Towards Individualised Persuasive Technology for Obesity

Prevention in Teenagers

Claudio L. Lafortuna

, Maurizio Caon

, Sarah A. Tabozzi

, Stefano Carrino

, Neil S. Coulson

José C. E. Serrano

, Marco Sacco

, Omar Abou Khaled

, Giovanna Rizzo

and Elena Mugellini

Istituto di Bioimmagini e Fisiologia Molecolare, Consiglio Nazionale delle Ricerche, via Cervi 93, Segrate Milano, Italy

HumanTech, Haute Ecole Spécialisée de Suisse Occidentale, Bd de Perolles 80, Fribourg, Switzerland

Division of Rehabilitation and Ageing, University of Nottingham, Queen's Medical Centre, Nottingham NG7 2UH, U.K.

NUTREN-Nutrigenomics, Dept Medicinal Experimental, Universidad de Lleida, Montserrat Roig, 2 – 25008 Lleida, Spain

Istituto di Tecnologie Industriali ed Automazione, Consiglio Nazionale delle Ricerche, Via Bassini 15, Milano, Italy

Keywords: Virtual Individual Model, Persuasive Technology, Obesity Prevention.

Abstract: Obesity is a major clinical problem for individuals and health care systems worldwide, alarmingly fuelled

by body mass excess in the juvenile age. In spite of its multi-factorial origin, unhealthy lifestyles relative to

alimentary behaviours and physical activity habits play a major causative role. Thus, an important

preventive action of this condition can be conducted by fostering motivation of young people towards

healthy lifestyles through engagement and inclusion. ICT technologies offer a powerful tool to address

effectively this serious medical and societal issue by the development of persuasive strategies based on an

accurate modelling of individual's characteristics. PEGASO is a technological multidisciplinary project

aimed at promoting healthy lifestyles among teenagers, through assistive technology enhancing motivation

to healthy lifestyles, empowered by a virtual individual model (VIM) for user characterisation. The VIM

intended for the PEGASO project, including functional, physical and psychosocial aspects profiling young

individuals' health status and behaviours relevant in alimentary and physical activity domain, will enable the

development of an individualised assistive technology expected to leverage motivation to healthy lifestyles

through implicit and explicit interaction.

1 INTRODUCTION

Obesity is a major public health challenge at all ages

in developed countries. According to the 2007 report

of EU Public Health Programme Project "Global

Report on the Status of Health in the European

Union - EUGLOREH" (EUGLOREH, 2007), across

the entire EU, overweight affects almost 1 out of 4

school age children/adolescents, in particular. This

number is likely to increase by more than 400,000

children a year.

Moreover, juvenile obesity is associated with a

number of serious medical conditions and can lead

to increased rates of non-communicable disease in

adulthood, such as cerebro-vascular disease,

diabetes, certain types of cancer, osteoarthritis, gall

bladder, endocrine disorders and premature death, in

relation with the high probability for obese children

to become obese adults; it is also recognized that

being overweight/obesity in young peole carries

within it a range of psycho-social consequences

including low self-esteem, depression and social

exclusion, all resulting in sizable economic impact

on health care and social systems (Speiser et al.,

2005) (Dent, 2010) (Trasande and Chatterjee, 2009).

From the literature it emerges that although also

for the juvenile age the cause of body mass excess is

multifactorial, including for a small portion genetic

background and neuroendocrine status, in the

greatest majority behavioural aspects related to

lifestyle and diet play an important causative role,

with a relevant interference of socioeconomic

factors (Speiser et al., 2005) (World Health

Organization, 2000) (Commission of European

Community, 2005). Therefore, in order to undertake

an effective action of prevention of the condition of

591

Lafortuna C., Caon M., Tabozzi S., Carrino S., Coulson N., Serrano J., Sacco M., Abou Khaled O., Rizzo G. and Mugellini E..

Towards Individualised Persuasive Technology for Obesity Prevention in Teenagers.

