W3C

- DRAFT -

Semantic Web Health Care and Life Sciences Interest Group Teleconference

17 Mar 2015

Agenda

See also: IRC log

Attendees

Present
Paul_Knapp, Darrell, David_Booth, Brian_Pech, Claude_Nanjo, Marc_Twagirumukiza, Tony_Mallia, Kerstin_Forsberg, Joshua, EricP, Rafael_Richards, Rob_Hausam, harryh, Guoqian
Regrets
Chair
David Booth and Paul Knapp
Scribe
Kerstin_Forsberg, kerfors

Contents


<trackbot> Date: 17 March 2015

<Claude> https://global.gotomeeting.com/join/157514853

<dbooth> scribe list: http://wiki.hl7.org/index.php?title=RDF_Group_Scribe_List

<dbooth> ScribeNick: Kerstin_Forsberg

<inserted> scribenick: kerfors

Approve Minutes of previous meetings

<dbooth> Last week's minutes: http://wiki.hl7.org/index.php?title=ITS_RDF_Concall_Minutes_20150310

<inserted> RESOLVED: Minutes of Mar 10 approved!

CCDA Examples from John

FHIR RDF project scope statement

<inserted> Claude: Will start drafting it this week

FHIR RDF and Validation/Translation Task force

<dbooth> Notes from TF: http://www.w3.org/2015/03/11-hcls-minutes.html

FHIR codes = FHIR RDF should use short code?

Eric: FHIR codes = FHIR RDF should use short code?

FHIR JSON-LD update

<dbooth> http://dbooth.org/2015/fhir/json-ld/fhir-in-json-ld.pdf

David: slide 2 - FHIR JSON-LD what @context should be handled? Best done direct as the other serialisations. Can't be anything missing or added outside
... Doing things more natural for RDF e.g. treating "id" as URIs is not a good way
... Require logic to create and roundtrip. So instead keep "id"
... Keep FHIR declarations of – fhir:resourceType as rdfs:subPropertyOf 
... slide 4 Most issues seems to be managaeble
... – • FHIR RDF will need custom serializationi to JSON-LD
... slide 5 Raw Observation -> Fully Understood Observation

Tony: Back to ID discussion (slide 2) - How about doing both - keep 'id' and create URIs? Is this a win/win with both raw and understood also for identifiers

David: FHIR uses a closed content model.
... slide 7 - Did find a "bad" thing: Element type depends on context. FHIR need the ordering of elements to be preserved. Can this be done with RDF lists to preserve the order?

Eric: Is it a help if the order is controlled by the content model instead?

David: However, downside with RDF lists - not good for SPARQL and inference
... RDF to RDF translations needed - can this be the place to use a ordered list ontology?
... slide 10 JSON-LD vs Turtle show that there needs always be lists also for single code e.g. fhir:code ( "3141­9" ) in the middle of slide 10

Tony: Underlying problem: FHIR uses local names e.g. code similar to name for person vs organisation... ... Discussion of alternative solutions between David, Tony, Eric P ... ...
... The profile give the context. Eric P: Needs clever parsers but this is not what you want to do.

Eric P: Push back to FHIR group to help developers

David: slide 11 with different options
... Add the push back as a fourtfh option
... Slide 11 - Different ways forward - Make FHIR JSON to be JSON-LD vs Custom RDF
... A third option - first a very direct RDF translation not ideal for inference as a base layer RDF. And then have other forms generated that are more friendly for inferenceing

Eric P: Hierarchies of code and resources seem to be the two core things to manage. Don't see that big step between them

Tony: Agree don't see that big gap either
... Evaluate the 3 options against the 14 requirements we have defined
... Terminology binding seems to be the core thing

David: Identify specific issues and test them against the three different options
... What do people need to see to do the review?

yes hear you

<dbooth> ACTION: Eric to suggest use cases to look at [recorded in http://www.w3.org/2015/03/17-hcls-minutes.html#action01]

<trackbot> Created ACTION-19 - Suggest use cases to look at [on Eric Prud'hommeaux - due 2015-03-24].

Eric P: Go to the meaningful use cases to nail down ... ... the inferencing we can do with RDF and not in XML

Tony: Patient safety use case ... ...

Eric: SNOMED has hierarchy and LOINC has 7 different dimensions but no hierarchy

<dbooth> ACTION: Rob Hausam to send example of multi-axial hierarchy in LOINC [recorded in http://www.w3.org/2015/03/17-hcls-minutes.html#action02]

<trackbot> Created ACTION-20 - Hausam to send example of multi-axial hierarchy in loinc [on Rob Frost - due 2015-03-24].

sorry have to leave

<dbooth> Rafael's use case: http://dbooth.org/2014/richards/Linked%20Vitals-20141112-v1.pdf

<dbooth> ADJOURNED

<rhausam> http://loinc.org/downloads

<rhausam> go to "LOINC Multiaxial Hierarchy File"

<dbooth> ScribeNick: kerfors

<ericP> dbooth, do you remember if i was supposed to deliver anything to the V&T group this week?

<ericP> i have a nagging feeling that i was, but it never materializes beyond that.

Summary of Action Items

[NEW] ACTION: Eric to suggest use cases to look at [recorded in http://www.w3.org/2015/03/17-hcls-minutes.html#action01]
[NEW] ACTION: Rob Hausam to send example of multi-axial hierarchy in LOINC [recorded in http://www.w3.org/2015/03/17-hcls-minutes.html#action02]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.140 (CVS log)
$Date: 2015/03/17 22:16:24 $

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: i/FHIR RDF and Validation/Translation Task force/Claude: Will start drafting it this week
Succeeded: s|Translation Task force/Claude|Claude|
Succeeded: i/Approve Minutes of previous meetings/scribenick: kerfors
Succeeded: i/CCDA Examples from John/RESOLVED: Minutes of Mar 10 approved!
Succeeded: s/Slide 10/Slide 11/
Succeeded: s/slide 10/slide 11/
Succeeded: s/slide 9/slide 10/g
Succeeded: s/slide 6/slide 7/g
Succeeded: s/What to people/What do people/
Found ScribeNick: Kerstin_Forsberg
WARNING: No scribe lines found matching ScribeNick pattern: <Kerstin_Forsberg> ...
Found ScribeNick: kerfors
Found ScribeNick: kerfors
WARNING: No scribe lines found matching ScribeNick pattern: <kerfors> ...
Inferring Scribes: Kerstin_Forsberg, kerfors
Scribes: Kerstin_Forsberg, kerfors
ScribeNicks: Kerstin_Forsberg, kerfors
Default Present: Pknapp, Darrell, DBooth, bpech, Claude, Marc_Twagirumukiza, Tony, Kerstin_Forsberg, +1.410.624.aaaa, Joshua, +1.410.905.aabb, ericP, rafael, +1.801.810.aacc, rhausam, +1.412.624.aadd, harryh, Guoqian, [IPcaller]
Present: Paul_Knapp Darrell David_Booth Brian_Pech Claude_Nanjo Marc_Twagirumukiza Tony_Mallia Kerstin_Forsberg Joshua EricP Rafael_Richards Rob_Hausam harryh Guoqian
Agenda: http://wiki.hl7.org/index.php?title=ITS_RDF_ConCall_Agenda
Found Date: 17 Mar 2015
Guessing minutes URL: http://www.w3.org/2015/03/17-hcls-minutes.html
People with action items: eric hausam rob

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


[End of scribe.perl diagnostic output]