Stream: shorthand
Topic: Running Sushi against other kinds of StructureDefinition
Keith Boone (Mar 03 2020 at 13:43):
Structured Definitions exist for both CDA (see ) and HL7 V2. I'm wondering if I can "Profile" against those SDs in FHIR Shorthand, or create examples against them?
Chris Moesel (Mar 03 2020 at 15:45):
Oh. Hmmm... that's a really interesting question. I'm assuming that you're not talking about CDA or HL7 v2 profiles on FHIR resources, but rather a set of new resource definitions and/or logical models. Is that right?
We certainly haven't tested that -- but in theory I think that a lot of the same processing would apply. For it to have any chance of working though, they'd at least have to use the FHIR data types (e.g., string, code, CodeableConcept, Quantity, etc). And currently, we can only pull dependencies that are in fhir.packages.org -- so they'd either need to be there or SUSHI would need to be updated to find dependencies elsewhere... That would be the starting point at which we could at least test the feasibility.
All that said, we (MITRE) are currently trying to focus on the core features -- so we likely won't have time to seriously look at something like this for a while. But it's an interesting thing to think about...
Keith Boone (Mar 03 2020 at 15:51):
I think the CDA SD (see http://build.fhir.org/ig/HL7/fhir-cda/index.html) uses new data types, and I'm certain that V2 Refactored (see http://www.hl7.eu/refactored/hl7.html)
Keith Boone (Mar 03 2020 at 15:52):
Sushi and the FHIR IG Builder solve a LOT of problems for HL7 if we can get it to work against those SDs...
Hmmm...
Chris Moesel (Mar 03 2020 at 16:09):
New data types is definitely a problem. A lot of what SUSHI does is pretty dependent on types. We need to know a string is a string, a Coding is a concept, a Quantity is a quantity, etc. Abstracting that might be messy. Not impossible (for sure), but quite potentially messy.
Last updated: Apr 12 2022 at 19:14 UTC
