Affiliations: INSERM UMR_S 872, Paris, F-75006 France. E-mails: {Audrey.Baneyx, Jean.Charlet}@spim.jussieu.fr | Assistance Publique, Hôpitaux de Paris, Paris, F-75014 France
Note: [] Corresponding author: A. Baneyx, INSERM UMR_S 872, eq. 20, Paris, F-75006 France. E-mail: Audrey.Baneyx@ spim.jussieu.fr.
Abstract: In the following paper, we describe OntoPneumo, our ontology of pneumology. We also focus on the method used to build it and discuss its validity. Indeed, we think that building an ontology in the biomedical field is very difficult or even impossible without considering the application system objectives. This is why we describe our goal in building this ontology, and our building methodology based on Natural Language Processing tools. We insist on the need to express precisely, in natural language, the meaning of each concept using differential principles inherited from the Differential Semantics theory. OntoPneumo is used in a software for pneumologists that represents medical knowledge to describe the medical diagnoses in the form of graphs and to propose the proper French DRG (Diagnosis Related Group) code. The middle ontology now contains 1460 primitive concepts. The top-ontology is based on the top-ontology of the MENELAS project and contains 400 concepts.