DOI: 10.5220/0004938805910598

In Proceedings of the International Conference on Health Informatics (SUPERHEAL-2014), pages 591-598

ISBN: 978-989-758-010-9

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

body mass excess, it is necessary to intervene on

adolescents’ behaviour, through education and

engagement, to enhance motivation towards healthy

diet and active lifestyle.

Thanks to the large diffusion and development of

ICT, time is ripe for a major employment of

persuasive technology to cope with healthcare

challenges. PEGASO is a technological project

aimed at promoting healthy lifestyles among

teenagers, through assistive technology fostering

motivation, enhanced by a virtual model of

individual's characteristics related to health

(PEGASO, 2013). The adoption of a Virtual

Individual Model (VIM) including functional,

physical and psychosocial aspects characterizing

individual’s health status and relevant behaviours,

will lead to a more individualised strategy for the

enhancement of motivation to engage in healthy

lifestyles, particularly with regards physical exercise

and dietary behaviour.

2 STATE OF THE ART

The deeper understanding recently achieved about

the biology of human being, introduces the idea of

the individual as a unique multiple organ system,

overtaking the traditional approach –in force in

medical practice- of the human body as a set of

independent sections Indeed most of the major

diseases affecting world population, one for all

obesity, are recognized to have multi-factorial

causation, spanning from physical to mental and

social factors (World Health Organization, 2000).

Thereof derives that in order to efficiently act on

complex health conditions it is necessary to analyse

the interaction of all the factors involved in that

phenomenon, with in mind that the more

information is provided, the more personalised and

exact the intended actions will be.

These premises led to the outline of the Virtual

Physiological Human (VPH), a methodological and

technological framework for integrated modelling of

a living human body (Fenner et al., 2008). Funded

by European Commission, the initiative hosted the

development of several projects in the recent years,

focused on the modelling of different human body

functions incorporating cross-disciplinary

knowledge from biochemistry, biophysics and

anatomy of cells, tissues and organs.

The models allowed by this framework collect

the results of multiple observations on organism’s

functionalities, disseminate them among experts

from multiple scientific disciplines to build a

collaborative analysis and develop systemic

hypotheses, and finally interconnect integrated data

in models that consolidate the original hypotheses.

The models so far realised in the context of VPH

concern different functional specialisms, including

projects in the cardiovascular, respiratory,

neurological, immunological and oncological

domains, which exploit the current knowledge about

physiological and pathological mechanisms finalised

to medical practice and/or tutorial simulation.

The many projects stemming from the

framework of VPH are substantially aimed at

describing the interaction of all the physiological

components of individuals - referred to as Physiome

-, from molecular to apparatus level. This systemic

approach inherent in VPH modelling conforms to

the holistic approach in the study of body function,

supplying the view of the body as a single multi

organ system.

However, such a holistic architecture of the

many projects, modelling physiology and pathology

of the different body functions to converge into the

Physiome solution, substantially describes the

human being from the perspective of biological

relations without accounting for the behavioural and

social externalities, which may interfere with and

determine the biological balance of functions in

health and disease.

By contrast, the PEGASO VIM aims to include

in individual's characterisation, both biological

specifications and relevantly related behaviour

factors spanning from the physical domain of body

structure to physiological description of functional

interaction, and psychosocial determinants of health

specifically involved in alimentary and physical

activity behaviours in young people.

VIM could be used as a robust support for

tailoring the computer-based multi-level persuasive

interventions. From this point of view in computer

science, research communities in the persuasive

technology domain have taken the first steps towards

exploiting mobile and wearable technologies, which

can gather data about the user’s activity and

behaviour in a real-time and long-term fashion, for

the purposes of developing tailored systems in order

to deal with obesity as a significant health-related

issue (Arteaga et al. 2009). In addition, based on

their wearability, these devices can be used for

active encouragement towards a healthier lifestyle

(Valentin and Howard, 2013). Several studies

demonstrate that a crucial aspect in the realization of

such a system is to design appealing applications for

the final users (Read et al., 2011), in this case young

people.

