Content uploaded by Vinícius Costa Lima
Author content
All content in this area was uploaded by Vinícius Costa Lima on Jan 06, 2019
Content may be subject to copyright.
ISBN 978-989-97433-9-7 E-book edition 2018 by SciKA
Book of industry papers, poster papers and abstracts of the
CENTERIS 2018 – Conference on Enterprise Information Systems /
ProjMAN 2018 – International Conference on Project MANagement /
HCist 2018 – International Conference on Health and Social Care Information Systems and Technologies
287
Development of a Knowledge Portal for Tuberculosis based
on Semantic Web
Ricardo Roberto de Limaa,
*
, Francisco Barbosa Juniorb, Felipe Carvalho
Pellisona, Vinicius Costa Limaa, Vinicius Tohoru Yoshiurab, Nathalia Yukie
Crepaldib, Lídia Maria Lourençön Rodriguesb, Rui Pedro Charters Lopes
Rijoc, Domingos Alvesb
aBioengineering Postgraduate Program, University of São Paulo, São Carlos, Brazil
bRibeirão Preto Medical School, University of São Paulo, Ribeirão Preto, Brazil
cSchool of Technology and Management, Polytechnic Institute of Leiria, Leiria, Portugal
Abstract
According to the World Health Organization (WHO) report published in 2017, tuberculosis (TB) is the ninth cause
of death worldwide and the main cause of death among infectious-contagious diseases. Brazil is ranked between
the top 20 with the highest estimated numbers of incident TB cases. In order to support health professionals in the
fight against TB, information technology becomes an important ally in the creation of strategies and tools that can
help in the management and application of local and global public policies recommended by WHO for TB patients
control. In this work, a support tool will be presented as a semantic web portal. The innovation of this portal in
relation to others TB portals in Brazil is the use of semantically tagged data that will enable the standardized
understanding of knowledge of TB data and TB indicators, both by humans and computers, on Brazilian scenario.
Keywords: Tuberculosis; Semantic Web; Health Information System.
* Corresponding author. Tel.: +55 16 99157 1529.
E-mail address: ricardo.usp@usp.br .
ISBN 978-989-97433-9-7 E-book edition 2018 by SciKA
Book of industry papers, poster papers and abstracts of the
CENTERIS 2018 – Conference on Enterprise Information Systems /
ProjMAN 2018 – International Conference on Project MANagement /
HCist 2018 – International Conference on Health and Social Care Information Systems and Technologies
288
1. Introduction
The large amount of data and the various methods of recording and storing data on the World Wide
Web increase the chances of redundancies, losses and inconsistencies of information, hampering
improvement actions and access to accurate information [1]. Thus, Berners-Lee, Hendler and Lassila
(2001) proposed the evolution of the current Web of Documents for a Programmable Web through the
application of semantics to the data to facilitate their understanding and interoperability. The Semantic
Web connects and associates information that were previously disconnected, establishing a standard for
communication based on several domains knowledge, defined through ontologies [2][3].
There are TB content portals worldwide, such as WHO/Tuberculosis, TB Alliance, REDE-TB, GEOTB, and others
[4][5][6][7]. However, no studies were found using the Semantic Web paradigm in the context of TB portals.
Therefore, the objective of this work is to develop a portal with semantic markup created as a tool to support health
professionals and to disseminate TB related content and indicators. Different from traditional TB portals, this one
will present information with aggregated knowledge in a standardize, optimized and direct way for consumption
by both humans and computers with best accuracy in the searches results performed on the web.
2. Background
The assistance to TB patients is carried out in different levels of the Brazilian public health system
and involves the use of different software that records data in different databases. This scenario
generates barriers to retrieve information, lack of data standardization and difficulties to produce health
indicators to support healthcare professionals and in the creation of new public policies to combat
tuberculosis.
In Brazil, the main databases used for TB control are: SisTB - Software for control and follow-up of
TB patients in the city of Ribeirão Preto; HygiaWeb - Public management software of the city of
Ribeirão Preto; TBWEB - Software used by the government of the State of São Paulo for
epidemiological surveillance; SINAM - National system of registration for notification of cases of
diseases and diseases of compulsory notification; and GAL - Software to manage the System of Public
Health National Laboratories of the National Networks of Laboratories of Epidemiologic Surveillance
and Environmental Surveillance Health.
Through Semantic web it is possible to connect these virtual repositories to share information and
knowledge using structured and formalized content through a set of vocabularies mapped and
standardized by ontologies [8]. Ontologies applied in the health field have been the object of interest of
several studies like ontologies to map population health indicators [9], as well as TB researches to map
TB ontologies [10][11]. In order to structure the content and the knowledge of the proposed portal, some
technologies are used: Microdata, Resource Description Framework (RDF), SPARQL Protocol and
RDF Query Language and Web Ontology Language (OWL).
ISBN 978-989-97433-9-7 E-book edition 2018 by SciKA
Book of industry papers, poster papers and abstracts of the
CENTERIS 2018 – Conference on Enterprise Information Systems /
ProjMAN 2018 – International Conference on Project MANagement /
HCist 2018 – International Conference on Health and Social Care Information Systems and Technologies
289
3. Research Methods
The research methodology chosen for this work is action research. Used across several research
areas, action research is a type of inquiry action applied in any process that investigates a practice and
acting, practically, to create improvements in the observed process [12]. This project presents an
investigative and practical character aligned to the methodology.
