MATCHuP: An mHealth Tool for Children and Young People Health

Promotion

Beatriz L

opez

, Sreynoch Soung

, Natalia Mordvanyuk

, Albert Pla

, Pablo Gay

and Abel L

opez-Bermejo

University of Girona, Girona, Spain

Institute of Technology of Cambodia, Phnom Penh, Cambodia

University of Oslo, Oslo, Norway

Biomedical Research Institute of Girona, Girona, Spain

Keywords:

Decision Support Systems, Telemedicine, Nursing Informatics, Wearable Health Informatics, eHealth

Applications.

Abstract:

The kids of European and occidental countries are threatened by obesity. They are potential persons to become

chronic patients. mHealth technology can help them to change their nutrition and physical activity habits. This

paper presents MATCHuP, a platform that involves several agents (kids, parents, healthcare providers) that

collaborate and compete by games in a social network in order to create a enjoyable environment to promote

a behavioural change towards a healthier life.

1 INTRODUCTION

Having care of our kids today is the best bid we can

make to have a healthy society tomorrow

. How-

ever, three main diseases threaten kids health: obe-

sity, diabetes and asthma. While parents are conscien-

tious that diabetes and asthma are serious diseases, it

seems that they are less aware about the harm of obe-

sity. Obesity can be cured, avoiding reaching the adult

age with related serious and chronic diseases with co-

morbidities.

Obesity is mainly caused by bad nutrition habits

and the absence of exercise. The development of

tools to support enhancing good nutrition and exer-

cise habits would result in beneﬁts for both, obesity

and diabetes, known as metabolic diseases.

Nowadays, information and communication tech-

nology (ICT) is offering a means to support follow up

of personal data, enabling the education on the right

habits of obese children, but also offers clinicians a

way to gather information about their patients, and

other data coming from social workers. In that regard,

Artiﬁcial Intelligence Techniques, including Machine

Learning, have shown to be cornerstone to transform

The MOCHA project (Models of Child Health Ap-

praised), http://www.childhealthservicemodels.eu/

gathered data to knowledge, so as to support decision

making towards a personalized health treatment (par-

ents, clinicians) (Herrero et al., 2016; L

opez et al.,

2013). Games is another important technology that

has been raised as a key issue for education, and can

be a key issue for kids to be compliant with their treat-

ment, especially if games are not designed as serious

but popular games are paid via exercise or good diet

instead of money.

This paper presents the MATCHuP platform cen-

tred on patient with metabolic diseases and their fam-

ilies to improve their education in the right habits

towards a healthy future society. In so doing,

MATCHuP aims to recommend actions connected to

the patients community, using collaboration strategies

to simplify input validation, and competition incen-

tives so as to award the patient with gaming.

This paper is organized as follows. First, some

related work is reviewed in Section 2. Next the de-

scription of the MATCHuP platform is presented in

Section 3. The status of the current implementation is

explained in Section 4. We end the paper on Section

5 with some conclusions and future work.

LÃ¸spez B., Soung S., Mordvanyuk N., Pla A., Gay P. and LÃ¸spez-Bermejo A.

MATCHuP: An mHealth Tool for Children and Young People Health Promotion.

DOI: 10.5220/0006143303130318

In Proceedings of the 10th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2017), pages 313-318

ISBN: 978-989-758-213-4

313

2 RELATED WORK

The use of mHealth approaches to child obesity has

been evaluated in (Tate et al., 2013). Among the

advantages, the authors highlight four issues. First,

that adolescent would prefer mobile-based and itera-

tive technology for treatment and prevention. Second,

that mHealth is a useful tool for monitoring adher-

ence. Third, that the reachability enhancement of the

population of a higher risk for obesity. And four, a

similar enhancement could be observed regarding of

the lowest educational level and income. However,

several challenges were detected, as the sedentary be-

haviour of screen technologies and the detected de-

creased ability to focus attention of screen users. In

that regard, wearables technology is arising as a new

tool for measuring the real activity of kids, enabling

the development of alternative mHealth platforms that

tackle such challenges. MATCHuP uses them.

Of course there are myriads of mHealth solutions.

In the recent report (Aitker, 2015) about 165,000

apps were identifyed. Among them, the authors con-

ﬁrmed 46,188 mHealth apps and they focus their

stydy on the English Apps from which 26,864 where

consumer/patient and only 8,965 apps were related

to healthcare providers. Our app includes health-

care providers as well as consumers, and it is multi-

language, currently in English, Spanish and Catalan.

