Automating lexical cross-mapping of ICNP to SNOMED CT

作者：Kim Tae Youn^*

来源：Informatics for Health & Social Care, 2016, 41(1): 64-77.

DOI：10.3109/17538157.2014.948173

• Terminology mapping
• UMLS
• ICNP
• SNOMED CT

Objectives: The purpose of this study was to examine the feasibility of automating lexical cross-mapping of a logic-based nursing terminology (ICNP) to SNOMED CT using the Unified Medical Language System (UMLS) maintained by the U.S. National Library of Medicine. Methods: A two-stage approach included patterns identification, and application and evaluation of an automated term matching procedure. The performance of the automated procedure was evaluated using a test set against a gold standard (i.e. concept equivalency table) created independently by terminology experts. Results: There were lexical similarities between ICNP diagnostic concepts and SNOMED CT. The automated term matching procedure was reliable as presented in recall of 65%, precision of 79%, accuracy of 82%, F-measure of 0.71 and the area under the receiver operating characteristics (ROC) curve of 0.78 (95% CI 0.73-0.83). When the automated procedure was not able to retrieve lexically matched concepts, it was also unlikely for terminology experts to identify a matched SNOMED CT concept. Conclusions: Although further research is warranted to enhance the automated matching procedure, the combination of cross-maps from UMLS and the automated procedure is useful to generate candidate mappings and thus, assist ongoing maintenance of mappings which is a significant burden to terminology developers.

• 出版日期2016-1-2