ONTO-MAMA

An Unified Ontology and 3D Graphic Model of the Female Breast Anatomy

P. B. L. Klavdianos

, M. Parente

, L. M. Brasil

and J. M. Lamas

Centre Universitaire Condorcet, uB – Université de Bourgogne, Le Creusot, France

Gama College (FGA), UnB - University of Brasília, Brasília, Gama, Brazil

Clínica Janice Lamas Radiologia (Radiology Clinic), Brasília, Brazil

Keywords: Ontology modeling, 3D graphic modeling, Female breast anatomy, Methontology.

Abstract: The science of ontology has been widely used in knowledge management, either to define and organize

concepts for the purpose of future preservation or to provide more efficient information processing by our

computers. From this, the use of ontology in medical field has become increasingly consolidated, either to

simply describe the correct meaning of technical terms or to completely depict the anatomical structure of

the human body or medical procedures. In addition, the use of ontology has been associated to artificial

intelligence and virtual reality in order to provide simulation of medical environments with the aim of better

understand the complexities of the human anatomy and the medical procedures. Accordingly, this article

presents the elaboration of an ontology and a 3D graphic model of the female breast anatomy to be used in a

virtual reality environment containing an intelligent tutor system which will eventually be able to assist

learners in the practice of the core needle biopsy. This article reports our experience so as to share

information about the process used, the artifacts generated and the systematic involved in the structuring of

such unified model.

1 INTRODUCTION

Information Technology (IT) has played an

important role in the process of improving medical

education by intensively using resources from

virtual reality, computer graphics, intelligent

tutoring systems and knowledge management

techniques and tools. The combination of conceptual

models which express the meaning of things and

graphical models which fulfill the need of a visual

representation for the objects being learning makes it

possible to completely replace the deteriorating

human anatomical parts from medical labs.

Moreover, models that were built by applying the

aforementioned computational techniques are easier

and cheaper to maintain. Such models, besides being

extremely reusable, are also likely to undergo

changes, or even a complete overhaul, if need be.

With regard to knowledge management, one of

the most relevant and complex goals of the area is

the appropriate transformation of implicit knowledge

into explicit, which implies ensuring that knowledge

about a specific subject must be correctly identified

and disseminated in order to preserve it. Knowledge

management is not different when applied to the

medical field. Expressing a concept or idea accepted

by different experts in a single way is not a trivial

matter, even when those concepts have an extensive

bibliography and dictionaries to address issues of

specific interest, as in the case of medical terms.

Another aspect that contributes to the difficulty of

expressing a concept of the real world through

written language involves the way one concept

relates to others. A concept is often not simply

defined by a word or word etymology. For a better

understanding of its meaning, it is necessary to

identify associations with other similar,

complementary or antagonistic idea, outright

concepts or even complete learning objects (videos,

images, sound, texts, animations, etc). The science

of Ontology acts in this context, and it has been used

as an alternative to define, organize and share

knowledge in a standardized way so that is

susceptible to be processed easily by a computer. As

a result, a more automatic process for extracting or

derivate meaningful information is created.

Nevertheless, the main issue of building ontology

models lies in the definition of the concepts under a

106

B. L. Klavdianos P., Parente M., M. Brasil L. and M. Lamas J..

ONTO-MAMA - An Uniﬁed Ontology and 3D Graphic Model of the Female Breast Anatomy.

DOI: 10.5220/0003796401060116

In Proceedings of the International Conference on Health Informatics (HEALTHINF-2012), pages 106-116

ISBN: 978-989-8425-88-1

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

coherent, complete and utterly understandable

fashion. After all, this is what will guarantee the

usability of the model and its connection to other

complementary resources such as a graphical model

or learning objects used for tutoring.

