Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Modeling » MDT (Model Development Tools) » [Announce] SBVR component proposal
[Announce] SBVR component proposal [message #378049] Tue, 15 January 2008 16:14 Go to next message
Kenn Hussey is currently offline Kenn Hussey
Messages: 1618
Registered: July 2009
Senior Member
A proposal to create a new SBVR component as part of the MDT subproject has
been posted to Eclipsepedia (see http://wiki.eclipse.org/MDT-SBVR-Proposal).
Technical discussions about the formation, scope, and direction of this new
component will take place on the eclipse.modeling.mdt newsgroup (please
prefix the subject with [SBVR]). If you have comments or ideas, or interest
in contributing to the component, please join in the discussion!
SBVR component proposal- interest in healthcare application [message #378136 is a reply to message #378049] Tue, 25 March 2008 14:51 Go to previous messageGo to next message
Eclipse User
Originally posted by: dgleet.bellsouth.net

I have an interest in participating in the MDT-SBVR project, with my focus
on the healthcare domain, specifically establishing metamodels and sample
implementations for RHIO(http://en.wikipedia.org/wiki/RHIO) plugins, both
in a domestic setting and in cross-language/governmental settings, such as
the Caribbean. Specific emphasis will be on "intelligent" plugins that
will provide early alerts for infectious disease and recommend/control
drug efficacy.
Re: SBVR component proposal- interest in healthcare application [message #378137 is a reply to message #378136] Wed, 26 March 2008 14:26 Go to previous messageGo to next message
Dave Carlson is currently offline Dave Carlson
Messages: 399
Registered: July 2009
Senior Member
Hi Duane,
I look forward to your participatin in the SBVR project. I also have a a
focus on healthcare applications -- indeed, that is one of the reasons why I
wrote the SBVR project proposal. Initially, I am starting to investigate
use of SBVR to represent vocabulary and rules for CDAr2 health record
conformance constraints. Thanks for the link to RHIO. I am providing some
tooling support to an architect working on the NHIN prototype, but had not
read anything on RHIO.

Your requirments and use cases for SBVR will be very helpful in starting
this new project. Our first priority is to create, refine, and test the
SBVR metamodel based on the adopted OMG specification. For details on this
discussion (starting soon...) subscribe to the mdt-sbvr.dev mail list.

Regards,
Dave Carlson

"Duane Leet" <dgleet@bellsouth.net> wrote in message
news:fab4fd81fa63aebfc357f92f38d7a9a7$1@www.eclipse.org...
>I have an interest in participating in the MDT-SBVR project, with my focus
>on the healthcare domain, specifically establishing metamodels and sample
>implementations for RHIO(http://en.wikipedia.org/wiki/RHIO) plugins, both
>in a domestic setting and in cross-language/governmental settings, such as
>the Caribbean. Specific emphasis will be on "intelligent" plugins that
>will provide early alerts for infectious disease and recommend/control drug
>efficacy.
Re: SBVR component proposal- interest in healthcare application [message #378138 is a reply to message #378137] Wed, 26 March 2008 16:21 Go to previous messageGo to next message
Eclipse User
Originally posted by: dgleet.bellsouth.net

I want to make sure I understand the background of the participants with
respect to healthcare IT First, google RHIO. You'll see much literature.
The defacto "implementation neutral" professional site is
http://www.himss.org/ASP/topics_rhio.asp. (The root url is the Health
Information and Management Systems Society.) Second, the defacto
information exchange xml standard is HL7 (http://www.hl7.org). There's a
metamodel (RIT) and a specific common implementation throughout the
industry which is NOT XML, but rather value/delimiter text based.
Industry implementations sometimes give lip service to ebXML and meta
models, but the practice is more "primitive." Nevertheless, the
implementations appear to be mappable to the ebXML architecture, etc.
There's some low hanging fruit to match the state of the industry to
meta-models. Funding?

I'm not aware of any UML use amongst the major providers of healthcare
clinical software. I'd like to know if there is. Maybe IBM?

I'm also not aware of any use of business rules components; they are all
data components, with some capability to represent security rules. I'd
like to know if there are any implementations using standard models.

In order for health care delivery and management to progress and use IT to
its full potential, rather than be a 1970s style terminal/database,
database analysis implementation, business rules metamodels specific to
healthcare must be derived. In my experience, funding for development at
the metamodel/metaschema level is difficult to get from the hospital,
pharma, insurance, and drugstore industries. Their collective noses are
too tactical. Is this your experience.

