Motor Rehabilitation and Biotelemetry Data Acquisition with Kinect
Francisco Marcelino Almeida de Ara
´
ujo
1 a
, Paulo Roberto Ferreira Viana Filho
1 b
,
Jesus Abra
˜
ao Adad Filho
1 c
, Nuno M. Fonseca Ferreira
2,3 d
, Ant
´
onio Valente
3,4 e
and Salviano F. S. P. Soares
4,5 f
1
Federal Institute of Piaui, Teresina, Brazil
2
ISEC - Institute of Engineering of Coimbra, Portugal
3
INESC-TEC Technology and Science, Portugal
4
University of Tr
´
as-os-Montes and Alto Douro, Portugal
5
IEETA - UA, Portugal
Keywords:
Rehabilitation, Kinect, Disabilities, Biotelemetry, Movement.
Abstract:
Accessibility and inclusiveness of people with disabilities is a recurring theme that is already perceived as an
issue in the field of human rights. Ramps, elevators, among other devices aim at the inclusion of these indi-
viduals with limited mobility. Various types of motor limitations, specially partial limitations, are linked to
corresponding physical-motor rehabilitation process, with the purpose of reducing or eliminating the patient’s
dependence on a caregiver or devices for adaptation. Patients with motor disabilities must practice physio-
therapeutical exercises along a physician in order to perform body and muscle analysis to ensure the patient’s
well-being. To reach a more accurate analysis, physiotherapists use a range of devices to acquire patient data,
such as the spirometer, to acquire the patient’s breath intensity and lung capacity. Similarly, there are other
technologies capable of acquiring motion data and quantifying them. This work aims to develop a system that,
paired together with an exercise game project (exergame), can acquire and transmit the motion data acquired
in-game for an easier and faster analysis of the patient’s growth, relying on graphs, tables, and other visual
indicators to improve the evaluation of physiotherapeutic treatments. The usage together with an exergame
also has benefits such as increased patient compliance with the treatment and improvements in well-being.
1 INTRODUCTION
Present as one of the strongest entertainment indus-
tries(Cummings, 2007), videogames have evolved
over the decades in many forms since its first appear-
ance in 1958 with Tennis for Two. Many of its im-
provements are reflected in the way the player is able
to input commands in order to play the game, ranging
from simple button presses to the usage of complex
technology. In order to innovate and gain their con-
sumer’s attention to buy their game consoles, some
companies invest in the way the player is able to in-
teract with the game and input these commands. One
a
https://orcid.org/0000-0001-8928-0077
b
https://orcid.org/0000-0002-4185-0613
c
https://orcid.org/0000-0001-6372-5541
d
https://orcid.org/0000-0002-2204-6339
e
https://orcid.org/0000-0002-5798-1298
f
https://orcid.org/0000-0001-5862-5706
of the most known and successful examples is Nin-
tendo’s Wii, which used motion capture technology
as a form of controlling the game.
Motion capture technologies are not limited only
for entertainment purposes, however. For example,
the Wii was used for physiotherapeutic rehabilitation
in Parkinson’s disease patients, where it was relevant
to the treatment as a supplementary activity, named
Wii-hab(Herz et al., 2013).
Interfaces that have human-computer interac-
tions that involve human motion, such as Nintendo
Wii
c
controllers, are known as Natural User Inter-
face (NUI). Another strong example of NUI usage
is the Kinect
c
, which uses only sensors to acquire
data and does not require any controller in the player’s
hand, unlike the previous case. That’s why it has be-
come a revolutionary (Martin-SanJose et al., 2017)
device for the video game industry. Like the Nin-
tendo Wii and other NUI devices, Kinect has also
been widely used for therapeutic purposes (Szykman
242
Araújo, F., Viana Filho, P., Adad Filho, J., Ferreira, N., Valente, A. and Soares, S.
Motor Rehabilitation and Biotelemetry Data Acquisition with Kinect.
DOI: 10.5220/0009157202420249
In Proceedings of the 13th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2020) - Volume 1: BIODEVICES, pages 242-249
ISBN: 978-989-758-398-8; ISSN: 2184-4305
Copyright
c
2022 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
et al., 2015)(Da Gama et al., 2015)(Dehkordi et al.,
2018), where it has gained notoriety in projects in-
volving stroke patients (Robertson et al., 2013)(Hsieh
et al., 2014)(Dukes et al., 2013), cerebral palsy,
Parkinson’s disease, people with cognitive disabilities
(Nazirzadeh et al., 2017), among other conditions that
limit physical movement.
2 JUSTIFICATION
There are about 46 million individuals in the Brazil-
ian territory. About 24% of the population claim to
have some form of disability, which may cause dif-
ficulties in performing many essential tasks of daily
life. Approximately 7% of the population declares
to have some motor disability that indeed causes dif-
ficulties in this type of tasks. Keeping that data in
mind and the various diseases and illnesses that lead
to these disabilities, it is possible to observe in many
situations that the systems designed to promote acces-
sibility to persons with disabilities are neglected and
often inoperative when they exist.
Research for motor disability treatment and reha-
bilitation commonly comes across a problem: An ap-
plied solution to one condition does not necessarily
work for another, even if similar. There is even the
possibility that solutions that could be successfully
applied to one patient may not serve another with the
same limitation. This could be due to different factors
such as the disease itself, the severity of each case,
and the area of the body affected. Thus, meeting this
demand for solutions for each case can become an ar-
duous and laborious task.
The benefits of physical therapy treatment for pa-
tients are many: (Costa et al., 2005) mentions, for ex-
ample, that an individual’s ability to stand up from
a wheelchair decreases their dependence on a care-
giver and reduces or eliminates problems deriving
from the condition, such as blood clots, muscle at-
rophy, bladder and urinary tract infection, and osteo-
porosis. More benefits are also seen in the psycholog-
ical part, with an improvement in the patient’s self-
esteem, family and social relationship, directly re-
flects on treatment.
3 TECHNOLOGY AND
VIDEOGAMES IN MOTOR
REHABILITATION
The employability of video games for physiothera-
peutic purposes is commonly researched in the aca-
demic field through the gamification of activities used
in the treatments. According to (Hamari et al., 2014),
gamification can be defined as a process that aims to
improve a service through motivational gains, creat-
ing game-like experiences. For the treatment of a
patient, a variety of technologies can be employed,
aside from commercial video game consoles aimed
at home use and commonly for entertainment pur-
poses. Since rehabilitation activities and exercises
aim to stimulate patient movement, the technologies
employed are usually reactive to the user’s movement
or action, such as the use of pressure sensors and sen-
sors that calculate the individual’s breathing strength
in lung exercises. NUI tools are included in this range
of applied technologies. They ensure user interaction
with the computer, which responds appropriately.
Game-creating software tools known as Game En-
gines provide the ability for people or groups to create
their own custom games. In combination with NUI
tools such as Kinect, this makes it possible to cre-
ate solutions that address both the treatment of peo-
ple with disabilities and the accessibility of players
with motor disabilities when using these devices. A
great example is Kinect Wheels(Gerling et al., 2013),
which used the C# library provided by the manufac-
turer, Microsoft, to create gestures so that wheelchair
users can replace movements they can’t do.
4 BIOTELEMETRY AND DATA
ACQUISITION
One aspect that can be explored for understanding
a patient’s situation regarding their treatment is data
acquisition. The act of acquiring measurements and
data at a distance is known as Telemetry (Cooke
et al., 2004). In turn, Biotelemetry represents the use
of this data acquisition within biology, medicine, and
health to acquire patient data such as vital signs (Kim
and Cho, 2001). An example of using biotelemetry
is to acquire and store a person’s heart rate, possibly
giving an alert when there is an abnormality.
(Cooke et al., 2004), in a paper analyzing
biotelemetry as a research tool for animals and
ecosystems, saw immense potential for researchers.