Hence, this article presents the elaboration of an

ontology model in conjunction with a 3D graphical

model for the anatomy of the female breast. Such

models are intended to be used in a virtual reality

environment having an intelligent tutor system

which aims at assisting the future doctor in the

practice of core needle biopsy. We aim to describe

how the two worlds (conceptual and visual) can be

constructed together in order to fulfill the

requirements of a virtual reality environment

idealized for tutoring purpose. The article is

structured as follows: section 2 presents a brief

review of the concepts associated to the science of

ontology, section 3 presents the work developed,

with emphasis on the construction process, the

artifacts generated and the internal structure of the

model. The article ends in Section 4, where we state

our conclusions and share our experiences and

difficulties in the carrying out of the task.

2 ONTOLOGY OVERVIEW

Ontologies define terms or concepts, the

relationships among them, the rules for the most

complex definitions and some extensions accepted

for purpose of vocabulary completeness and

understandability (Neches et al., 1999).

It refers to an abstract model of some

phenomenon or something that can be identified in

the real world. It is a formal technique, an explicit

specification of concepts that is intended to be

shared. It is formal because it is expressed in a

language which allows for easier processing through

a machine or computer. It has the characteristic to be

explicit, since the type of concepts expressed and the

rules and restrictions imposed must be categorically

defined. Finally, it is supposed to be shared because

it aims to capture and disseminate knowledge to

everybody, i.e. it should not be restricted to a few

individuals, but accepted and used by a group or

community (Studer et al., 1998). Ontology can be

manifested in several ways, but must necessarily

include a vocabulary of terms and specifications

detailing its meaning, which includes definitions of

rules and records of how concepts are interrelated

(Uschold and Jasper, 1999).

By providing a systematic framework to record

information, an ontology model creates enormous

possibilities for associations among concepts, which

broaden understanding on specific areas of expertise.

Consequently, new information can be extracted

from the model so that the initially imagined horizon

of knowledge is extended. For this reason, the

science of ontology highlights two types of models:

the conceptual model and the inference model. The

conceptual model is the one originally envisioned by

the creators of ontology. The inference model, on

the other hand, is that one which is generated by an

intelligent agent, i.e. a piece of software that

captures and extracts information not easily

identified during the idealization of the conceptual

model.

In the health field, ontologies have been

developed to translate texts related to medical

oncology (ONCOTERM), connect biomedical

vocabulary from various sources (UMLS - The

Unified Medical Language System) and to equalize

medical terms used in studies of anatomy,

physiology and pathology. Regarding to our project,

the creation of an ontology model was needed

because the existing ones couldn’t fulfill the

requirements of our project. The next sections will

present the process and methodology used to build

the ontology model (conceptual and inference

models) and the 3D model (visual model) in a

unified manner.

3 ONTO-MAMA PROJECT

3.1 Contextualization

The Onto-MAMA is part of a wider scientific

project called "3D Anatomical Atlas Applied to the

Breast". Such project is an initiative of the National

Laboratory of Scientific Computing of Brazil

(LNCC) and with the participation of the University

of Brasilia (UnB), the current co-executor of the

work. The project has the scientific and financial

support of the National Council for Scientific and

Technological Development (CNPq), an agency that

is connected to the Brazilian Ministry of Science and

Technology and has as its goal to promote scientific

and technological research and to create human

resources for research in Brazil.

The building project of a 3D Anatomical Atlas

Applied to the Breast began in July 2009, and its

goal is the search for new proposals for training

students in areas related to the anatomy of the

human breast, in opposition to the teaching of

classical anatomy using the dissection of cadavers.

ONTO-MAMA - An Unified Ontology and 3D Graphic Model of the Female Breast Anatomy

107

As a practical and more comprehensive result of

the project, it is expected that a web-based

educational environment can be established,

allowing for the practical-morphological learning of

the internal and external structures of the female

breast. By using artificial intelligence and virtual

reality techniques, we aim to seek new approaches to

the teaching of students of Health Sciences and

Human Biology, as well as to benefit teachers and

health professionals interested in receiving

continued education in teaching the anatomy of the

human breast. Another goal is to provide guided

surgical training to students and health professionals

interested in mastering the core needle biopsy

procedure.

Figure 1 illustrates the major components of the

project and identifies the key role played by the

ontology model for anatomy of the female breast,

which is the main subject in this paper.

Figure 1: 3D Anatomical Atlas Applied to the Breast.