HEALTHINF2014-InternationalConferenceonHealthInformatics

592

Figure 1 - The PEGASO virtual individual model assumes

that health status is primarily settled on elements of

physical status, physiological status and psychological

status. Body structure and functionality are influenced by

the individual's behaviours in the domains of alimentation

and physical activity (PA), which are driven by relevant

aspects of motivation. Social status, social behaviour and

psychological status (i.e. the psychosocial factors) are

considered as important determinants of motivation to

engage in healthy lifestyle behaviours. Stippled arrows

denote the presumptive relations among the model's

element which will be defined in the project.

Coupling of multi-domain VIM and persuasive

technology should represent a successful approach

to the user, for adolescents’ engagement in

behaviours preventing overweight and obesity (such

as healthy food and active life), starting from the

quantitative description of relevant individual’s

aspects.

3 VIRTUAL INDIVIDUAL

MODEL

The PEGASO VIM depicted in Figure 1 points to

integrating biological aspects of human functioning

with lifestyle behaviours and psychosocial

externalities that are relevant for the

development of overweight and obesity

conditions, especially in young people. PEGASO

VIM is based on the concept of the human body as a

single complex system, which already empowers the

several projects referring to VPH, funded by the

European Commission in the recent years (Fenner et

al., 2008).

As shown in Figure 1, PEGASO VIM considers

an individual's health as resulting from the balance

between physical, mental and social well-being,

according to the World Health Organisation

founding definition (World Health Organization,

1948), so that Health Status in the model is the

product of Physical, Functional and Psychological

Status. The elements characterising Physical Status

will be identified among the indicators of body

adiposity and risk factors for the development of the

disease conditions related to overweight and obesity,

as interpreted on the basis of standard reference

values (Dulloo et al., 2010). Similarly, parameters

relevant for the profiling of Functional Status will be

identified from evidence based literature accounting

for the metabolic derangements, which derive from

body mass excess (Zhu et al., 2003). Factors

determining the individual's exercise capacity which

are influenced by body mass excess and associated

conditions will be accounted as well (Lafortuna,

2013).

According to collective views of international

groups of experts (World Health Organization,

2000) (Commission of European Community, 2005),

the role of lifestyles as determinants of conditions

such as overweight and obesity has been thoroughly

evidenced, with particular focus to juvenile age.

Studies using motion sensors have shown that

children who spend less time in physical activity are

at higher risk of becoming obese during childhood

and adolescence. Television and video games have

contributed to more sedentary leisure activities and

are associated with the consumption of energy-dense

snacks and beverages, as well as encouraging

inappropriate food choices which are attributable to

television advertising. There is, in fact, a positive

correlation between hours of television viewing and

being overweight, especially in older children and

adolescents, regardless of their levels of physical

activity (Rey-Lopez et al., 2011). In fact, the

findings from the 2009/2010 survey in EU countries

from Health Behaviour in School-aged Children

(HBSC) international report indicate that young

people who are overweight/obese are more likely to

exhibit unhealthy alimentary patterns, are less

physically active and watch television more, an

increased prevalence of being overweight/obesity is

also significantly associated with low family

affluence (Currie et al., 2012). Therefore, the

PEAGASO VIM will identify parameters permitting

the characterisation of Alimentary and Physical

Activity Behaviours expected to have a direct

influence on both Physical and Functional Status.

