- From: Vipul Kashyap <kashyap.vipul@gmail.com>
- Date: Sat, 13 Dec 2014 13:52:25 -0500
- To: <John.E.Mattison@kp.org>, <grahame@healthintersections.com.au>
- Cc: <lloyd@lmckenzie.com>, <david@dbooth.org>, <public-semweb-lifesci@w3.org>, <its@lists.hl7.org>
- Message-ID: <00da01d01705$f0d683d0$d2838b70$@gmail.com>
I agree with John – the best way forward is to have “purpose based ontologies” and incrementally generalize them where there is clear value and need for doing so.
From: John.E.Mattison@kp.org [mailto:John.E.Mattison@kp.org]
Sent: Thursday, December 11, 2014 9:55 AM
To: grahame@healthintersections.com.au
Cc: lloyd@lmckenzie.com; kashyap.vipul@gmail.com; david@dbooth.org; public-semweb-lifesci@w3.org; its@lists.hl7.org
Subject: Re: Minutes of last week's (Dec 2) HL7 ITS RDF Subgroup / W3C HCLS COI call -- Review of FHIR ontology approaches (cont.)
One of my favorite tweets was that "Ontology is a one word oxymoron". Since ontologies are intended to be faithful to a specific set of both design principles and objectives for using the relevant data and data relationships, the "dream ontology" is in the eye of the beholder. What is best for clinicians is different than for researchers, or for decision support engines, or workflow engines, or super-subsetted or specialized purposes. There is a classic tradeoff between expressivity and reproducibility, and ontologies often represent that battleground, not because of political differences of opinion, but because of different priorities and focus for the intended use. While it is useful to think of a more generalized ontology to serve many purposes, it will be difficult to canonically define the relationships between any two or more concepts in ANY ontology and expect that those canonical relationship to survive either current or future knowledge about each of those concepts and how they are related.
It is more practical to think of coordinated "purpose-based ontologies" where there is cross-fertilization of SMEs and use cases, but there will unlikely be a time when a single universal ontology will survive the current, let alone the exponentially rising, growth in both data and knowledge about entities, concepts, and their "fundamental relationships".
just my two cents.
John
Sent from my iPad
On Dec 11, 2014, at 1:38 AM, "Grahame Grieve" <grahame@healthintersections.com.au> wrote:
If every FHIR element was mapped to a snomed term, then you could represent that in RDF no problems.
However the problem with this is that we already have a slot for mapping an element to it's snomed code, but there are hardly any snomed codes that are appropriate.
Ihtsdo are planning to have their own go at flapping these next year, I imagine that they'll have to create a lot of codes to are it useful.
Grahame
On Thursday, December 11, 2014, Lloyd McKenzie <lloyd@lmckenzie.com> wrote:
Hi Vipul,
To me, your first two points go together. The only reason to do an RDF representation is to support implementations that want to make use of semantic web technologies for inference purposes.
I wouldn't describe either as requirements for the syntax though. A requirement would be a particular type of inference that we wanted to enable (or wanted to optimize for).
In terms of your third point, there's two different ways we could interpret this. One would be to say that we'll convert both FHIR structures and FHIR vocabulary into RDF and include the linkage between them as part of the class model. I'm definitely on-side with that. The second way would be to say that we'll be able to easily integrate existing ontologies from other code systems such as SNOMED with the FHIR structure ontology. I think that's a much tougher thing to do, though it may be possible in some cases. I'd like to treat it as a secondary objective, but am ok with having it on the table.
(Feel free to add your thoughts to the wiki page.)
Lloyd
--------------------------------------
Lloyd McKenzie
+1-780-993-9501
Note: Unless explicitly stated otherwise, the opinions and positions expressed in this e-mail do not necessarily reflect those of my clients nor those of the organizations with whom I hold governance positions.
On Wed, Dec 10, 2014 at 2:59 PM, Vipul Kashyap <kashyap.vipul@gmail.com <javascript:_e(%7B%7D,'cvml','kashyap.vipul@gmail.com');> > wrote:
Good list, Lloyd. Would like to suggest some more additions (apologies if these have already been suggested).
