Recent organizational, disjointness axiom, and annotation updates to CPR
Chime Ogbuji
------------------------------------
Every laboratory test finding is a clinical finding that [sic] measurement of nothing but QualityValues and that [sic] output of nothing but laboratory tests and that represents nothing but qualities .
Everything that is had [sic] causally active participant by [sic] a morphologic alteration is a pathological disposition .
Every physical anatomical entity is a continuant that is an anatomical Structure or that is an immaterial anatomical continuant .
Every sign recording is a clinical finding that is the output of a clinical examination [and vice versa]
Every patient ' s record of a symptom [is] composed by a patient .
Every patient ' s record of a symptom is a clinical artifact .
Every patient ' s record of a symptom is the output of something that is a medical history screening act or that is a self examination .
Every patient ' s record of a symptom [is the] output of something that is a medical history screening act or that is a self examination .
Every patient ' s record of a symptom represents a bodily feature .
Every [muo] physical quality is a quality . Every quality is a physical quality .
---------- Other (disease-related) renderings via ACE View ------
Every pathological disposition actively participates in a pathological process that has ( as a part ) a morphologic alteration .
Every material pathological entity participates in a morphologic alteration .
Everything that is a part of a morphologic alteration is a pathological process .
Everything that is had causally active participant by a morphologic alteration is a pathological disposition .
Every pathological disposition is something that caused [sic] by an etiologic agent or that participates in a biological process that has ( as a participant ) an etiologic agent .
[1] http://code.google.com/p/cpr-ontology/source/detail?r=7
--
Chime Ogbuji
Sent with Sparrow
Alan Ruttenberg
Attempto is fun. I'll have to try it :) But it will be hard to forgive the annotation property names. And protege needs to be taught that search should find the attempto labels.
I recently added [1] Attempto Controlled English (ACE) annotations for use of the ACEView plugin for protege to render (with minor edits) recent logical changes:
------------------------------------
Every laboratory test finding is a clinical finding that [sic] measurement of nothing but QualityValues and that [sic] output of nothing but laboratory tests and that represents nothing but qualities .
Everything that is had [sic] causally active participant by [sic] a morphologic alteration is a pathological disposition .
Every physical anatomical entity is a continuant that is an anatomical Structure or that is an immaterial anatomical continuant .
Every sign recording is a clinical finding that is the output of a clinical examination [and vice versa]
Every patient ' s record of a symptom [is] composed by a patient .
Every patient ' s record of a symptom is a clinical artifact .
Every patient ' s record of a symptom is the output of something that is a medical history screening act or that is a self examination .
Every patient ' s record of a symptom [is the] output of something that is a medical history screening act or that is a self examination .
Every patient ' s record of a symptom represents a bodily feature .
Every [muo] physical quality is a quality . Every quality is a physical quality .
---------- Other (disease-related) renderings via ACE View ------
Every pathological disposition actively participates in a pathological process that has ( as a part ) a morphologic alteration .
Every material pathological entity participates in a morphologic alteration .
Everything that is a part of a morphologic alteration is a pathological process .
Everything that is had causally active participant by a morphologic alteration is a pathological disposition .
Every pathological disposition is something that caused [sic] by an etiologic agent or that participates in a biological process that has ( as a participant ) an etiologic agent .
Chime Ogbuji
See my response inline below.
--
Chime Ogbuji
Sent with Sparrow
On Tuesday, August 2, 2011 at 6:40 PM, Alan Ruttenberg wrote:
> Some unsats (see attached)
I think I have addressed them in my recent commit.
> Attempto is fun. I'll have to try it :)
Yes, I really like it for documentation and management purposes as well (it is much better than regular ManOWL sometimes).
> But it will be hard to forgive the annotation property names. And protege needs to be taught that search should find the attempto labels.
Absolutely
> > ------------------------------------
> > Every laboratory test finding is a clinical finding that [sic] measurement of nothing but QualityValues and that [sic] output of nothing but laboratory tests and that represents nothing but qualities .What about laboratory tests that measure times, e.g. clotting time?
The measurement of nothing but QualityValues disjunct should be removed (issue added) so it is becomes:
Every laboratory test finding is a clinical finding that [is a] measurement of nothing but QualityValues and that is the output of nothing but laboratory tests and that represents nothing but qualities [and vice versa].
> > Everything that is had [sic] causally active participant by [sic] a morphologic alteration is a pathological disposition .
> attempto: s/morphologic alteration/pathological morphologic alteration/.
> Else confusion about, e.g. puberty.
Changed (All ACE annotations in trunk).