A central issue in the development of VIM will

concern the specification of the psychosocial factors

TowardsIndividualisedPersuasiveTechnologyforObesityPreventioninTeenagers

593

Figure 2 - Automatic system architecture.

impacting upon young people’s motivation to

engage in healthy alimentary behaviour and active

lifestyles, which are a major determinant of the

Health Status. These factors have been shown to be

significantly associated with biomedical factors in

the genesis of body mass excess especially in

children and adolescents (Currie et al., 2012)

(Mikolajczyk and Richter, 2008) and obesity carries

with it a number of important psychosocial

consequences (Dent, 2010) (Trasande and

Chatterjee, 2009) (Gortmaker et al., 1993). For

example, body mass excess in children/adolescents

is associated with low self-esteem, depression and

social exclusion, with immediate consequences in

the psychological and social domain, possibly

leading also to concurrent or subsequent psychiatric

pathology. Specifically, social exclusion in

childhood has been associated with reduced

psychological functioning in adulthood, with an

expectation of lower educational attainment, less

money earning, experiencing higher rates of poverty

and having a lower likelihood of marriage

(Gortmaker et al., 1993). Thus, profiling of Social

Status and Social Behaviour along with

Psychological Status will be an important task in

VIM building during PEGASO project, in view of

the critical role played by these elements for the

characterisation of motivation to healthy behaviours.

Addressing the psychosocial determinants and

influences on these behaviours from a range of

domains (including cognitive, interpersonal,

familial, environmental) will in fact provide the

conceptual background for the development of the

assistive technology expected to foster lifestyles

preventing body mass excess in younger people.

4 AUTOMATIC SYSTEM BASED

ON PERSUASIVE

TECHNOLOGY

The multi-parametric PEGASO VIM is integrated in

an automatic system that helps experts in monitoring

users’ behaviour, and in tailoring system

functionalities. Smartphones and wearable devices

are responsible for interacting with the teenagers.

4.1 Architecture

The main components and their mutual relations

forming the automatic system architecture are

presented in Figure 2. The architecture is structured

in three levels: model, content and presentation.

1. Model level: is the base of the whole system and

contains the contribution brought by VIM,

translated into ontological form. The information

about the user, collected through the explicit and

implicit interaction of the individual with the

system, is entered into the generic VIM, and

processed to structure the Personalised Virtual

Individual Model (PVIM).

2. Content level: couples the user information

contained in PVIM with the system’s content, for

HEALTHINF2014-InternationalConferenceonHealthInformatics

594

shaping the intervention. It is also the means by

which the PVIM is updated. It contains four

blocks

 the alarm generator block provides instant

suggestions for short-term actions (e.g., "walk

until the next bus station while going to

school");

 the behaviour enhancer block aims at

encouraging long term changes in the user

behaviour, such as practicing physical activity

more regularly;

 the motivator block processes PVIM

information to adapt the task presented to the

user in the context of a serious game;

 finally, the VIM tailoring block analyses the

feedback resulting from implicit and explicit

interaction of the user with the system to

continuously adapt the PVIM (see section 5 for

a detailed explanation).

3. Presentation level: represents the system

interface to the user and manages dialogue and

interaction tasks.

One of the main advantages offered by the

inclusion of VIM in this architecture is the

possibility to personalize the model based on the

individual behaviour and behavioural changes: on

one hand, based on the VIM as fundamental element

of its architecture, the system provides different

types of information to the user: alarms, games, long

terms actions, etc.; on the other hand, information

acquired from the user is continuously used to

update the PVIM.

4.2 Tailoring

Tailoring is the process by which the automatic

system dynamically tracks the changes in user

parameters and, through the use of reinforcement

learning algorithms, allows the detection of users’

preferences concerning the favourite and most

effective intervention strategies.

Indeed, VIM tailoring has, as direct outcome, the

modification in the execution of actions planned in

the content level (alarm generator, behaviour

enhancer, and motivator blocks).

The tailoring is taken into account along

different axes and with different approaches:

 Personal axis

 Sociocultural axis

 Temporal axis

The personal axis uses the three aspects of the

VIM (functional, physical and psychosocial) to

understand and adapt the therein knowledge to the

user. For example, a PVIM describing a user

performing physical activity in an irregular fashion

can encourage the user to regularize his/her activity;

in contrast, users scarcely inclined to physical

exercise might be compensatively encouraged to

healthy eating. The previous simple scenarios show

how the PVIM can affect the type of feedback

provided to the user.

