EPIK

Virtual Rehabilitation Platform Devised to Increase Self-reliance of People with

Limited Mobility

Sonia Garrote

, Azael J. Herrero

1,2

, Miguel Pedraza-Hueso

, Carlos González-Gutiérrez

María V. Fernández-San Román

, F. J. Díaz-Pernas

, Héctor Menéndez

Cristina M. Ferrero

and Mario Martínez-Zarzuela

Faculty of Health Sciences, European University Miguel de Cervantes, Valladolid, Spain

Research Center in Physical Disability, ASPAYM Castilla y León Foundation, Valladolid, Spain

Imaging and Telematics Group, University of Valladolid, Valladolid, Spain

Keywords: Exergames, Rehabilitation, Stroke, Physical Therapy, Microsoft Kinect, Elderly.

Abstract: In this paper we describe a virtual rehabilitation platform designed to improve balance of people with

physical impairment using the Microsoft® Kinect® sensor. Different types of users can interact with the

platform: Administrators, therapists, and final users (patients), using their own interfaces and modules. Six

modules have been designed: Profile, Administrator, Evaluation, Therapist, Game and Results; but only four

have been implemented so far: Administrator, Evaluation, Therapist and Game. The Administrator’s module

is used to generate a database of exercises. The Therapist’s module allows therapists to configure the game

training session using combinations of exercises from the database. The patients’ or game module includes a

3D immersive environment, where they perform the prescribed rehabilitation exercises, previously

configured by a therapist. The platform is in its first beta version and ready to be tested.

1 INTRODUCTION

The live expectancy in countries like Spain is about

79 years for males and 85 years for females (World

Health Organization –WHO-, 2012). Different

factors like medical advances, allow a lot of people

to survive until their seventies or so far. However, in

many cases elderly people suffer chronic illness and

pathologies that reduce their mobility and make

them being dependent. Nowadays, one of the health

problems with a higher incidence in elderly subjects

is stroke. The risk of stroke doubles every decade

after age 55 (Mackay and Mensha, 2004). Patients

who suffer a stroke spend prolonged stays in

intensive care units and have a markedly higher

mortality rate (Hornero et al, 2013). These patients

usually suffer a paralysis that compromises their

ability to perform activities of daily living (ADLs)

(Nichols-Larsen et al., 2005). In order to maximize

patient outcomes and reduce disability it is essential

to carry out programs of physical rehabilitation

(Saunders, 2014).

Some programs applied to stroke patients during

the acute phase are based on Early Supported

Discharge (ESD). ESD consists on attending to the

acute stroke care and rehabilitation unit and pretend

an early discharge from hospital performing a

supervised rehabilitation program at home. These

programs can reduce long-term dependency, the

length of hospital stays and admission to

institutional care units (Fisher, 2011; Fearon and

Langhorne, 2012).

Nonetheless, what happens after the acute phase

of a stoke patient? Should patients continue the

rehabilitation programs in the chronic/late post

stroke period? The Evidence Based Review of

Stroke Rehabilitation states that there is strong

evidence about the effectiveness of balance training

programs to improve health-related outcomes. This

review also asserts that task-specific gait training

improves the gait after stroke, treadmill training

alone (without partial weight support) increases gait

velocity, and virtual reality training enhances gait

recovery (Teasell et al., 2013).

The Spanish healthcare system does not provide

these rehabilitation programs to stroke patients

188

Garrote S., Herrero A., Pedraza-Hueso M., González-Gutiérrez C., Fernández-San Román M., Díaz-Pernas F., Menéndez H., Ferrero C. and

Martínez-Zarzuela M..

EPIK - Virtual Rehabilitation Platform Devised to Increase Self-reliance of People with Limited Mobility.

DOI: 10.5220/0005484301880193

In Proceedings of the 1st International Conference on Information and Communication Technologies for Ageing Well and e-Health (ICT4AgeingWell-

2015), pages 188-193

ISBN: 978-989-758-102-1

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

during the chronic phase of the illness. Stroke

patients have to pay for these programs and have to

attend to special health centers to receive the

therapy. Nevertheless, what happens if patients live

too far from the rehabilitation institutions?

Moreover, how could patients continue and reinforce

their rehabilitation at home?

