HCLS/ClinicalObservationsInteroperability/12-18-coi-minutes
18 December 2007 Telcon Minutes
Attendees:
Action Items:
1) Vipul is working on obtaining data related to diabetes from an endocrinologist for mark-up as RDF.
He will contact everyone individually for their input on data requirements.
2) Due to the holidays, the next teleconference will by January 8, 2008.
Presentation by Dan Russler on "The HL7 RIM from the Perspective of the Semantic Web."
Major Points:
Propositions and Assertions
Propositions are statements about things that follow the structure of subject - predicate - object Assertions are propositions that are believed to be true by the author(s)
Eric Neumann emphasized the importance of this distinction
Ontologies
Collections of assertions
Most important part of an ontology is the first segregation of things into subclasses. Affects all subsequent distinctions
Measure of the value of ontologies is how well they work to do what they were intended to do. People can disagree about the truth value of ontologies.
HL7 RIM
Developed by physical scientists
Consequently makes fundamental distinction between mass (entities) and energy (acts) Next distinction between materials and labor
Eric Neumann asked if the entity-act distinction is another way to construct a proposition and can be rendered in RDF. Dan responded that this is true. There is a clear relationship between the HL7 RIM and RDF as models.
RIM Backbone: Entity (thing) - Act (events) - Associations (connections between entities and acts)
Associative Classes: Role (social) and Participation (relational)
Recursive Association: ActRelationship recursion models components of acts. These recursions can be infinite.
Vipul and Dan have a brief discussion of the nature of classes and instances without clear resolution. Agree to continue discussion offline.
HL7 and RDF
Goal is to map the subject-predicate-object propositions in HL7 to RDF triples. Dan believes that this will be fairly straight forward given the structure of HL7 RIM.