Biomedical Engineering through Ontologies
Seremeti Lambrini
and Kameas Achilles
School of Science and Technology, Hellenic Open University, Parodos Aristotelous 18, GR-26335, Patras, Greece
Keywords: Biomedical Engineering, Ontologies, Ontology Engineering.
Abstract: Biomedical engineering is the application of scientific and mathematical principles to practical ends in the
medicine and biology fields. It comprises many research directions including computational model of HIV
infection, integration of clinical and experimental data, etc. Much of the work in biomedical engineering
consists in providing solutions to the problem of securing an effective integration of biomedical content. To
this end, ontologies, as sharable, reusable and machine-readable artifacts capable of knowledge
representation, contribute to the interoperability between systems, the access of heterogeneous information
sources, and the reuse of voluminous and complex information. The aim of this paper is to present the
literature on biomedical ontologies in order to highlight how current research of the ontology field can be
brought to bear on the practical problems associated with biomedical engineering. Thus, we discuss the
fundamental role of ontologies in biomedical engineering, we review several methodologies suitable for
building specialized biomedical ontologies, and we present some peculiarities related to the creation of
biomedical ontologies that continue to constitute research challenges.
1 INTRODUCTION
Biomedical engineering aims at combining research
advances from disciplines such as biology,
medicine, computer science, mathematics,
bioinformatics and areas such as knowledge
representation, information retrieval, data mining,
reasoning and visualization, by attracting
contributions that are currently scattered in different
fields. Typically it involves the design,
implementation and operation of efficient
biomedical structures, processes and systems for
diagnostic, monitoring, or therapeutic purposes.
Currently, biomedical engineering focuses on
activities that aim at developing novel algorithmic
processes which lead to the creation of new
knowledge, by harmonizing basic types of
biomedical and clinical data, such as biomedical
images, biological sequences and biosignals. Their
successful application is strongly tied to the effective
adoption of semantic technologies, such as
terminologies, thesauri, ontologies, in order to
represent, acquire, process and manage knowledge
and data in the biomedical domain (Bodenreider,
2008).
Although various terminologies, such as MeSH
(Medical Subject Headings)
(http://www.nlm.nih.gov/mesh/) and UMLS
(Unified Medical Language System)
(http://www.nlm.nih.gov/research/umls/) have been
used in order to provide a sort of semantics in
biomedicine (Freitas et al., 2009), the need to
develop ontologies has accelerated in recent years,
due to the availability of huge biomedical data sets
that cannot manually be analyzed, interpreted or
processed to acquire inferred knowledge efficiently
(Saripalle, 2013). Unlike terminologies, which
collect the name of entities employed in the domain,
ontologies are concerned with the principled
description of classes of entities (i.e., substances,
qualities, processes) and the relations among them
(Bodenreider and Burgun, 2005).
While the number of biomedical ontologies
continues to increase, as new areas of biomedical
content become formalized, their creation still
remains a challenging task, due to the complex and
dynamic nature of the biomedical domain.
The objective of this paper is not to examine how
applications in biomedicine benefit from using
ontologies, but rather to present the critical issues
related with the construction of specialized
biomedical ontologies. In particular, we review the
fundamental role of ontologies in the biomedical
240
Lambrini S. and Achilles K..
Biomedical Engineering through Ontologies.
DOI: 10.5220/0005073802400247
In Proceedings of the International Conference on Knowledge Engineering and Ontology Development (KEOD-2014), pages 240-247
ISBN: 978-989-758-049-9
Copyright
c
2014 SCITEPRESS (Science and Technology Publications, Lda.)
engineering field, by studying the challenging issue
of creating a specialized biomedical ontology.
The rest of this paper is organized as follows:
Section 2 provides basic definitions and Section 3 a
brief overview of ontologies that semantically
support biomedical engineering. Section 4 refers to
peculiarities related to the creation of biomedical
ontologies that continue to constitute research
challenges. Finally, Section 5 draws some
conclusions.
