The automated capturing and summarization of medical consultations aims to reduce the administrative burden in healthcare. Consultations are structured conversations that broadly follow a guideline with a systematic examination of predefined observations and symptoms to diagnose and treat well-defined conditions. While the administrative burden could be relieved via automated conversation summarization systems, the availability of medical ontologies for the interpretation of the medical knowledge is still an obstacle. This paper introduces ontological conversation summarization by formalizing the representation of medical guidelines to develop a method for the systematic construction of ontologies from the human anatomy and medical guidelines. The well-known SNOMED CT nomenclature and medical guidelines from a medical authority in the Netherlands are used to develop: (i) a formalism for the target ontologies in the form of a Patient Medical Ontology (PMO), and (ii) a procedural method to develop such ontologies expressed in the form of a Process-Deliverable Diagram model. The PMO of the medical condition of ear canal inflammation (Otitis Externa) is created to validate the method.
Keywords: Ontology learning, knowledge extraction, method engineering · knowledge representation · SNOMED CT · patient medical ontology.
Download the publication here in pdf: