Page tree

Date & Time

Wednesday 30th March 2016, 20:00 UTC

GoToMeeting Details

Click here to see GoToMeeting joining information

Click here to see GoToMeeting recordings

Goals

To introduce the SPLG Confluence Space.

To discuss recent feedback about the SNOMED CT Expression Constraint Language.

To progress the SNOMED CT Template Syntax.

Agenda and Meeting Notes

ItemDescriptionOwnerNotesAction

1

Welcome, introductions and apologiesLinda Bird

SLPG meetings will be recorded and recordings will be accessible to SLPG members

  • Check attendance details and apologies
2Agenda reviewLinda Bird

Review proposed agenda for today's meeting

  • Introducing the SLPG Confluence Space
  • Discuss recent feedback about the Expression Constraint Language
    • Immediate children/parents
    • Comments
  • Review template syntax discussion
    • Scope and purpose of syntax
    • Simplifications to Finding context example
    • Cardinality in templates
    • Populating a template from a data structure
  • Review agenda
3SPLG Confluence SpaceLinda Bird

Make sure everyone has access to Confluence and the SPLG space

 

  • Introduce SPLG space
  • Make sure everyone has access to SPLG space
4SNOMED CT Expression Constraint LanguageLinda Bird

Discuss recent feedback about the ECL

  1. Immediate children/parents
    • FHIR Community have use case for these operators (user interface display)
    • Are these just 'syntactic sugar' for
        • < 404684003 | Clinical finding | : 116680003 |is a| = 404684003 | Clinical finding |
        • > 404684003 | Clinical finding | : R 116680003 |is a| = 404684003 | Clinical finding |
    • What syntax would we use - for example:
        • <! 404684003 | Clinical finding |  
        • >! 404684003 | Clinical finding |
  2. Comments
    • If we introduce comments into the ECL, should we also be introducing them into CG?
    • Should comments be part of the normative (brief) syntax included in interoperable sharing of ECs?
    • Or should comments be included in the non-normative (full) syntax?
    • If we introduce comments, what syntax should we use? For example:
      • /* ..... */

OUTCOMES

  1. The group agreed to add new operators to the ECL specification to support the FHIR community's use case for immediate children/parents. The preferred syntax was "<!" and ">!" .
  2. The group decided that comments should be added to the the brief syntax of the ECL using the "/*... */" syntax.
  • Discuss ECL feedback
  • Linda Bird to draft updates to ECL specification
5SNOMED CT Template SyntaxLinda Bird

Review discussion on optionality and populating attribute groups:

  1. Scope and purpose of syntax
    1. Extract/disentangle SNOMED CT (and SNOMED CT-relevant) content from a FHIR Condition resource (i) into a free-standing and ‘recognisable’ SNOMED CT expression, whilst (ii) ‘leaving nothing behind’ which may be of relevance to further processing
    2. Specify mappings from FHIR value sets (e.g. Condition.clinicalStatus) into SNOMED CT
    3. Transform the extracted expression into an ‘optimally-processable’ SNOMED CT expression (in particular grouping body site values with morphology)
    4. Specify constraints on what the extracted/disentangled SNOMED CT expression could or couldn’t contain (by e.g. cardinality instructions).

NOTE - The rest of this discussion will be carried over until the next meeting.

  1. (From a(ii) and b above) Simplify |finding context| refinement to either:
    • 408729009 |finding context| = [[ @findingContext ]]
    • 408729009 |finding context| = [[ findingContextTable ($clinicalStatus, $verificationStatus) ]]
  2. (From d above) How to specify cardinality in terminology binding when restricting valid values in an information model data element:
      • 62014003 |Adverse reaction to drug (disorder)|246075003 |Causative agent| = [[ [0..1] ^ 111115 | AMP reference set | ]]
      • 62014003 |Adverse reaction to drug (disorder)|: !! [0..1] !! 246075003 |Causative agent| = [[ ^ 111115 | AMP reference set | ]]
  3. (From c above) To indicate how the following data structure can be used to populate a template:
    • Data Structure A
      • Condition
        • Code: CodeableConcept [0..1]
        • MorphologyBS [0..*]
          • BodySite: CodeableConcept [0..1]
          • Morphology: CodeableConcept [0..1]
    • Possible template syntax examples:
      • [[ $code ]]{ 363698007 |finding site| = [[ $BodySite ]]116676008 |associated morphology| = [[ $Morphology ]] }
      • [[ $code ]]{ 363698007 |finding site| = [[ $MorphologyBS.BodySite ]]116676008 |associated morphology| = [[ $MorphologyBS.Morphology ]] }
      • [[ $code ]]!! For each M = $MorphologyBS !! { 363698007 |finding site| = [[ M.BodySite ]]116676008 |associated morphology| = [[ M.Morphology ]] }
    • Data Structure B
      • Condition
        • Code: CodeableConcept [0..1]
        • BodySite: CodeableConcept [0..*]
        • Morphology: CodeableConcept [0..1]
    • Possible template syntax examples:
      • To include the different finding sites within the same attribute group:
        • [[ $code ]]{ 363698007 |finding site| = [[ $BodySite ]]116676008 |associated morphology| = [[ $Morphology ]] }
      • To include an attribute group for each different finding site (with the same associated morphology):
        • [[ $code ]]!! For each BS = $BodySite !! { 363698007 |finding site| = [[ BS ]]116676008 |associated morphology| = [[ $Morphology ]] }
    • Other examples discussed by email (double scope):
      • |finding| : [[ {    [0..*] |findingSite| = $bodySite << 48566001 | Bone structure of extremity (body structure) |,
            [[ [0..1] |assocMorph| = $morphology < 72704001 | Fracture (morphologic abnormality) |]] } ]]
  • Review Template Syntax discussion
6Confirm next meeting date/timeLinda Bird

 

Confirm date and time of next SLPG meeting - Wednesday 27th April

  • Confirm date of next call

Meeting Files

No files shared here yet.

  • No labels