Document Type

Conference Paper


Available under a Creative Commons Attribution Non-Commercial Share Alike 4.0 International Licence


Information Science, Bioinformatics

Publication Details

A paper to summarise the experience of implementing a terminology service and the findings


The integration of terminology and EHR information models is an important step in the journey towards semantic interoperability. Archetypes and two-level models for EHRs provide a mechanism that not only applies constraints on clinical content but also ensures effective terminology binding. However the lack of a standardised mechanism to bind terminology to the EHR and the difficulty of systematically coding clinical content, has led to a number of possible implementation choices. This study presents a review of the problems that may occur when working with modern terminology systems and discusses some related state of art technologies. The paper aims to share the experience of prototyping a minimum terminology integration service. A set of tools utilising medical text processing and a customised SNOMED-CT data source are the output of prototyping that enables quick processing of archetypes and automatic link suggestions to SNOMED-CT. The elaboration of prototypes of this sort can be used as components of an integration engine.