Virtual reality systems, including video games,

can help to reach this purpose. In a recently

published review, it was concluded that video games

improve health outcomes in areas like psychological

therapy and physical therapy. However, the authors

suggest that more rigorous Random Clinical Trials

are needed to confirm this effectiveness (Primack et

al., 2012). Other researchers denote an increment in

the motivation of the patients towards physical

rehabilitation (Kato, 2010; Chang et al., 2011),

improvements in their motor skills and physical

condition (Jansen-Kosterink et al., 2013) or,

encouraging results for recovery of muscle force and

power of the lower limb for individuals with chronic

hemiparesis (Mirelman et al., 2010). These kinds of

developments are cost effective since they can

benefit a large number of patients and can improve

patients’ health.

Our multidisciplinary research group, which

includes physiotherapist, professionals in physical

activity and sport sciences, computer science

engineers and telecommunications engineers, aims

to design and implement virtual reality therapies

based on serious games.

This paper shows the technical specification of a

rehabilitation platform (EPIK) under development

by our research group. EPIK is similar to other

virtual rehabilitation systems like KiRes (Anton,

2013) that uses the Microsoft Kinect sensor to

interact with the users. In contrast to KiRes, the

administrator does not need to record his postures

using Kinect to create new exercises, so the time that

is needed to include a new exercises program into

the database is significantly decreased. Gamification

techniques such as an immersive 3D environment,

game levels, game modes, velocity, game session

time and victory points have been used to develop

EPIK. For this reason, the patient enjoys a “game

sensation” that increases his motivation, distracts

him from his illness and engages him to the system

decreasing the rate of neglect.

Other researchers (Su, 2013, Robertson et al.,

2013, Rajaratnam et al., 2013, Pirovano et al., 2012,

Muñoz et al., 2013), use Kinect based systems in the

physical rehabilitation area. However, those works

do not present a complete platform and do not either

include gamification nor have been tested using

random clinical trials. In a recent systematic review

(Da Gama et al., 2015) it is pointed that a complete

system must be developed to include aspects like

therapeutic configuration, different exercise

possibilities, guidance and feedback. EPIK has the

aim to pave the way to resolve all these lacks.

At this moment, we have designed a random

clinical trial (RCT), following the CONSORT

recommendations (Boutron et al., 2008) to test this

first prototype with users with reduced mobility

problems caused by a stroke. The RCT has been

performed in a physiotherapy research center. The

patients are in the chronic stage of their illness. The

results obtained in the tests will serve to improve the

platform and obtain the first operative release.

2 METHODS

2.1 Purpose of the System

The aim of the system is to improve the level of

independency of people with reduced mobility such

as elderly, subjects with physical disabilities or

physical damage, using a virtual platform for

physical training.

Although anyone can use this virtual platform as

a training tool, it has been designed with the aim of

improving the balance of users with reduced

mobility.

2.2 Technologies

The EPIK platform employs the Microsoft Kinect

v.2 sensor (http://www.microsoft.com/en-us/kinect

forwindows) to capture users’ movements and

includes a game module to train the users’ body.

The platform has been programmed with Unity

(http://unity3d.com/), a “game development

ecosystem” able to create interactive 3D and 2D

content and run it in multiplatform systems like

Desktop (PC, Mac and Linux), Web, iOS, Android

and consoles (i.e. Xbox, Wii, PS4).

The recommended hardware and software

configurations for EPIK are: a Microsoft Kinect

sensor v.2, a PC with Microsoft Windows v.8 or

later with the Kinect drivers installed and the

Microsoft recommended hardware configuration.

2.3 Main Features

The EPIK platform is comprised of these modules:

Profile module: it is used to maintain user’s basic

EPIK-VirtualRehabilitationPlatformDevisedtoIncreaseSelf-relianceofPeoplewithLimitedMobility

189

information: username, password, permissions,

patient’s clinical records. This module is under

development.

Administrator’s module: it is used to create a

database of exercises and classify them in levels. In

each level we can create groups of exercises –

specific human body positions that we call

‘silhouettes’–. Thus, a level can have different

groups. In beta version.

Evaluation module: it is used to evaluate the

range of movement (ROM) of the user joints. In beta

version.

Therapists’ module: the therapist can design the

game sessions for the user choosing groups of

exercises. In beta version.

Results module: to analyze the evolution of the

patient. This module is under development.

Game module: in the game module, two human

figures are showed to the patient. The first figure is

the virtual image of the user that is training, that is,

the user’s avatar. The second figure is a silhouette

moving towards the user’s avatar under a

configurable velocity. The user must adopt the pose

of the silhouette so that his avatar matches it in order

obtain different scores in the game (Figure 1). In

beta version.

