Logical analysis of mappings between medical classification systems

Abstract

Medical classification systems provide an essential instrument for unambiguously labeling clinical concepts in processes and services in healthcare and for improving the accessibility and elaboration of the medical content in clinical information systems. Over the last two decades the standardization efforts have established a number of classification systems as well as conversion mappings between them. Although these mappings represent the agreement reached between human specialists who devised them, there is no explicit formal reference establishing the precise meaning of the mappings. In this work we close this semantic gap by applying the results that have been recently reached in the area of AI and the Semantic Web on the formalization and analysis of mappings between heterogeneous conceptualizations. Practically, we focus on two classification systems which have received great widespread and preference within the European Union, namely ICPC-2 (International Classification of Primary Care) and ICD-10 (International Classification of Diseases). The particular contributions of this work are: the logical encoding in OWL of ICPC-2 and ICD-10 classifications; the formalization of the existing ICPC-ICD conversion mappings in terms of OWL axioms and further verification of its coherence using the logical reasoning; and finally, the outline of the other semantic techniques for automated analysis of implications of future mapping changes between ICPC and ICD classifications. © Springer-Verlag Berlin Heidelberg 2008.