In humans, biotelemetry can be used as a means of
monitoring rehabilitation patients, like acquiring a pa-
tient’s heart rate while performing a rehabilitation ex-
ercise(Kim and Cho, 2001). It can also be used to
check any abnormalities in an elderly person’s body
in case of emergency at home(Penhaker et al., 2007).
The data acquisition that biotelemetry provides
can be useful as an auxiliary tool in motor rehabilita-
tion exercises. Acquiring numerical data during treat-
Motor Rehabilitation and Biotelemetry Data Acquisition with Kinect
243
ment helps to measure, for example, treatment effec-
tiveness over time by comparing treatment initiation
data and the most recent data.
A project currently under development (Anto-
nio Valente, 2019) uses the Unreal Engine 4 game en-
gine to create a Kinect application that seeks to save
and display the time a patient takes to touch an ob-
ject when prompted. The intention of the project is
to analyze and provide data on the movement of a pa-
tient with motor disabilities, as well as allowing the
customization of the exercises as the physiotherapist
needs.
By analyzing the time it takes for a patient to per-
form a simple task, such as grasping an object in front
of him, it is possible to infer from real data the per-
formance and effectiveness of the treatment. In the
physiotherapeutic environment, the method used for
measurements and evaluations are dynamic, where
professionals make judgments according to observed
data and their experience (Gil, 2015). By receiving
the patient’s data via biotelemetry, it is possible to
obtain a more accurate and less empirical treatment
analysis, avoiding interpretation problems if, for ex-
ample, the physical therapist responsible for a patient
is changed. In the event of a change, the report left by
the previous manager may be interpreted differently
from the intentional one by the new physical thera-
pist. Given this, a system for managing this data can
prove to be a useful tool to remedy these and other
problems.
An online system enables quick and convenient
analysis of treatment data from anywhere, whether by
the physiotherapist or the patient himself. Depending
on the types of data that are obtained during the ses-
sion, they can be updated immediately after the ses-
sion ends for consultation. If the data is more complex
or needs longer treatment, the patient can be informed
and access the results online at home.
5 TIME-OF-FLIGHT CAMERAS
In order for data to be acquired from a patient’s body,
a sensor must be employed depending on the type of
data sought. For example, to obtain a patient’s heart
rate data, an electrode is placed to receive the signal.
Similarly, to acquire data from a person practicing
physical therapy exercises it is necessary that a sensor
that obtains the individual’s movements be connected
to obtain the movements for the computer.
The motion capture process, also known as Mo-
Cap is performed, according to Vital et al.(Vital et al.,
2017), acquiring the body segments’ acceleration, ve-
locity, time and position of an individual. This data
can be measured using specialized sensors that can
acquire them in different ways depending on the tech-
nology employed. In her work is also shown the dif-
ferent methods and apparatus for the acquisition of
this data, as well as their advantages and disadvan-
tages.
Time-of-flight cameras are optical devices that
have the ability to detect their surroundings by calcu-
lating the time light takes to reach an object and return
(Mutto et al., 2012). An example of using this tech-
nology is the later version of Kinect
TM
(Kinect V2),
also mainly used in the digital gaming business. Un-
like depth cameras, they offer 3D images at a good
frame rate per second (also called FPS), also bringing
a depth gauge for each pixel of the scenario (Vital,
2015). Although intended to be used as a form of con-
trol for digital games, it is a tool that produces, even
in its first version, acceptable results, but inferior to
a Vicon
TM
passive marker system, for example(Dutta,
2012).
Their advantages and disadvantages(Vital, 2015)
are:
Advantages: Depth at each pixel with high fram-
erate; light weight; small and compact design; self
lighting; and reduced power consumption. Disadvan-
tages: Susceptibility to illumination; Multiple reflec-
tions - ToF cameras illuminate an entire scene; and
interference. That means that if multiple ToF cam-
eras are running at the same time, they may disturb
each other’s measurements, depending on multiplex-
ing time and modulation frequencies.