· Clearly articulate the value of the new RDF/RDFS/OWL representation over the current XML/JSON representation
· Enablement of OWL/RDFS inference – so we could identify use cases that cannot be easily done based on the XML/JSON representation
· A common OWL/RDFS representation for information model elements and medical terminology concepts.
Thoughts. Suggestions?
---Vipul
From: Lloyd McKenzie [mailto:lloyd@lmckenzie.com <javascript:_e(%7B%7D,'cvml','lloyd@lmckenzie.com');> ]
Sent: Monday, December 08, 2014 1:36 PM
To: David Booth
Cc: w3c semweb HCLS; its@lists.hl7.org <javascript:_e(%7B%7D,'cvml','its@lists.hl7.org');>
Subject: Re: Minutes of last week's (Dec 2) HL7 ITS RDF Subgroup / W3C HCLS COI call -- Review of FHIR ontology approaches (cont.)
I think we need to define our objectives for the RDF representation. Mine are as follows:
1. It must be possible to round-trip from XML/JSON through RDF representation
* This includes retaining information about order of repeating elements
* Needs to allow for extensions where-ever they can appear, including simple types (date, boolean, etc.)
2. We want to be able to represent instances as RDF and Profiles as OWL/RDFS
3. Syntax needs to be "safe" when dealing with modifier extensions
4. Syntax should support vocabulary bindings to code, Coding and CodeableConcept - including dealing with extensible value sets and multi-code system value sets
5. Syntax should enforce constraints that are representable in RDF (i.e. schema constraints, regular expressions, etc.)
6. In the RDFS/OWL, should expose at least minimal annotation information for display
Lloyd
--------------------------------------
Lloyd McKenzie
+1-780-993-9501 <tel:%2B1-780-993-9501>
Note: Unless explicitly stated otherwise, the opinions and positions expressed in this e-mail do not necessarily reflect those of my clients nor those of the organizations with whom I hold governance positions.
On Mon, Dec 8, 2014 at 11:11 AM, David Booth <david@dbooth.org <javascript:_e(%7B%7D,'cvml','david@dbooth.org');> > wrote:
I'm so sorry I forgot to send these out last Tuesday, but here are draft minutes from our call, with Eric Prud'hommeaux reviewing his FHIR ontology approach:
http://www.w3.org/2014/12/02-hcls-minutes.html
and below in plain text.
Also, as a reminder, tomorrow's call (Tuesday) will continue with Claude Nanjo reviewing his FHIR ontology approach.
Thanks,
David Booth
--------------------------------------------------------
[1]W3C
[1] http://www.w3.org/
- DRAFT -
Semantic Web Health Care and Life Sciences Interest Group Teleconference
02 Dec 2014
See also: [2]IRC log
[2] http://www.w3.org/2014/12/02-hcls-irc
Attendees
Present
Bryn_Rhodes, Cati, Claude_Nanjo, David_Booth, EricP,
Guoqian, Hans_Cools, Ingeborg, Joshua_Phillips,
Kerstin_Forsberg, Marc_Twagirumukiza, Neda, Paul_Knapp,
TimW, Tony_Mallia, Charlie_Mead, egonw_(IRC_only?),
Scott_Marshall, Patricia, Rob_Hausam, Vassil_(IRC_only?)
Regrets
Chair
David Booth (and Paul Knapp)
Scribe
dbooth
Contents
* [3]Topics
1. [4]Approve Minutes of previous meetings
2. [5]Action Review
3. [6]FHIR Ontology Review
* [7]Summary of Action Items
__________________________________________________________
<trackbot> Date: 02 December 2014
<ericP> trackbot, start meeting
<trackbot> Meeting: Semantic Web Health Care and Life Sciences
Interest Group Teleconference
<trackbot> Date: 02 December 2014
<ericP> oops
<TimW> TimW is from +1.919.767...