Regarding apps for kids, Table 1 shows a list of

several apps. The elements that conﬁgure a mHealth

success regarding nutritional habits, as for example,

sugar ingesta, have been studied in (Sanders et al.,

2009).The key issues to have success tools involves

how the information is delivered. In that regards, sev-

eral agents are identiﬁed in order to improve the lit-

eracy on health: caregivers, health systems, family

health literacy skills, the educational system and the

community system. In our work, we involve most

of such actors: caregivers (endocrinologists), family

(parents), and the community.

3 PLATFORM DESCRIPTION

The goal of MATCHuP is to improve nutrition and

exercise habits of obese kids from 5 to 16 years.

To that end, the platform gathers information about

kids meals and physical activity, and according to

the healthy quality of the data entered, kids awarded

with some points, that are translated in skill scores

regarding a virtual game. The game is not played

in isolation but in teams. Therefore, kids should

collaborate among her mates in order to have a

competitive team that beats their adversaries. On the

other hand, the validation of the information entered

by the kids is performed in a collaborative way.

Therefore, several actors are involved in

MATCHuP. First caregivers set up healthy tar-

gets to the kids according to their progresses. Second,

the kids that self-monitor their progress toward

the targets. Third, the parents that collaborates in

the monitoring, by validating the inputs entered

by kids. And ﬁnally, all the community of users

(parents and kids) collaborate in different way inside

a social network implemented for their community.

Moreover, kids can set up teams in the community

which compete in virtual games, and the skills of the

avatars (virtual representation of the kid) depend on

their healthy progress. An overview of the platform

is shown in Figure 1. In the remaining of this section

the different roles of the agents involved in the

platform are described, including the social network

in which they collaborate, and how the games are

approached.

Table 1: Apps for healthy Children. Physical activity: v →

virtual exercise; t → teach about exercise.

App Name

Food

Nutrition

Physical Activity

Easy Eater 2 x x

Eat and Move-O-Matic x x v,t

Healthy Heores x x

Perfect Picnic x x

Smash Your Food x x

Veggie Circus Farm x x

Body Quest - Food of the Warrior. x x

Grow It-Know It x x

Catch the Carrot x x

Snack Planet x x v

Work It Off x x t

Max’s Plate x x

Frutas y verduras para nios x

Hora de Comer x x v

Emma breakfast - KIDS x x

EduKitchen-Kids Educational x x

Veggie Bottoms Lite x x

Sopa Hacedor x

Awesome Eats x x

Cocomong Season 2 x x

HEALTHINF 2017 - 10th International Conference on Health Informatics

314

Figure 1: MATCHuP overview.

3.1 Kids App

The kids app includes several modules in order to re-

spond to the following requirements:

• Nutrition: enable the entering of the different kid

meals

• Exercise: register the activity of the kid

• Validation: check and receive information about

the data entered as so far by other kids or by her-

self

• Assessment: acquaintance of the healthy be-

haviour progress

Every time the kid eats, she should register the nu-

trition information in the system. However, this infor-

mation is not entered manually neither with a text, or

by selecting photos in a library, as many other apps

in the market, but by making a photo of the dish he

is just eating. Next, the kid has three sliding buttons

to label the photo, according to his knowledge, which

is the amount of fruits and vegetables, carbohydrates,

and meat that contains the served meal (see Figure

2). From the sliding button, a percentage on nutrition

components is derived.

Figure 2: GUI for food labelling.

Of course, the information entered by the kid

should be validated. Validation is performed in a col-

laborative way. That is, parents and other kids in the

community validate the labels assigned to the photo

(see Figure 3). Once a day, every kid receives a set of

photos from other kids in the social network (see par-

ents and community validation on Sections 3.2 and

3.4 correspondingly). The owner of the photos re-

ceived for validation is unknown. They could come

either from kids in the same team or from adversary

teams. Therefore, the kid cannot manipulate the out-

come to favouring her mates. In order to incentive

kids in this validation process, some points are given

to the kids that actively participate in this process that

contribute to win the game match of the week (see

Section 3.5).

Figure 3: GUI for food validation.

Regarding physical activity, each kid deﬁnes a

proﬁle regarding her preferences about sports and the

timetable they use to practice (supported by parents

when under 12 years old). To validate the activity,

wearables are offering a smart way of capturing it. To

that end each type of activity, and its intensity is mea-

sured according to METs (the ratio of work metabolic

rate to resting metabolic rate) (see (Ainsworth et al.,

1993) and (Ainsworth et al., 2011) for further calcu-

lation descriptions). Some activities could come with

non-scheduled hours (as for example, playing soccer

in the school playground).