> > Every physical anatomical entity is a continuant that is an anatomical Structure or that is an immaterial anatomical continuant .
> attempto immaterial anatomical continuant -> site
Changed.
> > Every sign recording is a clinical finding that is the output of a clinical examination [and vice versa]
> Looks wrong, given that clinical examinations ought to be able to have other clinical finding typed things as output.
sign-recording should be a primitive (issue added)
> > Every patient ' s record of a symptom [is] composed by a patient .
> > Every patient ' s record of a symptom is a clinical artifact .Even if they record it in their diary?
Yes. It would still be clinical relevant, wouldn't it?
> > Every patient ' s record of a symptom is the output of something that is a medical history screening act or that is a self examination .
> What about stuff that their mom, or spouse observes?
anamnesis? I think SNOMED-CT refers those as Family history (SNOMED-CT clinical situation).
> > Every patient ' s record of a symptom [is the] output of something that is a medical history screening act or that is a self examination .
> same as above?
Using SNOMED-CT as a reference, I think this would correspond to the use of the SUBJECT RELATIONSHIP CONTEXT attribute (similar to cpr:subjectOfDescription) on an artifact composed by the mom, spouse, etc.
> > Every patient ' s record of a symptom represents a bodily feature .
> Even patients whose symptoms include being bothered by imaginary bugs?
Heh :) I hadn't considered mental illnesses comprehensively until I attended Jena's session. I have added an issue.
> > Every [muo] physical quality is a quality . Every quality is a physical quality .
> I think it would be more correct to say that BFO:quality is a subclass of MUO:physical quality. BFO doesn't think everything that can be computed is a distinct entity.
Issue added.
> > ---------- Other (disease-related) renderings via ACE View ------
> > Every pathological disposition actively participates in a pathological process that has ( as a part ) a morphologic alteration .In BFO we would have that the relation is realization, rather than participation.
Is (bfo2?:realized_in) different from ro:agent_in?
> > Every material pathological entity participates in a morphologic alteration .
> This one is a bit confusing because of the temporal thing.
Yes. :(
I guess one could use ro:derives_from with SOME anatomical entity, as a disjunct?
> It's not clear if you mean always. But certainly material pathological entity can be created in some process but then become passive.
Yes
> For example a scar. Should the relation be the other way around - that all morphologic alteration has_participant some material pathological entity?
Yes, that would probably better capture the intent. (issue added)
> > Everything that is a part of a morphologic alteration is a pathological process .
> I guess this could be definitional, but I'm thinking about, e.g. the movement of blood through vessels in a growing tumor. Is the blood circulation pathological? Or not part of the morphologic alteration?
The latter. I think the ACE rendering messed up the axiom. I can't find the source of that rendered ACE sentence.
[snip]
> > Every pathological disposition is something that caused [sic] by an etiologic agent or that participates in a biological process that has ( as a participant ) an etiologic agent .
> again the disposition participating.
This is one of those "etiology causal-chain via process/continuant RO agency role chain situations."
Alan Ruttenberg
> Hey Alan. Thanks for the edits
>
> See my response inline below.
I've snipped other than what I am responding to.
>> > ------------------------------------
>> > Every laboratory test finding is a clinical finding that [sic] measurement of nothing but QualityValues and that [sic] output of nothing but laboratory tests and that represents nothing but qualities .What about laboratory tests that measure times, e.g. clotting time?
> The measurement of nothing but QualityValues disjunct should be removed (issue added) so it is becomes:
>
> Every laboratory test finding is a clinical finding that [is a] measurement of nothing but QualityValues and that is the output of nothing but laboratory tests and that represents nothing but qualities [and vice versa].
Still says "represents nothing but qualities".
>> > Every patient ' s record of a symptom [is] composed by a patient .
>> > Every patient ' s record of a symptom is a clinical artifact .Even if they record it in their diary?
> Yes. It would still be clinical relevant, wouldn't it?
It would, but the definition is for clinical artifact, and the
documentation says it corresponds to HL7 CDA Act, which says something
about default: "A healthcare service.". Not everything clinically
relevant is a healthcare service.
BTW, I'd have the links be actual URIs, rather than xsd:anyURI. And
perhaps you can put (another) request in to the protege team to make
such links clickable.
>> > Every patient ' s record of a symptom is the output of something that is a medical history screening act or that is a self examination .
>> What about stuff that their mom, or spouse observes?
> anamnesis? I think SNOMED-CT refers those as Family history (SNOMED-CT clinical situation).
Doesn't have to be a history-taking. It might be a read report from 911.
>> > Every patient ' s record of a symptom [is the] output of something that is a medical history screening act or that is a self examination .