3.2 Methodology

The construction of an ontological model comprises

the following steps: (a) the adoption of a working

methodology, (b) the study of the field of action, (c)

the formal representation of knowledge and (d) the

distribution of the model for validation and use.

The elaboration of the ONTO-MAMA model

started in 2009 with a systematic study of existing

methodologies for the building of ontologies. The

project team defined five criteria for selecting a

methodology which could best adapt to the

characteristics of the project (Table 1). Additionally,

studies of the following existing and widely used

methodologies were conducted: CYC, USCHOLD

and KING, KACTUS, SENSUS, GRÜNINGER and

FOX, METHONTOLOGY and ON-TO-

KNOWLEDGE.

Table 1: Criteria for adoption of a working methodology.

Criteria Description

Scope of work

The methodology must describe the

complete creation cycle of an ontological

model (planning, creation and

maintenance)

Prototyping

The methodology must provide the

capacity to create intermediate models

throughout the development cycle

Adaptability

The methodology must be easily

adaptable by the team so as to

incorporate peculiarities inherent to the

project.

Documentation

The methodology must provide sufficient

documentation for its understanding.

Reputation

The methodology must be well renowned

and highly reputed among scientists.

The methodology which best suited the

requirements of the project was

METHONTOLOGY, since not only does it meet the

established criteria relatively well, but it also had the

advantage of having a cycle of evolutionary

development, based entirely on prototyping.

Prototyping was a requirement of great importance

in the context of our work, because we planned for

other sub-projects to use the models generated since

the beginning, either with the aim of experimenting

ideas and concepts, or as a simple mechanism of

immediate feedback to the ontology team.

Therefore, the ontology team started using

METHONTOLOGY as the working methodology,

as well as the tool called Protégé (version 4.1) for

the purpose of prototyping the model throughout its

whole development cycle. In addition, intermediate

artifacts were produced to facilitate communication

among members of the team and the medical

experts. These artifacts will be presented in the

subsequent sections.

Once the methodology was defined, the team's

next step was to decide whether the work would be

undertaken from an existing ontology (ontology

merge approach) or whether a new model should be

created specifically for the project (ontology

creation approach). On one hand, the creation of a

new ontology implied extra research efforts and

intense dedication by the breast anatomy experts. On

the other hand, the use of existing ontologies

required well-planned strategies for merge non-

existent concepts with the existent ones, not to

mention the need of specific arrangements for future

evolution of the model. Thus, the project team

initially opted for conducting an investigation of the

HEALTHINF 2012 - International Conference on Health Informatics

108

pre-selected ontologies in order to identify whether

some of them could actually be reused or not. The

evaluated ontologies were, therefore, classified

according to six categories shown in Table 2.

Table 2: Classification used for ontologies assessment.

Criteria Description

Degree of Detail

It indicates the level of description of the

terms contained in the ontology (low,

medium and high degree of detail).

Degree of

formalism

It describes the degree of formality used

by the ontology, which are: formal, semi-

formal and informal.

Technique used

for modeling

Description of the systematic used to

construct the ontological model, e.g. first

order logic, logic description or other

more specific techniques.

Licensing

It identifies the type of licensing of the

ontology.

Type of

ontological

model

Classification of ontologies according to

the criteria defined by Lassila and

McGuinness (2001), comprising

ontologies in: controlled vocabulary,

glossary, thesaurus, formal hierarchy,

informal hierarchy, frames, value

restriction and logical constraint.

The project team also decided to verify the

adherence of the ontologies to the principles of

Gruber T. R. (1993a), namely: (a) clarity, (b)

coherence, (c) extensibility, (d) consistency in the

use of common vocabulary and (e) non-dependence

of specific symbologies. Since this assessment is out

of the context of this paper, it is sufficient to

mention that among the pre-selected ontologies the

Foundation Model of Anatomy (FMA) was the one

which most closely represented the needs of the

project. However, the project team decided to build

a new model from scratch while keeping the FMA as

a reference ontology which could be integrated to

the ONTO-MAMA in future work if the need arises.