2 ROLE OF ONTOLOGIES IN
BIOMEDICAL ENGINEERING
Although many definitions of ontologies exist in the
scientific literature (Neches et al., 1991), (Gruber,
1993), (Guarino and Giaretta, 1995),
(Chandrasekaran et al., 1999), (Stevens et al., 2000),
(Fonseca, 2007) quite a few notions are common to
these definitions. Conclusively, an ontology can be
defined as an explicit representation of a shared
understanding of the important concepts in some
domain of interest. The role of an ontology is to
support knowledge sharing and reuse within and
among groups of agents (people, software programs,
or both). In their computational form, ontologies
often comprise definitions of terms organized in a
hierarchy lattice, along with a set of relationships
that hold between these definitions. These constructs
collectively impose a structure on the domain being
represented and constrain the possible interpretations
of terms. Ideally, an ontology should capture a
shared understanding of a domain of interest and
provide a formal and machine manipulable model of
the domain.
In the case of biomedical engineering,
ontologies, as computational knowledge resources
that capture the precise meaning of biomedical
terms, are used in order to enhance the design and
execution of experiments, data analysis, synthesis of
biomedical information, creation of research
hypotheses and discovery of new knowledge (Bard
and Rhee, 2004), (Blake, 2004).
Moreover, since ontologies are developed in
order to share common understanding of the
structure of the biomedical information among
people, or software agents, they are widely used in
order to resolve semantic conflicts that inhibit
interoperability among heterogeneous biomedical
systems (Horrocks, 2013).
According to Goh (Goh, 1997), there are three
main causes which are responsible for these
semantic conflicts: (a) Confounding conflicts, which
occur when different meaning is attributed to the
same term, according to the context in which it is
interpreted. For example, the Paget’s disease may be
interpreted, either as a rare type of breast cancer, or
as a chronic disorder affecting bones commonly
quoted as osteodystrophia deformans; (b) Scaling
conflicts, which refer to different reference systems
for measuring some kind of values. For example, the
measurement of normal glucose level in blood can
be represented either as 60-110 mg/dL, or as 3,5-6
mmol/L; and (c) Naming conflicts, which refer to
the use of synonyms in different knowledge
representation systems. For example, diabetes and
polygenic disorder, despite their lexicographic
incongruity, have the same meaning, that is, a
metabolic disease whereby a person has high levels
of blood sugar due to an inability to produce, or
metabolize sufficient quantities of the hormone
insulin.
In general, the main reason for developing
ontologies in the biomedical field is to capture the
exact meaning of biomedical terms, aiming mainly
at the integration and management of terminologies
in e-health and biomedicine, interoperability and
sharing, information retrieval, decision making and
automated logical inference (Cinimo, 2000),
(Rector, 2003), (Brand, 2003), (Rubin et al., 2007).
Although ontologies can have several
applications, such as representation of encyclopedic
knowledge, semantic search and query, data
exchange, data integration, and reasoning, the
creation of biomedical ontologies still remains a
critical issue, as discussed later on.
3 SPECIALIZED BIOMEDICAL
ONTOLOGY ENGINEERING
The construction of a specialized biomedical
ontology is difficult due to the dynamics and
complexity of the biomedical domain. It requires
methodological guidelines for the specification (i.e.,
identification of the scope and domain of the
ontology under construction), conceptualization (i.e.,
definition of the basic concepts structured in a
hierarchy, relations, instances and axioms),
implementation (i.e., selection of the ontology
development tool and the language in which the
ontology will be implemented) and evaluation (i.e.
the final outcome is evaluated against a gold
standard, or within an application) of the produced
artifacts.
BiomedicalEngineeringthroughOntologies
241
The establishment of international centers, such
as the National Center for Biomedical Ontology
(NCBO) (http://www.bioontology.org), the Open
Biomedical Ontologies (OBO) Foundry
(http://obofoundry.org), the Semantic Web Health
Care and Life Sciences Interest Group (HCLS IG)
(http://www.w3.org/blog/hcls/), the Stanford Center
for Biomedical Informatics Research
(http://bmir.stanford.edu/), etc. promote best
practices in the development of biomedical
ontologies. The Web Ontology Language (OWL)
(http://www.w3.org/TR/owl-features/) is widely
accepted as the standard language for representing
biomedical ontologies and Protégé
(http://protege.stanford.edu/) as the most suitable
and commonly used ontology development
environment in the biomedical field.
