W3C

- DRAFT -

Semantic Web Health Care and Life Sciences Interest Group Teleconference

02 Sep 2015

See also: IRC log

Attendees

Present
EricP, David_Booth, Tony_Mallia, Lloyd, Marc_T, Rob_Hausam
Regrets
Chair
DBooth
Scribe
dbooth

Contents


<trackbot> Date: 02 September 2015

ValueSets

http://wiki.hl7.org/index.php?title=File:Vocabulary_and_Terminology_in_RDFv2.pdf

<ericP> https://www.w3.org/wiki/images/7/72/Snomedct_extract_descriptive.ttl

Starting on line 195

<ericP> marc, lloyd, http://wiki.hl7.org/index.php?title=File:Vocabulary_and_Terminology_in_RDFv2.pdf

<ericP> (just in case -- i meant to paste insetad of the URL 3 lines up)

<Marc_Twagirumukiza> AFAIK in SNOMED each concept has its own code

<Marc_Twagirumukiza> I don't see imediatelly where we can have 2 codes for the same concept

At least two ways to model this: 1. ConceptAB is identified by code "A" or by code "B". 2. ConceptA is identified by code "A", ConceptB is identified by code "B", and ConceptAB is a superclass of ConceptA and ConceptB.

<ericP> http://piratepad.net/hcls

David: Why fhir:Null on line 223?

tony: To get around a tooling issue. A union of one class is otherwise treated as an equivalence relation.
... You're thinking of ValueSets from the perspective of sets of Concept. But there are other ValueSet definitions that are not concept driven.
... Need to be able to drive it by the string value, not the concept.
... Risky to use a named class for the concept with a specific code string, because people will assert relations between them that they shouldn't.

lloyd: FHIR defines ValueSets by enumeration and filters. Many filters are concept based, some are not.
... Concept URIs will typically be assigned by the terminologies, and won't be present in the ValueSet definition instance.
... Therefore we'll need to construct the ValueSet def from the information we have, which does not include the Concept URIs.

tony: Could compute Concept URI from the system and code, but that would be lossy, because there might be multiple codes for the same concept.

lloyd: We will create a class URI that is based on the combination of system+code, but we'll use that to refer to the concept -- not to the code. And we'll expect that when we tie our ignorant set of URIs that we gnerate from the instance, with the knowledgeable URIs from the terminology source, then there will be equiv class assertions made between them.

david: There are three ontologies involved: 1. the FHIR ont, which must be ignorant of all terminology-specific information. 2. Terminology-specific ontology, such as the SNOMED ont. 3. Bridge ontology, that link between the FHIR ont and the terminology-specific ont.

lloyd: looking at the piratepad: http://piratepad.net/hcls

<Lloyd> <filter>

lines 246-247

<Lloyd> <compose><include><system value="snomed.info/sct"/><filter><property value="concept"/><op value="is-a"/><value value="105590001"/></filter></include><include><system value="http://snomed.info/sct"/><property value="concept"/><op value="is-a"/><value value="373873005"/></filter></include></compose>

<ericP> pasted into pirate pad at line 27

<ericP> (added a missing <filter> start tag)

lloyd: Says 'Find the concept from snomed CT that corresponds to this code, then find all specializations for that concept, and allow all codes for those concepts (including the original concept).'

<Marc_Twagirumukiza> http://hl7.org/fhir/2015May/valueset-substance-type.html

eric: Cecil would probably be able to tell use how much computational load we're adding by having classes. Want to avoid being unable to process by off-the-shelf tools.

david: Should write examples to speed our progress.

eric: We could show Cecil examples and get his input.

<Marc_Twagirumukiza> I have to leave, see you next week

<scribe> ACTION: Tony to make expansion example using owl:hasValue [recorded in http://www.w3.org/2015/09/02-hcls-minutes.html#action01]

<trackbot> Error finding 'Tony'. You can review and register nicknames at <http://www.w3.org/2014/HCLS/track/users>.

<scribe> ACTION: lloyd to find example valueset definitions in FHIR that cover various aspects, and make up others for edge case [recorded in http://www.w3.org/2015/09/02-hcls-minutes.html#action02]

<trackbot> Error finding 'lloyd'. You can review and register nicknames at <http://www.w3.org/2014/HCLS/track/users>.

ADJOURNED

Summary of Action Items

[NEW] ACTION: lloyd to find example valueset definitions in FHIR that cover various aspects, and make up others for edge case [recorded in http://www.w3.org/2015/09/02-hcls-minutes.html#action02]
[NEW] ACTION: Tony to make expansion example using owl:hasValue [recorded in http://www.w3.org/2015/09/02-hcls-minutes.html#action01]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.140 (CVS log)
$Date: 2015/09/02 16:13:38 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.140  of Date: 2014-11-06 18:16:30  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/specific ontologies/specific ontology/
Succeeded: s/Bridge ontologies/Bridge ontology/
Succeeded: s/concepts./concepts.'/
Succeeded: s/those concepts/those concepts (including the original concept)/
No ScribeNick specified.  Guessing ScribeNick: dbooth
Inferring Scribes: dbooth
Present: EricP David_Booth Tony_Mallia Lloyd Marc_T Rob_Hausam
Found Date: 02 Sep 2015
Guessing minutes URL: http://www.w3.org/2015/09/02-hcls-minutes.html
People with action items: lloyd tony

[End of scribe.perl diagnostic output]