W3C home > Mailing lists > Public > public-rif-wg@w3.org > March 2006

RE: [RIF][UCR]Decision Support

From: Leora Morgenstern <leora@us.ibm.com>
Date: Wed, 8 Mar 2006 22:24:17 -0500
To: "Nichols, Deborah L." <dlnichols@mitre.org>, public-rif-wg@w3.org
Cc: "Chris Welty" <cawelty@frontiernet.net>, jpan@csd.abdn.ac.uk, aginsberg@imc.mitre.org
Message-ID: <OF9D3CC1CA.EFC4BE5A-ON8525712C.0011746F-8525712C.0012B4E0@us.ibm.com>

Thank you very much for catching the inconsistency re Diamicron. The issue 
is just one of American vs. European or Canadian names for medications. 
There was some inconsistency in previous versions, and I switched to all 
American names for medications. However, I had neglected to switch 
Diamicron to an American equivalent. I have done this now.

In addition, I have made sure to clarify that metformin is the generic 
equivalent of Glucophage. This required an extra line in the use case, but 
that will hopefully be okay.

The modifications are now at 
. I don't know whether the working draft has been frozen already, and 
whether these changes will get incorporated.

As to your other comment, I agree with you that a paragraph explaining how 
this example would generalize to other domains would enhance the strength 
of this use case. This was not, however, the opinion of those at the F2F 
meeting, who made the decision.

Thanks again for your help.


"Nichols, Deborah L." <dlnichols@mitre.org> 
Sent by: public-rif-wg-request@w3.org
03/08/2006 09:44 PM

Leora Morgenstern/Watson/IBM@IBMUS, <public-rif-wg@w3.org>
"Chris Welty" <cawelty@frontiernet.net>, <jpan@csd.abdn.ac.uk>
RE: [RIF][UCR]Decision Support

Overall I think the condensation of this use case turned out quite well. I 
would have liked if it had retained the mention of RIF-enabled decision 
support as having a wider field of application (because of MITRE's convoy 
support example), but it's easy enough for a reader to extrapolate. 
One sentence in the case raised questions for me, since I have no medical 
expertise:  Bob tells the neurologist that he was "taking Diamicron, ... 
forgetting that he had been switched to Metformin."
    (1) The sentence suggests that Bob was in fact previously taking 
Diamicron, but that is the first mention of Diamicron, at least by that 
name -- perhaps it is equivalent to something mentioned earlier?
    (2) The example stated that the doctor switched Bob to "Glucophage and 
Avandia."  True, earlier, a parenthesis mentioned that Glucophage is 
"(metformin, one of the biguanides)"; but it's not clear to me whether 
metformin is the generic equivalent of Glucophage, or a class of drugs to 
which Glucophage belongs.  The information on Bob that is pulled from the 
patient data base says he is "taking Metformin."  I might guess from this 
that Metformin is somehow equivalent to (or subsumes) Glucophage, but it 
could be clearer.
Perhaps clarifying those details would illustrate that ontological 
knowledge (here, about related or equivalent drug classes) is indeed 
essential if rules are really going to operate together.  In that 
connection, I liked the inclusion of Jeff's sentence referring to 
ontologies as one source of information for decision support.

From: public-rif-wg-request@w3.org [mailto:public-rif-wg-request@w3.org] 
On Behalf Of Leora Morgenstern
Sent: Tuesday, March 07, 2006 9:40 AM
To: public-rif-wg@w3.org
Cc: Chris Welty; jpan@csd.abdn.ac.uk
Subject: Re: [RIF][UCR]Decision Support

After the meeting in Mandelieu, I had been asked to shorten the exising 
Decision Support use case. The resulting use case is at 
. Since Jeff's extensions essentially subsume that use case (modulo some 
minor issues of style/grammar/presentation), I propose that I incorporate 
Jeff's changes into the use case. 

Leora Morgenstern 

"Jeff Z. Pan" <jpan@csd.abdn.ac.uk> 
Sent by: public-rif-wg-request@w3.org 
03/06/2006 07:36 PM 

Leora Morgenstern/Watson/IBM@IBMUS, Chris Welty <cawelty@frontiernet.net> 
[RIF][UCR]Decision Support

Dear RIFers,

> Jeff Pan has taken an action to propose a slight extension to the 
example for
Decision Support.

I have added a new wiki page to address this action:


The extended example is based on discussions with Christine Golbreich, 
Stamou and Giorgos Stoilos. The motivation of slightly changing the 
scenario is
to use some real-world medical recommedation (thanks to Christine's 
in our example.



Dr. Jeff Z. Pan (http://www.csd.abdn.ac.uk/~jpan/)
Department of Computing Science, The University of Aberdeen
Received on Thursday, 9 March 2006 03:24:33 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:47:37 UTC