The European dimension of our study will allow

us to take into account also the different ethnological

specificities impacting on adolescents’ lifestyle. This

will open the possibility to define the socio-cultural

axis of our tailoring approach. During the PEGASO

project, three pilots in different countries will take

place (Italy, Spain and United Kingdom). These

pilots will allow us to examine the cultural

differences that may impact on teenagers' lifestyle

and to adapt the VIM accordingly. Unlike the

personal axis, the socio-cultural tailoring is a static

process.

Finally, the temporal axis will take into account

the evolution of the user behaviour over time. The

goal of this analysis is twofold. Firstly, it will be

possible to associate particular events to the teenager

lifestyle. For example, we can associate eating habits

and stressful events (such as those related with

demanding school duties) or to a specific period of

the year (summer or winter vacations). Combining

this information with the personal axis will also

allow us to explore how the social life of the user

influences such habits. For example, the system

should be able to understand if the user engages in

healthier dietary practices when eating alone or with

peers. Secondly, the temporal axes will allow

analysing the response of the user to the system

clues.

The temporal and the personal axes will evolve

with the system and the continuous interaction with

the user. The adaptation of the PVIM is, therefore,

performed dynamically. Machine learning

approaches will use the dynamical information about

the user to adapt such model. Such valued

information is provided by mobile and wearable

technologies as presented in the next paragraph.

4.3 Mobile and Wearable Technologies

Nowadays, smartphones and other wearable devices

have promising sensing and processing capabilities,

also based on the integration of different sensors

with mobile functionalities.

The wearability of sensor devices permits us to

gather data about the user’s activity and behaviour in

a real-time, long-term fashion (Martín, et al., 2013).

TowardsIndividualisedPersuasiveTechnologyforObesityPreventioninTeenagers

595

If compared with classical approaches, based on

sporadic clinical testing, our approach has well

defined advantages. In particular, while real-time

information can be exploited to conceive instant

feedback to the user, long-term data can be

integrated with conventional information more

sporadically acquired in a clinical setting thus

leading to a more precise estimation of the overall

trends concerning physical and functional

conditions, as well as about changes in behaviour

liable to influence the health status.

In addition, due to the intrinsic possibility to be

always-on and always-connected, smartphones have

the potentiality to become the most appropriate

technological companion (Siewiorek, 2012).

Wearable and mobile technologies allow also

developing pervasive applications, such as pervasive

games, which “are no longer confined to the virtual

domain of the computer, but integrate the physical

and social aspects of the real world” (Magerkurth et

al. 2005). That means that users can interact and

play while moving freely in the real world, thus

fostering physical activity. This concept created a

rift with the traditional concept of playing games

that usually confined the players on their chairs in

their rooms.

Moreover, such devices can also be used to

anonymously gather information about the everyday

life context of the users (such as social habits, meal

frequency, exercise volume and modality). By

gathering these apparently independent data, it is

also possible to obtain information about the

relationship between different parameters of the

VIM model, (e.g., to evaluate if and how the social

context interferes with the user’s alimentary habits

or physical activity behaviours)

However, sensors are not the only source of

information in the PEGASO system. In particular, a

direct interaction of the user with the system will

gain further information to better refine PVIM as

explained in the next section.

5 INTERACTION

In the traditional approaches of human-computer

interaction research, user modelling was a well-

known approach for the design of a usable interface

(Fischer, 2001). In PEGASO, the automatic system

will integrate a VIM that does not aim at only

improving the usability but that will enable the

system to choose the best motivational mechanisms

for an effective intervention on the users’ life style.

The system aims at providing personalised

interventions that take into account the user’s

individual characteristics in order to obtain the

greatest effectiveness. The VIM characterizes the

user’s nutritional habits, physiological status, and

socio-psychological status to provide personalised

motivational mechanisms to help the adoption of a

healthy life-style. Obviously, the interaction between

the system and the user plays a crucial role in the

tailoring process described in the previous section.