<RHausam> RHausam is 801.949.1556
<Claude> [8]https://global.gotomeeting.com/join/157514853
[8] https://global.gotomeeting.com/join/157514853
<Claude> Please join our GoToMeeting
Approve Minutes of previous meetings
Nov 18:
[9]http://wiki.hl7.org/index.php?title=ITS_RDF_Concall_Minutes_
20141118
[9] http://wiki.hl7.org/index.php?title=ITS_RDF_Concall_Minutes_20141118
Nov 25:
[10]http://wiki.hl7.org/index.php?title=ITS_RDF_Concall_Minutes
_20141125
[10] http://wiki.hl7.org/index.php?title=ITS_RDF_Concall_Minutes_20141125
Nov 18 minutes unanimously approved.
Nov 25 minutes unanimously approved.
Action Review
<scribe> ACTION: ericP to set up tracker [recorded in
[11]http://www.w3.org/2014/11/25-hcls-minutes.html#action01 --
PENDING
[11] http://www.w3.org/2014/11/25-hcls-minutes.html#action01
<trackbot> Created ACTION-2 - Set up tracker [recorded in
[12]http://www.w3.org/2014/11/25-hcls-minutes.html#action01 --
pending [on Eric Prud'hommeaux - due 2014-12-09].
[12] http://www.w3.org/2014/11/25-hcls-minutes.html#action01
<scribe> ACTION: Tony to find out more details about how iCat
handles ICD-11 ont and report back [recorded in
[13]http://www.w3.org/2014/11/18-hcls-minutes.html#action01 --
PENDING
[13] http://www.w3.org/2014/11/18-hcls-minutes.html#action01
<trackbot> Error finding 'Tony'. You can review and register
nicknames at <[14]http://www.w3.org/2014/HCLS/track/users>.
[14] http://www.w3.org/2014/HCLS/track/users
<scribe> ACTION: Guoqian to figure out whether he can share URI
conventions for ICD-11 [recorded in
[15]http://www.w3.org/2014/11/25-hcls-minutes.html#action07 --
PENDING
[15] http://www.w3.org/2014/11/25-hcls-minutes.html#action07
<trackbot> Created ACTION-3 - Figure out whether he can share
uri conventions for icd-11 [recorded in
[16]http://www.w3.org/2014/11/25-hcls-minutes.html#action07 --
pending [on Guoqian Jiang - due 2014-12-09].
[16] http://www.w3.org/2014/11/25-hcls-minutes.html#action07
<Claude> For those who just joined IRC, we also have a
GoToMeeting at:
<Claude> [17]https://global.gotomeeting.com/join/157514853
[17] https://global.gotomeeting.com/join/157514853
<scribe> ACTION: Kerstin and Ingeborg to prepare a status and
future state ideas for PhUSE-FDA work [recorded in
[18]http://www.w3.org/2014/11/18-hcls-minutes.html#action05 --
PENDING
[18] http://www.w3.org/2014/11/18-hcls-minutes.html#action05
<trackbot> Error finding 'Kerstin'. You can review and register
nicknames at <[19]http://www.w3.org/2014/HCLS/track/users>.
[19] http://www.w3.org/2014/HCLS/track/users
<scribe> ACTION: Eric to establish/make a wiki page for C-CDA
RDF representations work [recorded in
[20]http://www.w3.org/2014/11/18-hcls-minutes.html#action06 --
PENDING
[20] http://www.w3.org/2014/11/18-hcls-minutes.html#action06
<trackbot> Created ACTION-4 - Establish/make a wiki page for
c-cda rdf representations work [recorded in
[21]http://www.w3.org/2014/11/18-hcls-minutes.html#action06 --
pending [on Eric Prud'hommeaux - due 2014-12-09].
[21] http://www.w3.org/2014/11/18-hcls-minutes.html#action06
<scribe> ACTION: Eric and Joshua to report on C-CDA RDF
representations work plan [recorded in
[22]http://www.w3.org/2014/11/18-hcls-minutes.html#action07 --
PENDING
[22] http://www.w3.org/2014/11/18-hcls-minutes.html#action07
<trackbot> Created ACTION-5 - And joshua to report on c-cda rdf
representations work plan [recorded in
[23]http://www.w3.org/2014/11/18-hcls-minutes.html#action07 --
pending [on Eric Prud'hommeaux - due 2014-12-09].
[23] http://www.w3.org/2014/11/18-hcls-minutes.html#action07
Joshua: Sent email describing a use case. Will add to wiki
page.
