Clustering clinical models from local electronic health records based on semantic similarity

Autor: Ronald Cornet, Stig Kjær Andersen, Kirstine Rosenbeck Gøeg
Přispěvatelé: Other Research, Medical Informatics
Rok vydání: 2014
Předmět:
Zdroj: Gøeg, K R, Cornet, R & Andersen, S K 2015, ' Clustering clinical models from local electronic health records based on semantic similarity ', Journal of Biomedical Informatics, vol. 54, pp. 294-304 . https://doi.org/10.1016/j.jbi.2014.12.015
Journal of biomedical informatics, 54, 294-304. Academic Press Inc.
ISSN: 1532-0480
1532-0464
DOI: 10.1016/j.jbi.2014.12.015
Popis: Display Omitted We propose a method for clustering clinical models based on SNOMED CT.Semantic similarity and aggregation techniques facilitate hierarchical clustering.We evaluate the method using templates from local electronic health record systems.Dendrograms provide an overview of semantic similarity of templates.The clustering method can be used to compare and summarize multiple clinical models. BackgroundClinical models in electronic health records are typically expressed as templates which support the multiple clinical workflows in which the system is used. The templates are often designed using local rather than standard information models and terminology, which hinders semantic interoperability. Semantic challenges can be solved by harmonizing and standardizing clinical models. However, methods supporting harmonization based on existing clinical models are lacking. One approach is to explore semantic similarity estimation as a basis of an analytical framework. Therefore, the aim of this study is to develop and apply methods for intrinsic similarity-estimation based analysis that can compare and give an overview of multiple clinical models. MethodFor a similarity estimate to be intrinsic it should be based on an established ontology, for which SNOMED CT was chosen. In this study, Lin similarity estimates and Sokal and Sneath similarity estimates were used together with two aggregation techniques (average and best-match-average respectively) resulting in a total of four methods. The similarity estimations are used to hierarchically cluster templates. The test material consists of templates from Danish and Swedish EHR systems. The test material was used to evaluate how the four different methods perform. Result and discussionThe best-match-average aggregation technique performed better in terms of clustering similar templates than the average aggregation technique. No difference could be seen in terms of the choice of similarity estimate in this study, but the finding may be different for other datasets. The dendrograms resulting from the hierarchical clustering gave an overview of the templates and a basis of further analysis. ConclusionHierarchical clustering of templates based on SNOMED CT and semantic similarity estimation with best-match-average aggregation technique can be used for comparison and summarization of multiple templates. Consequently, it can provide a valuable tool for harmonization and standardization of clinical models.
Databáze: OpenAIRE
Externí odkaz: