AN ONTOLOGY IN ORTHOPEDIC MEDICAL FIELD

Harold Vasquez, Ana Aguilera

Departamento de Computación, FACYT, Universidad de Carabobo, Valencia, Venezuela

Leonid Tineo

Departamento de Computación, Universidad Simón Bolívar, Caracas, Venezuela

Keywords: Ontology, Medical Data Base, Orthopedic, SQLf.

Abstract: At present time, Ontology is a powerful Knowledge Representation tool for Web Information Retrieval and

Mining. In particular, lots of medical applications in the field of diagnosis and tele-health would take

advantage of information sharing and publication on the Web with the use of Ontology. We are especially

interested in the field of orthopedic pathologies of human march. There are several laboratories in the world

working in this topic but they are not exploiting potential of the Web in theirs works, they are almost isolate

in information management. We have already carried out mining in database from the Venezuelan Hospital

Ortopédico Infantil, nevertheless, it is necessary to build an ontology in order to query and mine

information from different laboratories in the world. We would like in the near future to apply fuzzy logic

techniques and fuzzy querying over such information. In this paper, we present the building of an

orthopedic medical ontology, to the best of our knowledge, has not been reported other in this specific field

that we are interested on.

1 INTRODUCTION

Several investigations aiming at optimizing the

handling of medical information so that it becomes

both useful and agreeing with the user requirements,

in the sense of extracting significant knowledge to

be used in the detection and cure of diseases, have

been reported.

In the sense of allowing both manipulation and

access to medical information with the purpose of

satisfying exigent user requirements, it is important

to have computational tools to perform this.

Currently, the relational database querying has

answered to some extent these needs.

Queries could be improved using ontologies,

since they are very useful to represent knowledge

within a specific domain. This representation tool

has desirable features, such us: re-usability,

interoperability and formality. Therefore, there are

many current researches in the area of knowledge

representation and manipulation based on

ontologies.

Ontologies have demonstrated to be very

advantageous tools to formalize, maintain, reuse,

share, generalize and communicate knowledge in a

specific dominion. The ontologies provide a

common vocabulary, in relation with an area of

knowledge and define, at different formal levels, the

meaning of the terms and relations among them

(Valencia, 2005).

We are interested in querying medical

information, in particular for orthopedics, employing

SQLf, a fuzzy querying language (Goncalves &

Tineo, 2006), (Bosc & Pivert, 1995). It obeys to the

fact that fuzzy sets are powerful manners to

represent user preferences, imperfect data and weak

relationships. We would like our project to be not

limited to a single database but capable of accessing

information from diverse health centers.

We would like to build a fuzzy Ontology, based

on knowledge, in the medical database of the March

Laboratory at Hospital Ortopédico Infantil (HOI) in

Caracas, Venezuela.

Nevertheless, before to obtaining this Fuzzy

Ontology, it is necessary to build a classic ontology

and subsequently convert it with fuzziness. There

are a lot of mechanism to get this done but not all

them it are useful to our proposal, so we need to

372

Vasquez H., Aguilera A. and Tineo L. (2008).

AN ONTOLOGY IN ORTHOPEDIC MEDICAL FIELD.

In Proceedings of the Third International Conference on Software and Data Technologies - PL/DPS/KE, pages 372-375

DOI: 10.5220/0001888503720375

 SciTePress

study and apply some of these strategies that accord

fit to our necessity.

This paper shows the process of making this (up

to our knowledge very first) Ontology of knowledge

about medical studies over patients with Orthopedic

diseases in pathological walking.

2 SOURCE DATA BASE

Venezuelan March Laboratory at Hospital

Ortopédico Infantil (HOI) is one of the biggest and

most advanced laboratory for diagnostics, research

and treatment of march pathologies in Latin

America. HOI uses an advanced mediation system

for both diagnostic and treatment of the illness

related with the locomotive and neuromuscular

system. This system allows to collect simultaneously

both data and images in three dimensions, and after

an adequate treatment. HOI also has a database that

keeps tracks of all its patients’ cases since 1997. For

the purpose of this research work, HOI has provided

a copy of this database with registers until December

2006 (Vasamon & León, 2007).

HOI database was developed in Ms.Access™.

This database has the logic scheme show in Figure

1. Names in he scheme are in Spanish, because this

is the Venezuelan current language. Semantics of

tables is as follows:

Paciente. This table has basic information of

patients. – ID_paciente, primary key code. –

Nombre, given name. – Apellido, family name. –

Sexo, gender. – Fecha_Nac, date of birth, –

Historia_HOI_I, medical history identification

number. – ID-DIAGNOSTICO, foreign key to the

main diagnostics disease for the patient. –

Fundational/Privado, indicates whereas the patients

comes from a beneficial foundation or a private

consultation. – Lado, Tono, Lado 1, Tono 1,

indicators for the side leg (left/right) and a measure

of muscular tone of this side. – Nivel, regarding to

the level of the main disease in the patient.