The OBO Foundry proposes a set of principles
for ontology development that guide the modular
creation of biomedical ontologies. These principles
refer to the design to which biomedical ontologies
should adhere, such as openness (i.e., the ontology
must be open and available to be used by all without
any constraint), the employment of a shared syntax
(i.e., the ontology must be expressed in the OBO
syntax, or OWL for facilitating shared software
implementations), orthogonality (i.e., the ontology
must be orthogonal to other ontologies already
lodged within OBO in order to allow two different
ontologies to be combined through additional
relationships), the inclusion of textual definitions
(i.e., the ontology should include textual definitions
for all terms, so that their precise meaning will be
clear to a human reader), and the application of
standardized relations (i.e., the ontology must use
relations which are unambiguously defined in the
OBO Relation Ontology) (Smith et al., 2005).
Hereafter we present a set of techniques for
constructing specialized biomedical ontologies and
indicative real world applications.
In (Valarakos et al., 2006) a methodology is
proposed for the design and implementation of
a formally defined ontology on allergens
consisting of the exploitation of existing
taxonomies and documents that describe the
allergen nomenclature, the semi-automatic
maintenance of the allergens ontology by
discovering knowledge from domain specific
corpora using machine learning techniques,
and the evaluation of the maintenance process
through the study of the factors that may
affect its performance;
In (Sahay et al., 2007), an ontology for the
Nuclear Cardiology domain is built
automatically from text using existing medical
resources, as well as parsing and extracting
methods. First, a set of abstracts from the
Journal of Nuclear Cardiology is collected.
Then, from these abstracts, by using statistical
natural language processing techniques,
concepts and relations are extracted in the
form of Subject-Verb-Object and thereby a
semantic network is created;
In (Baneyx et al., 2007) an ontology of
pulmonary diseases is created, by applying a
methodology which combines two approaches
to enrich ontology building: (1) a method
which consists in building terminological
resources through distributional analysis and
(2) a method based on the observation of
corpus sequences, in order to reveal semantic
relationships;
In (Satria et al., 2012) the development of a
medical ontology in the domain of tropical
diseases is discussed. The ontology building
process is based on the reuse of existing
biomedical ontologies with overlapping
content, such as the Infectious Disease
Ontology (Cowell and Smith, 2010), the
Human Malaria Control Ontology (Daramola
and Fatumo, 2009) and the Disease Ontology
(Schulz et al., 2006);
In (Warrender, 2013) the Mitochondrial
Disease Ontology is constructed by applying
the following steps: (1) term capture in order
to acquire knowledge of the domain of
interest, (2) definition of competency
questions which the ontology should
reasonably be able to answer, (3) refinement
in order to filter the terms and competency
questions for discarding the non-relevant ones,
(4) combination of manual and automated
construction in order to incorporate data that
already exists in autonomous structured
formats, and (5) evaluation of the terminology
and the taxonomy of the ontology, as well as,
the ontology’s fitness for purpose;
In (Xiang and He, 2013) the Human Interaction
Network Ontology (HINO) is generated by
extending the BFO (Basic Formal Ontology)-
aligned Interaction Network Ontology (INO)
and by importing terms from ontologies, such
as CHEBI and GO. It represents various
human molecular interaction pathways and
networks as classes.
From analyzing the various techniques used
during the construction of the specialized biomedical
ontologies, it is evident that there appears to be no
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
242
strict interrelationship between the technique used
during a specialized biomedical ontology building
and the special characteristics of the corresponding
biomedical subfield. A general conclusion could be
that, when the biomedical subfield in question is too
specialized (e.g., allergens, nuclear cardiology,
pulmonary diseases), then the construction of the
corresponding ontologies is based on automated
analysis of text, while in the case of fields, such as
tropical diseases, or human molecular interactions
and pathways, where there exist many available
ontologies with overlapping content, their reuse is
the common technique used. There are also some
exceptions (e.g. Mitochondrial Disease Ontology),
whose generation is based on the combination of
two techniques (reuse and construction from
scratch); these are cases where the corresponding
biomedical subfield is too specialized thus there do
not exist ontologies related to it.
4 PECULIARITIES IN
CREATING BIOMEDICAL
ONTOLOGIES
In this section, we present specific issues related to
the creation of biomedical ontologies that continue
to constitute research challenges.
