=Paper=
{{Paper
|id=Vol-222/paper-13
|storemode=property
|title=Experience with an Ontology of Pediatric Electrolyte Disorders in a Developing Country
|pdfUrl=https://ceur-ws.org/Vol-222/krmed2006-a13.pdf
|volume=Vol-222
|dblpUrl=https://dblp.org/rec/conf/krmed/ChaichanamongkolT06
}}
==Experience with an Ontology of Pediatric Electrolyte Disorders in a Developing Country==
Experience with an Ontology of Pediatric Electrolyte Disorders
in a developing country
Vorapong Chaichanamongkol, B.Sc., M.D., Wanwipa Titthasiri, M.Sc., Ed.D.
Department of Information Technology, Rangsit University, Bangkok, Thailand
Email:vorapongch@yahoo.com
Motivation prototype of the Pediatric Electrolyte Disorder Ontol-
There are few pediatric nephrologists in Thailand and ogy. There is a reliable ADSL network in the capital
physicians in rural Thailand have limited access to Bangkok. Protégé and Swoop installed on a web
up-to-date biomedical information such, as biomedi- server are used to publish and share the ontology. The
cal journals. Moreover, biomedical information is same applications are also installed on client com-
often compiled in developed countries and may not puters and can be used both online and offline.
appropriate for use in developing countries. For ex-
ample, the guidelines established by the World Health Usability study
Organization (WHO) for the treatment of acute diar- In order to evaluate the ontology, questionnaires were
rhea in children are not always applicable in the case sent to the 25 end users, i.e., general pediatricians and
of rotavirus gastroenteritis, because the concentration family physicians in rural areas. 24 responses were
of sodium in oral saline solutions is too high for in- received and analyzed. Most physicians involved with
fants. In this context, we believe ontologies can play the study were young (age 30-40).
an important role in patient management. Semantic The Jambalaya plug-in in Protégé was found to pro-
Web ontologies foster sharing and reuse of knowl- vide good visualization support, displaying 2D inter-
edge and facilitate collaboration between pediatri- active representations of the domain of the electrolyte
cians and consultant pediatric nephrologists. Such disorders. Some pediatricians liked Swoop publish-
ontologies can be part of the telemedicine arsenal and ing, because it is easy to understand, especially in
help physicians in rural areas of Thailand to better Text mode.
manage difficult cases. In this paper, we report our Speed was sometimes an issue in those areas with a
experience in developing and using an ontology of large number of ADSL users. Even lower connec-
pediatric electrolyte disorders. tivity was available in rural areas. In some cases, the
ontology had to be sent by email in several pieces.
Developing and publishing the ontology However, once downloaded, the OWL ontology can
In the knowledge elicitation phase, we used concept be exploited offline, using Swoop or Protégé.
maps to formalize the knowledge of a small group of
eleven experts. Knowledge was contributed by pedi- Conclusions
atric nephrologists, pediatricians, general practitio- The development of our ontology of pediatric electro-
ners, as well as extracted from clinical practice guide- lyte disorders took more than one year. It was moti-
lines, text books and the medical pediatric literature. vated by the need for providing up-to-date therapeutic
Some 500 concepts were identified in the domain of to general practitioners in this specialized domain,
pediatric electrolyte disorders. and to tailor this information to the particular patient
These concepts were then organized into an ontology population. Preliminary results show that the ontol-
and related to other concepts. Textual definitions ogy has helped physicians better manage pediatric
were created. For example, the concept severe hypo- patients, especially in the rural areas of Thailand.
natremia is defined as “Sodium concentration is Despite limited connectivity in some areas and lim-
below 125 mEq/l” and is a subclass of the concept ited performance of computer systems, the experience
disease. In addition to subclass relations, we use the was globally successful, in both creating the ontology
relationship “look for” between diseases and symp- from expert knowledge and making it available to
toms. Another example is the concept Urine Sodium physicians in rural areas. Ontologies such as the one
concentration, subclass of Urine test, and for which we created for pediatric electrolyte disorders will play
an important property is “more than or less than 20 an increasing role in telemedicine.
mEq/l”. The Web Ontology Language OWL-DL was In future work, we plan to build a larger Semantic
selected for representing the ontology. In this phase, Web Ontology for Pediatric Nephrology. Rule lan-
we used Protégé-OWL (http://protege.stanford.edu/, guages such as SWRL – the Semantic Web Rule
Stanford University and University of Manchester) Language – may be used in addition to OWL in order
and SWOOP (http://www.mindswap.org/2004/- to represent clinical guidelines.
SWOOP/, University of Maryland) for building a
99