The main reason for such decision stems from the

degree of detail desired and from the need to be in

compliance with the principles established by

Gruber T. R. (1993a). Even a quite extensive and

complete ontology such as FMA did not present, at a

high level of detail, the concepts needed to describe

the macro and micro structures of the female breast,

as required in our project. Besides, the evaluated

models provided little or no documentation in order

to clarify the concepts and its usage, making it

difficult to extend the model.

The METHONTOLOGY will be described in the

next section under the point of view of ONTO-

MAMA project as well as by considering a unified

methodology which promotes the integration of a

ontology model and a 3D graphic model.

3.3 A “Unified Ontology Model”

3.3.1 METHONTOLOGY: Building the

Ontology Model

METHONTOLOGY is a methodology for building

ontologies created by the Universidad Politécnica de

Madrid by the Ontological Engineering Group.

Figure 2 illustrates the METHONTOLOGY’s

phases and stages (or tasks). The most important

tasks or stages, located in the development phase,

are the followings: (a) Specification, (b)

Conceptualization, (c) Formalization, (d)

Implementation and (e) Maintenance. Additionally,

METHONTOLOGY describes management and

support activities, both necessary to complete the

cycle of construction, evolution and maintenance of

the ontology model.

Figure 2: METHONTOLOGY Process.

The first step of METHONTOLGY is the

specification. It defines the boundaries of the model

and clearly states, to all participants, the objectives

that are supposed to be achieved, the desired degrees

of formality, the scope of the ontology and the

desired degree of granularity.

The next step involves the organization and

capture of the knowledge, which is called

Conceptualization stage in the METHONTOLOGY

process. This step consists on the conversion of tacit

knowledge obtained from experts into explicit and

semi-formal knowledge. It is an intermediate record

of knowledge, a leveling mechanism of concepts

among all team members. Gomez-Perez A., Corcho

O. and Fernandez-Lopez M. (2004) suggest some

intermediate representations, for example, glossaries

of terms, tables and data dictionaries to be used at

this stage.

The formalization and implementation stages

ONTO-MAMA - An Unified Ontology and 3D Graphic Model of the Female Breast Anatomy

109

consist of describing the model using a formal

language. In the case of this project the language

chosen was Web Ontology Language (OWL)

generated by Protégé. During the formalization and

implementation stages the ontology model can be

submitted for validation by experts as well as be

used by others sub-projects members in order to

provide immediate feedback. The implementation

stage may also include the construction of libraries

or reusable softwares which aims to use the model

so that usability, performance and portability can be

measured beforehand.

The METHONTOLOGY development cycle is

concluded with the maintenance stage which

includes the validation and the delivery of the model

for the scientific community. One may understand

this phase as a beta test of the ontology model.

3.3.2 METHONTOLOGY: Building the 3D

Graphic Model

The first multimedia model incorporated to our

unified ontology model was the 3D graphic model.

This model was chosen due to the importance of

representing the anatomical parts in a more

appealing manner, including the possibility of

interaction and animation directly performed on the

3D structure (for simulation purpose) and also

because the 3D modeling brought to us some

challenges to overcome in order to represent the

anatomic parts in a more realistic way.

Some of the challenges we had to overcome can

be summarized as follows: (a) the lack of a specific

3D modeling methodology, (b) the divergences

found among different anatomical atlases used as

reference, (c) the difficulty of setting standards for

the 3D modeling such as the size of the 3D objects,

proportionality of the model compared with a real

breast, color and texture issues and grouping and/or

linkage of the 3D objects to the ontology model.

The lack of a 3D modeling methodology was

solved by adapting the METHONTOLOGY process

to the 3D modeling tasks. This way we could

provide a better interaction between the ontology

team and the 3D modeling team which resulted in a

better validation process since the graphical artifacts

was frequently used to check the comprehension of a

concept described by the ontology model. The

divergences among the references were solved by

using the opinion of the experts and scientific

community whom were in participation of the

ontology model construction. Finally, the need of

setting standards were addressed in the beginning of

the 3D modeling process so that all the people

involved already new and understood our goals and

limitations for modeling the female breast in a more

fidelity and truthiness manner.

