- From: Matt Williams <matthew.williams@cancer.org.uk>
- Date: Sat, 14 Jun 2008 23:42:21 +0100
- CC: W3C HCLSIG hcls <public-semweb-lifesci@w3.org>
Dear All,
Have been through the minutes. Very stimulating, as usual.
Just one plea (form someone who is neither a member of a W3C body NOR an
IE)....please don't restrict the mailing list to these people in these
groups.....
There's seemed to be a reasonable amount of discussion about trying to
capture clinical trial results. The wok on AD was very interesting, but
from a clinical POV, a little too removed from clinical practice.
Possible sources would be to use a subset of Pubmed (see the string at
the end) or else some est of conference proceedings (I've looked at the
European Breast Cancer Conference, which was earlier this year). Both
would give you a good corpus of info. The advantage of the Pubmed search
is that you can already export the info in XML, and their are tages that
pick out things of particular relevance to the trial, which might make
an IE approach easier.
HTH,
Matt
"Breast Neoplasms"[Mesh] AND ("antineoplastic agents" [Mesh] AND
"Prognosis"[Mesh]) AND (methods[Title/Abstract]) limited to English
language, Female, Adults, RCTs gives >400. The use of "Methods" aims to
pick out more structured abstracts
Dr. M. Williams
Clin. Onc. SpR
Clin. Research Fellow, CR-UK/ UCL
Received on Saturday, 14 June 2008 22:43:08 UTC