Versions Compared


  • This line was added.
  • This line was removed.
  • Formatting was changed.


  • Record this call
  • Agenda review
  • Mixins and fragments"Mixins" or "Fragments" that would be common building blocks for provenance, roles, participations, etc
    • Reguirements
      • Layers of information - layer realm specific information on top of a base model
      • Ability to use common structures with different context/semantics
        • Reusing the core part of a condition description as reason for order, final diagnosis, chief complaint, description of complication, etc.
        • It is difficult to represent the common data by profiling on the condition resource
      • Want to be able to share content across different projects/domains: sharing across cardiology, imagining
        • Reuse and specialization of bodyLocation for different kinds of breast lesions: cyst, mass, wound, etc.
      • Reusable data types
      • Keeping attributes consistent when the attributes are contained in some children of a general type but not in all children
        • One option is choosing the qualifying attributes from a library of qualifiers
      • Support as many levels of composition as are required by the use case
  • Updates on a number of different fronts - Kurt and Richard
    • Have made good progress on the high level model (template) for things like breast radiology, anatomic pathology, etc.
    • Mark - may want to use abstract profile for some items (take up this topic in some "next" meeting)
      • Quantitative lab, coded lab, panel, etc. (helper labs, patterns)
      • Also show examples of specific patterns
      • Also show the tables for detailed content
    • Next week we will do a deep dive on high level structures - Richard and Kurt
      • Need to have a "meaining" code for extensions - Mark has submitted a tracker request to FHIR-I
    • Next actions
      • Say how our requirements map to FHIR resources
        • We should make a position paper on this need, including good examples - Claude can make a first draft
        • Socialize the proposal with other HL7 Working Groups
        • Need to engage in the current conversion on chat and Julip - everyone should help
  • What do we ballot?
    • Motion: Our policy will be to ballot high level patterns/structures/styles of models (i.e. quant lab, codeable lab, titre lab) and any classes that are subject to inclusion in requlation, and the we not ballot the high volume leaf level models (hct, hemoglobin, glucose level, etc.).  Stan moved, Susan second.  Vote: passes 4, 0, 0.
    • We need to socialize this policy with the other Work Groups that we commonly work with.
  • Any other business

Conference Call Information