Estudio. this contains information about studies

performed to patients, some of the main columns are

the following. – Estudio_ID, primary key code. –

Paciente_ID, foreign key to the patient. – Pre_Post,

indicates whereas the study is made as a requisite

previous or posterior to a surgical intervention. –

Fecha_estudio, date of the study, – Master_Video,

number of the video register for the exam. –

Tipo_Estudio_ID, foreign key to the study kind

study’s analysis resulting diagnostics disease.

Tipo_estudio. This part of the database gives

identification and name of different kinds of studies.

Interpretador. This table gives identification and

name of physicians that interpret studies.

Referente. Here give identification and name of

physicians that reference patients and ordain studies

for patients.

Referente 1. As the previous one, it gives

identification and name of physicians that reference

patients and ordain studies.

Diagnostico. This table gives identification and

name of different diseases or pathologies

diagnostics.

Figure 1: Logical schema of HOI Data Base.

3 BULDING THE ONTOLOGY

There are many definitions of ontologies, but the

most accepted is the one given (Gruber, 1993) that

states: “an ontology is an explicit, formal

specification of a shared conceptualization of a

domain of interest”. However, since we are going to

use the method described in (Men et al, 2005), we

adopt the following definition:

An Ontological Structure O is a 5-tuple

O = {C, R, H

, rel, A

}

where C is a finite set of concepts; R is a finite

set of relations; H

is called concept hierarchy or

taxonomy, which is a directed relation H

⊆ C×C,

Paciente

Id_Paciente

Nombre

Apellido

Sexo

Fecha_Nac

Historia_HOI_ID

ID_Diagnostico

Fundacional/Privado

Lado

Lado1

Tono1

Tono

Nivel

Estudio

Estudio_ID

Paciente_ID

Pre_Post

Fecha_estudio

Master_Video

Interpetrador_ID

Tipo_Estudio_ID

Notas

Combo

EDAD

TALLA

PESO

Referente_ID

ID-Diagnostico

Diagnostico_referencia

EXAMINADO POR

Diagnostico

ID-DIAGNOSTICO

DIAGNOSTICO

Referentes

Referente_ID

Referente

Referentes_1

Referente_ID

Referente

Interpretador

Interpretador_ID

Interpretador

Tipo_Estudio

Tipo_Estudio_ID

Tipo_De_Estudio

1 ∝

AN ONTOLOGY IN ORTHOPEDIC MEDICAL FIELD

373

for example, H

, C

) specifies that C

is a

subconcept of C

; rel relates concepts non-

taxonomically, for example, rel (R) = (C

, C

)

specifies that C

and C

have relation R.; A

is a set

of axioms, which is expressed in an appropriate

logical language, e.g. First Order Logic (FOL).

3.1 Language of Representation

Typically, an ontology can be generated from

various data types such as textual data, dictionaries,

knowledge-based semi structured schemata and

relational schemata (Quan et al, 2006). However, in

all those cases, the ontology can be represented with

different languages, but the most commonly used

today are XML, OWL and RDF. Because OWL is

one of the most used languages in the field of

ontologies and it is standardized by the W3C, it is

going to be used to model the medical ontology that

will use in this investigation. The following figure

presents the evolution of the different languages

employed to represent Ontologies.

Figure 2: The architecture of Levels of T. Berners Lee.

3.2 Methodologies Building

In (Men et al, 2005), they propose an approach of

learning an OWL ontology from data in a relational

database. Compared with existing methods, the

approach can acquire ontologies from relational

databases automatically by using a group of learning

rules instead of using a middle model. However, this

proposal has the inconvenient that an ontology is

created in OWL directly without an intermediate

model, so if the database is poor in knowledge, then

Ontology is poor too. Furthermore, it’s important

this intermediate model, that makes rich the learning

produced, is important.

There are so many methodologies for learning

ontologies (Garcia, 2005), but only few are helpful

for our purpose. Among them, it is the so-called

Methontology (Fernández et al, 1997 and Gómez-

Pérez, 1998). This methodology guides the ontology

building by specifying a set of intermediate

representations (IRs) at the knowledge level. These

IRs bridge the gap between how people think about

a domain and the languages in which ontologies are

formalized, that it allows us both to extend and

enrich our ontology, previously generated with the

HOI database. We sketch this methodology,

followed by our study case, in Figure 3.