In conclusion, we decided to keep and adapt the

METHONTOLOGY as the working methodology

for the 3D modeling tasks.

From this, the first step of 3D modeling

consisted of planning the work. During the planning

the following tasks were executed: a) study of the

METHONTOLOGY; b) structuring of

METHONTOLOGY stages to the 3D modeling

needs; c) production of schedule; d) initial

preparation of requirements.

Next the planning phase, the stages of

METHONTOLOGY were applied directly to the 3D

modeling tasks with some adaptations.

In the Specification stage the 3D team worked

closely to the ontology team by gathering the

requirements of the model from the experts.

Consequently, the use of knowledge acquisition

(KA) techniques is imperative at this point as

defined in Brasil L. M (1994). Some of the KA

techniques applied in this project are: a) dynamic

reading of documents; b) observations; c)

interviews; d) definition of protocols.

In the Conceptualization stage the 3D team

provided support to the ontology team by initially

prototyping visual presentations of the concepts as

well as by validating such concepts with the experts

in mastology. It is also part of the work to define the

technique to be applied in the final version of the 3D

modeling. Regarding to this, the method of polygons

has proved effective due to its compatibility with

complementary techniques such as edge loops and

face projection which improved our results.

The Formalization and Implementation stages of

METHONTOLOGY represent the construction of

the 3D model for an anatomic structure as well as

the connection of such graphical object to the

ontology. At this stage, the 3D modeling team may

also need to compare graphical models constructed

by using different techniques which improves the

accuracy of the work.

It is important to point out that due to modeling

easiness and the already acquired knowledge of our

3D modeling team we have opted for using Maya

Software 2010. Maya has a system of particles and

fluids and the facility of texturization, composition,

renderization and animation which have proved to

be handy in our project. Besides, this software is

widely used in the biomedical area and was the

object of study in Gu S. (2006) and Sharpe J.,

Lumsden C. J. and Woolridge N. (2008) whom

presented it as an accessible alternative to capture

HEALTHINF 2012 - International Conference on Health Informatics

110

biomedical properties. Maya also allows the 3D

modeling by using polygons, NURBS (Non-

Uniform Rational B-Spline) and subdivisions which

provided us a way of modeling at different levels of

complexity. In our project we also opted for

applying complementary 3D techniques in order to

modify or simply polish the final 3D model (Maraffi

C., 2003).

The connection with the ontology model is done

by grouping the 3D structures (vertices, edges and

faces) according to its semantic meaning. After a

long period of research, it was found that there were

no visual ontologies for 3D models in the health

field. Therefore, this characteristic represents a great

contribution of this work. In this regard, Maya

proved to be very efficient in storing the simplified

version of the identification structure defined

primarily in our ontology model with the purpose of

tagging each 3D object. As a result, from the

generated 3D model, in VRML or OBJ format, we

were able to retrieve not only the descriptions of our

3D objects (vertices, edges and faces), but also the

metadata for their identification, which further on

could be linked to our entire ontology model or be

manipulated by the intelligent tutoring system.

The stage of Maintenance is the endpoint and is

represented basically by the validation process. At

this point, both models (ontology and 3D) and

intermediate artifacts go through a systematic

verification and validation by experts in the area of

radiology and mastology.

3.3.3 The “Unified Model”

Since our goal in the ONTO-MAMA was to

construct an ontology model to be used as a

reference for learning medical concepts related to

the female breast anatomy and to the core needle

biopsy, the primary requirement of the project

became the representation of the knowledge in

several formats or medias, not only in textual mode.

Therefore, our building process as showed before

provided the possibility for combination of the

ontology model and a 3D model. We have no doubt

that such a process can also accommodate others

digital models in a more general and unified view.

This unified view, then, will be considered the most

complete representation of the knowledge because it

will contain different points of view and also the

possibility of interchangeability among them.

