You are on page 1of 6

2008 First International Workshop on Ontologies in Interactive Systems

An ontology for intelligent e-therapy for obesity


Irene Zaragoz
Human Centred
Technology Laboratory,
Universidad Politcnica de
Valencia
irzaal@upvnet.upv.es

Jaime Guixeres
Human Centred
Technology Laboratory,
Universidad Politcnica de
Valencia
jaiguipr@doctor.upv.es

Abstract

In the last few years the use of medical and


biomedical ontologies has increased considerably. It is
very common to find applications and semantic webs
using this kind of ontologies and a large number of
papers have been written explaining why ontologies
are useful in these fields. In this paper the use of
ontologies for psychology and its benefits is discussed
and a first ontology for obesity treatment is presented.

An ontology defines (specifies) the concepts,


relationships, and other distinctions that are
relevant for modelling a domain.

The specification takes the form of the definitions


of representational vocabulary (classes, relations,
and so forth), which provide meanings for the
vocabulary and formal constraints on its coherent
use. [7]
Although the objective of this paper is not to study
in depth the ontology concept it is very interesting to
emphasize some advantages of the design of
ontologies:
Share common understanding of the structure of
information among people or software agents
To enable reuse of domain knowledge
To make domain assumptions explicit
To separate domain knowledge from the
operational knowledge
To analyze domain knowledge
The advantages above mentioned are the reasons
why ontologies are been used in wide knowledge areas
such as medicine, bio-informatics, bio-medicine, etc.
An example of how useful is the design of
ontologies is the Gene Ontology project [8]. This
project arises to avoid biologists waste of time and
effort in searching for all of the available information
about each small area of research. This is hampered
further by the wide variations in terminology that may
be common usage at any given time, which inhibit
effective searching by both computers and people.

1. Introduction
Ontologies have become the knowledge
representation medium chosen in recent years for a
range of science areas including medicine [1], biomedicine [2], bio-informatics [3], semantic web [4],
agents [5], etc. However until the moment the design
of ontologies for psychology is not very common. In
this paper the ontological needs of psychology are
investigated and an ontology design for obesity
treatment is presented.
The term ontology was first defined by T. R.
Gruber in 1992 as a formal specification of a
conceptualization [6] which is the objects, concepts,
and other entities that are presumed to exist in some
area of interest and the relationships that hold among
them."
While the terms specification and
conceptualization have caused much discussion, the
essential points of this definition of ontology are:

978-0-7695-3542-5/08 $25.00 2008 IEEE


DOI 10.1109/ONTORACT.2008.10

Mariano Alcaiz
Human Centred
Technology Laboratory,
Universidad Politcnica de
Valencia
malcaniz@degi.upv.es

27

In the Knowledge base it is possible to have more


than one ontology, since ontologies allows inheritance.
Sometimes is useful to have small ontologies (an
ontology for a very specific topic) and to use
inheritance and composition to design more complex
ontologies instead of tackle directly the design of
complex ontologies.

As said before until the moment the design of


ontologies in not a habitual practice in psychology. All
the benefits previously mentioned, such as reusing and
sharing knowledge that made ontologies useful for
disciplines as medicine, biology, bio-informatics etc.
can also be applied in a similar way to psychology. It
could be very interesting for psychological treatments
to use a common vocabulary and to share what other
people is doing in different parts of the world.

Intelligents
fabrics

Mobile
devices

Wirelees
Communications

UBIQUITOUS COMPUTING
Nanotechnology

Body Area network


Ultra-Low power

H
e
a
l
t
h

2. Intelligent e-therapy
P
a
t
i
e
n
t

All Spain and Europe are demanding new ways for


treating chronic disorders in a potential population of
millions of people. Our group intends to develop the
knowledge and technology to get new Ambient
Intelligence Scenarios for the next decade: E-Therapy
concept.
This new intelligent e-therapy aims as an evolution
of the Computer Aided Psychotherapy (CAP) [9], [10],
[11] to solve the need of a continuous and ubiquitous
customization for each patient. The e-therapy must to
guarantee an infrastructure that:
Therapy must adapt to patients lifestyle
It must offer patient-therapy interaction 24/7,
every moment, every place
It must be global. (Accessible for everyone)
It must not only treat, it must teach the patient
Intelligent e-therapy (e-IT) can be applied to many
different disorders:
Mental health : depression, anxiety, PTSD,, ED
Enhance adherence for non-mental health: cancer,
obesity,
stroke
rehabilitation,
cardiac
rehabilitation.
Solutions to social problems: elderly, child
violence, immigrants insertion
Figure 1 shows the functional diagram of the e-IT
concept.
As shown in the figure 1 there is a Knowledge base
block that contains all the information related to the
therapy. For this knowledge base due to the benefits
that ontologies provide, previously mentioned, we have
adopted an ontology schema that:
Specifies a set of constraints, which declare what
should necessarily hold in any possible world
(patient variables)
Any possible world (patient) should conform to
the constraints expressed by the ontology