Kid assessment about her progress is provided by

plots in which the differences between the current

state and the targets is shown following a colour code

(see Figure 4). The nutrition information is not taken

from the kid’ labelling, but from the outcome of the

validation process. An aggregation method is used

to combine the information of the kid (self informa-

tion and the validation data from other users), giving

a higher importance to the information coming from

parents. The ﬁnal nutrition ﬁtness is provided in a

MATCHuP: An mHealth Tool for Children and Young People Health Promotion

315

scale from A to E, being A the best value. A similar

outcome is obtained for the physical activity, obtain-

ing a second value deﬁned in the same scale. Both ﬁt-

ness values, nutrition and physical activity, are ﬁnally

aggregated, obtaining the kid current healthy state.

Figure 4: Kid assessment.

3.2 Parents App

Parents role is mainly focused on providing reliable

validation information. In that regard, parents receive

once a day a set of labelled photos that they need to

revise. As in the kid case, they are not aware about the

provenance of the photos (i.e. whenever they belong

to their kid or do not).

3.3 Healthcare Professionals Web

Service

Healthcare professionals are in charge of setting up

the nutrition and exercise targets for the kids. They

can also follow the kid progresses thanks to a visu-

alisation screen that shows the distance between the

target and the achieved results, in a colour code (see

Figure 5).

Healthcare professionals access to the platform

has been designed as a web service, instead of an mo-

bile app because this facilitates the integration of the

tool in the current Healthcare Information Systems of

our region.

3.4 Social network

The community of users is managed by means of a

social network, where collaborative and competitive

events take place.

Regarding competition, users in the network are

identiﬁed according to their sportive preferences (soc-

cer, basket, dance, etc.), and her healthy status. This

Figure 5: GUI for healthcare providers.

data enables the conﬁguration of sport teams and the

corresponding game competitions (see Figure 6). In

so doing, two conditions should be fulﬁlled:

• There should be enough teams in each sport to set

up a game (sport competition matches)

• There should be a certain satisfaction degree

among the kids preferences and the team as-

signed.

Figure 6: Physical activity teams.

On the other hand, collaboration arises in two di-

rections:

• Help team mates to achieve their healthy targets

(see Figure 8)

• Validate food photos from other users (see Figure

7 )

Regarding the validation of photos,

• 25% of the photos of a kid are validated by mem-

bers of the same team

• 50% of the photos of a kid are validated by mem-

bers of other teams of the same sport

• 25% of the photos of a kid are validated by mem-

bers of other sports

• 50% of the photos of a kid are validated by par-

ents, selected at random

Therefore, the number of validations obtained per

photo is expected to be higher than 1. The worse case

HEALTHINF 2017 - 10th International Conference on Health Informatics

316

scenario would be when no validation is achieved for

a given photo. In that situation, the photo is scored

neutral. Future work should include a monitoring

module to facilitate the assessment and control such

situations.

The validation feedback can help to understand

the kid about the real contents of the meals, and learn

about nutrition.

Figure 7: Food validated by the community.

Figure 8: Chat functionalities for kid collaboration.

3.5 Games

In the social networks, there are n games, according

to the kids sportive preferences, although it depend

on the number of users, too. For example, there could

be a soccer game, or a skate game (see Figure 6). For

each game, there are a given number m of sport teams.

Once a weak (e.g.Sunday),all teams play a match.

Ideally, the match should correspond to a market

available game integrated in the platform. For exam-

ple, there are soccer games which teams could be con-

ﬁgured by the user. However, in the ﬁrst approach of

MATCHuP, the game is a simple rank of scores: the

winner of the match is the team with the highest score

(see Figure 9). The team score (deﬁned in ℑ)is ob-

tained by the aggregation of the healthy state values

of all of its members.

Figure 9: Competition outcome.

4 FIRST VERSION

The system has been developed using the Spring Java

environment. The mobile application is deployed in

Android (Figure 10). The ﬁrst version deployed with

all of the involved agents, but physical activity en-

ter manually. Current languages are English, Spanish

and Catalan.

The design goals of MATCHuP are simplicity and

easy to use. Thus, photos favours usability and collab-

orative validation is simply. Of course, some image

processing engines could be used to obtain nutrition

components from photos, for example, but the state

of the art of such engines are still under research for

such purpose.

Next step will consider the inclusion of the exer-

cise activities by means of a smartwatch or activity

band. Other technological advances that are ready to

use and that could be incorporated in a near future

are Artiﬁcial Intelligence techniques to handle prefer-

ences to set up teams, as well as to aggregate infor-

mation.

