Representação do conhecimento biomédico: uma investigação sobre a teoria da terminologia e a teoria da ontologia aplicada no domínio do sangue humano

Abstract

Modern information systems are capable of handling a large volume of data, they automatically process them and provide new insights for researchers and professionals. There are several initiatives of information and knowledge representation oriented to build information systems with such capabilities. This sort of system is fully justified in the current informational environment because of the countless benefits it can provide in fields like Medicine, which are characterized by the presence of large volume of sensitive data. In the context of information and knowledge representation, on the one hand, there are traditional and well-known theories; on the other hand, there are some theories that just recently have been disseminated as effective alternatives for representation. An example of the first kind of theory is the Theory of Terminology; an example of the second kind is the Theory of Applied Ontology. The present work discusses aspects of both theories, as well as verifying their potential in the development of formal ontologies for information systems. This works goal is to compare theoretical principles that ground the Theory of Terminology and the Theory of Applied Ontology, by approaching the creation of definitions in both theories. In order to accomplished this, after presenting some of the main principles of each theory and examples of their application in the healthcare field, we have developed a set of procedures to create definitions for biomedical entities in the scope of the Blood Project, which is a research project for knowledge organization in the domain of hematology and blood transfusion. As contribution, we describe our findings about the similarities and differences, advantages and disadvantages, dialects and applications of both theories, as well as a list of methodological steps to create well-formed definitions for biomedical ontologies.