The best bet is government or medical research institution sponsorship,
which can be coupled with demonstration implementations. I'd certainly
like to know of any financing opportunities and team with others.
Re: SBVR component proposal- interest in healthcare application [message #378139 is a reply to message #378138] Thu, 27 March 2008 02:04 Go to previous messageGo to next message
Dave Carlson is currently offline Dave Carlson
Messages: 399
Registered: July 2009
Senior Member
"Duane Leet" <dgleet@bellsouth.net> wrote in message
news:149438f693850dcf558f3d78bf6f3729$1@www.eclipse.org...
>I want to make sure I understand the background of the participants with
>respect to healthcare IT

This SBVR project will not include any functionality that is specific to
healthcare. However, look at the new Open Health Tools (OHT) organization,
see www.openhealthtools.org I will be at the next OHT board meeting in
mid-April and will ask about RHIO involvement. OHT is duplicating the
governance structure of Eclipse (in fact, the Eclipse Foundation, OMG, and
HL7 are all charter members of OHT). There is no software available yet,
but several projects are getting started. Any work that I do which builds
on this SBVR project (and UML2) to add healthcare-specific functionality
will probably be contributed to OHT.

Watch for discussion forums starting at OHT, hopefully within a month or so.

> information exchange xml standard is HL7 (http://www.hl7.org). There's a
> metamodel (RIT) and a specific common implementation throughout the
> industry which is NOT XML, but rather value/delimiter text based.

I have been deeply involved in UML modeling with HL7 for the past 18 months.
But only for the newer v3 standards, and they do use XML. There are some
adopters for parts of this, mostly CDA and CCD specifications, including
clinical data interoperability between VA and DoD. Also applications in
Canada and the U.K. Tools that I have been developing that use UML for HL7
modeling will be contributed to OHT within the next few months. Thougthts
about how/where to use SBVR are still in the research stages.
Re: SBVR component proposal- interest in healthcare application [message #378280 is a reply to message #378139] Thu, 27 March 2008 14:32 Go to previous message
Eclipse User
Originally posted by: dgleet.bellsouth.net

I am excited at the prospect of participating in this. I'm retired, so I
can
spend time doing what I want to do :-).

Please bring me up the learning curve on your activities.

The RHIO market is very competitive, with IBM and GE among the major
players.

I'm in South Carolina, where there is a large active and retiree military
population; consequently, there's a large VA and DoD presence, with
associated medical records. If appropriately swift management of
infectious disease (and terrorist activity) is to be a reality, there will
have to be an interchange between state, private and government health
facilities. Just evaluating potential state and private RHIO information
will not be adequate.

Such a mechanism does not exist today; governmental rules stand in the
way. It's possible the SBVR component activity may suggest modifications
to the government's rules plus design characteristics of the health
records systems that would be acceptable to Congress.

Of course, all privacy concerns will need to be addressed.

RHIO implementations include rules and agents for security management.
There are some tools for managing protocols, but, as far as I know, they
are targeted at a research setting. If your are interested, I can dig up
the url of a company I know - it may be relevant at the SBVR level because
it indicates a boundary of functionality, which would lead to a piece of
the SBVR semantics.

RHIOs need to contain "business = infectious disease" rules for
identifying outbreaks, current treatments, and recommending the most
effective (cost, efficacy) countermeasures. Such population variables as
age, race, sex, and overall medical conditions must be considered in the
rules. (In a multinational setting, such as the Caribbean, or ,where
there is a country with semiautonomous, cultural different regions, such
as Iraq, language, culture, and government regulations must be
considered.)

Another use case is this: during a patient visit, the physician
accumulates the patient complaint and test results data, establishes a
(tentative) diagnosis, and then, in real time, this information is passed
to a local agent of the primary agent. The local agent establishes the
best treatment option for that patient, either autonomously, given a local
copy of the recommendations from the primary agent, or, in real time
dialog with the primary agent. Since the local agent is implemented by
different EMR (Electronic Medical Record) vendors, I think the SBVR
component is relevant in guiding implementation.

Within the primary agent, is a calculating engine, which may be
proprietary by company, to establish the effectiveness metrics and
prioritize the treatments.

A key critical success factor, independent of how the process is
implemented- it can even be manual- is that medical practice must be
changed to insure follow up and record treatment effectiveness; ie, did
the patient recover + any other relevant data. In terms of implementation
and the SBVR component, patient reporting might be done through a medical
web portal that results in information back to both the local EMR and,
eventually, to the primary agent. The SBVR component might be relevant in
establishing the semantics for the dialog.

