...
- Automatic translation
- [Thilo] I agree with Eric that term binding is (currently!) very questionable for the purpose of automatic translation. But IMO there are other use cases for term binding (see 2. and 3.)
- Export into non-openEHR formats
- [Thilo] An archetype is self contained model and the meaning of its nodes is defined within the context that the archetype provides. I don't think an external multipurpose terminology can be more accurate. Thus, decision support should be developed based on archetypes and/or templates.
But many non-openEHR formats are less semantically rich e.g. vanilla CDA (i.e. without a constraining template). In order to provide the best possible (most semantically rich) exports into the non-openEHR world the meaning that can be derived from the terminology could be helpful.
- [Thilo] An archetype is self contained model and the meaning of its nodes is defined within the context that the archetype provides. I don't think an external multipurpose terminology can be more accurate. Thus, decision support should be developed based on archetypes and/or templates.
- Terminology-enriched archetype-based decision support
- [Thilo] Although most decision support will be based on the information in the archetype and/or template I think sometimes addional information (e.g. 'calf' is part of 'lower extremity' via ISA relation) can be gained from the terminology.
- Combining archtyped and non-/pre-archetyped data for research
- [Sam, cited from list discussion] A final part of the equation is the area that David Markwell has been working on in the NHS in the UK. He is investigating how to generate computable terminology code phrases from an archetype: that is, how to post-coordinate information captured in an archetype for inferencing in the terminology space. This has benefit in linking with the pre-archetype data and may allow complex research to be undertaken in the future using ontological tools and engines (full source: see mailing list thread).