With its approximate price of $200, it is greatly
advantageous compared to other tools available on the
market.(Steward et al., 2015).
How to choose which equipment to use is a crit-
ical topic that underpins the development of a motor
rehabilitation solution for patients. To achieve good
results, you should consider the following aspects of
the tool: Price; Data capture quality; Market acces-
sibility; and value in money.
Comparing the tools currently available to each
other is essential to understanding each one’s
strengths and weaknesses. This comparison proce-
dure is called Benchmarking.
Vital et al.(Vital, 2015) for example, benchmarks
the main aspects of the tools available in the market,
comparing with the proposed solution in her work:
the FatoXtract
TM
.
Using the data raised by Vital et al.(Vital, 2015)
and through analysis of the results of the review ar-
ticles presented by Da Gama et al.(Da Gama et al.,
2015), Szykman et al.(Szykman et al., 2015), and
Dekhordi et al.(Dehkordi et al., 2018), it is noticed
that the Kinect is able to detect and replicate user
BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices
244
movements in a portable way, achieving good results,
and costing less than 10% of the value of the other
high-end tools.
Between the Kinect V1 and V2 cameras, which
have a low price difference, the most reliable and
recent one is the Kinect V2 for Xbox One
TM
, us-
ing time-of-flight technology to map the body and
ensuring better performance. The responsible com-
pany (Microsoft) offers a developer tools package,
also known as Software Development Kit (SDK), al-
lowing the tool not only to be used for MoCap, but for
anything the developer may need to create an applica-
tion.
About the accuracy of the data acquired,
Dutta(Dutta, 2012) reports that Kinect V2 has accept-
able accuracy, but less than that of a Vicon system by
at least one degree of magnitude. It also reports that
some may choose to sacrifice some accuracy in ex-
change for portability. Another point to note is the
availability of Kinect, which can be purchased online
or in physical stores with little effort and few inven-
tory or delivery issues.
6 OBJECTIVES
6.1 General Objectives
Seeking a better quality of life for people with phys-
ical disabilities, this work seeks to create a tool that
demonstrates the acquisition of data from patients un-
dergoing motor rehabilitation. The purpose of this
project is also to help standardize the evaluation of
acquired data and give both patient and physiothera-
pist support and convenience in accessing this data.
6.2 Specific Objectives
This work aims to produce a software using the
Django platform and its libraries to program in
Python language an API (Application programming
interface) that will perform data acquisition, stor-
age and request operations using the Django REST
Framework library. Along with the API, an applica-
tion, also in the same Django project, must be created
as an interface for navigating its users who wish to
view the data.
Data acquisition should be done through a pre-
existing application(Antonio Valente, 2019) made us-
ing Unreal Engine 4.20 integrated with Kinect V2 and
its computer adapter. This application is responsible
for acquiring patient movement data and communi-
cating with the Django API via the VaREST plugin
for requesting, packaging, and sending data in JSON
format.
7 THEORETICAL BASIS
Several researchers have conducted studies in the area
of motion capture. Many use motion capture tools for
healthcare applications, and others research their use
to automate processes with gesture recognition.
Despite being a fairly recent technology, many
applications and research have already been done
achieving great results. The result of these researches
were works that generated important data for the cre-
ation of this article.
The increasing use of motion capture technol-
ogy, as previously mentioned, has grown significantly
since its first appearances and public availability, es-
pecially in healthcare and medicine. With this growth,
a review article is necessary for a better follow-up of
these results obtained in the academic and scientific
community. Szykman(Szykman et al., 2015), Dehko-
rdi(Dehkordi et al., 2018), and Da Gama(Da Gama
et al., 2015) conducted research including the use of
Kinect for physical therapy rehabilitation, including
patients with autism, cerebral palsy, multiple sclero-
sis, and stroke, among other conditions.
Dehkordi et al.(Dehkordi et al., 2018) explore
the use of applications with Kinect, demonstrating
in some of the results the successful use of the tool
with patients with autism, children with psychologi-
cal or emotional disorders and their monitoring. Most
studies included in the research refer to stimulating
physical movements and motor skills, some relying
on some kind of gesture recognition for functioning.
Dehkordi et al. points out that the research and
its implementation are partly employed in the school
environment or at home, but according to the results
obtained, the use of the Kinect system can also be ap-
plied in many other environments, including hospitals
and clinics, where new applications are being devel-
oped and tested.
However, some limitations on Kinect were discov-
ered during the analysis:
Limitation on movement capture of participants
due to lack of space;
The height of adult participants and objects be-
tween player and researcher or Kinect does not
allow the system to correctly detect participant
movement;
Kinect Systems Unable to Support Multiple Dis-
abled Participants
Motor Rehabilitation and Biotelemetry Data Acquisition with Kinect
245
The Kinect sensor must be fixed in one place, and
has a range of around ten meters, meaning that
the movement should occur only in front of the
sensors.
In addition, the study proves a beneficial relation-
ship between Kinect systems and training games, and
findings that suggest facilities for caregivers such as
teachers, therapists, physicians, and family members.
Szykman et al.(Szykman et al., 2015) expresses
how video games have strayed from being not just en-
tertainment gadgets, but also valuable tools that, using
Natural User Interface (NUI), could be used to heal
and treat metabolic, neurological, and vascular com-
plications.
The authors use a automatic search system to
acquire data and keywords from various scientific
databases. It was found that research involving the
development of games for people with disabilities is
increasing, as well as reporting significant improve-
ments in patients’ well-being and the data doctors can
obtain to conduct a better treatment
One of the clearest benefits found was the motiva-
tion generated by games and applications that persist
during treatment, where many researchers take advan-
tage of this element as a way to enhance the treatment
performed.
8 METHODOLOGY
For the creation of a software that fulfills its intended
purpose, a mechanism for storing and manipulating
data is seen as a key to its operation. To achieve
this functionality, the Django platform was used along
with the Django REST Framework plugin to develop
a system that has the ability to function online. Work-
ing online is also an essential attribute for the most
convenient display of this data to the target audience,
ie physiotherapists and patients, enabling them to ac-
cess the data at any time.
For a demonstration of the data that the sys-
tem can manage, this software will initially process
the data obtained from the previous project at Un-
real(Antonio Valente, 2019), which will be developed
along with the program described in this work. The
software (called ’KinectAPI’) consists of a Django
project with a single app called ’kinect’
Using Django and the Django REST Framework
plugin, you able to create an API that can create, store,
and display custom objects in a database. This means
that we can abstract real data into a form that the sys-
tem understands and is free to manipulate. In Django,
these entities are abstracted to objects of a model.
Therefore, we can think of a physiotherapist as an
object model that has important data from the actual
physiotherapist. This object would have, for example,
the name of the physiotherapist, his clinic, his CRM
(Brazilian Physician Credentials), e-mail, telephone
number, among other relevant data.
Like the physical therapist, other entities need to
be abstracted before they can be stored. These other
entities also need to be relevant in the treatment aspect
of the patient. Are they:
Physiotherapist;
Patient;
Treatment;
Treatment Session;
Exercises;
Time.
In Django, to implement this abstraction, you need
to create a model for each of these entities. With mod-
els created in the ’models.py’ file, Django itself gen-
erates the database automatically for immediate use.
Models, like their real entities, have relationships
with each other. For example: A treatment is made
up of a physical therapist, a patient, and has a series
of sessions of an exercise. All of these relationships
must be represented and implemented as per the ac-
tual cases. These relationships can most easily be
seen in the class diagram:
Figure 1: Class Diagram.
Having now all models put and associated as
needed, now you have to create and manipulate this
data. For the correct registration of objects, an initial
interface is required before they can be created. This
can be done through Django’s Forms, coupled with a
View class and an HTML website template.