Figure 1: The user’s avatar (in first plane) must to adopt

the position of the silhouette to imitate (in second plane).

The silhouette to imitate comes towards the user's avatar

with a predefined velocity.

2.4 Account Types

Depending on the type of user account, the user is

able (or not) to access a given module in the EPIK

platform, with more or less permissions.

Patient’s account: if you have a patient account,

you are the person who uses the platform for your

physical training. You can access to the next

modules: profile module with read permission,

evaluation module with execution permission, game

module to play the game sessions configured for you

and results module with read permission.

Administrator’s account: the administrator has

full access to all modules with full permissions. His

main job is to create the silhouettes by configuring

their positions and save them in the database. The

therapist uses this database to design the game

sessions of a patient.

Therapist’s account: a therapist has full access to

the Therapist’s module. In this module the therapist

can see all the silhouettes of the database classified

into levels and groups by the administrator using the

Administrator’s module. The therapist selects groups

of silhouettes for each patient’s game session. The

therapist can access the rest of the modules with full

permissions, except to the Administrator’s module.

2.5 Modules of the System

2.5.1 Administrator’s Module

In the Administrator’s module silhouettes for a given

level of difficulty can be created. Silhouettes can be

generated by rotating the skeletal joints of an avatar

up to the desired position (Figure 2). Silhouettes

being generated can be grouped to ease the

configuration of a rehabilitation session afterwards.

The Administrator’s module does not restrict the

type of silhouettes that the administrator can create.

It is the administrator who is responsible of

following a previously agreed protocol about the

type of exercises to be included on a specific level.

Figure 2: Window to create groups of silhouettes in a

level.

This protocol has, for each level, the type of

recommended movements. In Table 1 we can see the

types of exercises that protocol has for level 1. A

higher level should imply a greater difficulty. Our

physiotherapists and professionals in physical

activity and sport sciences have created this protocol

ICT4AgeingWell2015-InternationalConferenceonInformationandCommunicationTechnologiesforAgeingWelland

e-Health

190

based in their experience with physically disabled

patients. The protocol has six levels of difficulty.

Here, we show only the first one, as an example.

In case that the protocol needs to be changed in

the future, it would not be necessary to change any

line of source code of the Administrator’s module,

because there are no restrictions to create whatever

silhouette that the administrator desires.

Table 1: Type of movements for level 1.

Level 1

Description Postures with trunk movements and

unilateral upper limb movements

Movements Shoulder abduction, horizontal

abduction, shoulder flexion, shoulder

extension, lateral trunk tilt, trunk flexion

and extension

Detail The silhouettes may adopt combinations

of these movements always with bipodal

support, legs hip width apart and knee

totally extended. Body weight will be

evenly distributed between both feet or

more weight on one leg than the other.

2.5.2 Evaluation Module

The evaluation module measures the maximum

ROM of each joint when certain movements are

performed. These movements are defined in the

evaluation protocol designed by our physiotherapists

and professionals in physical activity and sport

sciences (Table 2).

The patient stands opposite to the Microsoft

Kinect sensor and can see his own representation in

the screen (his avatar) on the right and an avatar

performing an exercise that he or she has to imitate

on the left (Figure 3).

Figure 3: Evaluation module.

Two performances of the same movements

(Table 2) have to be carried out by the patient. The

user performs the movements in real time while they

are showed in the screen. When the two sequences

of movements are finished, the system computes the

maximum ROM that he or she reaches, using the

information collected during the two performances.

All the movements start from a base position: the

body erect with the arms at the sides and the palms

forward. In each movement, the user tries to reach

his maximum ROM and then, returns to the base

position.

Table 2: Sequence of movements during the evaluation

phase.

Abduction of the right shoulder, then left and finally

both bilaterally

Flexion of the right shoulder, then left and finally both

bilaterally

Horizontal abduction of the right shoulder, then left and

finally both bilaterally

Extension of the right shoulder, then left and finally

both bilaterally

Trunk tilt left and right. Trunk flexion and extension

Abduction right hip and then left (only exercise with

unilateral support, not for hip abduction exercises in

bilateral support)

Right hip flexion and then left (only for unilateral

support exercises and knee flexion)

Bending both knees (squat)

Step to the right side. Step to the left side. Step forward

with the right foot. Step forward with the left foot

2.5.3 Therapist’s Module

In this module the therapist can configure the game

session for each user. A game session represents the

groups of exercises. It is possible to choose groups

of silhouettes from different levels. For example: the

therapist can choose two groups of silhouettes from

level one, and one group of silhouettes from level

two; or he can choose one group from level two, and

one group from level five. There is total flexibility to

select whatever the therapist wants.

