W3C home > Mailing lists > Public > semantic-web@w3.org > June 2007

Re: Evidence

From: Matt Williams <matthew.williams@cancer.org.uk>
Date: Wed, 13 Jun 2007 07:36:42 +0100
Message-ID: <466F907A.7060509@cancer.org.uk>
To: William Bug <William.Bug@DrexelMed.edu>, kskinner@nida.nih.gov, Semantic Web <semantic-web@w3.org>

Dear Karen,

I think you raise an interesting question, and I'm going to take a 
different tack to William in replying.

You raise the question of study inclusion/ exclusion criteria, and 
(correctly) point out that we would need to know when people meet the 
criteria, which means we need definitions of the criteria to check against.

I've mainly been concentrating on trials in the BM domain where 
eligibility criteria are fairly transparent (e.g. ER +ve tumours), 
although even here there are evidential considerations over which test 
one uses to measure ER receptors.

Ontologically, one *could* probably define a "smoker" as person who had 
a relation such as hasFagerstromScore to some integer score (or you 
could use subclasses, so a HighlyDependentSmoker would have a certain 
set of values). As far as this goes, it doesn't seem too complex 
(although I suspect becomes so when wants to do it for hundreds of 
different tests) and the important thing seems to be to make sure that 
the terms we use match-up with those used by others, and I would hope 
that the approach that William outlines will work well.

My worry is different to that; I am interested in how we represent the 
evidential basis of this inference. While we could do ontologically (as 
above), this seems to have some problems; we can't, for example, easily 
represent the fact that we might want to dispute the basis of such an 
inference (e.g. we think the assessment is rubbish). It is the potential 
to do things like this that I think are important.


> We're working on an expression of standard cognitive assessments and 
> behavior tests in BIRNLex, as many of the functional imaging studies 
> covered in the Function BIRN test bed make use of a whole battery of 
> such tests - including the Fagerstrom Test for Nicotine Dependence.
> Our goal is to express these using BFO and other OBO Foundry ontologies 
> in such a way that the ultimate result can be incorporated into OBI.  To 
> do so, requires we create a great many other entities, such as "Smoking 
> Behavior".   Working through a few examples so that we can put these out 
> on the BFO list for review is one of the current tasks on our plate.  In 
> fact, I've got a meeting coming up soon with the FBIRN Project Manager 
> Jessica Turner (http://www.nbirn.net/about/personnel.shtm) who along 
> with Christine Fennema-Notestine from the Morphometry BIRN and Angie 
> Laird from the Brainmap.org group (Fox/Lancaster) has been spear-heading 
> this effort to compile a comprehensive list of the assessments and 
> behavioral features we need to represent.  We'll also ultimately want to 
> vet this with the folks at the fMRI Data Center (www.fmridc.org).
> The ultimate goal is - built on a BFO + OBO-Relations foundation - to 
> express evidence associated with executing these assessments using OBI 
> and the PATO-associated Phenotype assertion formalism under development 
> by NCBO+GO investigators.
> We'd hoped to also make use of relevant terms in the CVs created by 
> PsychInfo, but there have been licensing issues that have got in the way 
> to date.
> Cheers,
> Bill
> On Jun 12, 2007, at 4:41 PM, Skinner, Karen (NIH/NIDA) [E] wrote:
>> This discussion on "evidence" makes me wonder if "inclusion" or 
>> "exclusion" criteria for a study are considered as "evidence" or are 
>> they something else?  For example, if "smokers" are to be excluded 
>> from a study, a definition of a "smoker" must exist, and then an 
>> interpretation of whether the subject satisfies the definition  must 
>> exist. In the  
>> "Fagerstrom Test for Nicotine Dependence", the subject is evaluated 
>> with respect to a series of questions, each of which requires its own 
>> "evaluation."  
>> http://ww2.heartandstroke.ca/DownloadDocs/PDF/Fagerstrom_Test.pdf
>> Is there a definition for "evidence" that applies to any aspects of 
>> this situation, and if so, what would it be?
>> Karen Skinner
>> -----Original Message-----
>> From: samwald@gmx.at [mailto:samwald@gmx.at] 
>> Sent: Tuesday, June 12, 2007 3:53 PM
>> To: Waclaw Kusnierczyk
>> Cc: phismith@buffalo.edu <mailto:phismith@buffalo.edu>; 
>> public-semweb-lifesci@w3.org <mailto:public-semweb-lifesci@w3.org>
>> Subject: Re: Evidence
>> Hi Waclaw,
>>> Matthias, if you look carefully at BFO, you'll see that roles are 
>>> entities.  This means that evidences, as roles, are entities.
>> Of course. I just wanted to differentiate that an experiment is not an 
>> instance of any class called 'evidence' (in other words, an experiment 
>> 'is not' evidence). Instead, it should be associated with an 
>> 'evidence-role'.
>> cheers,
>> Matthias
>> cheers,
>> Matthias Samwald
>> ----------
>> Yale Center for Medical Informatics, New Haven / Section on Medical 
>> Expert and Knowledge-Based Systems, Vienna / http://neuroscientific.net
>> --
>> Psssst! Schon vom neuen GMX MultiMessenger gehört?
>> Der kanns mit allen: http://www.gmx.net/de/go/multimessenger
> Bill Bug
> Senior Research Analyst/Ontological Engineer
> Laboratory for Bioimaging  & Anatomical Informatics
> www.neuroterrain.org
> Department of Neurobiology & Anatomy
> Drexel University College of Medicine
> 2900 Queen Lane
> Philadelphia, PA    19129
> 215 991 8430 (ph)
> 610 457 0443 (mobile)
> 215 843 9367 (fax)
> Please Note: I now have a new email - William.Bug@DrexelMed.edu 
> <mailto:William.Bug@DrexelMed.edu>

+44 (0)7834 899570
Received on Wednesday, 13 June 2007 06:36:56 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:41:57 UTC