Views in Django are responsible for managing re-
quests for the software and their responses to each.
In this case, it displays the site with the data to be
filled in by the user, and when submitting, properly
acquires and processes the data received. With this,
we can create a screen for the user to register, be it a
patient or a physical therapist.
BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices
246
Figure 2: Register screen.
Like the registration screen, a login screen has
been implemented. With registered users, they must
be given actions that can be performed with the data.
These actions should also include the registration of
subsequent objects, exercise and treatment. And en-
compassing all the necessary actions, you can create
a use case diagram:
Figure 3: Use case Diagram.
The physiotherapist must be able to perform all
necessary operations before using the application on
Unreal. Therefore, he/she should first register the
treatment, with all necessary data, including the pa-
tient’s id (can also be selected from a list of patients)
and the prognosis of the problem to be solved. After
the treatment has been registered, the Unreal applica-
tion may be used.
To perform the exercise in the application, the
physiotherapist must enter the username and pass-
word along with the registered treatment ID. Thus, it
is not necessary to enter patient data or write about
the prognosis, all these data were previously entered
and processed. As soon as the physiotherapist enters
the required data, he is taken to a menu screen where
the corresponding exercise can be selected.
Any operation involving data transfer between the
application in Unreal and the Django API is per-
formed through JSON requests. The API has a se-
ries of URL addresses where the application can make
these requests to perform a desired operation. When
logging in, for example, the credentials obtained are
Figure 4: Training selection screen.
registered and sent to the API. It then checks the en-
tered credentials and, if authorized, returns a Token.
This token is stored so that it can be used in place
of the username and password for subsequent opera-
tions.
Once the exercise is selected, a session is created
and assigned to the treatment whose id was provided,
along with the exercise date, time, and name. Times
are sent each time the patient touches an object when
asked. In the Unreal app, the exercise is performed
from a series of static objects floating over the patient,
all semi-translucent in color, which turn golden when
the patient must move and stretch to reach them. Only
one of the objects activates and turns golden at a time.
At touch, the name of the touched object as well as the
limb used to touch are sent to the database. Like login,
session creation and time recording are also requests
with JSON.
With the interaction between the two systems im-
plemented, it now remains to create an interface for
users to see the data acquired, and in the case of the
physiotherapist, create and manage this data. In the
Django API itself it is possible to provide this inter-
face in the same way as the register. A view is created
to manage each operation made. When logging in, the
user must have operations available in a menu accord-
ing to their type: Physiotherapist or Patient.
The patient can check his treatments in a list,
and check each exercise session, from the same point
of view of the physiotherapist. The physiotherapist
has more operations, such as recording a treatment,
adding an evaluation to a treatment, recording a new
exercise, among other views that can be seen in the
use case diagram (Figure 3).
After selecting a treatment, information about it
can be viewed on the screen as either a small table
sorted with the time taken to perform the exercise ac-
tion and the limb used.
As in the session details, a graph is also shown
when selecting a treatment, with performance data per
session. All graphs are generated using the Graph.js
extension, enabling automatic generation of an inter-
active graph using the JavaScript language. These
Motor Rehabilitation and Biotelemetry Data Acquisition with Kinect
247
Figure 5: Performance Graph per Limb (Right/Left Arm).
Figure 6: Performance Graph per Session.
charts can display or omit information according to
your user’s needs. So if the user wishes, only one or
all sessions can be displayed at a time.
9 CONCLUSIONS
From the results obtained from the integration be-
tween the two platforms, it is possible to demonstrate
the ability of both systems to communicate and ex-
change information with each other, enabling the ac-
quisition of patient movement data through the Un-
real application and the processing and display of data
obtained using a Django system. In addition to com-
municating and receiving application information, the
API is also capable of using graphical systems and ta-
bles for proper data display, facilitating and ensuring
better and faster understanding by both the physio-
therapist in charge and the patient that wants to check
their own performance.