From this idea of building a unified ontology

model we defined a more complete structure for

identification of the objects in our ontology model

by considering different types of media and external

resources. This same identification or linkage

structure will be used in future versions in order to

add references to a diversified set of digital media so

that an ontology engine could make the connection

among several different types of knowledge

representation. Furthermore, the linkage structure in

our ontology model serves to determine the details

of the relationship among the ontology description

and any external media resource.

The next section describes the results obtained so

far on each phase of our building process by

considering both models: the ontology and the 3D

model.

3.4 Results

3.4.1 Stage 1: Specification

The Specification stage of the ONTO-MAMA

project, was formalized in a document called

"Ontology Requirements Specification" that follows

the pattern suggested by Gomez-Perez A., Corcho

O. and Fernandez-Lopez M. (2004).

Basically, as the goal of the ontology we’ve

defined that the model express the vocabulary of the

female breast anatomy for the purpose of performing

the core needle biopsy procedure. In the future the

description of these medical procedures will be also

incorporated to the model.

Regarding to the characteristics of the model,

we’ve decided for building a formal ontology

described in Resource Description Framework

(RDF) or Web Ontology Language (OWL) format

and doted of a medium degree of granularity which

means to say that we won’t intend to represent and

describe the very microscopic structure of the breast

anatomy, not relevant to the performance of the

medical procedure.

As we can notice the specification phase

provided the general guidelines for both ontology

and 3D graphic modeling.

3.4.2 Stage 2: Conceptualization

The outcomes generated during the

Conceptualization stage are the three types of

glossaries of terms, as well as diagrams which define

the structure of the model in a high-level view called

Model View, and intermediate visions of the model,

which were named Meta data view and

Implementation view.

Figure 3 illustrates the structure of all artifacts

that compose the ONTO-MAMA. The Glossary of

Terms and the Model View (Diagrams) can be

ONTO-MAMA - An Unified Ontology and 3D Graphic Model of the Female Breast Anatomy

111

optionally used to guide the modeling of any type of

multimedia representation, for instance the 3D

graphic model. The ontology model, on the other

hand, is considered a compulsory artifact which

serves as the main guidance for the multimedia

modeling if we consider the unified ontology model

described in previous section.

In fact, it is the ontology model that will

determine which objects can be represented by

multimedia artifacts and what characteristics and

degree of detail (anatomical and visual aspect) it is

expected or accepted.

Figure 3: Intermediate Representation in the

conceptualization phase.

Table 3, in turn, illustrates the structure of the

glossaries of terms created in the project that broadly

follows the recommendations of Gomez-Perez A.,

Corcho O. and Fernandez-Lopez M. (2004). An

example of the glossary of terms and the meta data

structure for the ontology model created for this

project can be found in the appendix section at the

end of this paper.

Table 3: Structure of glossaries of term.

Sections of the document

- Term or concept name

- Term or concept description

- Source of reference for term or concept

- Copyright statement for using the term/concept

description (optional)

- Examples of multimedia representation (optional)

- Source of reference for multimedia representation

(optional)

- Copyright statement multimedia usage (optional)

From the point of view of the 3D modeling, the

team provided some intermediate 3D objects for

conceptualization, understanding and future

improvements. Some examples are shown in Figure

Figure 4: a) External view of the breast: triangulated

mesh; b) Side view of the Ductal System – Current model

mammary glands Part 1; c) Side view of the Ductal

System – Current model mammary glands Part 2.

3.4.3 Stage 3 and 4: Formalization and

Implementation

As stated earlier, beside the artifacts suggested by

the methodology itself, the project team decided for

building complementary diagrams which are able to

express the building block structures of the unified

model in a more understandable way.

This decision took into account the different

profiles of people involved in the project, which

include professionals with little or no knowledge

whatsoever of concepts involving ontology. Thus,

these diagrams facilitated communication within the

team in order to equalize levels of knowledge among

all participants. Furthermore, these artifacts were

shown to be quite useful to experts as a source of

validation and verification regarding to the evolution

of the work over time.

Figure 5 illustrates the so-called Model View

diagram that describes the high-level ONTO-

MAMA structure.

Figure 5: Model view.

From this Model view of ONTO-MAMA, it is

possible to verify that the model was constructed

from basic structures existing in the female breast.