Modularization: The biomedical ontologies
tend to be complex and large in size, so it is
difficult to maintain them or incorporate
changes to them (Pathak et al., 2008). A
desirable feature of biomedical ontologies is
to be organized into discrete units (modules),
in order to be easily manipulable and
extensible (Rector, 2003), (Schulz and Lopez-
Garcia, 2011). The basic problem in building
modular biomedical ontologies is the need of a
predefined set of relationships among the
different modules (Bittner and Smith, 2003);
Reusability: Biomedical ontologies should be
reusable in order to adhere to their general
purpose, that is, to support interoperability
issues. For example, the construction of a
coronary artery disease ontology may be
facilitated by reusing the Disease Ontology;
the ontology of genetic susceptibility factor
can be constructed by reusing the Basic
Formal Ontology (Lin and He, 2013). This
feature requires the use of a standard
representation language, so that the structure
and semantics of the ontology are fully
understood;
Size: As biomedical ontologies become larger,
their development and maintenance become
more difficult and error prone. To address this
challenging issue human judgment is still
necessary (Mortensen et al., 2013). Moreover
the large size of biomedical ontologies makes
the generation of correct alignments a time-
consuming process (Ba and Diallo, 2013);
Uncertainty in knowledge representation: In
the biomedical domain, the use of certain
indicators such as HDL (High Density
Lipoprotein) and LDL (Low Density
Lipoprotein) is commonplace. The values of
these indicators give a threshold between two
states (normal and abnormal). The
representation of this kind of medical
knowledge requires the use of fuzzy logic,
which allows representing the imprecision that
is inherent to the definition of some
biomedical concepts (Kayaalp, 2005),
(Hudelot et al., 2008);
Alignability: Biomedical ontologies should be
alignable (Musen, 2006), that is, semantic
correspondences among different biomedical
ontologies should be easily generated. This
contributes to the evaluation of biomedical
ontologies, since the more alignments an
ontology has, the more reliable it is, because
this means that it has been reused multiple
times by different applications. The
alignability of ontologies is mainly achieved
by using terms from other commonly accepted
ontologies;
Perdurants and endurants: A consistent
representation of biomedical knowledge
requires the clear distinction between the
entities for representing endurants (or
continuants) and perdurants (or occurrents)
(Yu, 2006). Endurants are those entities that
continue to exist through time as a whole (e.g.
an organ), while perdurants are entities that
unfold themselves in successive temporal
phases (e.g. breathing). A clear
interdependence of these two types of entities
arises, since perdurants depend on the
existence of endurants. For example, a
molecular protein, which is an endurant, tends
to perform a specific function (e.g. binding),
which is a perdurant;
Semantic ambiguity: Resolving semantic
ambiguity is a very challenging issue which
refers to the existence of polysemy and
homonymy during the knowledge
representation process (Beisswanger et al.,
BiomedicalEngineeringthroughOntologies
243
2008). Polysemy terms may have the same
lexicographic form but refer to totally
different concepts, even in the same field (e.g.,
the term “gene” in Biomedicine may encode
either the concept of “the recorded segment of
DNA that translates protein” or the concept of
“biological interest DNA region that carries a
genetic phenotype”. In contrast, synonyms
usually have different lexicographic form but
refer to the same concept (e.g., the terms
“cell” and “closet” may be both interpreted as
“one of the parts into which an enclosed space
is divided”, depending on the context);
Multidimensionality: In the biomedical
domain a concept exhitits diverse
characteristics, so during its ontological
representation it may be structured in different
ways (e.g., an organ may be considered either
as an anatomical entity, or as a functional
entity). The case in which a node in a tree
hierarchy appears to have more than two
parent-nodes is called multidimensionality
(Madsen et al., 2005). For example, the term
“pulmonary tuberculosis” can be classified
either as a respiratory, or as an infectious
disease;
Diversity: Diversity in biomedical ontologies
occurs when a concept may be expressed in
many variants of the same term. The most
common cases involve morphological (e.g.,
gene vs genes), orthographic (e.g., hemolysis
vs Hemolysis), lexical (e.g., cancer vs
carcinoma), structural (enzyme activity vs
activity of enzyme), and semantic diversity
(e.g., genetic disease vs hereditary disease)
(Wächter, 2010);
Abbreviations and acronyms: The increasing
number of neologisms in Biomedicine for
gene names, diseases, etc. has led to the use of
abbreviated terms and acronyms, which has