<scribe> ACTION: Tony and Rob to report their plan on
High-level concept mapping to RDF work [recorded in
[24]http://www.w3.org/2014/11/18-hcls-minutes.html#action08 --
PENDING
[24] http://www.w3.org/2014/11/18-hcls-minutes.html#action08
<trackbot> Error finding 'Tony'. You can review and register
nicknames at <[25]http://www.w3.org/2014/HCLS/track/users>.
[25] http://www.w3.org/2014/HCLS/track/users
<scribe> ACTION: Tony and all to decide on a wiki for Term Info
work [recorded in
[26]http://www.w3.org/2014/11/18-hcls-minutes.html#action09] --
PENDING
[26] http://www.w3.org/2014/11/18-hcls-minutes.html#action09]
<trackbot> Error finding 'Tony'. You can review and register
nicknames at <[27]http://www.w3.org/2014/HCLS/track/users>.
[27] http://www.w3.org/2014/HCLS/track/users
<scribe> ACTION: Guoqian to figure out whether he can share URI
conventions for ICD-11 [recorded in
[28]http://www.w3.org/2014/11/25-hcls-minutes.html#action07]
[28] http://www.w3.org/2014/11/25-hcls-minutes.html#action07]
<trackbot> Created ACTION-6 - Figure out whether he can share
uri conventions for icd-11 [recorded in
[29]http://www.w3.org/2014/11/25-hcls-minutes.html#action07]
[on Guoqian Jiang - due 2014-12-09].
[29] http://www.w3.org/2014/11/25-hcls-minutes.html#action07]
<scribe> [PENDING]
FHIR Ontology Review
<Claude> Please use GoToMeeting only for video
[30]https://global.gotomeeting.com/join/157514853
[30] https://global.gotomeeting.com/join/157514853
Access Code: 157-514-853
Eric's slides: ->
[31]http://www.w3.org/2014/Talks/1125-fhir-rdf-egp/ FHIR-RDF
[31] http://www.w3.org/2014/Talks/1125-fhir-rdf-egp/
Eric: Our goal was to let anything in FHIR XML be translated
into RDF.
... We wrote something that read the XML definition files and
produced something that maps the XML to RDF.
... Python code reads the json definition files and spits out
XML, being the parts that we need of the FHIR spec.
... "subs" is what is allowed inside a FHIR resource.
... effectively the attributes of a resource.
... Then we embedded that in an XSLT script, and hand edited
the foot of it.
... The stuff we wrote into the footer of the XSLT says that
for a particular construct in the XML instance data, there's
special handling, such as URLs for identifiers.
David: How does the fact that it is an atom feed affect the
interpretatinon of the XML?
Eric: It leaves some other graph stuff superimposed on it.
Marc: Re datatypes, you used FHIR value, but the value itself
looks like it is a string.
... If it is a string, then reasoners cannot do much with it.
Eric: Datatype values are showing up as literals.
Marc: Another is that a date is just a string, not an xsd:date.
... On slide 2
Eric: Ideally it should have the datatype stuck on the end of
it. There's a tension between having the datatype on the
fhir:value and having it on a blank node.
... The natural RDF-ish way to do it would be using
xsd:datetime.
... Need to decide which of these ways will be most palatable
to RDF folks versus native FHIR folks.
... Sometimes these things are not just datetimes in FHIR. Need
to make it as simple as possible and no simpler.
Guoqian: Is a different URI being used for the namespaces? Will
it cause problems when you use SPARQL across different models?
Eric: I tested some of this. I stuck a bunch of these
namespaces together, and I believe we've addressed your
concern.
David: Is the datatype issue due to the fact that datatypes are
extensible in FHIR?
Eric: Maybe. RDF needs to be monotonic.
Paul: Modifying extensions can appear only in the roots of
classes (considering the complex type as a class)
Eric: The XSLT footer has the custom code for extensions.
Paul: We did that to limit what you need to examine, to find
out if there is a modifying extension.
Claude: I think you can have modifying extensions on the root
type and anything that is defined from the root type.