In fact, the reinforcement learning algorithm needs

the user’s feedback in order to provide the best

personalised motivational mechanisms. Two kinds

of user’s feedback exist: implicit and explicit.

5.1 Implicit Interaction

Albrecht Schmidt formalised the concept of implicit

interaction and provided its definition: the implicit

human-computer interaction is “an action performed

by the user that is not primarily aimed to interact

with a computerised system but which such a system

understands as input” (Schmidt, 2000). The mobile

and wearable technologies can sense the user’s

activity in an unobtrusive way. The sensors that will

be integrated in mobile devices and in clothes will

collect information while the user is wearing them.

The user can be monitored during his/her everyday

activity and the personal parameters present in the

VIM can be dynamically updated in order to adapt

the system also when the user’s habits change. The

sensed data are used to interpret the user’s activities

and personal behavioural trends, especially about the

nutritional habits and physical exercise.

The implicit interaction is based on the activity

recognition and context-awareness (Lukowicz et al.,

2010). In fact, not only the activity is important but

also the context where it is performed acquires a

very important value for choice of the opportune

motivational mechanism. Knowing that a user is not

doing his/her regular physical exercise in the fixed

time should trigger a memorandum message on the

user’s mobile; but if the GPS localizes the user in

another city or country, the system can interpret this

as the user being on holidays and can avoid the

message since it could be perceived as annoying.

Moreover, the behavioural trends recognition allows

the system to recognize whether the selected

mechanisms actually influenced the user’s

behaviour.

5.2 Explicit Interaction

Users perceive smartphones as personal life

companions (Siewiorek, 2012). For this reason, the

HEALTHINF2014-InternationalConferenceonHealthInformatics

596

PEGASO system will focus the explicit interaction

through the smartphone. The explicit interaction is

the conscious command and information that the

user gives to the system. This is important for many

reasons. For example, when the system provides a

message to the user, he/she has also the possibility to

communicate directly with his/her companion; in

particular, the user can show to the companion

whether the message and its content have been

appreciated or not. The explicit interaction can be

performed through gestures. One gesture expresses

the appreciation of the feedback (like caressing the

smartphone), another one the dislike (like tapping it

twice). In this way the system can learn how to

provide the best personal message possible thanks to

the context information and the monitored activity.

In this way, the system can adapt not only the

content of the message, but it can choose the best

moment and the best modality thanks to a context

aware reasoning engine.

The feedback provided by the user is important

for the modelling of tailored interventions and to

make the information personally relevant. In fact,

research has demonstrated that computer-tailored

health education is more effective in motivating

people to make dietary changes (Brug et al., 2003)

and to promote physical exercise (den Akker et al.,

2011). The motivational mechanisms will be of

different types. For example, there will be

interventions for short-term efficacy as alarm

messages proposed by the companion when the user

is eating something unhealthy or forget to take

physical exercise. Other long-term mechanisms aim

at changing the user behaviour through serious

gaming and the implication of the social community,

e.g., friends and family (Johnston et al. 2011).

6 CONCLUSIONS

In this paper, we presented the concept of PEGASO

Virtual Individual Model designed for individualised

persuasive technology for obesity prevention in

teenagers.

The adoption of a Virtual Individual Model

(VIM) including functional, physical and

psychosocial aspects allows the development of a

more individualised strategy for the enhancement of

healthy lifestyles through increasing motivation. The

PEGASO VIM considers an individual's health as

resulting from the balance among physical, mental

and social well-being.

For the twofold goal of encouraging and

monitoring the teenager activity, we chose to adopt

mobile and wearable technologies. PEGASO aims at

increasing the encouragement of healthier

behaviours through the use of a tailored ensemble of

alerts, serious games applications developed

following a gamification approach.

