This wiki has undergone a migration to Confluence found Here
<meta name="googlebot" content="noindex">

Difference between revisions of "2017-04-04 TermInfo Project Call"

From HL7Wiki
Jump to navigation Jump to search
Line 116: Line 116:
  
 
* Further review or answer questions regarding FHIR IG publishing process and tools (if time permitting)
 
* Further review or answer questions regarding FHIR IG publishing process and tools (if time permitting)
 +
 +
==Minutes==
 +
<!---================================================================
 +
|                                                                  |
 +
|  MM    MM  IIIIII  NN    NN  UU  UU  TTTTTT  EEEEEE  SSSSS    |
 +
|  MMM  MMM    II    NNN  NN  UU  UU    TT    EE      SS        |
 +
|  MM MMM MM    II    NN NN NN  UU  UU    TT    EEEE    SSSS      |
 +
|  MM  M  MM    II    NN  NNN  UU  UU    TT    EE          SS    |
 +
|  MM    MM  IIIIII  NN    NN  UUUU    TT    EEEEEE  SSSSS      |
 +
|                                                                  |
 +
=================================================================--->
 +
<!-- **** Delete instructions  and fill in minutes ON NEXT LINES  ******-->
 +
 +
 +
'''Initial discussion/review of reasoning across FHIR and SNOMED CT using OWL'''
 +
* Rob presented "Tutorial" on blending FHIR and RDF in Protégé. Work from RDF group by Harold Solbrig. Mechanisms for crossing the information model and terminology. Agreement this is very interesting, and an approach that could facilitate "computable TermInfo".
 +
 +
Questions: How have FHIR resources been bound to the SCT condept model? The translation from creating OWL from SNOMED is not based on IHTSDOs Pearl-script. 
 +
 +
    [https://github.com/BD2KOnFHIR/BLENDINGFHIRandRDF Link to tutorial]
 +
 +
 +
'''Contunued discussion on document "Using SNOMED CT with HL7 FHIR Resources" - Condtion resource'''
 +
* Condition resource/ manifestation or symptom. Is not fully compatible with the SCT concept model. Value set is all of descendents of Clinical Finding – should remove some subhierachies.
 +
* Body site value – expansion of anatomical structure. Should cell structure be included? In almost all use cases that would not be relevant. Body structure can be precoordinated into the condition itself > semantic overlap. > use of RDF for crossover.
 +
 +
''HL7 value sets in the Condition resource that should be considered converted to SCT:''
 +
*Condition category. Definition of category. Clinical finding or disorder or findings? Ex preganancy? This field is not useful unless it restricts the Category across the board in FHIR – simple set of category values to do something with the data (without even understanding the code itself) you can still figure out what to do with the data – bucket. They are however treated in different ways. This one is CodeableConcept.
 +
*Problem-list-item/ encounter diagnosis
 +
*Clincal status and verification status  also not using SNOMED CT.
 +
 +
How can we maintain terminology binding over time? SCT term amended to FHIR – how to make this formal and maintainable. Terminfo artefact? 
 +
 +
 +
Should we extend the call 90 min?

Revision as of 19:59, 4 April 2017

Meeting Information

TermInfo Project Conference Call

GTM and call-in details:

1. Please join my meeting. https://global.gotomeeting.com/meeting/join/605884653

2. Use your microphone and speakers (VoIP) - a headset is recommended. Or, call in using your telephone.

Australia: +61 2 8355 1040 Canada: +1 (647) 497-9391 Denmark: +45 (0) 69 91 88 61 Sweden: +46 (0) 852 503 498 United Kingdom: +44 20 3535 0624 United States: +1 (646) 982-0002

Date: 2016-03-28
Time: 4:00-5:00pm ET
Chair Rob Hausam Scribe Øyvind Aassve
Attendee Name Affiliation


Rob Hausam Hausam Consulting LLC
Linda Bird SNOMED International
Yongsheng Gao SNOMED International
Peter Jordan HL7 New Zealand
Øyvind Aassve Norwegian Directorate of e-health
Quorum Requirements Met: yes

Agenda

Agenda Topics
Announcements

  • Minutes (as available)
  • Still need to make progress on this. Will post minutes on Wiki. The project group minutes should not require a formal vote for approval
  • Suggest post to the Wiki and send a notification to the list and allow folks to update/correct (or suggest updates/corrections) as needed?
  • Review action items (send any updates in advance!)
  • Review and update TermInfo PSS
  • Update TermInfo Wiki and HL7 TermInfo project site information
  • Initial discussion/review of reasoning across FHIR and SNOMED CT using OWL

"Tutorial" on blending FHIR and RDF from Harold Solbrig

  • Continue FHIR SNOMED CT Model Meaning Binding and Value Set Review
  • The "Using SNOMED CT with HL7 FHIR Resources" document was distributed on Feb. 13


  • Further review or answer questions regarding FHIR IG publishing process and tools (if time permitting)

Minutes

Initial discussion/review of reasoning across FHIR and SNOMED CT using OWL

  • Rob presented "Tutorial" on blending FHIR and RDF in Protégé. Work from RDF group by Harold Solbrig. Mechanisms for crossing the information model and terminology. Agreement this is very interesting, and an approach that could facilitate "computable TermInfo".

Questions: How have FHIR resources been bound to the SCT condept model? The translation from creating OWL from SNOMED is not based on IHTSDOs Pearl-script. 

   Link to tutorial


Contunued discussion on document "Using SNOMED CT with HL7 FHIR Resources" - Condtion resource

  • Condition resource/ manifestation or symptom. Is not fully compatible with the SCT concept model. Value set is all of descendents of Clinical Finding – should remove some subhierachies.
  • Body site value – expansion of anatomical structure. Should cell structure be included? In almost all use cases that would not be relevant. Body structure can be precoordinated into the condition itself > semantic overlap. > use of RDF for crossover.

HL7 value sets in the Condition resource that should be considered converted to SCT:

  • Condition category. Definition of category. Clinical finding or disorder or findings? Ex preganancy? This field is not useful unless it restricts the Category across the board in FHIR – simple set of category values to do something with the data (without even understanding the code itself) you can still figure out what to do with the data – bucket. They are however treated in different ways. This one is CodeableConcept.
  • Problem-list-item/ encounter diagnosis
  • Clincal status and verification status also not using SNOMED CT.

How can we maintain terminology binding over time? SCT term amended to FHIR – how to make this formal and maintainable. Terminfo artefact?


Should we extend the call 90 min?

Retrieved from "https://wiki.hl7.org/w/index.php?title=2017-04-04_TermInfo_Project_Call&oldid=137105"