FHIR Chat · HL7 Connectathon January 2022 · smart/health-cards

Stream: smart/health-cards

Topic: HL7 Connectathon January 2022


view this post on Zulip Paul Denning (Jan 11 2022 at 15:13):

https://github.com/dvci/shc-terminology

view this post on Zulip Paul Denning (Jan 11 2022 at 15:13):

http://fhir.cerner.com/millennium/r4/foundation/other/health-cards/

view this post on Zulip Paul Denning (Jan 11 2022 at 15:14):

https://confluence.hl7.org/display/FHIR/2022-01+SMART+Health+Cards+for+Vaccination+and+Testing

view this post on Zulip Paul Denning (Jan 11 2022 at 15:22):

https://jira.hl7.org/browse/UP-271 Add Value Set for Specimen Collection Supervision Status

view this post on Zulip Paul Denning (Jan 11 2022 at 15:30):

https://github.com/HealthIntersections/australian-fhir-support/blob/main/fhir/ConceptMap/covid-cvx-air-mapping.json

view this post on Zulip Max Masnick (Jan 11 2022 at 15:35):

CI build of the IG for the Connectathon: http://build.fhir.org/ig/HL7/fhir-shc-vaccination-ig/branches/2022-01-connectathon/index.html

This has all the open PRs merged in. I didn't read through the whole IG in this branch, so it's possible the merges did something weird...if you see something that looks out of place please let me know.

view this post on Zulip Neelima Karipineni (Jan 11 2022 at 16:03):

Sample VCI Directory formatted file including Connectathon test endpoints - https://raw.githubusercontent.com/dvci/vci-directory/connectathon_2022_01/vci-issuers.json.

view this post on Zulip Paul Denning (Jan 11 2022 at 16:18):

https://github.com/cqframework/ecqm-content-r4 example of a "content IG"

view this post on Zulip Neelima Karipineni (Jan 11 2022 at 16:44):

Seems to be based on https://github.com/cqframework/sample-content-ig

view this post on Zulip Max Masnick (Jan 11 2022 at 17:02):

@Grahame Grieve and @Josh Mandel , we were playing with the idea of putting the shc-terminology content inside a lightweight IG to get a better human-readable version.

https://dvci.github.io/shc-terminology/ is what the first pass looks like (no CI build yet -- we can set this up if we decide to continue)

Code is in this branch: https://github.com/dvci/shc-terminology/tree/fhir-ig

What do you think?

view this post on Zulip Max Masnick (Jan 11 2022 at 17:03):

Direct links to the two vocab resources as there's no other content:

view this post on Zulip Josh Mandel (Jan 11 2022 at 18:06):

I find this very helpful!

view this post on Zulip Josh Mandel (Jan 11 2022 at 18:07):

In general, does anything prevent us from populating expansion for any VS we publish this way?

view this post on Zulip Max Masnick (Jan 11 2022 at 20:14):

In general, does anything prevent us from populating expansion for any VS we publish this way?

I think just terminology server support?

view this post on Zulip Max Masnick (Jan 11 2022 at 20:19):

The other thing I'd like to do is tweak the styling for the human display of the value sets and concept maps. Perhaps we can contribute this back to the IG Publisher proper.

view this post on Zulip Neelima Karipineni (Jan 11 2022 at 20:24):

The page says 'Expansion based on http://hl7.org/fhir/sid/cvx version 20210406', which is a bit old now.

view this post on Zulip Max Masnick (Jan 11 2022 at 20:25):

@Rob Hausam can you please pull the lever on updating CVX for tx.fhir.org?

view this post on Zulip Rob Hausam (Jan 11 2022 at 20:43):

Yes - probably can start to look at that tomorrow - after the Connectathon. :)

view this post on Zulip Rob Hausam (Jan 11 2022 at 20:44):

But let me know if you actually need it for tomorrow.

view this post on Zulip Josh Mandel (Jan 11 2022 at 23:05):

Max Masnick: I think just terminology server support?

