RE: Multi-layered Knowledge Representations for Healthcare (was RE: An argument for bridging information models and ontologies at the syntactic level)

Vipul,
 
One more point.  I agree with most everything you have stated.  One distiction that we have found valuable is between high level domain models (level 2 Ontologies) and specific detailed classes of information (Example: Procedures vs the class of patients who have had a laproscopic Cholecystectomy with intra-operative cholangiogram), level 3 Ontologies.  Some of our rationale is that often this crosses the information model with the terminology model, the later for example being formed from a compositional terminological construction.
 
With warm regards,
 
Peter
 
 

________________________________

From: Kashyap, Vipul [mailto:VKASHYAP1@PARTNERS.ORG]
Sent: Sat 5/31/2008 7:00 AM
To: Elkin, Peter L., M.D.; dan.russler@oracle.com
Cc: Samson Tu; public-semweb-lifesci@w3.org; public-hcls-coi@w3.org
Subject: RE: Multi-layered Knowledge Representations for Healthcare (was RE: An argument for bridging information models and ontologies at the syntactic level)


Dear Peter,
 
Apologies for the delay in responding  There'a a lot of stuff going around right now and I needed some "think" time. Responses to your questions are included inline.


	In order to not confuse the Ontology classification with First Order / Second Order / Higher Order logics, we use Level 1 Ontologies to be domain independent (EAV just being a representational mechanism for a logical system), 
	
	[VK] Metamodels may be viewed as domain independent and based on your descriptions so far Level 1 Ontologies do appear to be similar to the Meta Model layer
	 
	 level 2 Ontologies are domain dependent (e.g. CDA), and level three contain defined instances as well as class based definitions.  We have been able to make these distinctions work across multiple projects.  If there is a level zero 
	 
	[VK]  Our Level 2 proposal seem to be similar with the difference being that my proposal seeks to spearate instances into Layer 3 and keeps classes in Layer 2. I believe that there is value in distinguishing between classes and instances. Level zero may be viewed as a simple subject-predicate-object representation as in RDF. Upper level ontologies can also be represented in this either at Level 1 or the MetaModel layer.
	 
	I believe we need a final single formal representational schema where constructions defined across Information Models and Terminological Models can be validated.  
	[VK] Absolute agreement there. There is no need to represent Infromation Models and Terminologies separately, in fact that could lead to confilcts such as the same piece of knowledge represent inconsistently in an Information Model and Terminology.
	 
	 This interlingua should be defined from transforms from all other valid logical languages and should empower all those SMEs familiar with any valid logical system to work as they are comfortable.  I
	[VK] This is a critical requirement which is important to make it all work. There have to be well defined and sound transformations from and into the interlingua from any logical system. Of course we would need to work through issues of varying expresiveness of all these formalisms.
	 
	 n the end, that work product must be validated through the common interlingua to ensure that meaning is preserved and therefore we are not creating unrecognized ambiguity.
	[VK] Agreed!
	 
	Peter: Do you have a document summarizing these thoughts which could be viewed as requirements or desiderata which you could share with us?
	 
	Thanks and Regards,
	 
	---Vipul 

The information transmitted in this electronic communication is intended only
for the person or entity to whom it is addressed and may contain confidential
and/or privileged material. Any review, retransmission, dissemination or other
use of or taking of any action in reliance upon this information by persons or
entities other than the intended recipient is prohibited. If you received this
information in error, please contact the Compliance HelpLine at 800-856-1983 and
properly dispose of this information.