Hi David,
I would like to say yes,they would be of use but I don't know for sure. At
the moment our processing is just to recognise SCT terms (no deeper
sematnics is implied, i.e. we're not doing deeper semantic analysis),
however it also most important to understand how variable clinical notes
can be and as well to start opening the question of "localisation", that
is how does any automatic system tackle the problem of the writer
idiosyncrasies.
So my answer is yes we do want your notes so that even if we can't find any
SCT terms we learn something about why we can't find them and adjust our
search strategies accordingly.
thanks
jon
Quoting dguest <[EMAIL PROTECTED]>:

> Jon Patrick wrote:
>
> >Dear List members,
> >You've heard me discuss previously our need for clinical notes to test
> our
> >work on recognising SNOMED CT terms. I appreciate that everyone is
> cautious
> >about violating the confidentiality of their patients. Hence rather
> than try
> >to get a large collection of notes from one location I am requesting a
> very
> >small sample from many locations. If 50 people would be prepared to
> package up
> >20 sentences from any notes  that contain terms that you would expect
> to be
> >present in a large terminology like SNOMED we would then have sample of
> 1000
> >sentences to do our trials on.  It is my hope that 20 sentences is not
> such a
> >burden that people would feel they could vet the material and send it
> without
> >undue time cost.
> >yours in hope
> >
> >
> Sounds good, John.
>
> On problem I foresee is that my computer notes are mostly fragments of
> history and examination findings that are annotated by me in a highly
> idiosyncratic way. I don't know if I am atypical but I suspect most of
> my colleagues do the same thing in their own highly idiosyncratic ways.
>
> Would this note format be of any use to you?
>
> David
>
>
> _______________________________________________
> Gpcg_talk mailing list
> [email protected]
> http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
>


--
Jon Patrick
Chair of Language Technology
School of Information Technologies
University of Sydney
Sydney, NSW, 2006, Australia

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.
_______________________________________________
Gpcg_talk mailing list
[email protected]
http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk

Reply via email to