These structures, in turn, were categorized

according to the point of view of an observer that

may be external or internal to the breast. The model

also proposes to describe some additional

anatomical structures which contribute to a better

HEALTHINF 2012 - International Conference on Health Informatics

112

understanding of the female breast, namely:

lymphatic, venous and arterial structures. Such

structures are classified in the model as

complementary structures. The Model View also

suggests that any structure in the model make use of

a documentation element which is implemented as

notes, properties, attributes, relationships, rules and

axioms.

As for the annotations and properties / attributes,

they are considered elements of metadata in the

ONTO-MAMA model because they define the

detailing mechanisms of the concepts presented in

the model. For these elements, the project team has

defined a systematic representation that is shown in

Figure 6.

The systematic definition of the metadata from

the ontology model primarily defines that it must

contain annotations, which we called "Ontology

annotations".

The conceptual elements that represent the

anatomical parts of the female breast, in turn,

receive annotations and also other specific attributes

which were defined by medical experts.

Additionally, a special structure based in

annotation or notes is also provided to fulfill the

need of identifying the multimedia objects in others

associated models.

Figure 6: Meta data view.

This identification structure takes into account

that a metadata for identifying one object can be

expressed basically according the following formats:

URI locator, text-based and coded-based which

matches a predefined pattern. This way we provide a

flexible mechanism for identifying an object of

interest in particularly any type of media.

After defining the metadata structure, the project

team opted for direct prototyping (implementation)

of the models on Protégé and Maya, which brought

agility to the construction and review tasks.

For future work we plan the construction of a

software library which will facilitate the use of the

ONTO-MAMA models by the scientific community.

This library will consist of a set of APIs

developed in Java with the ability to interpret and/or

reference the model and its constituent parts.

In addition, the library to be built will serve as a

basis for the building of a website in which the

concepts proposed by the models can be consulted in

order to provide a faster and easier access for

medical practitioners, students and others interested

in the knowledge of breast anatomy and associated

medical procedures.

4 CONCLUSIONS

Our experience in the project showed that the

success in the creation of an ontology depends on

the team's level of commitment, on the level of the

experts' knowledge and on the working methods

applied.

Furthermore, also deeply connected with the

working method is the team's ability to understand

the requirements of their project, plan actions, and

work on the elaboration of products, which may not

have been necessarily explicit in the methodology in

use, but that can bring agility and improvement to

the communication between those involved.

Regarding to the construction of a unified

ontology-multimedia model our project shows up

that the teams must be involved since the beginning

of the tasks and definition of goals and principals.

Also, it proved very handy to use the artifacts

produced by the multimedia team as a support for

validation and check points with experts.

Additionally, it is important to reinforce the need of

having efficient and flexible mechanisms for

concepts identification so that these can be reused in

the identification of the multimedia objects and

further yet as a complementary identification of

learning objects if we are considering the

construction of intelligent tutoring systems.

Currently, the model is defined only in

Portuguese (Brazilian), but the translation work into

ONTO-MAMA - An Unified Ontology and 3D Graphic Model of the Female Breast Anatomy

113

English has already begun. The ontology model as

well as the most important anatomic parts of our 3D

model is under validation of experts working in the

area of radiology and mastology. We plan to build a

complementary ontology as part of the project with

the focus on the description of the core needle

biopsy procedure. In addition, we also intended to

build libraries and supporting tools for others being

able to use our models efficiently and easily. Finally,

it is also important to mention that the virtual reality

environment and the intelligent tutor system are

under development in order to validate our models

with regard to its usefulness, easiness and

importance.

ACKNOWLEDGEMENTS

This work has been supported by the National

Council for Scientific and Technological

Development (CNPQ) and the National Laboratory

of Computer Science (LNCC). It also counted with

the participation of one student from the Université

de Bourgogne, Centre Universitaire Condorcet,

France who is enrolled in the second year of the

Master program in Computer Vision.

REFERENCES

Gomez-Perez A., Corcho O. and Fernandez-Lopez M.