The presented automatic system will interact

with the user providing implicit and explicit

interactions. The first approach, which is activity-

driven, does not require an active participation

facilitating the gathering of significant quantity of

data over a long period of time since it does not

annoy the user. The second approach takes into

account the direct, conscious interaction between the

user and the system. On the one hand, explicit

interaction allows retrieving information that is not

possible to interpret through the mere sensing; on

the other hand, it allows a direct connection with the

user that, thanks to a tailored interaction, can

establish an emotional relationship with the

companion.

ACKNOWLEDGEMENTS

The PEGASO project is co-funded by the European

Commission under the 7th Framework Programme.

The project is compliant with European and National

legislation regarding the user safety and privacy, as

granted by the PEGASO Ethical Advisory Board.

The authors of the paper wish to thank all the

project partners for their contribution to the work.

REFERENCES

Arteaga, S., M., Kudeki M, and Woodworth, A.. 2009.

Combating obesity trends in teenagers through

persuasive mobile technology. SIGACCESS Access.

Comput. 94 (June 2009), 17-25.

Brug, J., Oenema, A., and Campbell, M., 2003. Past,

present, and future of computer-tailored nutrition

education.. Am. J. Clin. Nutr., vol. 77, no. 4 Suppl, p.

1028S–1034S.

Commission of European Community, 2005. Green Paper

- Promoting healthy diets and physical activity: a

European dimension for the prevention of overweight,

obesity and chronic diseases. COM(2005) 637,

Brussels, December 2005.

Currie, C., Zanotti, C., Morgan, A., Currie, D., de Looze,

M., Roberts, C., Samdal, O., Smith, O.R.F.,

Barnekow, V., 2012. Social determinants of health and

well-being among young people. Health Behaviour in

School-aged Children (HBSC) study: international

report from the 2009/2010 survey. Copenhagen, WHO

Regional Office for Europe, 2012 (Health Policy for

Children and Adolescents, No. 6).

TowardsIndividualisedPersuasiveTechnologyforObesityPreventioninTeenagers

597

den Akker, H.O., Moualed, L.S., Jones, V.M., and

Hermens, H.J., 2011. A self- learning personalized

feedback agent for motivating physical activity,” In

Proc. 4th Int. Symp. Appl. Sci. Biomed. Commun.

Technol. - ISABEL ’11, pp. 1– 5.

Dent, M., 2010. The economic burden of obesity. Oxford:

National Obesity Observatory.

Dulloo, A.G., Jacquet, J., Solinas, G., Montani, J.P.,

Schutz, Y., 2010. Body composition phenotypes in

pathways to obesity and the metabolic syndrome. Int J

Obes (Lond). 2010; 34 Suppl 2: S4-17.

EUGLOREH 2007 project The Status of Health in the

European Union: Towards a Healthier Europe,

available at: http://euglorehcd.eulogos.it.

Fenner, J.W., Brook, B., Clapworthy, G., Convey, P.V.,

Feipel, V., Gregersen, H., Hose, D.R., Khol, P.,

Lawford, P., McCormack, K.M., Pinney, D., Thomas,

S.R., Van Sint Jan, S., Waters, S., Viceconti, M.,

2008. The EuroPhysiome, STEP and a roadmap for

the virtual physiological human. Phil Trans R Soc A.

2008; 366: 2979–2999.

Fischer, G., 2001. User modeling in human–computer

interaction. User modeling and user-adapted

interaction 11, no. 1-2: 65-86.

Gortmaker, S.L., Must, A., Perrin, J.M., Sobol, A.M.,

Dietz, W.H., 1993. Social and economic consequences

of overweight in adolescence and young adulthood. N

Engl J Med. 1993; 329: 1008-1012.

Johnston, H., Whitehead, A., Pose presentation for a

dance-based massively multiplayer online exergame.

In: Entertainment Computing, Volume 2, Issue 2,

2011, Pages 89-96, ISSN 1875-9521.

Lafortuna, C.L., 2013. Physiological Bases of Physical

Limitations During Exercise. In: Disabling Obesity -

From Determinants to Health Care Models,

Capodaglio P, Faintuch J, Liuzzi A, eds., Springer-

Verlag, Heidelberg, 2013, pp. 21-38.

