Mapping ICPC-2 PLUS to SNOMED CT
Author Information
Author(s): Wang Yefeng, Patrick Jon, Miller Graeme, O'Hallaran Julie
Primary Institution: University of Sydney
Hypothesis
Can automated mapping methods effectively link ICPC-2 PLUS terminology to SNOMED CT?
Conclusion
Automated mapping of medical terminologies can enhance data exchange and interoperability, but requires human validation for accuracy.
Supporting Evidence
- 80.58% of ICPC-2 terms were successfully mapped to SNOMED CT.
- UMLS mapping achieved a precision rate of 96.46%.
- Post-coordination mapping accounted for 20.24% of mappings.
Takeaway
This study shows how computers can help connect different medical terms so doctors can share information better, but people still need to check the results.
Methodology
The study used three automated mapping methods: UMLS metathesaurus mapping, computational linguistic mapping, and post-coordination mapping.
Potential Biases
Potential bias exists due to the reliance on automated methods that require human validation.
Limitations
The accuracy of post-coordination mapping has not been evaluated yet, and some mappings may be ambiguous due to differences in terminology structures.
Participant Demographics
The study involved terminology from Australian general practice, specifically ICPC-2 PLUS used by approximately 1,500 GPs.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website