I
m
p
l
i
o
c
a
t
i
o
n

Biometric

Detect Patient Aims

Voice
Recognition

NATURAL INTERACTION

Contextual
Information

Affectice
Computing

Personalized
PsicoEducation
Medical
Reporitory

Adaptive
Treatment
EHR

Artificial Intelligence

KNOWLEDGE BASE

Semantic Web
Ontologies

Virtual
Reality

Standards
Interactive
Tools

CLINICAL DIAGNOSE & TREATMENT


Real-time
Data

Tele-therapy

p
r
o
f
e
s
s
i
o
n
a
l
s
d
e
s
i
g
n

Real
World

Figure 1: e-IT functional diagram

A first application of intelligent e-therapy for


obesity is being developed. For this application a
knowledge base is required, as shown in Figure 1, for
this reason an obesity ontology has been designed.

3. Obesity Ontology
The Obesity is a problem that is increasing in an
alarming way among the population of all the ages, so
much in children population [12], [13] , as in the adult
age [14], up to the point of being considered as a new
epidemic. In fact, obesity is the most frequent
nutritional disorder in developed countries, reason why
the WHO (World Health Organization) declared it in
1998 [15] as a global epidemic, since it affects at least
300 million people who can be diagnosed like obese,
number that rises up to a billion of people in the world
if we speak about overweight [16]. For this reason we
have chosen obesity as an appropriate disorder to be
treated with intelligent e-therapy.
The language used to design the ontology is OWL
(Web Ontology Language). OWL is the most recent
development in standard ontology languages, endorsed
by the World Wide Web Consortium (W3C) to
promote the Semantic Web vision.
"An OWL

28

ontology may include descriptions of classes,


properties and their instances. Given such an ontology,
the OWL formal semantics specifies how to derive its
logical consequences, i.e. facts not literally present in
the ontology, but entailed by the semantics. These
entailments may be based on a single document or
multiple distributed documents that have been
combined using defined OWL mechanisms".[17]
There are lot of available tools for designing and
managing ontologies [18]. We have decided to use free
tools, concretely we have chosen Protg as an editor
for the design of the ontology and Jena framework for
managing the designed ontology (querying, reading
data, writing data, persistent storage, rule-bases
inference engine, etc.).
As said before we have used Protg to design an
ontology for obesity treatment. Following are
described the main concepts (entities) we have found
and their relationships.
In our ontology we can find 4 main entities:

3.1. Agent
Any user of the system is considered as an agent.
Each agent has a name, a surname, a login and a
password. We can distinguish three different kinds of
agent: patient, familiar or professional. There are also
two types of professional: doctor or psychologist.
A patient is an obesity user that is going to be
treated. For each patient lot of data is needed.
Following are described the most important
information related to the patient:
Profile, has general information about the patient.
This information includes the address, academic data,
psychological data (test related to personality, selfrespect, life habits, etc.), background, etc.
Familiar, each patient can have one or more
familiar. In psychological treatments it is very
important to have one person that can help the patient
with the treatment at home.
Doctor and psychologist, each patient has a doctor
assigned that is going to see him during the treatment.
In the same way a patient has a psychologist assigned.
Medical and psychological diagnosis, after
studying the patient the doctor gives a medical
diagnosis and the psychologist gives a psychologist
diagnosis.
Medical and psychological treatment, the
professionals establish a specific treatment for the
patient. The medical treatment includes medicines
prescribed to the patient and the objectives of the
treatment. The psychological treatment includes
activities to be done by the patient alone or with the
psychologist.
Evaluation, include the measurements and
variables under control collected during the treatment.
A familiar is a user that is going to help a patient
during the treatment. Since a familiar is an agent has a
name, surname, login and password. Familiars are
really important when the patient is a child; in this case
the familiar (mother or father) participation is essential
(since the mother is the one that cooks for the child).
The professionals are the doctors or psychologists
that are going to treat the patient. As in the rest of cases
as are agents have a name, surname, login and

Figure 2: Main entities of obesity ontology

Alarms: This concept represents warnings to the


patient or the professional (doctor or psychologist)
that a variable under control is out of the allowed
range. Some of the variables under control
(represented in the Evaluation entity) have defined
possible values, if one of these variables gets a non
possible value it is stored indicating the time when
it happened.

Agent: Any user of the system. Each agent needs a


login and a password to access to the system and
to be able to read or write data. We have different
kinds of agents (subclasses): doctor, psychologist,
patient, familiar.
Evaluation: This concept represents the variables
(contextual, physiological, etc.) obtained from the
patient in different moments. There are many
different variables under control. The information
can be collected by the doctor, the psychologist,
the own patient, devices connected to the patient,
etc.
Treatment: This concept represents the planning
done by the psychologist to treat the patient. It also
includes the general objectives of the treatment
and the monitoring of the patient.

29

the patient or asking directly to the patient. This


information is analysed by doctors and psychologist.
Psychological measurements are important because
obesity can have implications in the state of the people
who suffer it [19]. For this reason psychologist done
several test to the patients to know how they felt. These
tests are done at the beginning of the treatment and
repeated during the treatment and several months after
the treatment is finished. Self-respect, anxiety,
impulsivity, personality, weight history, body image,
depressed mood are some of the tests that are being
considered.

password. Professionals have also information about


the patients that are treating (their profiles, evaluation,
diagnosis, treatment, etc.).

Figure 3: Agent overview

3.2. Evaluation
The Evaluation entity represents all the variables
under control and measurements obtained from the
patient. We have different type of measurements:
psychological, obtained from the tests done by the
psychologist, physiological obtained by the doctor or
by a device connected to the patient and contextual
related to the environment of the patient (activity done
by the patient, position, etc.).
Physiological measurements are important due to
the risk that obese people have to suffer diseases of
diverse type, especially cardiovascular [19]. These
measurements are collected by a device connected to
the patient and analysed by the doctor to evaluate
patients situation and his evolution. At the moment
the variables measured are: respiratory rate, arterial
pressure, heart rate, body temperature. As well as these
variables the doctor also measures in each consultation:
tanita, tanner, height, weight and waist perimeter.
Contextual measurements give information about
the habits of the patient. In the obesity treatment is
very important the activity done by the patient and the
alimentary habits. Some of the variables considered as
contextual measurements are: caloric consumption,
physical activity level, position (sat down, standing up
and lied down), how many times the patient open the
fridge, what has eaten the patient, etc. These
measurements are collected by a device connected to

Figure 4: Evaluation overview

3.3. Treatment
This entity represents the planning done by the
psychologist to treat the patient. The psychologist
needs the doctor establishes general objectives for a
patient to plan his treatment. Once the general
objectives are defined the psychologist schedule the
sessions of the treatment.
A treatment is defined as a set of modules. A
completed obesity treatment consists of 8 different
modules. After analysing the psychological
information obtained from the tests initially done to the
patient the psychologist decided the appropriate
modules for this patient (not all the patients need all
the modules, sometimes is useful to focus only in some
specific modules).
A number of sessions has to be specified for each
selected module and for each session specific
objectives are defined. As well as objectives a session
also has information about what is going to be done in

30

and minimum possible value, in other words a range is


defined for them.
When a new measurement is obtained its value is
checked and if this value is not in the defined range an
alarm is launched. Alarms are stored in the system.
The alarm is a warning that the professional (doctor
or psychologist) can consult. For them it is really
useful to know in which moments something wrong
has happened.
It is possible to associate an action to the alarm, in
this case at the same time that the alarm is stored and
the action is fulfilled. Some of the actions considered
are: sending an e-mail to the patient or to the
professional or sending a message to the mobile of the
patient or the professional.

this session. In each session the psychologist defines


some tasks that the patient should do during the week.
The results of these tasks can be consulted by the
psychologist at any moment.

Figure 5: Treatment overview

There are three types of tasks the psychologist can


order to the patient: games, information and selfregisters.
The games have different levels and scenarios and
depending on the evolution of the patient the
psychologist can recommend the most appropriate
game.
The psychologist gives information to the patient
during the consultation, but this information should be
remind to the patient during the week, for this reason to
consult information is another type of task. The
information can be: images, text, audio, video, etc.
There are some measurements that should be done
by the own patient and report to the system, these are
the self-registers. The psychologist indicates the patient
the kind of information that should be reported (what
has eaten, how many exercise has done, etc).

Figure 6: Alarm overview

4. Conclusions and future work


During the last few years the design and use of
ontologies in medicine, bio-medicine, bio-informatics
and semantic web has risen noticeably due to the fact
that the design of an ontology allows reusing and
sharing knowledge defining a common vocabulary.
This vocabulary allows people to share what other
people is doing in different parts of the world in the
same research area.
Recently a new concept of psychological therapy
has appeared. This new intelligent e-therapy (e-it)
adapts itself to the patients lifestyle, offering a 24/7
monitoring to the patient. E-it can be applied to many
disorders, and it is based in a knowledge base that
includes all the knowledge related to the disorder and
its treatment.
The benefits provided by the use of ontologies
made ontologies the best option for the design of the
knowledge base of the intelligent e-therapy.

