Re: A problem-oriented medical record OWL ontology

Hi All,

I would also add that the effort amongst OBO Foundry participants to  
create a Clinical Trials ontology (http://www.bioontology.org/wiki/ 
index.php/Workshop_on_Clinical_Trial_Ontology) is focussed on  
expressing (2) that Vipul highlights below.  They are also looking to  
efforts such as CDISC for guidance on what entities MUST be covered  
and exploring the issue of how this ontology effort relates to data  
modeling oriented efforts such as HL7.

Kerstin is registered as one of the participants - as are a large  
number of people contributing to the OBI effort (Ontology of  
Biomedical Investigation).

NCBO and the OBO Foundry are also working on a disease ontology and  
an ontology + formalism for expressing phenotype observations as  
ontology-based annotations.  The meetings for these ontology efforts  
just took place:
	disease ontology (Nov 6-7, 2006): http://www.bioontology.org/wiki/ 
index.php/Workshop_on_Ontology_of_Diseases
	PATO (Phenotype Attribute and Trait Ontology] (Nov 30-Dec 1, 2006):  
http://www.bioontology.org/wiki/index.php/PATO_Meeting

Each of these is directly relevant to the use of the CTO being  
developed by this group.

There is also an informal effort underway in this same context for  
associating "speech acts" with the annotations, which relates both to  
the "acts" as represented in HL7 RIM and similar contextual  
qualifiers as represented in ontologies such as DOLCE & GALEN.

Cheers,
Bill


On Dec 6, 2006, at 10:15 AM, Kashyap, Vipul wrote:

>
>
>
>
> Hi Chimezie,
> I have just briefly looked through the introduction to your Problem  
> Oriented Medical Record ontology and I find it very interesting. In  
> the same way as HL7 RIM Acts has a "ontological failings" in mixing  
> up the "concept of the action of recording clinical information and  
> the phenomena being described by such recordings", CDISC (Clinical  
> Data Interchange Standards Consortium) fundamental Observations has  
> a similar problem.
>
> [VK] This reminds me of a discussion I had with Dan Russler and  
> similar to the above sentiment, the design choices seem to be  
> between one of the following:
>
> 1.   Modeling the action of recording a clinical information; or  
> ordering a drug; or making a diagnosis, …
>
> 2.   Modeling the information required for performing the action;  
> and the information as a result of the outcome of the action…
>
> The argument appears to be that the use case in requires metrics  
> such as cost and efficiency of hospital operations. etc. That said,  
> one still needs to differentiate the two and chances are
> you may want to model (2) as opposed to (1)
>
> It will be great if Kerstin can point us to the “use cases” and  
> metrics in the Pharma context which would drive the modeling efforts.
>
> Cheers,
>
>      ---Vipul
>
>

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