At the end, we will evaluate our ontology,

generated by using quality assuring metrics,

according to pervious work what is proposed in

(Lozano, 2002).

Schema

DB-HOI

Middle Building

Ontology

Methontology

IRs

Orthopedic Medical

Ontology in OWL

Method described

in Men et al, 2005

Figure 3: Model to building the Orthopedic Medical

Ontology.

3.3 Editing and Browsing Tool

Instead of using the Methontology tool which

supports the ontology maker during the entire life

cycle of the ontology development process, called

ODE (Ontology Design Environment); we prefered

to use another environment to carried out this. The

reason to do that was the fact that ODE has the

inconvenient that it does not work with the OWL.

ICSOFT 2008 - International Conference on Software and Data Technologies

374

Because these reason of that, we are

experimenting using Protege or OntoWeb with those

two mechanisms to build our ontology instead.

These tools are well known, free and with a profuse

support.

4 CONCLUSIONS AND FUTURE

WORKS

In this work we combine several methodologies in

order to build an Ontology of knowledge regarding

medical studies that have been performed over

patients with Orthopedic diseases in pathological

walking or march. The work has been undertaken on

the basis of the Venezuelan Hospital Ortopédico

Infantil (HOI) March Laboratory database. The

result of this work would be an original contribution

to the research for this kind of laboratories, because

there is no precedent ontology in this field. Several

institutions could take advantage of this as a support

tool for the diagnosis and treatment of these

diseases. This work would allow the publication and

sharing of this kind of information for querying and

mining purposes. In future works we will apply

fuzzy logic tools for mining and querying medical

information from march laboratories. In order to do

so, we will extend such techniques with the use of

Ontology. We hope to reach a benefit for humanity.

ACKNOWLEDGEMENTS

We acknowledge that the inspiration and all that we

need for working an living come from Heaven, from

our Lord and Eternal Father. For this work, we have

the financial aid of Venezuelan National Foundation

for Science, Technology and Innovation FONACIT

by means of the project G-200500278. We also give

thanks to the Hospital Ortopédico Infantil

Foundation and especially to the staff of the March

Laboratory who have cooperate in this research

proposing valuable data and knowledge. Finally, we

want to express acknowledgement to Dr. Miriam

Rodríguez, physician, specialist in rehabilitation and

physiatrists, medical advisor of our project.

REFERENCES

Goncalves, M., Tineo, L.: SQLf vs. Skyline - Expressivity

and Performance. Proceedings of FUZZY-IEEE 2006.

Vancouver, Canada (2006).

Bosc, P., Pivert, O.: SQLf: A Relational Database

Language for Fuzzy Querying. IEEE Transactions on

Fuzzy Systems. 3(1). (1995)

Valencia R.: Un Entorno para la Extracción Incremental

de Conocimiento desde Texto en Lenguaje Natural.

Departamento de Ingeniería de la Información y las

Comunicaciones. Universidad de Murcia. Murcia,

España (2005)

Gruber T. A translation approach to portable ontology

specifications. Technical Report KSL 92-71.

Knowledge System Laboratory. (1993)

Man Li, Xiao-Yong Du and Shan Wang: Learning

Ontology from Relational Database. Proceedings of

the Fourth International Conference on Machine

Learning and Cybernetics, Guangzhou. pp. 18-21.

(2005)

Vasamon, D. y León, Y. Aplicación de Minería de Datos

en Bases de Datos Médicas. Departamento de

Computación. Universidad de Carabobo. Carabobo,

Venezuela (2007)

Quan Thanh Tho, Siu Cheung Hui, Fong and Tru Hoang

Ca: Automatic Fuzzy Ontology Generation for

Semantic Web. In: IEEE transactions on knowledge

and data engineering, 18(6). pp. 842--856. (2006)

García V., Rafael. Un Entorno para la Extracción

Incremental de Conocimiento desde Texto en

Lenguaje Natural. Tesis Doctoral. Universidad de

Murcia. Departamento de Ingeniería de la Información

y las Comunicaciones. (2005).

Fernández, M.; Gómez-Pérez, A.; Juristo, N.

METHONTOLOGY: From Ontological Art Towards

Ontological Engineering. Spring Symposium Series.

Stanford. PP: 33-40. (1997).

Gómez-Pérez. A. Knowledge Sharing and Reuse. The

Handbook of Applied Expert Systems. Edited by

Liebowitz. CRC. (1998).

Lozano T., Adolfo. Metrica de Idoneidad de Ontologias.

Tesis Doctoral. University of Extremadura.

Departamento de Informatica. (2002)

AN ONTOLOGY IN ORTHOPEDIC MEDICAL FIELD

375