2.5.4 Game Module

Once the therapist has created the game sessions for

a given patient, the game module can be ran.

First, we select the patient that is going to play.

Second, we select the avatar.

Third, we select one of the sessions configured

for this user.

Fourth, we set the next parameters (Figure 4):

Speed: time it takes for the silhouette to reach the

user’s avatar (Figure 1). A higher speed implies a

greater difficulty.

Mode: this parameter is used to dynamically

adapt the silhouettes to be played using a percentage

of the maximum ROM that the user obtained in the

EPIK-VirtualRehabilitationPlatformDevisedtoIncreaseSelf-relianceofPeoplewithLimitedMobility

191

evaluation module. It is used to adjust the difficulty

of the game session.

Table 3: Types of Game Modes.

Mode % of the maximum ROM

Beginner 80%

Official 90%

Master 100%

Doctor 105%

Expert 110%

Suppose that, in the evaluation module, the user

performed an abduction movement of the arm and

his maximum ROM was 60º. If the therapist sets the

game session mode to beginner, the user will need to

rise his arm up to the 80% of 60º -48º- to obtain the

best score.

Waiting time between silhouettes.

Length of session: duration of a training session.

It is measured in minutes and seconds. The user

scores when he adopts the correct posture, in other

case there is no penalization. The concept of "life"

does not exist. Users do not loss a "life" when they

perform incorrect movements. The training session

Figure 4: Setting the session parameters in the game

module.

Figure 5: The first round of the game starts.

Figure 6: The user playing.

continues until the end or until it is paused. The

length of the session is a key configuration

parameter from the clinical point of view.

In the last step, the game starts (Figure 5), takes

place (Figure 6), finishes and saves results.

3 CONCLUSIONS

This paper presents the beta version of the EPIK

platform. A new Microsoft Kinect sensor based

system developed to improve balance through

training sessions played in a virtual 3D environment.

EPIK has a serious exergame that can be easily

configured to include exercises oriented to improve

patient’s movements velocity, coordination,

flexibility, etc.

All the EPIK modules have been designed with

the support of physical rehabilitation professionals.

Moreover, EPIK is being tested with stroke patients

today. The results of these tests will be used to

further improve the system and to have, in the near

future, an effective rehabilitation-assisted

technology with a real user centered design.

The aim is to construct a complete system using

a low cost sensor like Microsoft Kinect that will

address the evaluation and treatment cycle of a

patient in a simple and inexpensive way.

REFERENCES

Antón D., Goñi A., Illarramendi A., Torres-Unda J.J.,

Seco J., 2013. KiRes: A Kinect-based

telerehabilitation system. IEEE 15th International

Conference on e-Health Networking, Applications and

Services, 456-460.

Boutron I., Moher D., Altman D.G., Schulz K.F., Ravaud,

P., 2008. Extending the CONSORT statement to

ICT4AgeingWell2015-InternationalConferenceonInformationandCommunicationTechnologiesforAgeingWelland

e-Health

192

randomized trials of nonpharmacologic treatment:

explanation and elaboration. Annals of Internal

Medicine, 148(4), 295-309.

Chang Y.-J., Chen S.-F., Huang J.-D., 2011. A Kinect

based system for physical rehabilitation: A pilot study

for young adults with motor disabilities. Research in

Developmental Disabilities, 32(6), 2566–2570.

Da Gama A., Fallavollita P., Teichrieb V., Navab, N.,

2015. Motor rehabilitation using Kinect: A systematic

review. Games for Health Journal, 4(2), 123-135.

Fearon P., Langhorne P., 2012. Services for reducing

duration of hospital care for acute stroke patients.

Cochrane Stroke Group, 9, 1-97.

Fisher R.J., Gaynor C., Kerr M., Langhorne P., Anderson

C., Bautz-Holter E., Indredavik B., Mayo N.E., Power

M., Rodgers H., Rønning O.M., Widén-Holmqvist L.,

Wolfe C.D., Walker M.F., 2011. A consensus on

stroke: early supported discharge. Stroke, 42(5), 1392-

1397.

