FHIR Chat · CT Australian extension · terminology

Stream: terminology

Topic: CT Australian extension


view this post on Zulip François Macary (Nov 17 2016 at 13:12):

The medication content of SNOMED CT Australian extension has its own root concept standing by side of the SNOMED CT root concept. Thus this extension is only visible in Shrimp. Not in IHTSDO browser. Can someone explain why it was built this way?

view this post on Zulip Grahame Grieve (Nov 17 2016 at 13:42):

@Dion McMurtrie - ping

view this post on Zulip Michael Lawley (Nov 18 2016 at 01:04):

Here's the detailled technical reason:
The problem really is that the SNOMED CT root concept is defined (in my opinion) in the wrong module. The Australian National Release Centre (NRC) has (had?) a need to distribute AMT as a stand-alone extension (independent of the normal SNOMED content), but the root concep is defined in that module rather than in the metadata module. We've (myself and the NRC) been working with the IHTSDO to sort out an alternative but it's taken longer than we hoped.

view this post on Zulip Michael Lawley (Nov 18 2016 at 01:07):

However, I'm not entirely sure that this is the reason why the extension is not visible in the IHTSDO browser. I'm not sure what's required to get additional releases supported by their browser, but I imagine that SNOMED CT-AU's use of concrete domain content (for medication strengths etc) is a more likely reason.

view this post on Zulip Michael Lawley (Nov 18 2016 at 06:46):

@François Macary I've just checked the IHTSDO browser - if you search for the AMT content it's all there. For example, a search for the other root concept (30561011000036101) takes you to http://browser.ihtsdotools.org/?perspective=full&conceptId1=30561011000036101&edition=au-edition&release=v20160930&server=https://browser-aws-1.ihtsdotools.org/api/snomed&langRefset=32570271000036106 and you can drill down in the concept details panel

view this post on Zulip François Macary (Nov 18 2016 at 11:07):

Thanks Michael for your two replies. I see it in IHTSDO browser, now.
The other reason I asked is that we (Phast company) are currently working on translating and extending the medication content for France, and our natural intent is to reuse as much as possible the content from the core edition (substances, medicinal products, qualifiers), and plug our extended contents onto these core hierarchies. As I understand the 'module' management in CT, we can still publish these extensions as a single "France extension" module, even though the content is scattered across multiple hierarchies. Correct?
In other words I hope we won't have the module issue that you had.

view this post on Zulip Michael Lawley (Nov 20 2016 at 01:03):

Yes, that is correct. The only reason for the AU issue is the desire to be able to publish AMT standalone.
This will naturally go away (not be possible) once the AU content is integrated with core substances, but this integration raises a number of problematic modelling issues relating (largely) to salt forms where the is_a hierarchy in core is either 1) broken or 2) the meanings of the substance codes in AMT vs core are subtly different


Last updated: Apr 12 2022 at 19:14 UTC