created uncertaintly in modeling knowledge,
since for any given acronym or abbreviation,
there are often multiple possible long form
expansions. For example, the acronym TTP
can refer, either to “thiamine triphosphate”, or
to “thymidine triphosphate” (Melton et al.,
2010);
Ordinary statements: Statements such as
“Lmo-2 interacts with Elf-2” which refer to
the behavior of biomedical entities (e.g., about
the interaction between two proteins) occur in
biomedical literature. However, from an
ontological perspective, these statements may
have more than one possible interpretations
(e.g., this interaction did actually happen, or
the molecules Lmo-2 and Elf-2 have the
tendency to interact in such a way). Thus,
such statements reveal a kind of ambiguity
that has to be taken into account in the
practice of biomedical ontology engineering
(Schulz and Janses, 2009);
Multi-word terms and nested terms: The
majority of concepts used in the biomedical
domain consist of multiword terms (e.g., bel-2
protein level, or RA mediated tumor cell
invasion), or nested terms (e.g., Leukemic [T
cell] line Kit255), where the brackets denote
nested terms in a multi-word term (Ananiadou
et al., 2000). Although, their interpretation is
obvious to a domain expert, it remains a
challenging issue for a machine;
Reverse transformations: Pathological
transformations in Biomedicine, such as the
transformations of erythrocytes into
echinocytes and back again, or of a healthy
organ into an unhealthy organ and back again
complicate the ontological engineering
process, since it is unknown in advance
whether all instances of a class have been
transformed into pathological ones and
whether all instances of a pathological entity
return to the original normal state, for example
after a medical intervention (e.g., 5 of 100
patients with lung cancer survive and keep
their lungs) (Keet, 2009);
Mereotopological and location relations:
Explicitly representing general truths such as
“a given clump of tissue is part of the left
gyrus frontalis medius, which in turn, is part
of the brain, which is part of the nervous
system, which is part of the body”, “the
stomach and the small intestine are connected
but they have no parts in common”, “in males,
the urinary system and the reproductive
system overlap”, “a myocardial parasite is
located in a hole in the heart tissue”, etc.
requires the definition of parthood and
location relations (Donnelly, 2004). Parthood
relations describe the spatial arrangement of
parts of the human body at different levels of
granularity. The location relations are used to
describe the location of body parts within the
body and the location of foreign occupants in
the body. This knowledge is important for
biomedical informatics, because it can be used
for example in automated reasoning about
parts of the body affected by disease (Schulz
and Hahn, 2001);
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
244
Updating biomedical knowledge: Although
the use of ontologies in biology and medicine
for the semantic integration of heterogeneous
data receives increased attention, problems
occur due to the changing and evolving nature
of biomedical knowledge (McGarry et al.,
2006). Advances in molecular biology,
genetics and genomics through the use of new
technologies, such as microarrays, provide
vast quantities of experimental data. Thus,
existing biomedical ontologies require review
and updating of their contained knowledge.
For example, an ontology for diabetes should
be updated on any new advances regarding the
effects of insulin resistance on protein
expression and insulin regulated protein
trafficking in fat cells (McGarry et al., 2007);
Searching and selecting biomedical
ontologies: Nowadays, the number and
variety of biomedical ontologies is so large,
that selecting one for an annotation task, for
the construction of a new ontology, or for
designing a specific application, is a
challenging issue. It mainly relies on
evaluation and comparison of the available
ontologies (Tan and Lambrix, 2009).
Although the National Center for Biomedical
Ontology through its web portal offers
integrated access to a library of biomedical
ontologies (Rubin et al., 2006), the process of
choosing the most suitable one is often a time
consuming task. Therefore, during the
ontology engineering process, the use of an
ontology recommendation system to facilitate
the identification of the ontology that is best
suited to a specific application is needed
(Jonquet et al., 2010);
The dynamics of relationships within the
construction team: Care should be taken to
the efficient communication among ontology
experts, knowledge engineers and domain
experts during the biomedical ontology
engineering process, in order to assure that the
ontology is logically consistent, it adequately
represents the domain of biomedicine under
consideration and it explicitly aligns to
ontological principles. The opposite may lead
to wrong design decisions in biomedical
ontologies (Boeker et al., 2012).
