Anne Randorff Rasmussen

SNOMED CT implementation: implications of choosing clinical findings or observable entities

Internationally, it is a priority to develop and implement semantically interoperable health information systems.[1] One required technology is the use of standardised clinical terminologies. The terminology, SNOMED CT, has shown superior coverage compared to other terminologies in multiple clinical fields. The aim of this paper is to analyse SNOMED CT implementation in an Electronic Health Record (EHR). More specifically, differences and consequences of applying clinical findings (CFs) as an alternative to observable entities (OEs) is analysed. Results show that CFs represents the content of the templates with better coverage, with more parent concepts and with a higher degree of fully defined terms than the OEs. We discuss the possibility to further evaluate the observable entity hierarchy to overcome a potential overlapping use of the two hierarchies.

Conceptual Graphs as Algebras - With an Application to Analogical Reasoning

The first part of this paper presents a logico-algebraic reconstruction of conceptual graph fundamentals using an appropriately extended binary relation algebra. The algebraisation comprises axioms which in a straightforvifard and instructive fashion admit deductive reasoning by term rewriting. As an application of the described logicoalgebraic framework, the second part of the paper addresses formalisation of analogical reasoning. It is suggested that analogical reasoning be carried out as clausal resolution inference using axioms of the binary relation algebra together with the apphcation specific axioms corresponding to the domains under consideration. This makes the analogical reasoning amenable to implementation in logic programming languages.

Methods for testing SNOMED CT in a defined clinical domain.

This paper presents the results of a study in which SNOMED CT has been tested in clinical practice by combining quantitative and qualitative methods. An application has been developed for registration of vascular diseases and procedures using SNOMED CT concepts. We conclude that it requires clinical and technical insight to adapt the terminology to a defined clinical domain.