9.1 Testing with Wheelchair Users
As stated by Da Gama(da Silva et al., 2007), a com-
mon problem seen in many studies related to motor
rehabilitation with Kinect is the lack of actual patient
testing. However, for successful and properly tested
tests, the approval and follow-up of the Brazilian Re-
search Ethics Committee (CEP) with human beings
is required and to be in accordance with international
ethical guidelines (Declaration of Helsinki, Interna-
tional Guidelines for Biomedical Research involv-
ing Human Beings - CIOMS) and Brazilian (Resolu-
tion CNS 466/12 and complementary). This follow-
up process requires an extensive process that takes
months to complete before testing procedures.
That said, during project development, the objec-
tives and prototypes were delivered and tested with
one person, both wheelchair user and physical ther-
apy professional. During the tests, exercise proto-
cols were presented and the data acquisition proposal
validated, besides participating in proposed exercise
sessions. However, for more extensive research with
more concise data, more physical therapists and vol-
unteer wheelchair users are required, which again re-
quires the presence of the Research Ethics Commit-
tee.
10 FUTURE WORKS
With both functional systems in hand, there remains,
in addition to wheelchair research and testing, the im-
provement of the exercises to be used and their ap-
plication to different parts of the body and in dif-
ferent environments. Another progress to be imple-
mented in the project is the expansion of data obtained
via biotelemetry and its processing, which may range
from the patient’s heart rate to the angle of the limb in
interest.
Another necessary implementation is the direct
experimental use of the system in a hospital environ-
ment and in physiotherapy clinics, for a better anal-
ysis of the strengths and weaknesses of the system
and their appropriate corrections. With appropriate
changes made, the system should be retested until it
is ready for actual use in rehabilitation clinics.
REFERENCES
Antonio Valente, Salviano Soares, F. M. d. A. N. F. F. P. F. J.
A. F. (2019). A new approach of developing games for
motor rehabilitation using microsoft kinect. Proceed-
ings of SeGAH 2019, IEEE 7th International Confer-
ence in Serious Games and Applications for Health,
Kyoto, Japan.
Cooke, S. J., Hinch, S. G., Wikelski, M., Andrews, R. D.,
Kuchel, L. J., Wolcott, T. G., and Butler, P. J. (2004).
Biotelemetry: a mechanistic approach to ecology.
Trends in ecology & evolution, 19(6):334–343.
BIODEVICES 2020 - 13th International Conference on Biomedical Electronics and Devices
248
Costa, N., Brown, M., Hutchins, G., and Caldwell, D.
(2005). Design of human-friendly powered lower limb
rehabilitation orthosis. In Workshop on Human Adap-
tive Mechatronics, pages 69–75.
Cummings, A. H. (2007). The evolution of game controllers
and control schemes and their effect on their games.
In The 17th annual university of southampton multi-
media systems conference, volume 21.
Da Gama, A., Fallavollita, P., Teichrieb, V., and Navab, N.
(2015). Motor rehabilitation using kinect: a system-
atic review. Games for health journal, 4(2):123–135.
da Silva, M. A. S. R., de Lima, E. V., and de Carvalho, F.
A. S. (2007). A relev
ˆ
ancia do tempo de reac¸
˜
ao em
modalidades esportivas.
Dehkordi, S. R., Ismail, M., and Diah, N. M. (2018). A
review of kinect computing research in education and
rehabilitation. International Journal of Engineering
and Technology (UAE), 7(3):19–23.
Dukes, P. S., Hayes, A., Hodges, L. F., and Woodbury, M.
(2013). Punching ducks for post-stroke neurorehabil-
itation: System design and initial exploratory feasi-
bility study. In 2013 IEEE Symposium on 3D User
Interfaces (3DUI), pages 47–54. IEEE.
Dutta, T. (2012). Evaluation of the kinect
TM
sensor for 3-
d kinematic measurement in the workplace. Applied
ergonomics, 43(4):645–649.
Gerling, K. M., Kalyn, M. R., and Mandryk, R. L. (2013).
Kinect wheels: wheelchair-accessible motion-based
game interaction. In CHI’13 Extended Abstracts on
Human Factors in Computing Systems, pages 3055–
3058. ACM.
Gil, J. A. N. (2015). Medic¸
˜
ao e avaliac¸
˜
ao em fisioterapia.
Sa
´
ude & Tecnologia, (6):5–9.
Hamari, J., Koivisto, J., Sarsa, H., et al. (2014). Does gami-
fication work?-a literature review of empirical studies
on gamification. In HICSS, volume 14, pages 3025–
3034.
Herz, N. B., Mehta, S. H., Sethi, K. D., Jackson, P., Hall, P.,
and Morgan, J. C. (2013). Nintendo wii rehabilitation
(“wii-hab”) provides benefits in parkinson’s disease.
Parkinsonism & related disorders, 19(11):1039–1042.
Hsieh, W.-M., Chen, C.-C., Wang, S.-C., Tan, S.-Y.,
Hwang, Y.-S., Chen, S.-C., Lai, J.-S., and Chen, Y.-
L. (2014). Virtual reality system based on kinect for
the elderly in fall prevention. Technology and health
care, 22(1):27–36.
Kim, Y. and Cho, J. M. (2001). Development of wireless
bio-telemetry system using fm stereo method for ex-
ercising rehabilitation patients. In 2001 Conference
Proceedings of the 23rd Annual International Confer-
ence of the IEEE Engineering in Medicine and Biol-
ogy Society, volume 4, pages 3338–3340. IEEE.
Martin-SanJose, J.-F., Juan, M.-C., Moll
´
a, R., and Viv
´
o,
R. (2017). Advanced displays and natural user inter-
faces to support learning. Interactive Learning Envi-
ronments, 25(1):17–34.
Mutto, C. D., Zanuttigh, P., and Cortelazzo, G. M. (2012).
Time-of-flight cameras and microsoft kinect (TM).
Springer Publishing Company, Incorporated.
Nazirzadeh, M. J., C¸ agiltay, K., and Karasu, N. (2017). De-
veloping a gesture-based game for mentally disabled
people to teach basic life skills. International Associ-
ation for Development of the Information Society.
Penhaker, M.,
ˇ
Cern
`
y, M., Martinak, L., Spi
ˇ
s
´
ak, J., and
Valkova, A. (2007). Homecare—smart embedded
biotelemetry system. In World Congress on Medi-
cal Physics and Biomedical Engineering 2006, pages
711–714. Springer.
Robertson, C., Vink, L., Regenbrecht, H., Lutteroth, C.,
and W
¨
unsche, B. C. (2013). Mixed reality kinect mir-
ror box for stroke rehabilitation. In 2013 28th Inter-
national Conference on Image and Vision Computing
New Zealand (IVCNZ 2013), pages 231–235. IEEE.
Steward, J., Lichti, D., Chow, J., Ferber, R., and Osis, S.
(2015). Performance assessment and calibration of the
kinect 2.0 time-of-flight range camera for use in mo-
tion capture applications. FIG Working week 2015,
pages 1–14.
Szykman, A. G., Gois, J. P., and Brand
˜
ao, A. L. (2015).
A perspective of games for people with physical dis-
abilities. In Proceedings of the Annual Meeting of the
Australian Special Interest Group for Computer Hu-
man Interaction, pages 274–283. ACM.
Vital, J. P., Faria, D. R., Dias, G., Couceiro, M. S.,
Coutinho, F., and Ferreira, N. M. (2017). Combining
discriminative spatiotemporal features for daily life
activity recognition using wearable motion sensing
suit. Pattern Analysis and Applications, 20(4):1179–
1194.
Vital, J. P. M. (2015). An
´
alise do movimento humano:
classificac¸
˜
ao temporal de ac¸
˜
oes humanas.
Motor Rehabilitation and Biotelemetry Data Acquisition with Kinect
249