Development of specialized biomedical
ontologies is still quite an empirical process. Due to
the complex and dynamic nature of the biomedical
domain, a flexible methodology is required which
should take into account all the above mentioned
peculiarities related to the explicit representation of
the biomedical knowledge.
5 CONCLUSIONS
Biomedical engineering involves activities that
mainly require effective integration of biomedical
content. This paper highlighted the fundamental role
of ontologies in addressing this critical issue, by
reviewing related work on biomedical ontology
building, which is essential in designing next
generation biomedical applications, such as
magnetic nanoparticles, comparative analysis of
human genome sequences, advanced imaging
technologies, etc. Its main focus is on presenting the
peculiarities emerged during biomedical ontology
engineering.
ACKNOWLEDGEMENTS
This research has been co-financed by the European
Union (European Social Fund – ESF) and Greek
national funds through the Operational Program
“Education and Lifelong Learning” of the National
Strategic Reference Framework (NSRF) – Research
Funding Program: THALES. Investing in
knowledge society through the European Social
Fund.
REFERENCES
Ananiadou, S., Albert, S., Schumann, D., 2000. Evaluation
of automatic term recognition of nuclear receptors
from MEDLINE. In Genome Informatics, Vol. 11, pp.
450-451.
Ba, M., Diallo, G., 2013. Large-scale biomedical ontology
matching with ServOMAP. In IRBM, Vol. 34, Issue 1,
pp. 56-59.
Baneyx, A., Charlet, J., Jaulent, M., 2007. Building an
ontology of pulmonary diseases with natural language
processing tools using textual corpora. In International
Journal of Medical Informatics, Vol. 76, No. 2, pp.
208-215.
Bard, J., Rhee, S.Y., 2004. Ontologies in biology: Design,
applications and future challenges. In Nature Reviews
Genetics, Vol. 5, pp. 213-222.
Beisswanger, E., Porpat, M., Hahn, U., 2008. Lexical
properties of OBO ontology class names and
synonyms. In Foreword to the 3
rd
International
Symposium for semantic mining in Biomedicine
(SMBM), pp. 13-20.
BiomedicalEngineeringthroughOntologies
245
Bittner, T., Smith, B., 2003. A theory of granular
partition. In M. Duckham, M. Goodchild, M. Worboys
(eds), Foundations of geographic information science,
London, Taylor & Francis Books, pp. 117-151.
Blake, J., 2004. Bio-ontologies – fast and furious. In Nat
Biotechnol, Vol. 22, pp. 773-774.
Bodenreider, O., Burgun, A., 2005. Biomedical ontologies.
In H. Chen, S.S. Fuller, C. Friedman, W. Hersh (eds),
Medical Informatics: Integrated Series in Information
Systems, Vol. 8, pp. 211-236, Springer.
Bodenreider, O., 2008. Biomedical ontologies in action:
Role in knowledge management, data integration and
decision support. Yearbook of Medical Informatics,
Issue 1, pp. 67-79.
Boeker, M., Schober, D., Raufie, D., Grewe, N., Röhl, J.,
Janses, L., Schulz, S., 2012. Teaching good
biomedical ontology design. In R. Cornet, R. Stevens
(eds), ICBO, Proceedings of the CEUR Workshop,
Vol. 897.
Brand, J., 2003. Ontologies formalizing biological
knowledge for bioinformatics. In Bioessays, Vol. 25,
No. 5, pp. 501-506.
Chandrasekaran, B., Josephson, J.R., Benjamins, V.R.,
1999. Ontologies: What are they? Why do we need
them? In IEEE Intelligent Systems and Their
Applications, Special Issue on Ontologies, Vol. 14,
No. 1, pp. 20-26.
Cinimo, J.J., 2000. From data to knowledge through
concept-oriented terminologies: experience with the
Medical Entities Dictionary. In J Am Med Inform
Assoc, Vol. 7, No. 3, pp. 288-297.