ISSUE: FHIR Modifying extensions and monotonicity
<trackbot> Created ISSUE-1 - Fhir modifying extensions and
monotonicity. Please complete additional details at
<[32]http://www.w3.org/2014/HCLS/track/issues/1/edit>.
[32] http://www.w3.org/2014/HCLS/track/issues/1/edit
Eric: Example here may not be up to date. Anyone know?
Paul: the dev site has DSTU 2.
Eric: Another script generates the ShEx definitions from the
FHIR spec.
... and the ShEx is used to generate FHIR XML back again from
FHIR RDF.
... Slide 7 is showing how to take C-CDA in RDF and producing
FHIR RDF.
Guoqian: The result prefix is fhir but should be patient.
Eric: Yes, you're right. That can be fixed in the ShEx.
Tony: Sort of verbatim translation. The difference is maybe
target audience style -- a style that is good for the general
tools used in RDF. Only difference is the end style of RDF.
... Establishing closure between FHIR ontology and SNOMED-CT
would need to happen. Main difference is in the final RDF
style.
<pknapp> Zakim: Have to leave for another meeting, thx, great
work.
Eric: Will need to need to be a balance.
... Once we have transliterated, what we map to our dream
ontology. May want to balance how much is wedged into the XSLT
and how much into the interpretive dance later.
Claude: Might want a low level FHIR ont and a higher level FHIR
ont both.
David: I kind of like the auto-generated aspect of this
approach.
<vassil> чуит
ADJOURNED
<vassil> quit
<ericP> [33]coding mapping example
[33] https://www.w3.org/wiki/HCLS/ClinicalObservationsInteroperability/FDATherapeuticAreaOntologies#codeAndSystemToIRI
<ericP> (for next week)
<vassil> exit
Summary of Action Items
[PENDING] ACTION: Eric and Joshua to report on C-CDA RDF
representations work plan [recorded in
[34]http://www.w3.org/2014/11/18-hcls-minutes.html#action07]
[PENDING] ACTION: Eric to establish/make a wiki page for C-CDA
RDF representations work [recorded in
[35]http://www.w3.org/2014/11/18-hcls-minutes.html#action06]
[PENDING] ACTION: ericP to set up tracker [recorded in
[36]http://www.w3.org/2014/11/25-hcls-minutes.html#action01]
[PENDING] ACTION: Guoqian to figure out whether he can share
URI conventions for ICD-11 [recorded in
[37]http://www.w3.org/2014/11/25-hcls-minutes.html#action07]
[PENDING] ACTION: Kerstin and Ingeborg to prepare a status and
future state ideas for PhUSE-FDA work [recorded in
[38]http://www.w3.org/2014/11/18-hcls-minutes.html#action05]
[PENDING] ACTION: Tony and all to decide on a wiki for Term
Info work [recorded in
[39]http://www.w3.org/2014/11/18-hcls-minutes.html#action09]
[PENDING] ACTION: Tony and Rob to report their plan on
High-level concept mapping to RDF work [recorded in
[40]http://www.w3.org/2014/11/18-hcls-minutes.html#action08]
[PENDING] ACTION: Tony to find out more details about how iCat
handles ICD-11 ont and report back [recorded in
[41]http://www.w3.org/2014/11/18-hcls-minutes.html#action01]
[34] http://www.w3.org/2014/11/18-hcls-minutes.html#action07
[35] http://www.w3.org/2014/11/18-hcls-minutes.html#action06
[36] http://www.w3.org/2014/11/25-hcls-minutes.html#action01
[37] http://www.w3.org/2014/11/25-hcls-minutes.html#action07
[38] http://www.w3.org/2014/11/18-hcls-minutes.html#action05
[39] http://www.w3.org/2014/11/18-hcls-minutes.html#action09
[40] http://www.w3.org/2014/11/18-hcls-minutes.html#action08
[41] http://www.w3.org/2014/11/18-hcls-minutes.html#action01
[End of minutes]
__________________________________________________________
Minutes formatted by David Booth's [42]scribe.perl version
1.140 ([43]CVS log)
$Date: 2014-12-02 17:30:25 $
__________________________________________________________
[42] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm
[43] http://dev.w3.org/cvsweb/2002/scribe/
Scribe.perl diagnostic output
[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.140 of Date: 2014-11-06 18:16:30
Check for newer version at [44]http://dev.w3.org/cvsweb/~checkout~/2002/
scribe/
[44] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/
Guessing input format: RRSAgent_Text_Format (score 1.00)
No ScribeNick specified. Guessing ScribeNick: dbooth
Inferring Scribes: dbooth
Default Present: DBooth, +1.919.767.aaaa, ericP, rhausam, TimW, Joshua_P
hillips, +1.202.528.aabb, charlie, +1.469.226.aacc, Tony, Neda, patricia
, Mark_Twagirumukiza, Kerstin_Forsberg, Cati, Kerstin, +1.801.368.aadd,
+1.604.250.aaee, Bryn_Rhodes, +31.62.427.aaff, mscottm, Guoqian, +1.608.