Unlike other portals that address information about TB, this portal also addresses TB-related issues,
but includes in its model the use of semantic structures. These structures create in the portal an
intelligence that enables the extraction of knowledge based on the consumption of ontologies for the
general content. In addition, the portal includes the presentation of some health indicators defined on
the Manual of Recommendations for the Control of Tuberculosis in Brazil to facilitate the quantification
and evaluation of the information generated for this purpose [13][14]. Furthermore, these indicators will
be semantically marked.
The stages that precede the development include the understanding of the main Semantic Web
technologies and tools, study of health ontologies repositories and selection of ontologies associated
with tuberculosis. The project management is based on agile management methodology through
Scrumban [15]. The logical modeling uses Unified Modelling Language (UML) [16]. Drupal, an open
source content management system (CMS) that provides a native functionality for semantic markup and
allows users to easily manage the portal, was the tool chosen to develop the Web Portal [17]. An
interoperability framework provides functional and semantic access to all databases, such as SisTB,
HygiaWeb, TBWEB, SINAM and GAL, used in the portal. Some of the purposes of this framework are
to provide a set of standardized ontologies and to retrieve standardized semantic data to be used in
semantically tagged HTML pages [18]. Basic Formal Ontology (BFO) will be used as the formal
reference to map tuberculosis knowledge within the domain proposed by the work [19].
4. Conclusion
The innovation presented in this work is a web portal developed with a user friendly interface that
addresses TB content as well as related indicators with capacity to manage knowledge through content
semantic markup. This view is aligned with the Semantic Web proposal that defines that both humans
and machines must have the same understanding and knowledge about things. Google searches show
results in organic order and so the user still needs to filter the search content. Unlike Google, a
semantically structured portal knows exactly what to show to the user according to the search query
reducing the need of content filtering by the user. For computers, semantic structures allow the
extraction of relevant information from the portal content, enabling the reuse and exchange of
information through a machine-friendly interface.
ISBN 978-989-97433-9-7 E-book edition 2018 by SciKA
Book of industry papers, poster papers and abstracts of the
CENTERIS 2018 – Conference on Enterprise Information Systems /
ProjMAN 2018 – International Conference on Project MANagement /
HCist 2018 – International Conference on Health and Social Care Information Systems and Technologies
290
References
[1] Arens, Y., Chee, C. Y., Hsu, C. N., & Knoblock, C. A. (1993). Retrieving and integrating data from multiple information
sources. International Journal of Intelligent and Cooperative Information Systems, 2(02), 127-158.
[2] Berners Lee, T., Hendler, J., & Lassila, O. (2001). The semantic web. Scientific American, 284(5), 2837.
[3] Gruber, T. R. (1993). A translation approach to portable ontology specifications. Knowledge acquisition, 5(2), 199-220.
[4] WHO, World Health Organization [homepage]. Tuberculosis (TB) [accessed in May 11, 2018]. Available at:
http://www.who.int/tb/en/.
[5] TB ALLIANCE [homepage]. 2018 [accessed in May 11, 2018]. Available at: https://www.tballiance.org/.
[6] REDE-TB [homepage]. 2018 [accessed in May 11, 2018]. Available at: http://www.redetb.org.br/.
[7] GEOTB [homepage]. 2018 [accessed in May 11, 2018]. Available at: http://www.eerp.usp.br/geotb.
[8] Baofu, Peter. (2015). The Future of Post-Human Meta-Data: Towards a New Theory of Structure. Brisbane, Australia: The
Primrose Hall Publishing Group.
[9] Shaban-Nejad A et al. (2017). PopHR: a knowledge-based platform to support integration, analysis, and visualization of
population health data. Annals of the New York Academy of Sciences. https://doi.org/10.1111/nyas.13271.
[10] Abhishek, K., & Singh, M. P. An Ontology based Decision support for Tuberculosis Management and Control in India.
[11] Pandiyan, M., El-Hassan, O., Khamis, A. H., & Rajasekaran, P. (2016). Ontology with SVM Based Diagnosis of
Tuberculosis and Statistical Analysis. International Journal of Medical and Health Sciences Research, 3(3), 37-43.
[12] Tripp, David. (2005). Action research: a methodological introduction. Educação e Pesquisa, 31(3), 443-466.
https://dx.doi.org/10.1590/S1517-97022005000300009.
[13] Rondelli, JPP (2017). Desenvolvimento de uma Ferramenta de Geração de Relatórios para Auxiliar o Acompanhamento do
Tratamento de Tuberculose.
[14] Brasil. Ministério da Saúde. Secretaria de Vigilância em Saúde. Departamento de Vigilância Epidemiológica. Brasília;
Ministério da Saúde; 2011. 284 p.
[15] C. Ladas. (2008) Scrumban: Essays on Kanban Systems for Lean Software Development. Seattle, WA: Modus Cooperandi
Press.
[16] Rumbaugh J, Jacobson I, Booch G. (2004). Unified Modeling Language Reference Manual, The. Pearson Higher
Education.
[17] Drupal (2015). Drupal’s Content Management Features. Accessed in March 16, 2015. Available at:
https://www.drupal.org/features.
[18] Pellison FC et al. (2017). Development and evaluation of an interoperable system based on the semantic web to enhance
the management of patients’ tuberculosis data. Procedia Computer Science, 121, 791–796.
https://doi.org/10.1016/j.procs.2017.11.102.
[19] Arp R, Smith B, Spear AD (2015) “Building Ontologies with Basic Formal Ontology” London, England, MIT Press.