Cowell, L.G., Smith, B., 2010. Infectious disease
ontology. In V. Sintchenko, Infectious Disease
Informatics, New York: Springer, pp. 373-395.
Daramola, O., Fatumo, S., 2009. Developing ontology
support for human malaria control initiatives. In
International Conference on Biomedical Ontology.
Donnelly, M., 2004. On parts and holes: The spatial
structure of the human body. In M. Fieschi et al.,
(eds), MEDINFO2004, IOS Press, pp. 351-356.
Fonseca, F., 2007. The double role of ontologies in
inflrmation science research. In Journal of the
American Society for Information Science and
Technology, Vol. 58, No. 6, pp. 786-793.
Freitas, F., Schultz, S., Moreas, E., 2009. Survey of
current terminologies and ontologies in biology and
medicine. In RECIIS, Electronic Journal of
Communication Information & Innovation in Health,
Vo. 3, No. 1.
Goh, C.H., 1997. Representing and reasoning about
semantic conflicts in heterogeneous information
sources. MIT.
Gruber, T.R., 1993. Towards principles for the design of
ontologies used for knowledge sharing. In R. Poli, N.
Guarino (eds), International Workshop on Formal
Ontology.
Guarino, N., Giaretta, P., 1995. Ontologies and knowledge
bases: Towards a terminological classification. In
Towards Very Large Knowledge Bases: Knowledge
Building & Knowledge Sharing, IOS Press.
Horrocks, I., 2013. What are ontologies good for? In B.O.
Küppers et al. (eds), Evolution of Semantic Systems,
pp. 175-188, Springer-Verlag Berlin Heidelberg.
Hudelot, C., Atif, J., Bloch, I., 2008. Fuzzy spatial relation
ontology for image interpretation. In Fuzzy Sets and
Systems, Vol. 159, pp. 1929-1951.
Jonquet, C., Musen, M.A., Shah, N.H., 2010. Building a
biomedical ontology recommender web service. In
Journal of Biomedical Semantics, Vol. 1.
Kayaalp, M., 2005. Why do we need probabilistic
approaches to ontologies and the associated data? In
AMIA Annual Symposium Proceedings.
Keet, M.C., 2009. Constraints for representing
transforming entities in bio-ontologies. Knowledge
Representation Centre Technical Report, Free
University of Bozen-Bolzno, Italy.
Lin, Y., He, Y., 2013. Ontology modelling of genetic
susceptibility to adverse events following vaccination.
In Proceedings of the International Conference on
Biomedical Ontology (ICBO).
Madsen, B.N., Thomsen, H.E., Vinker, C., 2005.
Multidimensionality in terminological concept
modelling. In Proceedings of the 7
th
International
Conference on Terminology and Knowledge
Engineering, pp. 161-173.
McGarry, K., Garfield, S., Morris, N., 2006. Recent trends
in knowledge and data integration for the life sciences.
In Expert Systems: The Journal of Knowledge
Engineering, Vol. 23, No. 5, pp. 337-348.
McGarry, K., Garfield, S., Morris, N., Wemster, S., 2007.
Integration of hubrid bio-ontologies using baeysian
networks for knowledge discovery. In Proceedings of
the 3
rd
International Workshop on Neural-Symbolic
Learning and Reasoning.
Melton, G.B., Moon, S., McInnes, B., Pakhomov, S.,
2010. Automated identification of synonyms in
biomedical acronym sense inventories. In Proceedings
of the NAACL HLT2010 Second ouhi Workshop on
Text and Data Mining of Health Documents, pp. 46-
52.
Mortensen, J.M., Musen, M.A., Noy, N.F., 2013.
Crowdsourcing the verification of relationships in
biomedical ontologies. In AMIA Annual Symposium.
Musen, M.A., 2006. Don’t know much about philosophy:
The confusion over bio-ontologies. In Proceedings of
the Computational Systems Bioinformatics (CSB), pp.
3-5.
Neches, R., Fikes, R., Finin, T., Gruber, T., Patil, R.,
Senator, T., Swartout, W.R., 1991. Enabling
technology for knowledge sharing. In AI Magazine,
pp. 36-56.