In general, it seems to me any SBVR implementation will need to satisfy a
collection of use cases based on the above description.

My bottom line assertion is that a necessary condition for SBVR
specification validation is that it be validated through appropriately
robust medical disease control use cases and trial implementations.
SBVR component proposal- interest in healthcare application [message #584949 is a reply to message #378049] Tue, 25 March 2008 14:51 Go to previous message
Eclipse User
Originally posted by: dgleet.bellsouth.net

I have an interest in participating in the MDT-SBVR project, with my focus
on the healthcare domain, specifically establishing metamodels and sample
implementations for RHIO(http://en.wikipedia.org/wiki/RHIO) plugins, both
in a domestic setting and in cross-language/governmental settings, such as
the Caribbean. Specific emphasis will be on "intelligent" plugins that
will provide early alerts for infectious disease and recommend/control
drug efficacy.
Re: SBVR component proposal- interest in healthcare application [message #584963 is a reply to message #378136] Wed, 26 March 2008 14:26 Go to previous message
Dave Carlson is currently offline Dave Carlson
Messages: 399
Registered: July 2009
Senior Member
Hi Duane,
I look forward to your participatin in the SBVR project. I also have a a
focus on healthcare applications -- indeed, that is one of the reasons why I
wrote the SBVR project proposal. Initially, I am starting to investigate
use of SBVR to represent vocabulary and rules for CDAr2 health record
conformance constraints. Thanks for the link to RHIO. I am providing some
tooling support to an architect working on the NHIN prototype, but had not
read anything on RHIO.

Your requirments and use cases for SBVR will be very helpful in starting
this new project. Our first priority is to create, refine, and test the
SBVR metamodel based on the adopted OMG specification. For details on this
discussion (starting soon...) subscribe to the mdt-sbvr.dev mail list.

Regards,
Dave Carlson

"Duane Leet" <dgleet@bellsouth.net> wrote in message
news:fab4fd81fa63aebfc357f92f38d7a9a7$1@www.eclipse.org...
>I have an interest in participating in the MDT-SBVR project, with my focus
>on the healthcare domain, specifically establishing metamodels and sample
>implementations for RHIO(http://en.wikipedia.org/wiki/RHIO) plugins, both
>in a domestic setting and in cross-language/governmental settings, such as
>the Caribbean. Specific emphasis will be on "intelligent" plugins that
>will provide early alerts for infectious disease and recommend/control drug
>efficacy.
Re: SBVR component proposal- interest in healthcare application [message #584978 is a reply to message #378137] Wed, 26 March 2008 16:21 Go to previous message
Eclipse User
Originally posted by: dgleet.bellsouth.net

I want to make sure I understand the background of the participants with
respect to healthcare IT First, google RHIO. You'll see much literature.
The defacto "implementation neutral" professional site is
http://www.himss.org/ASP/topics_rhio.asp (The root url is the Health
Information and Management Systems Society.) Second, the defacto
information exchange xml standard is HL7 (http://www.hl7.org). There's a
metamodel (RIT) and a specific common implementation throughout the
industry which is NOT XML, but rather value/delimiter text based.
Industry implementations sometimes give lip service to ebXML and meta
models, but the practice is more "primitive." Nevertheless, the
implementations appear to be mappable to the ebXML architecture, etc.
There's some low hanging fruit to match the state of the industry to
meta-models. Funding?

I'm not aware of any UML use amongst the major providers of healthcare
clinical software. I'd like to know if there is. Maybe IBM?

I'm also not aware of any use of business rules components; they are all
data components, with some capability to represent security rules. I'd
like to know if there are any implementations using standard models.

In order for health care delivery and management to progress and use IT to
its full potential, rather than be a 1970s style terminal/database,
database analysis implementation, business rules metamodels specific to
healthcare must be derived. In my experience, funding for development at
the metamodel/metaschema level is difficult to get from the hospital,
pharma, insurance, and drugstore industries. Their collective noses are
too tactical. Is this your experience.