OK, I'd suggest that we should include a full expansion to make these useful.

view this post on Zulip Michael O'Hanlon (Jan 12 2022 at 18:01):

https://connectathon.vci.org/ is now live for anybody who would like to test!

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:36):

@Josh Mandel

OK, I'd suggest that we should include a full expansion to make these useful

Since the expansion is a full expansion, what are you talking about?

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:37):

@Max Masnick a human readable version is a good thing for sure, and this one works if it gets navigable

view this post on Zulip Josh Mandel (Jan 13 2022 at 19:37):

When I looked at the files the other day I saw a definition but not an expansion. Maybe I was looking the wrong place multitasking

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:38):

agree the rendering could be improved, though some of the issues are with the content itself

view this post on Zulip Josh Mandel (Jan 13 2022 at 19:38):

https://dvci.github.io/shc-terminology/ValueSet-shc-covid-cvx.json.html this is where I was expecting to see a populated expansion

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:38):

well, it is, but it's got code problems. I think that's something that would be fixed by upgrading cvx

view this post on Zulip Josh Mandel (Jan 13 2022 at 19:38):

I understand that all the concepts may already be present, but this is much harder for a consuming client to know and understand, and the contract should just be here's a bag of codes in a flat list

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:49):

umm I don't follow that at all

view this post on Zulip Josh Mandel (Jan 13 2022 at 19:49):

As a developer I want to download a value set and iterate over an expansion. I don't want to try to understand things like compositional definitions...

view this post on Zulip Josh Mandel (Jan 13 2022 at 19:50):

Like, I might not have good support for processing the full semantics of value set definitions (handling: compose, include, filter, exclude etc) but it's easy for me to support processing an expansion.

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:52):

which expansion?

view this post on Zulip Josh Mandel (Jan 13 2022 at 19:52):

Ideally the one that would exist at https://dvci.github.io/shc-terminology/ValueSet-shc-covid-cvx.json

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:53):

but there's lots of valid expansions. Depends on your context which is the right one.

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:54):

the general expansion will be pretty much identical to the composition rules.

view this post on Zulip Josh Mandel (Jan 13 2022 at 19:54):

That's what I'm asking for, pretty much identical. Just available under the contract of here is an expansion and you don't need to worry about handling compose and other features (filter, exclude...) that could slip into the definitions over time and make them more complex to process

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:55):

the point of the infrastructure is that we can't decide what the right expansion is at publication time, so you can get a free high performance terminology server that does this on the fly for you

view this post on Zulip Josh Mandel (Jan 13 2022 at 19:55):

But the point of my wanting to create a GitHub repository for this specific purpose was to have a nice flat JSON file that I could just drop into my software:-)

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:55):

because what you're asking for is 'I want all the complexity to go away, and the publication to just work for me alone'

view this post on Zulip Josh Mandel (Jan 13 2022 at 19:55):

As we start moving that into a formal IG, these other issues are coming up

view this post on Zulip Josh Mandel (Jan 13 2022 at 19:56):

If we are saying the right way to use this content is to download it, then feed it into a high performance terminology server, then generate an output that you can load into your software... That's a different level of complexity and developer familiarity with FHIR than I think we can expect here

view this post on Zulip Grahame Grieve (Jan 13 2022 at 19:58):

well, like i said, you can wish complexity away, but that won't actually make it go away

view this post on Zulip Oliver Egger (Jan 13 2022 at 20:08):

Could we make it for the 80% of the valueset's which can be expanded not the same way it is done for html presentation of the expanded ValueSet? I was very astonished that the representation of the expansion in the ValueSet html, but it is not included in the ValueSet resource itself in the expansion.

view this post on Zulip Grahame Grieve (Jan 13 2022 at 20:42):

the vocab committee has tried to insist before the the html not include an expansion, and the template is supposed to include a note to the effect that this is just an example of an expansion that could occur. Note that it's 'an expansion' not 'the expansion' since there's an infinite number of correct expansions for any non-trivial value set.

And this particular value set is a non-trivial value set


Last updated: Apr 12 2022 at 19:14 UTC