Ontology Matching
Associating Clinical Archetypes Through UMLS Metathesaurus Term Clusters
Journal of Medical Systems
Hi-index | 0.00 |
Integrating clinical data models, such as the European open-EHR Archetypes and standard terminologies, such as SNOMED CT, is important to reduce medical errors and to get interoperability between health information systems. In this study, we propose an automated approach to mapping observation archetypes to SNOMED CT. Our approach applies a sequential combination of several basic matching methods, classically used in ontology matching. First, different lexical techniques identify similar strings between the observation archetypes and SNOMED CT. Second, a structure-based technique traverses two types of SNOMED CT relationships: IS A and interprets, searching for possible mappings not found with lexical techniques. The method was applied to the mapping of 20 observation archetypes. In total, 94% precision and 69% recall of SNOMED CT concepts was reached. Our method has revealed a degree of semantic similarity between some relationships in the tree representing observation archetypes and the relationships IS A and interprets in SNOMED CT.