Analyzing differences in disease definitions using ontological modeling

Linda Peelen, Michel C.A. Klein, Stefan Schlobach, Nicolette F. De Keizer, Niels Peek

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review


In medicine, many diseases cannot be denned unequivocally by etiology or anatomical localization, but are instead described by a combination of signs and symptoms that are common in patients believed to be suffering from that disease. In the communication between health care professionals, this is generally not problematic. In biomedical research, however, crisp definitions are required to distinguish patients with and without the disease unambiguously. In practice, this results in different operational definitions being in use for a single disease. Comparing those definitions, e.g, for trial design, statistical analysis, or guideline development, is complicated by their complex structure. Operational definitions of disease often consist of large disjunctions of conjunctions, or have a polythetic structure ('at least n out of m'). This paper presents an approach to compare different operational definitions of a single disease in a systematic manner using formal ontological modeling and reasoning. The concept of an operationalizution hierarchy is introduced, which is subsequently used to model and analyze operational disease definitions. The approach is illustrated with a case-study in the area of severe sepsis.

Original languageEnglish
Title of host publicationAnalyzing differences in disease definitions using ontological modeling
Number of pages8
Publication statusPublished - 2007
Event19th Belgian-Dutch Conference on Artificial Intelligence, BNAIC 2007 - Utrecht, Netherlands
Duration: 5 Nov 20076 Nov 2007

Publication series

NameBelgian/Netherlands Artificial Intelligence Conference


Conference19th Belgian-Dutch Conference on Artificial Intelligence, BNAIC 2007

Cite this