Pathak, J., Johnson, T.M., Chute, C.G., 2008. Modular
ontology techniques and their applications in the
biomedical domain. In IEEE IRI2008, pp. 351-357.
Rector, A., 2003. Medical Informatics. In F. Baader, D.
Calvanese, D. McGuinness, D. Nardi, P. Patel-
Shneider (eds), The description logic handbook,
Cambridge University Press, pp. 406-426.
Rubin, D.L., Lewis, S.E., Mungall, C.J., Misra, S.,
Westerfield, M., Ashburner, M., Sim, I., Chute, C.G.,
KEOD2014-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment
246
Solbriq, H., Storey, M.A., Smith, B., Day-Richter, J.,
Noy, N.F., Musen, M.A., 2006. National center for
biomedical ontology: Advancing biomedicine through
structured organization of scientific knowledge. In
Omics, Vol. 10, No. 2, pp. 185-198.
Rubin, D., Shah, N., Noy, N., 2007. Biomedical
ontologies: A functional perspective. In Briefings in
Bioinformatics, Vol. 19, No. 1, pp. 75-90.
Sahay, S., Li, B., Garcia, E.V., Agichtein, E., Ram, A.,
2007. Domain ontology construction from biomedical
text. In International Conference on Artificial
Intelligence (ICAI), pp. 28-34.
Saripalle, R.K., 2013. Current status of ontologies in
biomedical and clinical informatics. In J.J. Cinimo,
E.H. Shortliffe, (eds), Biomedical Informatics:
Computed Applications in Health Care and
Biomedicine (Health Informatics.
Satria, H., Priya, R.S., Ismail, L.H., Supriyanto, E., 2012.
Building and reusing medical ontology for tropical
diseases management. In International Journal of
Education and Information Technologies, Vol. 6, Issue
1, pp. 52-61.
Schulz, S., Hahn, U., 2001. Mereotopological reasoning
about parts and (w)holes in bio-ontologies. In C.W.B.
Smith (ed), Formal Ontology in Information System,
ACM Press, pp. 210-221.
Schulz, S., Kumar, A., Bittner, T., 2006. Biomedical
ontologies: What part-of is and isn’t. In Journal of
Biomedical Informatics, Vol. 39, Issue 3, pp. 350-361.
Schulz, S., Jansen, L., 2009. Molecular interactions: On
the ambiguity of ordinary statements in biomedical
literature. In Applied Ontology, Vol. 4, No. 1, pp. 21-
34.
Schulz, S., Lopez-Garcia, P., 2011. Modularity issues in
biomedical ontologies. In Proceedings of the 5
th
International Workshop (WoMo), IOS Press, pp. 7-9.
Smith, B., Ceusters, W., Klagges, B., Köhler, J., Kumar,
A., Lomax, J., Mungall, C., Neuhaus, F., Rector, A.L.,
Rosse, C., 2005. Relations in biomedical ontologies.
In Genome Biology, Vol. 6.
Stevens, R., Goble, C., Becchofer, S., 2000. Ontology-
based knowledge representation for bioinformatics. In
Briefings in Bioinformatics, Vol. 1, No. 4, pp. 398-
414.
Tan, H., Lambrix, P., 2009. Selecting an ontology for
biomedical text mining. In Proceedings of the
Workshop on BioNLP, pp. 55-62.
Valarakos, A., Karkaletsis, V., Alexopoulou, D.,
Papadimitriou, E., Spyropoulos, C., Vouros, G., 2006.
Building an allergens ontology and maintaining it
using machine learning techniques. In Computers in
Biology and Medicine, Vol. 36, No. 10, pp. 1155-
1184.
Wächter, T., 2010. Semi-automated ontology generation
for biocuration and semantic search. PhD dissertation,
Technischen Universität Dresden.
Warrender, J.D., 2013. Building a logical model of
mitochondrial disease. School of Computing Science,
Newcastle University.
Xiang, Z., He, Y., 2013. HINO: BFO-aligned ontology
representation of human molecular interactions and
pathways. In Proceedings of the 4
th
International
Conference on Biomedical Ontology.
Yu, A., 2006. Methods in biomedical ontology. In Journal
of Biomedical Informatics, Vol. 39, pp. 252-266.
BiomedicalEngineeringthroughOntologies
247