>> same as above?
> Using SNOMED-CT as a reference, I think this would correspond to the use of the SUBJECT RELATIONSHIP CONTEXT attribute (similar to cpr:subjectOfDescription) on an artifact composed by the mom, spouse, etc.
Well, I don't know what SNOMED is thinking about, but the point
remains that "Every patient ' s record of a symptom [is the] output of
something that is a medical history screening act or that is a self
examination" (aside from the fact that the possessive on "patient"
isn't clear) is that it doesn't include mom.
>> > ---------- Other (disease-related) renderings via ACE View ------
>> > Every pathological disposition actively participates in a pathological process that has ( as a part ) a morphologic alteration .In BFO we would have that the relation is realization, rather than participation.
> Is (bfo2?:realized_in) different from ro:agent_in?
I don't know what agent_in means. Ask Werner. But I don't believe they
are considered the same in any interpretation.
>> > Every material pathological entity participates in a morphologic alteration .
>> This one is a bit confusing because of the temporal thing.
> Yes. :(
>
> I guess one could use ro:derives_from with SOME anatomical entity, as a disjunct?
Or as the sole axiom on this side, with the inverse on the process
side (as in submitted issue).
>> It's not clear if you mean always. But certainly material pathological entity can be created in some process but then become passive.
> Yes
>> For example a scar. Should the relation be the other way around - that all morphologic alteration has_participant some material pathological entity?
> Yes, that would probably better capture the intent. (issue added)
>> > Everything that is a part of a morphologic alteration is a pathological process .
>
>> I guess this could be definitional, but I'm thinking about, e.g. the movement of blood through vessels in a growing tumor. Is the blood circulation pathological? Or not part of the morphologic alteration?
> The latter. I think the ACE rendering messed up the axiom. I can't find the source of that rendered ACE sentence.
I think it is 'Process of physiological or granularity-agnostic
anatomical alteration as a result of the realization
of a pathological disposition (a disease)' part_of only 'Pathological process'
>> > Every pathological disposition is something that caused [sic] by an etiologic agent or that participates in a biological process that has ( as a participant ) an etiologic agent .
>
>> again the disposition participating.
> This is one of those "etiology causal-chain via process/continuant RO agency role chain situations."
Yah. Agree that something needs to be done here.
-Alan
Chime Ogbuji
On Thursday, August 4, 2011 at 12:37 AM, Alan Ruttenberg wrote:
> Every laboratory test finding is a clinical finding that [is a] measurement of nothing but QualityValues and that is the output of nothing but laboratory tests and that represents nothing but qualities [and vice versa].
>
> Still says "represents nothing but qualities".
> It would, but the definition is for clinical artifact, and the
> documentation says it corresponds to HL7 CDA Act, which says something
> about default: "A healthcare service.". Not everything clinically
> relevant is a healthcare service.
> BTW, I'd have the links be actual URIs, rather than xsd:anyURI. And
> perhaps you can put (another) request in to the protege team to make
> such links clickable.
> > anamnesis? I think SNOMED-CT refers those as Family history (SNOMED-CT clinical situation).
>
> Well, I don't know what SNOMED is thinking about, but the point
> remains that "Every patient ' s record of a symptom [is the] output of
> something that is a medical history screening act or that is a self
> examination" (aside from the fact that the possessive on "patient"
> isn't clear) is that it doesn't include mom.
( cpr:subjectOfDescription ONLY Mom and
cpr:composedOf ONLY cpr:patient ) ?
> > > > ---------- Other (disease-related) renderings via ACE View ------
> > > > Every pathological disposition actively participates in a pathological process that has ( as a part ) a morphologic alteration .In BFO we would have that the relation is realization, rather than participation.
> > Is (bfo2?:realized_in) different from ro:agent_in?
I'm not sure, but, actually, I think of the latter as having more to do with causality than potentiality .
> > > > Every material pathological entity participates in a morphologic alteration .
> > > This one is a bit confusing because of the temporal thing.
> > Yes. :(
> >
> > I guess one could use ro:derives_from with SOME anatomical entity, as a disjunct?
They now (derives_from some 'Anatomical Structure') or (transformation_of some 'Anatomical Structure')
> I think it is 'Process of physiological or granularity-agnostic
> anatomical alteration as a result of the realization
> of a pathological disposition (a disease)' part_of only 'Pathological process'
currently the relationship is about the same (I believe) but defined form the perspective of the 'disease entity':
agent_in some is a 'Pathological process' AND ('has (as a) part' SOME 'morphologic alteration'))
-- Chime