Stream: implementers
Topic: Terminology Servers
Kesara Liyanage (Oct 14 2018 at 22:47):
I'm a bit confused on the function of a terminology server. Do you connect with one to retrieve appropriate value sets for a given code system definition found in a structure definition? If so, are there open source options?
Grahame Grieve (Oct 14 2018 at 22:52):
yes, and there are plenty of options. we haven't made a formal list, but see
Grahame Grieve (Oct 14 2018 at 22:53):
actually:
Grahame Grieve (Oct 14 2018 at 22:53):
see http://wiki.hl7.org/index.php?title=Open_Source_FHIR_implementations for a start, but no one has assembled a complete list. I'd list Ontoserver, Apelon DTS, Terminz, Art Decor, VSAC, HAPI, fhir.loinc.org and my own server at http://healthintersections.com.au/FhirServer - but these are all different sorts of servers that are adaptable to some degree or other. Ontoserver is the best general purpose server
Kesara Liyanage (Oct 14 2018 at 23:35):
Thanks Grahame very helpful
Anil Patel (Feb 19 2019 at 21:06):
Hi, does anyone know of anyone has implemented a terminology server solution that takes a Lab V2.4 message inbound and then validates it using a FHIR terminology server and sends a outbound HL7 V2 message to the backend database to store. I have an interesting problem that requires replacing a HUB solution that validated terminology, location and provider information, the problem is that the HUB does a translation from a local code to a LOINC code, so the HUB must maintain mappings.
Grahame Grieve (Feb 19 2019 at 21:07):
this has not happened yet. We haven't described how to handle table 396 rather than fhir code system URIs
Lloyd McKenzie (Feb 19 2019 at 21:53):
We sort of have. The NamingSystem resource can capture the v2 codes as well as the FHIR URIs for a given code system and could be used as the basis for transformation. What we haven't done is publish NamingSystems for the code systems in table 396 that show the codes, OIDs and URIs for each.
Grahame Grieve (Feb 19 2019 at 22:03):
What we haven't done is defined what it looks like on the terminology service API
Anil Patel (Feb 20 2019 at 02:06):
It would seem to me that defining V2 in a terminology service API would be something everyone is looking for. Are there other counties or organisations working on this.? Can I help to contribute to this effort? what is needed?
Yunwei Wang (Feb 20 2019 at 20:54):
If the inbound and outbound are V2 messages, why do you need a FHIR terminology server sit in the middle to do validation?
Jim Steel (Feb 21 2019 at 05:17):
Depends what you're validating, I guess. I suppose if you're doing code mapping on the way through, you might want to validate that a code from the incoming message is valid (against a ValueSet) before you translate into a code for inclusion in the outgoing message
Jim Steel (Feb 21 2019 at 05:17):
(Just guessing)
Grahame Grieve (Feb 21 2019 at 06:26):
I don't know why code validation is less important in the v2 world than the FHIR world
Anil Patel (Feb 21 2019 at 16:32):
Yes @Jim Steel yes you would want to validate the code from the valueset, but if the terminology server had a namespace for each sending system, with maps from the local code to a valueset, then it might be possible to do translation, but it would still need to forward the message as V2 to the backend.
Anil Patel (Feb 21 2019 at 16:32):
Code Validation is important
Anil Patel (Feb 21 2019 at 16:33):
@Lloyd McKenzie Do you know of any organisations that are doing these for a Lab V2 inbound message
Lloyd McKenzie (Feb 21 2019 at 16:35):
I don't, but if they were, I probably wouldn't - it's not a world I'm terribly tied to.
Jim Steel (Feb 21 2019 at 22:36):
The code translation we (Ontoserver) do - I'm sure other TS products do too - but the decoding/reencoding as V2 is (IMHO) outside the scope of FHIR terminology services (which doesn't mean a product won't do it)
Grahame Grieve (Feb 22 2019 at 01:20):
right. but where is impacts the terminology service API is how the systems are identified
Last updated: Apr 12 2022 at 19:14 UTC
