search

oral-health-and-disease-ontologies / ohd-ontology

The OHD is an ontology for representing the diagnosis and treatment of dental maladies.
Creative Commons Attribution 4.0 International
7 stars 6 forks source link

date relations #26

Closed GoogleCodeExporter closed 8 years ago

GoogleCodeExporter commented 8 years ago 
Email from Bill

Currently we have the data properties "occurrence data" and
"birth_date" in OHD. These represent:

1. the date on which a procedure occured
2. the birth date of the patient

An alternative to having these as data properties is to represent this
information using ICEs. That is, have as a subtype of ICE something
like "patient birthday" and "procedure occurrence date".

There is an example of what these ICEs would be about. Would they be about the
patient/procedure or about the temporal region during which the birth
of the patient or the procedure occurred. For now, I think it would be
fine for them to be about the patient/procedure. Keeps of out of the
metaphysical deep water :(

Making this change feels ontologically right, but do we gain added
practical advantage? Are there queries that we can now do? One
possible advantage is that we can more cleanly represent case in which
such information changes. For example, a patient corrects an incorrect
information about his birthday. Is this enough motivation?

Original issue reported on code.google.com by alanruttenberg@gmail.com on 22 Dec 2014 at 5:31

GoogleCodeExporter commented 8 years ago 
Leave as is for now.

Original comment by wddun...@gmail.com on 6 Jan 2015 at 7:33

GoogleCodeExporter commented 8 years ago 
<< Making this change feels ontologically right, but do we gain added
practical advantage? Are there queries that we can now do? One
possible advantage is that we can more cleanly represent case in which
such information changes. For example, a patient corrects an incorrect
information about his birthday. Is this enough motivation? >>

That would be a weak justification. If the patient birthday and date of the 
procedure are correctly recorded, that information never changes. Yes, it is 
possible that either one of those  dates  is wrong sometimes, but  couldn't we, 
for the purposes of the ontology,  assume that we are representing the "true"  
dates, even if they have been corrected in the source data?

Original comment by titus.sc...@gmail.com on 7 Jan 2015 at 2:18