310.aagg, vassil, +41.78.847.aahh, [IPcaller]
WARNING: Replacing previous Present list. (Old list: Bryn_Rhodes, Cati,
Claude_Nanjo, David_Booth, EricP, Guoqian, Hans_Cools, Ingeborg, Joshua_
Phillips, Kerstin_Forsberg, Marc_Twagirumukiza, Neda, Paul_Knapp, TimW,
Tony_Mallia, Charlie_Mead, egonw, (IRC, only?), Scott_Marshall, Patricia
, Rob_Hausam, Vassil, (IRC, only?))
Use 'Present+ ... ' if you meant to add people without replacing the lis
t,
such as: <dbooth> Present+ Bryn_Rhodes, Cati, Claude_Nanjo, David_Booth,
EricP, Guoqian, Hans_Cools, Ingeborg, Joshua_Phillips, Kerstin_Forsberg
, Marc_Twagirumukiza, Neda, Paul_Knapp, TimW, Tony_Mallia, Charlie_Mead,
egonw_(IRC_only?), Scott_Marshall, Patricia, Rob_Hausam, Vassil_(IRC_on
ly?)
Present: Bryn_Rhodes Cati Claude_Nanjo David_Booth EricP Guoqian Hans_Co
ols Ingeborg Joshua_Phillips Kerstin_Forsberg Marc_Twagirumukiza Neda Pa
ul_Knapp TimW Tony_Mallia Charlie_Mead egonw_(IRC_only?) Scott_Marshall
Patricia Rob_Hausam Vassil_(IRC_only?)
Found Date: 02 Dec 2014
Guessing minutes URL: [45]http://www.w3.org/2014/12/02-hcls-minutes.html
People with action items: all eric ericp guoqian ingeborg joshua kerstin
rob tony
[45] http://www.w3.org/2014/12/02-hcls-minutes.html
[End of [46]scribe.perl diagnostic output]
[46] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm
***********************************************************************************
Manage subscriptions - http://www.HL7.org/listservice
View archives - http://lists.HL7.org/read/?forum=its
Unsubscribe - http://www.HL7.org/tools/unsubscribe.cfm?email=lloyd@lmckenzie.com <http://www.HL7.org/tools/unsubscribe.cfm?email=lloyd@lmckenzie.com&list=its> &list=its
Terms of use - http://www.HL7.org/myhl7/managelistservs.cfm?ref=nav#listrules
***********************************************************************************
Manage your subscriptions <http://www.HL7.org/listservice> | View the archives <http://lists.HL7.org/read/?forum=its> | Unsubscribe <http://www.HL7.org/tools/unsubscribe.cfm?email=grahame@healthintersections.com.au&list=its> | Terms of use <http://www.HL7.org/myhl7/managelistservs.cfm?ref=nav#listrules>
--
-----
http://www.healthintersections.com.au / grahame@healthintersections.com.au / +61 411 867 065
***********************************************************************************
Manage your subscriptions <http://www.HL7.org/listservice> | View the archives <http://lists.HL7.org/read/?forum=its> | Unsubscribe <http://www.HL7.org/tools/unsubscribe.cfm?email=john.e.mattison@kp.org&list=its> | Terms of use <http://www.HL7.org/myhl7/managelistservs.cfm?ref=nav#listrules>
Received on Saturday, 13 December 2014 18:52:59 UTC