See also: IRC log
scribenick Bob
<Amit> #HealthIT HHS inv wkshp on Next Gen Interop for Health attended by key policy makers, PCAST http://bit.ly/dIaki0 & #semanticweb on agenda
Amit: Workshop organized by HHS
<Amit> My talk w/ Ora on #semanticweb for 360deg health: http://slidesha.re/health360 at HHS inv workshop for Health Interop #knoesis #HealthIT
Michel: Home stretch for the paper
<Amit> is there a draft to look at - just for my fyi
Christi and Joanne discuss Latex etc
Jim McCusker is suggesting to look at http://docs.latexlab.org/
Michel: Response to reviewers
Joanne: All addressed
Christi can relax now.
Michel: Great job everybody! Papers are good motivators for getting concrete done.
<Amit> even if it is too late for me to contribute for this one, can I get a sneak peek?
Michel: Requirements for AMIA preparation
Joanne: No deadline. Challenge is
how to demo an ontology
... talk is March 8, session at 10:30AM
... perhaps show what queries you can answer w/wOut
Michel: Longitudinal
queries
... db or recorded side effects. Sider may not have
timelines
... at granularity of query, presentation on a timeline, what
data will appear?
<Joanne> check this out: http://sparql.tw.rpi.edu/?page_id=57
Michel: timelines more interesting than spatial data
<Joanne> http://sparql.tw.rpi.edu/
Joanne: We have started some visualizations here
<michel> http://sparql.tw.rpi.edu/?p=32
<michel> timeline demo
Scott: Two datasets have come: one from IO Informatics, kidney transplant data
<michel> scott: triplemap - sneak preview
Scott: includes microarray
data
... also proteomics data for those patients
... Another source of data is UPitt, gotten approval for 1000s
records of clincial reports
... gathered for NLP research
<michel> http://nlp.dbmi.pitt.edu/nlprepository.html
relates also to Terminology concern w. clinical reports
scribe: thru Richard there is a whole load of data available to uus
Michel: Relfinder(?) tool
<mscottm> Fullname: Richard Boyce, of Univ. of Pittsburgh
Joanne: i2b2 challenge is really
an NLP challenge
... interesting if we could show how semweb fits here
<Joanne> https://www.i2b2.org/NLP/DataSets/Main.php
<Joanne> https://www.i2b2.org/NLP/Coreference/Call.php
Michel: Amit, how should we present integrating patient data w. LOD?
<Joanne> it deals with discharge summaries
Amit: Do you have a handle on the quality?
Michel: There are corresponding references into LODD
Amit: Need to look at if from the
perspective of the clinician
... see it in a doctor's use case scenario
... why am I paying the doctor if he's just googling? :)
... need a new, very specific use case.
... and also handle the quality of the outcome, plus
timeliness
... otherwise hazard: CS telling us what to do!
Michel: How does this IT really help
<Amit> http://knoesis.wright.edu/library/download/S+2008-ASEMR-chapter6.pdf
Amit: Deployed Jan 2006, still being used
<Joanne> They are complementary, the clinician and the computer
<Joanne> that is the fear - being told what to do, which is one reason that decision SUPPORT is more accepable
Amit: Clinicial wants to see a result in 15 min, time is the cost driver
<Joanne> You're paying the doctor to look at the patient. Because the clician's eyes are trained - they can see if you look sick, feel your energy. computers are no supbstitue for that (- though I can imagine some skin sensors and combination of biometrics)
Amit: Start out w. something that grabs them
Michel: We seem to be focussed
more for clinical research rather than clinical practice
... perhaps we are in a mid-range rather than in clinical
practice
Amit: Demonstrate on
chemo-genetics.
... number of adverse events or number of drugs to be
considered can be drastically reduced
... look at the quality of results (better Type 1
performance?)
<Amit> http://knoesis.org/library/demos/
Amit: Ex of how semantic solution improved the quality
Joanne: AMIA would be
awesome
... demonstration! would be awesome
Michel: Scott and I want to work
on SNPs
... URIs for APOE.
Scott: Matthias is also interested
Michel: We should have a
discussion re SNP. Previously did some work re depression
... Adrian Coulet also has a model for SNPs
... start collecting models for SNP representation
... AOG
AOB
Scott: What to do w. new data
Michel: Maybe start wiki pages
for better focus
... will create page for SNPs
<Joanne> You should have access to the presentation google doc now- pls check
Joanne: Partner-up w. a hospital, more direct involvement?
Michel: Chime.
... CPR will re-start shortly
Joanne: Look at i2b2 challenge too
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. Default Present: +1.317.433.aaaa, Bob_Powers, michel, Christi, Amit, Joanne_Luciano, +62427aadd, +1.781.431.aaee, +1.302.598.aaff Present: +1.317.433.aaaa Bob_Powers michel Christi Amit Joanne_Luciano +62427aadd +1.781.431.aaee +1.302.598.aaff 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: 10 Feb 2011 Guessing minutes URL: http://www.w3.org/2011/02/10-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]