W3C

- DRAFT -

SV_MEETING_TITLE

15 Oct 2013

See also: IRC log

Attendees

Present
EricP, Tony, Kerstin_Forsberg, Mike_Denny, EmoryFry
Regrets
Chair
SV_MEETING_CHAIR
Scribe
ericP

Contents


http://www.w3.org/2013/ShEx/SimpleShExDemo

<scribe> scribenick: ericP

US FDA Therapeutic Areas

ericP: [FDA TA summary]

eafry: what of the TAs will the OWL capture?
... data, workflow, everything?
... how does that differ from BRIDG

ericP: when people work in BRIDG, what's their wire protocol?

eafry: mostly UML. emphasis is on data capture and persistance for aggregation across studies

kerstin: i think many people talk about BRIDG but few have done anything with it
... as i understand the TA, they will work on model but also units and data elements.
... overlaps with SNOMED-CT
... CDISC is doing a concept map which they can connect to BRIDG and data pipes
... the ISO21090 data pipes is in focus

ericP: example of a q from CRF?

kerstin: a bunch of initiatives are talking about tying this all together

eafry: i asked because i'm tyring to see how the concept maps fit into this
... OWL can be used to model lots of levels
... BRIDG is, in my mind, a domain model
... in concepts maps, you're trying to represent cognitive processes

ericP: as i understand it, they're using concept maps as a gui tool to interface with domain experts.
... who'm do we tap?

<eafry> Brian Moon <brian@perigeantechnologies.com>

eafry: Robert Hoffman, forefather of concept maps
... Brian helped us with pragmatic issues
... they're working on a tool to map to OWL.
... they're working on the CMaps to OWL. could help there

Guoqian: work with CIMI
... CIMI is building common formalismms that will be converted to other syntaxes, like OWL
... people are working on a common model
... currently OpenEHR archetype tooling
... the other direction is UML models
... talk to Harold and Dave Carlson about the "Archetype Modeling Language"
... there are existing tool to convert archetytpes to OWL
... is IBM going to build the TA from scratch? work from some base doc? grab codes from e.g. SNOMED-CT?

ericP: my impression is that they are using a standard in-house methodology to extract knowledge from domain experts

FHIR as RDF

ericP: [outlined work, asked for volunteers]

<Mike> for Cmap OWL = COE http://www.ihmc.us/groups/coe/

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2013-11-05 16:01:18 $