Showing interest in RDF

Eric: I think Cognitive is planning to use RDF.

s|RDF/|FHIR/RDF

harold: Got email from someone wanting to link FHIR into the human phenotype ontology
... Difference between RDF model and Turtle, and JSON-LD working group intends to make RDF usable for mortals.
... Seamless of integration of FHIR: ability to view FHIR space as LOD, to traverse from Observation to Patient to DiagnosticReport, to Reporter seamlessly.
... If we do it right there would be plenty to show, if we put together a triplestore where we have a big chunk of SMART-on-FHIR structured data we can traverse.

eric: I stuck the FHIR examples into SOLID.

<hsolbrig> synthea

harold: synthea is a FHIR data generator.

<hsolbrig> FHIR as a linked ecosystem

<hsolbrig> FHIR + terminology as a linked ecosystem

david: Harold demonstrated that one nicely in his Yosemite webinar several months ago.

<hsolbrig> FHIR + genomics space (OBO / Translator) as a linked ecosystem

harold: Biggest problem with genomics use case is the code systems. They talk different code systems.

<hsolbrig> FHIR is LOINC + ICD-10 + RxNorm + SNOMED. Genomics is HPO + MonDO + ChemBL + ...

eric: If you use RDF you get to do SPARQL queries across it, and validate across the data that spans FHIR resources.

harold: Yes, and I think Grahame recognized that a while back. Once you go into a triplestore the notion of Resource falls by the wayside.

eric: Another 4th use case: TermInfo problem, they've been tackling bit by bit, by binding resource properties to SNOMED. You can microparse the SNOMED term. That sets the precedence to other info models that are bound to SNOMED.
... SNOMED expressions are a candidate for expressing clinical decision support (CDS).

harold: CDS hooks package, an implementation of what Claude has been working on, is partially implemented in EPIC and other places. Based on references to element definitions. They use FHIR PATH down in the guts.

david: Most of the FHIR/RDF still seem aspirational -- things that people *want* to do.
... Interop use case is hard to make. Too early to be able to push it on a large scale.
... But secondary use cases, and CDS use cases are more viable right now.

eric: FHIR release 1 didn't have much uptake, but release 3 does. All the people who used CCDA would use FHIR/RDF quite happily. But it has to exist for 3 years for grant processed to go through their cycles. So we should ask for as long as JSON and XML had before they had wide adoption. And we can push the things that are on the cusp: Personal Health Train and what Cognitive are doing.
... Personal Health Train is what Louise Bonino and Berend Mons are pushing along with FAIR.
... I'm supposed to work on SOLID for PhD use cases in Feb.

david: We can cite that if it is definitely planned.

https://www.go-fair.org/implementation-networks/overview/go-fair-personal-health-train/

eric: Could we help Michael Van Der Zel's student get some attention for his work.
... SHould also ask Davide Sottaro and Claude Nanjo.

david: I'll ask Mark Musen what he's doing, and any use of FHIR/RDF.

<scribe> ACTION: Eric to ask Mark Musen about FHIR/RDF.

david: Guoqian at Mayo still using FHIR/RDF?

harold: Yes, we have a grant proposal for it.

david: I think we should make a list of FHIR/RDF projects.

<scribe> ACTION: david to reach out to people about FHIR/RDF usage.

ADJOURNED

This is scribe.perl Revision: 1.154  of Date: 2018/09/25 16:35:56  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

FAILED: s|RDF/|FHIR/RDF|
Succeeded: s/grant/grant proposal/
Succeeded: s/Quoxian/Guoqian/

WARNING: No "Present: ... " found!
You can indicate people for the Present list like this:
        <dbooth> Present: dbooth jonathan mary
        <dbooth> Present+ amy
        <amy> Present+

No ScribeNick specified.  Guessing ScribeNick: dbooth
Inferring Scribes: dbooth

WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: david eric

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)