Hornero .F, Martín E., Rodríguez R., Castellà M., Porras

C., Romero B., Maroto L., Pérez De La Sota E, 2013.

A multicentre Spanish study for multivariate

prediction of perioperative in-hospital cerebrovascular

accident after coronary bypass surgery: the PACK2

score. Interact Cardiovasc Thorac Surg, 17(2), 353-

358.

Jansen-Kosterink S.M., Huis in ’t Veld R., Schönauer C.,

Kaufmann H., Doz Mag P., Hermens H.J.,

Vollenbroek-Hutten M., 2013. A Serious Exergame

for Patients Suffering from Chronic Musculoskeletal

Back and Neck Paint. (Pilot Study). Games for Health

Journal: Research, Development, and Clinical

Applications, 2(5), 299-307.

Kato P.M., 2010. Video games in health care: Closing the

gap. Rev. Gen Psychol, 14(2), 113-121.

Mackay, J., Mensha G. A., 2004. The Atlas of Heart

Disease and Stroke. Risk factors. World Health

Organization (WHO), in conjunction with the USA's

Centers for Disease Control and Prevention (CDC) of

the U.S. Department of Health and Human Services,

and is strongly supported by NGOs such as the World

Heart Federation (WHF). Available: http://www.who.

int/cardiovascular_diseases/en/cvd_atlas_03_risk_fact

ors.pdf?ua=1 (Last visited Feb. 4, 2015).

Mirelman A., Patritti B. L., Bonato P., Deutsch J. E.,

2010. Effects of virtual reality training on gait

biomechanics of individuals post-stroke. Gait &

Posture, 31(4), 433-437.

Muñoz J., Henao O., López J., Villada J., 2013. BKI:

Brain Kinect Interface, a new hybrid BCI for

rehabilitation. Games for Health, 233-245.

Nichols-Larsen D.S., Clark P.C., Zeringue A., Greenspan

A., Blanton S., 2005. Factors influencing stroke

survivors' quality of life during subacute recovery.

Stroke, 36(7), 1480-1484.

Pirovano M., Mainetti R., Baud-Bovy G., Lanzi P.L.,

Borghese N.A., 2012. Self-adaptive games for

rehabilitation at home. Computational Intelligence and

Games (CIG), 2012 IEEE Conference on IEEE, 179-

186.

Primack B. A., Carroll M.V., McNamara M., Klem M.L,

King B., Rich M., Chan C.W., Nayak S., 2012. Role

of Video Games in Improving Health-Related

Outcomes A Systematic Review. American Journal of

Preventive Medicine, 42(6), 630-638.

Rajaratnam B., Gui KaiEn J., Lee JiaLin K., SweeSin K.,

Sim FenRu S., Enting L., Ang YiHsia E., KeatHwee

N., Yunfeng S., Woo YingHowe W., 2013. Does the

Inclusion of Virtual Reality Games within

Conventional Rehabilitation Enhance Balance

Retraining after a Recent Episode of Stroke?.

Rehabilitation research and practice, vol. 2013,

Article ID 649561, 6 pages.

Robertson C., Vink L., Regenbrecht H., Lutteroth C.,

Wünsche B.C., 2013. Mixed reality Kinect Mirror box

for stroke rehabilitation. 2013 28

International

Conference of Image and Vision Computing New

Zealand (IVCNZ), 231–235.

Saunders D.H., Greig C.A., Mead G.E., 2014. Physical

activity and exercise after stroke: review of multiple

meaningful benefits. Stroke, 45(12), 3742-3747.

Su C.J., 2013. Personal rehabilitation exercise assistant

with kinect and dynamic time warping. International

Journal of Information and Education Technology,

3(4), 448-454.

Teasell R., Foley N., Salter K., Richardson M., Allen L.,

Hussein N., Bhogal S., Jutai J., Speechley M., 2013.

Evidence - Based Review of Stroke Rehabilitation.

Executive Summary (16th Edition). EBERSR.

Available: http://www.ebrsr.com/ (Last visited Feb. 4,

2015).

World Health Organization (WHO), 2012. Global Health

Observatory Data Repository. Available:

http://apps.who.int/gho/data/node.main.688. (Last

visited Feb. 4, 2015).

EPIK-VirtualRehabilitationPlatformDevisedtoIncreaseSelf-relianceofPeoplewithLimitedMobility

193