: review. Cyberpsychol Behav. 2005;8(3):
241–262.
4 Iosa M, Morone G, Fusco A, et al. Seven capital devices for the
future of stroke rehabilitation. Stroke Res Treat. 2012;2012:
187965.
5 Duncan PW, Goldstein LB, Matchar D, Divine GW, Feussner
J. Measurement of motor recovery after stroke. Outcome
assessment and sample size requirements. Stroke. 1992;23:
1084–1089.
6 Paolucci S, Bragoni M, Coiro P, et al. Quantification of the
probability of reaching mobility independence at discharge
from a rehabilitation hospital in nonwalking early ischemic
stroke patients: a multivariate study. Cerebrovasc Dis. 2008;
26:16–22.
7 Carr J, Shepherd R. Neurological Rehabilitation:
Optimizing Motor Performance. Oxford: Butterworth-Heine-
mann; 1998.
8 Masiero S, Carraro E, Ferraro C, Gallina P, Rossi A, Rosati G.
Upper limb rehabilitation robotics after stroke: a perspective
from the University of Padua, Italy. J Rehabil Med. 2009;
41(12):981–985.
9 Morone G, Bragoni M, Iosa M, et al. Who may benefit from
robotic-assisted gait training? A randomized clinical trial in
patients with subacute stroke. Neurorehabil Neural Repair.
2011;25(7):636–644.
10 Morone G, Iosa M, Bragoni M, et al. Who may have durable
benefit from robotic gait training?: a 2-year follow-up random-
ized controlled trial in patients with subacute stroke. Stroke.
2012;43(4):1140–1142.
11 Poli P, Morone G, Rosati G, Masiero S. Robotic technologies
and rehabilitation: new tools for stroke patients’ therapy.
Biomed Res Int. 2013;2013:153872.
12 Fusco A, Iosa M, Venturiero V, et al. After vs. priming effects
of anodal transcranial direct current stimulation on upper
extremity motor recovery in patients with subacute stroke.
Restor Neurol Neurosci. 2014;32(2):301–312.
13 Laver KE, George S, Thomas S, Deutsch JE, Crotty M. Virtual
reality for stroke rehabilitation. Cochrane Database of Syst Rev.
2011; (2):CD008349.
14 Belda-Lois JM, Mena-del Horno S, Bermejo-Bosch I, et al.
Rehabilitation of gait after stroke: a review towards a top-
down approach. J Neuroeng Rehabil. 2011;8:66.
15 Pomeroy V, Aglioti SM, Mark VW, et al. Neurological
principles and rehabilitation of action disorders: rehabilitation
interventions. Neurorehabil Neural Repair. 2011;25(5 Suppl):
33S–43S.
16 Paolucci S, Di Vita A, Massicci R, et al. Impact of participation
on rehabilitation results: a multivariate study. Eur J Phys
Rehabil Med. 2012;48(3):455–466.
17 Lange B, Chang CY, Suma E, Newman B, Rizzo AS, Bolas M.
Development and evaluation of low cost game based balance
rehabilitation tool using the Microsoft Kinect sensor. Conf
Proc IEEE Eng Med Biol Soc. 2011;2011:1831–1834.
18 Lucca LF. Virtual reality and motor rehabilitation of the upper
limb after stroke: a generation of progress? J Rehabil Med.
2009;41(12):1003–1006.
19 Fluet GG, Deutsch JE. Virtual reality for sensorimotor rehabi-
litation post-stroke: the promise and current state of the field.
Curr Phys Med Rehabil Rep. 2013;1(1):9–20.
20 Saposnik G, Teasell R, Mamdani M, et al. Effectiveness of
virtual reality using Wii gaming technology in stroke rehabilita-
tion: a pilot randomized clinical trial and proof of principle.
Stroke. 2010;41(7):1477–1484.
21 Morone G, Tramontano M, Iosa M, et al. The efficacy of
balance training with video game-based therapy in subacute
stroke patients: a randomized controlled trial. Biomed Res
Int. 2014;2014:580861.
22 Bragoni M, Broccoli M, Iosa M. Influence of psychological
features on rehabilitation outcomes in patients with subacute
Topics in Stroke Rehabilitation 2015 VOL . 22 NO . 4 315
Iosa et al. Leap motion controller for stroke rehabilitation
stroke trained with robotic-aided walking therapy. Am J Phys
Med Rehabil. 2013;92(10):e16–e25.
23 Bernhardt J, Dewey H, Thrift A, Donnan G. Inactive and
alone: physical activity within the first 14 days of acute stroke
unit care. Stroke. 2004;35(4):1005–1009.
24 Weichert F, Bachmann D, Rudak B, Fisseler D. Analysis of the
accuracy and robustness of the leap motion controller. Sensors.
2013;13(5):6380–6393.
25 Lenze EJ, Munin MC, Quear T, et al. The Pittsburgh Rehabi-
litation Participation Scale: reliability and validity of a clini-
cian-rated measure of participation in acute rehabilitation.
Arch Phys Med Rehabil. 2004;85(3):380–384.
26 Bohannon RW, Smith MB. Interrater reliability of a modified
Ashworth scale of muscle spasticity. Phys Ther. 1987;67(2):
206–207.
27 Oxford Grice K, Vogel KA, Le V, Mitchell A, Muniz S,
Vollmer MA. Adult norms for a commercially available Nine
Hole Peg Test for finger dexterity. Am J Occup Ther. 2003;
57(5):570–573.
28 Iosa M, Morone G, Ragaglini MR, Fusco A, Paolucci S.
Motor strategies and bilateral transfer in sensorimotor learning
of patients with subacute stroke and healthy subjects.
A randomized controlled trial. Eur J Phys Rehabil Med. 2013;
49(3):291–299.
29 Bohannon RW, Andrews AW. Interrater reliability of handheld
dynamometry. Phys Ther. 1987;67:931–933.
316 Topics in Stroke Rehabilitation 2015 VOL . 22 NO . 4
30 Penta M, Tesio L, Arnould C, Cancan A, Thonnard JL.
The ABILHAND questionnaire as a measure of manual ability
in chronic stroke patients: Rasch-based validation and relation-
ship to upper limb impairment. Stroke. 2001;32(7):1627–1634.
31 Roorda LD, Houwink A, Smits W, Molenaar IW, Geurts AC.
Measuring upper limb capacity in poststroke patients: develop-
ment, fit of the monotone homogeneity model, unidimensional-
ity, fit of the double monotonicity model, differential item
functioning, internal consistency, and feasibility of the stroke
upper limb capacity scale, SULCS. Arch Phys Med Rehabil.
2011;92(2):214–227.
32 Green CS, Bavelier D. Learning, attentional control, and action
video games. Curr Biol. 2012;22(6):R197–R206.
33 Borghese NA, Bottini G, Sedda A. Videogame based neglect
rehabilitation: a role for spatial remapping and multisensory
integration? Front Hum Neurosci. 2013;7:116.
34 Yang YR, Tsai MP, Chuang TY, Sung WH, Wang RY. Virtual
reality-based training improves community ambulation in
individuals with stroke: a randomized controlled trial. Gait Pos-
ture. 2008;28:201–206.
35 Jaffe DL, Brown DA, Pierson-Carey CD, Buckley EL, Lew HL.
Stepping over obstacles to improve walking in individuals with
poststroke hemiplegia. J Rehabil Res Dev. 2004;41(3A):283–292.
36 Crosbie JH, Lennon S, Basford JR, McDonough SM. Virtual
reality in stroke rehabilitation: still more virtual than real.
Disabil Rehabil. 2007;29(14):1139–1152.