Lukowicz, P., Amft, O., Roggen, D., and Cheng, J., 2010.

On-Body Sensing: From Gesture-Based Input to

Activity-Driven Interaction. Computer, vol. 43, no. 10,

pp. 92–96.

Magerkurth, C., Cheok, A. D., Mandryk, R. L. and Nilsen,

T., 2005. Pervasive games. Computers in

Entertainment, vol. 3, no. 3, p. 4.

Martín, H., Bernardos, A.M., Iglesias, J., Casar, J.R. 2013.

Activity logging using lightweight classification

techniques in mobile devices. Personal Ubiquitous

Comput. 17, 4 (April 2013), 675-695.

Mikolajczyk, R.T., Richter, M., 2008. Associations of

behavioural, psychosocial and socioeconomic factors

with over- and underweight among German

adolescents. Int J Public Health. 2008; 53: 214-220.

PEGASO - PErsonalised GuidAnce Services for

Optimising lifestyle management in teenagers through

awareness, motivation and engagement, FP7-ICT-

2013-10, Grant Agreement 610727.

Read, J., Fitton, D., Cowan, B., Beale, R., Guo, Y., and

Horton, M. 2011. Understanding and designing cool

technologies for teenagers. In CHI '11 Extended

Abstracts on Human Factors in Computing Systems

(CHI EA '11). ACM, New York, NY, USA, 1567-

1572.

Rey-López, J.P., Vicente-Rodríguez, G., Répásy, J.,

Mesana, M.I., Ruiz, J.R., Ortega, F.B., Kafatos, A.,

Huybrechts, I., Cuenca-García, M., León, J.F.,

González-Gross, M., Sjöström, M., de Bourdeaudhuij,

I., Moreno, L.A., 2011. Food and drink intake during

television viewing in adolescents: the Healthy

Lifestyle in Europe by Nutrition in Adolescence

(HELENA) study. Public Health Nutr. 2011; 14:

1563-1569.

Schmidt, A., 2000. Implicit human computer interaction

through context. Pers. Technol., vol. 4, no. 2–3, pp.

191–199.

Siewiorek, D., 2012. Generation smartphone. IEEE

Spectr., vol. 49, no. 9, pp. 54–58.

Speiser, P.W., Rudolf, M.C.J., Anhalt, H., Camacho-

Hubner, C., Chiarelli, F., Eliakim, A., et al., 2005.

Consensus statement: childhood obesity. J Clin

Endocrinol Metab; 90:1871-87.

Trasande, L., Chatterjee, S., 2009. The impact of obesity

on health service utilization and costs in childhood.

Obesity. (Silver Spring). 2009; 17: 1749-1754.

Valentin, G.; Howard, A.M., 2013, Dealing with

childhood obesity: Passive versus active activity

monitoring approaches for engaging individuals in

exercise, Biosignals and Biorobotics Conference

(BRC), 2013 ISSNIP , vol., no., pp.1,5, 18-20 Feb.

2013.

World Health Organization, 1948. Preamble to the

Constitution of the World Health Organization as

adopted by the International Health Conference, New

York, 19-22 June, 1946; signed on 22 July 1946 by the

representatives of 61 States (Official Records of the

World Health Organization, no. 2, p. 100) and entered

into force on 7 April 1948.

World Health Organization, 2000. Obesity: Preventing and

Managing the Global Epidemic. Report of a WHO

Consultation. WHO Technical Report Series no. 894.

WHO: Geneva, 2000.

Zhu, S., Wang, Z., Shen, W., Heymsfield, S.B., Heshka,

S., 2003. Percentage body fat ranges associated with

metabolic syndrome risk: results based on the third

National Health and Nutrition Examination Survey

(1988-1994). Am J Clin Nutr. 2003; 78: 228-235.

HEALTHINF2014-InternationalConferenceonHealthInformatics

598