W3C home > Mailing lists > Public > public-semweb-lifesci@w3.org > November 2009

Re: [TMO] rdf for patients 1 and 2

From: Eric Prud'hommeaux <eric@w3.org>
Date: Sun, 22 Nov 2009 09:40:41 -0500
To: public-semweb-lifesci@w3.org
Message-ID: <20091122144039.GA12731@w3.org>
* Eric Prud'hommeaux <eric@w3.org> [2009-11-22 02:38-0500]
> I'm not super-confident about the modeling of test results, and I
> didn't find that our data lined up fabulously (except for
> demographics) with indivo, but here are some files to throw rocks
> at. Near indivo misses are documented in comments near the approp-
> riate lines in the .rdf files.
> 
>   http://www.w3.org/2009/11/19-TMO-xslt/AD_PCHR_1_Final.rdf
>   http://www.w3.org/2009/11/19-TMO-xslt/AD_PCHR_2_Final.rdf
> 
> I believe Colin, Bosse and Michel all expressed an interest and lining
> this up with TMO and other existing ontologies (though I sampled pretty
> well from foaf and a couple others).

I forgot to mention, I also don't feel I understand the boundries on
tests. Should a patient undergo multiple tests with the same IDC9 code
and with different parameters recorded? Monica (patient 2)
  line 731 of the hand xml:
  http://esw.w3.org/topic/HCLSIG/PharmaOntology/Data/FakePatients?action=AttachFile&do=get&target=AD_PCHR_2_Final.doc
  or line 726 of the machine xml:
  http://www.w3.org/2009/11/19-TMO-xslt/AD_PCHR_2_Final.xml
undergoes six tests on her 2004-10-01 visit, three of them tagged
<ICD9 Procedure>94.08</ICD9 Procedure>, and each reporting different
parameters, summarizable as:
1st instance:
  classic         8.67
  85 point        13.67
2nd instance:
  total score     25
3rd instance:
  total score     47
  total correct   26
  recog score     12
As you can see, the recorded parameters differ, and I don't know how
to code that or subsequently use them.

I played around with the triangle for Colin. I have provided two
relationships btween an encounter and the patient:
  direct:
  trans:Encounter trans:patient <patient> ;
        trans:physician "Henry No" .

  triangle:
  trans:Encounter trans:participant _:pat .
  _:pat a trans:Participation ;
        trans:role trans:Patient .
  trans:Encounter trans:participant _:doc .
  _:doc a trans:Participation ;
        trans:role trans:Physician ;
        trans:persona _:d2 .
  _:d2 a foaf:Person ;
       foaf:name "Henry No"


The following triples relate the other data to the patient:

  trans:Drug trans:prescribedto <patient> .
  <patient> trans:hasCondition trans:Condition .
  trans:Procedure trans:performedOn <patient> .
  <patient> trans:hasAllergy trans:Allergy .



-- 
-ericP
Received on Sunday, 22 November 2009 14:41:17 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 18:00:57 GMT