The best bet is government or medical research institution sponsorship,
which can be coupled with demonstration implementations. I'd certainly
like to know of any financing opportunities and team with others.
Re: SBVR component proposal- interest in healthcare application [message #584987 is a reply to message #378138] Thu, 27 March 2008 02:04 Go to previous message
Dave Carlson is currently offline Dave Carlson
Messages: 399
Registered: July 2009
Senior Member
"Duane Leet" <dgleet@bellsouth.net> wrote in message
news:149438f693850dcf558f3d78bf6f3729$1@www.eclipse.org...
>I want to make sure I understand the background of the participants with
>respect to healthcare IT

This SBVR project will not include any functionality that is specific to
healthcare. However, look at the new Open Health Tools (OHT) organization,
see www.openhealthtools.org I will be at the next OHT board meeting in
mid-April and will ask about RHIO involvement. OHT is duplicating the
governance structure of Eclipse (in fact, the Eclipse Foundation, OMG, and
HL7 are all charter members of OHT). There is no software available yet,
but several projects are getting started. Any work that I do which builds
on this SBVR project (and UML2) to add healthcare-specific functionality
will probably be contributed to OHT.

Watch for discussion forums starting at OHT, hopefully within a month or so.

> information exchange xml standard is HL7 (http://www.hl7.org). There's a
> metamodel (RIT) and a specific common implementation throughout the
> industry which is NOT XML, but rather value/delimiter text based.

I have been deeply involved in UML modeling with HL7 for the past 18 months.
But only for the newer v3 standards, and they do use XML. There are some
adopters for parts of this, mostly CDA and CCD specifications, including
clinical data interoperability between VA and DoD. Also applications in
Canada and the U.K. Tools that I have been developing that use UML for HL7
modeling will be contributed to OHT within the next few months. Thougthts
about how/where to use SBVR are still in the research stages.
Re: SBVR component proposal- interest in healthcare application [message #592355 is a reply to message #378139] Thu, 27 March 2008 14:32 Go to previous message
Eclipse User
Originally posted by: dgleet.bellsouth.net

I am excited at the prospect of participating in this. I'm retired, so I
can
spend time doing what I want to do :-).

Please bring me up the learning curve on your activities.

The RHIO market is very competitive, with IBM and GE among the major
players.

I'm in South Carolina, where there is a large active and retiree military
population; consequently, there's a large VA and DoD presence, with
associated medical records. If appropriately swift management of
infectious disease (and terrorist activity) is to be a reality, there will
have to be an interchange between state, private and government health
facilities. Just evaluating potential state and private RHIO information
will not be adequate.

Such a mechanism does not exist today; governmental rules stand in the
way. It's possible the SBVR component activity may suggest modifications
to the government's rules plus design characteristics of the health
records systems that would be acceptable to Congress.

Of course, all privacy concerns will need to be addressed.

RHIO implementations include rules and agents for security management.
There are some tools for managing protocols, but, as far as I know, they
are targeted at a research setting. If your are interested, I can dig up
the url of a company I know - it may be relevant at the SBVR level because
it indicates a boundary of functionality, which would lead to a piece of
the SBVR semantics.

RHIOs need to contain "business = infectious disease" rules for
identifying outbreaks, current treatments, and recommending the most
effective (cost, efficacy) countermeasures. Such population variables as
age, race, sex, and overall medical conditions must be considered in the
rules. (In a multinational setting, such as the Caribbean, or ,where
there is a country with semiautonomous, cultural different regions, such
as Iraq, language, culture, and government regulations must be
considered.)

Another use case is this: during a patient visit, the physician
accumulates the patient complaint and test results data, establishes a
(tentative) diagnosis, and then, in real time, this information is passed
to a local agent of the primary agent. The local agent establishes the
best treatment option for that patient, either autonomously, given a local
copy of the recommendations from the primary agent, or, in real time
dialog with the primary agent. Since the local agent is implemented by
different EMR (Electronic Medical Record) vendors, I think the SBVR
component is relevant in guiding implementation.

Within the primary agent, is a calculating engine, which may be
proprietary by company, to establish the effectiveness metrics and
prioritize the treatments.

A key critical success factor, independent of how the process is
implemented- it can even be manual- is that medical practice must be
changed to insure follow up and record treatment effectiveness; ie, did
the patient recover + any other relevant data. In terms of implementation
and the SBVR component, patient reporting might be done through a medical
web portal that results in information back to both the local EMR and,
eventually, to the primary agent. The SBVR component might be relevant in
establishing the semantics for the dialog.

In general, it seems to me any SBVR implementation will need to satisfy a
collection of use cases based on the above description.

My bottom line assertion is that a necessary condition for SBVR
specification validation is that it be validated through appropriately
robust medical disease control use cases and trial implementations.
Previous Topic:[Announce] UML2 Tools at EclipseCon 2008
Next Topic:Derive constraints
Goto Forum:
  


Current Time: Thu Oct 23 09:01:10 GMT 2014

Powered by FUDForum. Page generated in 0.02325 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software