(2004). Ontological Engineering: with examples from

the areas of knowledge management, e-commerce and

semantic web. 1

edition, Springer-Verlag

Neches R., et al., (1991) Enabling technology for

knowledge sharing. AI Magazine 12(3): 36-56.

Studer R., Benjamins V.R. and Fensel D. (1998)

Knowledge Engineering: Principles and Methods.

IEEE Transactions on Data and Knowledge

Engineering 25(1-2) 161-197.

Uschold M. and Jasper R. (1999) A Framework for

Understanding and Classifying Ontology Applications.

In: Benjamins V. R. (ed) IJCAI’99 Workshop on

Ontology and Problem Solving Methods: Lessons

Learned and Future Trends. Stockholm, Sweden.

CEUR Workshop Proceedings 18:11.1- 11.12.

Amsterdam, The Netherlands.

Gruber T. R. (1993a) A translation approach to portable

ontology specification. Knowledge Acquisition 5(2):

199-200.

Borst W. N. (1997) Construction of Engineering

Ontologies. Centre for Telematica and Information

Technology, University of Tweenty. Enschede, The

Netherlands.

Mizoguchi R., Vanwelkenhuysen J. and Ikeda M. (1995)

Task Ontology for Reuse of problem solving

knowledge. In: Mars N. (ed) Towards Very Large

Knowledge Bases: Knowledge Building and

Knowledge Sharing (KBKS’95). University of

Twente, Enschede, The Netherlands. IOS Press,

Armsterdam, The Netherlands, pp 46-57.

Van Heijst G., Schreiber A. T. and Wielinga B. J. (1997)

Using explicit ontologies in KBS development.

International Journal of Humam-Computer Studies 45:

183-292.

Lassila O. and McGuiness D. (2001) The Role of Frame-

Based Representation on the Semantic Web. Technical

Report KSL-01-02. Knowledge Systems Laboratory.

Stanford University. Stanford, California.

Guarino N. and Giaretta P. (1995) Ontologies and

Knowledge Bases: Towards a Terminological

Clarification. In: Mars N. (ed) Towards Very Large

Knowledge Bases: Knowledge Building and

Knowledge Sharing (KBKS’95). University of

Twente, Enschede, The Netherlands. IOS Press,

Armsterdam, The Netherlands, pp 25-32.

Guarino N. (1998) Formal Ontology in Information

Systems. In: Guarino N. (ed) 1

International

Conference on Formal Ontology in Information

Systems (FOIS’98). Trento, Italy. IOS Press,

Amsterdam, pp 3-15.

Brasil L. M. (1994) Aquisição de conhecimento aplicada

ao diagnóstico de Epilepsia. Dissertação de Mestrado

em Engenharia Biomédica, Departamento de

Engenharia Elétrica, UFSC, Florianópolis – SC.

Gu S. (2006) Investigating the suitability of Maya for

Biomedical Modeling and Simulation. Retrieve June

12, 2011, from http://www.cs.auckland.ac.nz

/~burkhard/Reports/2006_SimonGu.pdf.

Sharpe J., Lumsden C. J. and Woolridge N. (2008) In

Silico: 3D Animation and Simulation of Cell Biology

with Maya and MEL (The Morgan Kaufmann Series in

Computer Graphics). Elsevier, Burlington, MA, USA.

Maraffi C. (2003) Maya Character Creation: modeling

and animation control. New Riders Publishing, USA.

HEALTHINF 2012 - International Conference on Health Informatics

114

APPENDIX

ONTO-MAMA - An Unified Ontology and 3D Graphic Model of the Female Breast Anatomy

115

ONTO-MAMA Sample of the Unified Ontology-Multimedia Model.

(ontology model associated to the 3D graphic model)

Composed

3D Object

Identification Structure

(

Ontolo

-Multimedia Inte

ration

)

Simple External

3D Ob

ect

Identification Structure

(

Ontolo

-Multimedia Inte

ration

)

Simple Internal

3D Ob

ect

Identification Structure

(

Ontolo

-Multimedia Inte

ration

)

HEALTHINF 2012 - International Conference on Health Informatics

116