3.4. Alarms
As previously mentioned an evaluation entity,
representing all the variables under control and
measurements obtained from the patient, has been
defined. Most of these measurements have a maximum

31

Also during these last years obesity problem is


increasing in an alarming way among the population of
all the ages, up to the point of being considered as a
new epidemic (the WHO (World Health Organization)
declared it in 1998 as a global epidemic).
For all these reasons obesity ontology to be
integrated in an obesity intelligent e-therapy has been
developed.
The objective of the obesity intelligent e-therapy is
to improve the efficiency of the obesity treatment,
since until the moment the treatments that are being
applied for this disorder are not having good results.
As soon as the developments of the system have
finished a comparison among this system and the
habitual treatments in order to check out its efficiency
will be done.

[9] Baos, R.M., Botella, C., Perpi, C., Alcaiz, M.,


Lozano, J.A., Osma, J., Gallardo, M. , 2002. Virtual reality
treatment of flying phobia. IEEE Transactions on
Information Technology in Biomedicine v.6 (3) pp. 206-212.
[10] Botella, C., Baos, R.M., Perpia, C., et al., 1998.
Virtual reality treatment of claustrophobia: a case report.
Behaviour Research & Therapy 36 pp. 239-246.
[11] Alcaiz, M., Baos, R., Botella, C., Rey, B., 2003. The
EMMA Project: Emotions as a Determinant of Presence.
PsychNology Journal, 1(2) pp. 141-150.
[12] Hedley, A.A., Ogden,C.L., Johnson, C.L., Carroll, M.D.,
Curtin, L.R., Flegal, K.M., 2004. Overweight and obesity
among U.S. children, adolescents and adults, 1999-2002.
JAMA 291 pp. 2847-2850.
[13] Ogden,C.L., Flegal, K.M., Carroll, M.D., Johnson,
C.L., 2002. Prevalence and trends in overweight among
U.S. children and adolescents. JAMA 288 pp. 1772-1773.

5. References
[1] Hu, B., Dasmahapatra, S., Dupplaw, D., Lewis, P.,
Shadbolt, N., 2007. Reflections on a medical ontology.
International Journal of Human- Computer Studies 65(2007)
pp.569-582.

[14] Srinvasan, S.R., Bao, W., Wattigney, W.A., Berenson,


G.S., 1995. Adolescent overweight is associated with adult
overweight and related multiple cardiovascular risk factors:
The Bogalusa heart study. Metabolism, Clinical and
Experimental, 107 pp.782-787.

[2] Rubin,D.L., Shah, N.H., Noy, N.F., 2007. Biomedical


ontologies: a functional perspective. Briefings in
bioinformatics v. 9 (1) pp. 75-90.

[15] WHO, 1998. World Health Organization. Obesity:


Preventing and managing the global epidemic. Geneva,
Switzerland.

[3] Stevens, R., Egaa Aranguren,M., Wolstencroft, K.,


Sattler, U., Drummond, N., Horridge, M., Rector, A., 2007.
Using OWL to model biological knowledge. International
Journal of Human- Computer Studies 65(2007) pp.583-594.

[16] WHO, 2003. World Health Organization. Global


strategy on diet; physical activity and health. Geneva,
Switzerland.

[4] Park, S., Lee, J.K., 2007. Rule identification using


ontology while acquiring rules from Web pages.
International Journal of Human- Computer Studies 65(2007)
pp. 644-658.

[17] OWL, 2004. OWL Web Ontology Language Guide.


Retrieved 2008-04-07 from <http://www.w3.org/TR/owlguide/>

[5] Clark, K.L., McCabe, F.G., 2007. Ontology schema for


an agent belief store. International Journal of HumanComputer Studies 65(2007) pp. 625-643.

[18] Synthetic Biology, 2008. Semantic web ontology


software.
Retrieved
2008-04-07
from
http://syntheticbiology.org/Semantic_web_ontology/Softwar
e.html

[6] Gruber, T. R. , 1993. A Translation Approach to


Portable Ontology Specifications. Knowledge Acquisition,
5(2) pp. 199-220, 1993.

[19] Franco, C., Bengtsson, B., Johannsson, G., 2006. The


GH/IGF-1 Axis in Obesity: Physiological and Pathological
aspects. Metabolic syndrome and Related Disorders, 4 pp.
51-56.

[7] Gruber, T. R. , 2007. Ontology.


Retrieved
2008-04-02,
from
<http://tomgruber.org/writing/ontology-definition-2007.htm>
[8] GeneOntology, 2008,
Retrieved
2008-04-02,
<http://www.geneontology.org/index.shtml>

from

32

You might also like