See also: IRC log
scribenick Bob
Michel: OSCAR presentation
Bosse: Very interesting, more and
more as he talked about features
... interesting to see a long history in the project
... not developed toward research interests
Michel: Interesting re charts,
lots of numbers in red on the chart
... goes to show that HC pros are accustomed to a very busy
interface
... oscar hones in on what they want
... also put in pictures and videos, as of patient trying to
stand up
Bosse: Do they track lifestyle info for research?
Michel: Not sure
Bosse: This is a need in research: to know about normal life situation before disease
Michel: OSCAR is growing and is open source?
OSCAR *is* open source.
Matthias and Elgar have seen the slides
Trish: Interesting as it went on,
seem open
... have sent an email to them
Michel: Doubt that patient data
are available; onus would be on us to add functionality
... might be able to test our stuff w. participating
institutions
Michel's brother is lead developer on OSCAR
Trish: Seems like a good opp, given the nature of groups and resources
Michel: We could write grant proposal to extend oscar
Trish: How do we take the next step; opp. to take TMO to something working in the real world
Michel: Don't quite want to
commit right away on this.
... would like to get more vendors in to describe the lay of
the land from their viewpoint.
... we don't want to re-write a core in order to do something
interesting
... oscar, etc, already deployed where we can evaluate our
contribution
... Would like to survey different softwares; write proposals
to extend
... OSCAR terminology, it looked as though user had to type in
specific codes
... auto-complete might be good there
... uses LOINC, SNOMED, ICD. None of the physicians want to
type in codes
... autocomplete functionality w semantic
Trish: These would be easy ways to start using terminology in a more user-friendly way
OSCAR based on Indivo
Michel: This is good, given
ericP's work
... depends on their stuff being encoded. Indivo does not
require that everything be encoded.
... Indivo would be a connection point
... Who should we get next?
Trish: Human Studies Database
<michel> http://hsdbwiki.org/index.php/HSDB_Wiki
Trish: AMIA, may have interaction w. Larry Hunter's group
Scott: OSCAR prime candidate for
RDF-ification
... brings a lot to the table, but not immediately
semantifiable(?) perhaps. Needs more work
... David would be interested in accessing RDB via SPARQL
Indivo would be the place to plug in
Michel: Also semantic
auto-completion
... HSDB based on Ontology of Clinical Research
... already using ontologies, so interesting
Trish: Have other cats to herd.
Scott will contact HSDB
Trish: i2b2 has hives, cells;
There is a patient component in there.
... Redcap
<michel> http://www.bmhi.umn.edu/ctsi_bmi/initiatives/redcap/home.html
<whetzel> http://www.project-redcap.org/
<michel> http://www.bmhi.umn.edu/ctsi_bmi/initiatives/portal/home.html
Scott: We have looked at i2b2 in
recent months. Visited IO Informatics last week
... bumped into Xavier(?) from Oracle. Maybe get Oracle drivers
for SWOjects, thru IO
Trish: Shawn Murphy from i2b2
<michel> http://lcs.mgh.harvard.edu/projects/rpdr.html
Michel: Research Patient Data Registry
ericP: Children's, but grant didn't come through
Scott: Xavier Lopez from Oracle
ericP: Dan has access to de-id,
Children's has access to non-de-id
... ODBC always painful.
... Children's guy might have leads for funding or data. They
are working on SHRINE pediatric, 50 sites
Scott: SHRINE looks like a tremendous resource
ericP: Could try to talk to shrine/i2b2
Michel: Paper opportunity, invitation from J. Pharmacogenomics
<mscottm> yes, pharmacogenomics
<whetzel> Has there been any follow-up with the EMERGE or Marshfield Clinic?
Michel: approaches to mark-up. We
can write review of existing approaches DB --> formal
representation
... take the best, leave behind the worst, say this is how we
would do it.
... representaton + queries, using TMO
... need more review of other HC systems, how they do this
Michel: want to look at OCR and OGMS
<mscottm> Trish: Good question. Marshfield was in motion last year. I don't recall anything about EMERGE.
Michel: Chime CPR Clincial
Patient Record
... look at resources like OMIM
Bosse: Can do some on this
<mscottm> Bob - that was Matthias, not Bosse. :)
ericP: Have a landscape of patient encounter, tests, demographics; to a map that displays all of this.
Matthias: Can do some of this (not Bosse :-)
ericP: "Landscape browser"
Example Indivo, CPR have similar concepts
... not in same universe model, but they do talk about the same
stuff
<michel> section about mappings between different approaches
<michel> analogous structures
ericP: needs to be browsable
Michel: Registry of these kinds of mappings?
ericP: Not sure how far down the
machine this should go
... allow a person to navigate. Would be awesome to provide
semantic functionality.
Michel: Can be more conceptual,
need not be fully operational
... can leave real mappings for follow-on paper
... focus can be around pharmacogenomics
ericP: Neither Indivo nor i2b
have structures around this
... XML encourages people to get stuck in microparsing
... so RDF would be good here
Michel and Scott will start a paper framework
ericP: Ben Adida would be an excellent resource for this
This is scribe.perl Revision: 1.135 of Date: 2009/03/02 03:52:20 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) No ScribeNick specified. Guessing ScribeNick: Bob Inferring Scribes: Bob WARNING: No "Topic:" lines found. WARNING: No "Present: ... " found! Possibly Present: Bosse Matthias Michel P10 P2 Scott aaaa aabb bbalsa epichler ericP https matthias_samwald mscottm trish whetzel You can indicate people for the Present list like this: <dbooth> Present: dbooth jonathan mary <dbooth> Present+ amy WARNING: No meeting title found! You should specify the meeting title like this: <dbooth> Meeting: Weekly Baking Club Meeting WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Got date from IRC log name: 24 Mar 2011 Guessing minutes URL: http://www.w3.org/2011/03/24-hcls2-minutes.html People with action items: WARNING: No "Topic: ..." lines found! Resulting HTML may have an empty (invalid) <ol>...</ol>. Explanation: "Topic: ..." lines are used to indicate the start of new discussion topics or agenda items, such as: <dbooth> Topic: Review of Amy's report[End of scribe.perl diagnostic output]