Semantic integration and analysis of clinical data

Abstract

There is a growing need to semantically process and integrate clinical data from different sources for Clinical Data Management and Clinical Decision Support in the healthcare IT industry. In the clinical practice domain, the semantic gap between clinical information systems and domain ontologies is quite often difficult to bridge in one step. In this paper, we report our experience in using a two-step formalization approach to formalize clinical data, i.e. from database schemas to local formalisms and from local formalisms to domain (unifying) formalisms. We use N3 rules to explicitly and formally state the mapping from local ontologies to domain ontologies. The resulting data expressed in domain formalisms can be integrated and analyzed, though originating from very distinct sources. Practices of applying the two-step approach in the infectious disorders and cancer domains are introduced.