5 CONCLUSION

Obesity is a main issue for many people, specially

children. It is becoming a big concern in European

and occidental countries. Obese persons are not con-

sidered patients (as diabetic persons are), and there-

fore, they are very difﬁcult to motivate for bring-

ing them to healthier states. This paper presents the

MATCHuP platform with the aim of helping obese

MATCHuP: An mHealth Tool for Children and Young People Health Promotion

317

Figure 10: Home page for kids.

children and young persons (under 16) to reach a

healthier state with the supervision of healthcare prac-

titioners, families, and other users in a similar situa-

tion.

Along the paper, the system has been described,

and the ﬁrst version of MATCHuP presented. Next

steps include the integration of smartwatches or sim-

ilar wearable able to automatically detect the physi-

cal activity, as well as artiﬁcial intelligence tools to

improve the aggregation methods and additional ad-

vices for further personalization and fast adaptation.

Moreover, a the evaluation of the tool in for medical

evidence is also required.

In that regard, the main challenge is to keep kids

engaged in the platform. Clinicians argue that about 6

month of using the platform could be sufﬁcient for ob-

taining some behaviour change. However, some stud-

ies have shown that having a kid engaged in a game

more than 3 months is a great success. The long trial

of the tool will provide inputs to that concern, and

work for alternative artifaxts (Hevner et al., 2004) .

A secondary, technological challenge, is the fact that

sensitive data is stored in mobiles. The recent study

(Blenner et al., 2016) highlights the necessity of con-

sider privacy implications before using health apps.

ACKNOWLEDGEMENTS

This project has received funding from the grant of

the University of Girona 2016-2018 (MPCUdG2016),

ans has been developed with the support of

the research group SITES awarded with distinc-

tion by the Generalitat de Catalunya (SGR 2014-

2016). Sreynoch Soung received ﬁnancial support

from the European Commission (Erasmus Mundus

project Techno II, ref. 372228-1-2012-1-FR-ERA

MUNDUS-EMA21).

REFERENCES

Ainsworth, B. E., Haskell, W. L., Herrmann, S. D., Meckes,

N., Bassett, D. R., Tudor-Locke, C., Greer, J. L., Vez-

ina, J., Whitt-Glover, M. C., and Leon, A. S. (2011).

2011 Compendium of Physical Activities: a second

update of codes and MET values. Medicine and sci-

ence in sports and exercise, 43(8):1575–81.

Ainsworth, B. E., Haskell, W. L., Leon, A. S., Jacobs, D. R.,

Montoye, H. J., Sallis, J. F., and Paffenbarger, R. S.

(1993). Compendium of physical activities: clas-

siﬁcation of energy costs of human physical activi-

ties. Medicine and science in sports and exercise,

25(1):71–80.

Aitker, M. (2015). Patient Adoption of mHealth. Use, Ev-

idence and Remainding Barriers to Mainstream Ac-

ceptance. Technical Report September.

Blenner, S. R., K

ollmer, M., Rouse, A. J., Daneshvar,

N., Williams, C., Andrews, L. B., J, W., and AM,

B. (2016). Privacy Policies of Android Diabetes

Apps and Sharing of Health Information. JAMA,

315(10):1051.

Herrero, P., Lopez, B., and Martin, C. (2016). Pepper:

Patient empowerment through predictive personalised

decision support. In ECAI Workshop on Artiﬁcial In-

telligence for Diabetes, pages 8–9.

Hevner, A. R., March, S. T., Park, J., and Ram, S. (2004).

Design science in information systems research. MIS

Quarterly, 28(1):75–105.

opez, B., Coll, J., Gamero, F., BargallL

o, E., and L

opez-

Bermejo, A. (2013). Intelligent systems for support-

ing premature babies healthcare with mobile devices.

In Mobilemed, page 4 pages.

Sanders, L. M., Shaw, J. S., Guez, G., Baur, C., and Rudd,

R. (2009). Health literacy and child health promotion:

implications for research, clinical care, and public pol-

icy. Pediatrics, (124 Suplement 3):306–14.

Tate, E. B., Spruijt-Metz, D., O’Reilly, G., Jordan-Marsh,

M., Gotsis, M., Pentz, M. A., and Dunton, G. F.

(2013). mHealth approaches to child obesity preven-

tion: successes, unique challenges, and next direc-

tions. Translational behavioral medicine, 3(4):406–

15.

HEALTHINF 2017 - 10th International Conference on Health Informatics

318