PCD RTM 2008-12-17 WebEx
Meeting Purpose
IHE PCD Rosetta Terminology Mapping (RTM) Ventilator Task Group regular discussion meeting. See the Proposed Agenda below for specific topics.
WebEx Information
Topic: IHE PCD RTM Vent TG
Date: Wednesday, December 17, 2008
Time: 12:00 pm, Eastern Time (GMT -05:00, New York)
Duration: 120 Minutes
Note: Specific web & phone informaiton will be provided via e-mail to group members.
Contact Manny Furst for more information.
Proposed Agenda
- 1 Review agenda
- 2 Review Discussion Summary from 2008.12.03 meeting
- 3 Review Work Program
- 4 Review Core Semantic Architecture MindMap
- 5 Identify the first order numeric parameters to be defined
- 6 Finalize the method and documents and assignments for completing these definitions
- 7 Planning
Attachments / Materials
- (no attachments)
Minutes
Participants
- Chair/Host: Todd Cooper (Breakthrough Solutions Foundry)
- Steven Dain (ISO TC121/SC4), Joel Halle (Cardinal), Ken Marks (GE), John Rhoads (Philips), Paul Schluter (GE), Jan Wittenber (Philips)
Discussion
Item Topic Discussion 1 Introductions & Agenda Review
- ChairStatus/Discussion: Decisions/Issues:
- NOTE: The sessions started with some difficulty due to WebEx issues
Action(s):
2 Discussion Summary Review
- ChairStatus/Discussion: - Reviewed vent TG discussion summary from 2008.12.03.
Decisions/Issues:
Action(s):
3 General Vent Work Program Review
- <tbd>Status/Discussion: - Todd reviewed the group "status":
- - Previously reviewed models
- - Top priority is to focus on filling the monitored numeric parameter gaps
- - Initial target is to have a start on this by 2009.01.07
- Paul also indicated the need to finalize the vent group tags in the Rosetta tables.
- - Secondarily, there is the need to review the current group mappings for consistency
Decisions/Issues:
Action(s):
- (Paul) Add in a canonical UoM column to facilitate group definitions
4 Modeling Reviews
- GroupStatus/Discussion: - Vent terminology "MindMap" development:
- - Jan is working on one (primarily for ICE-PAC), but a database will probably also be needed
- - Paul indicated that given the XML / mmap tooling support and availability, a DB may not be required
- - The MM provids a good metamodel that describes different views of the concepts (e.g., clinical vs. deployment architecture vs. state, etc.)
- - Todd has also been working on a MM for the vent terminology
- General discussion regarding the use of various tools (mmap, open XML, SQL database, etc.) for capturing the new term definitions. As demonstrated in
Decisions/Issues:
Action(s):
- (Jan & Todd) Coordinate on development of a MindMap model for use by the group
- (Jan & Todd) Evaluate mmap v8 to see if the tool facilitates XML usage better than previous versions
5 Ontology / Terminology Architecture Discussion
- GroupStatus/Discussion: - Steven Dain mentioned a recent discussion with an "ontology expert" regarding formal term definition and existing standards. This led into a general discussion about how best to architect the vent term definitions:
- - We should name our terms with what they are first, e.g., <vent> <observable> <...> (e.g., VENT_SETTING_....)
- - General discussion about the various ways of organizing the semantic components, esp. w.r.t. ontology / sequential precedence
- - Also the need to factor in the way that a given term is usually displayed is preferable, though this is often (usually?) different from ontological precedence
- - Is the sequence arbitrary or should it be more constrained? Should a primary VIEW be selected, but allow additional perspectives / views
- - What we do need to do is come up with a grammar that will allow consistent composition of the definitions and aggregation of related terms into groups
- - There was additional discussion regarding the definition of parameters where at an abstract level they are the same but as more detail as to the measurement / control modality is included, then those concepts get added on at the end … the “devil is in the details” …
- - Grammar…
- MCD_ grouping category (e.g.., VMD or body organ system/physiology… “ECG_” OR “VENT_” …)
- Observation / setting / event … <may be multiple tokens>
- PEEP <core semantic … physio variable>
- … <measurement / control modality>
- - Though this may be further differentiated by, for example, types of “SET” such as limit, configuration, therapeutic parameter … all of which have regulatory implications
- - From the CORE SEMANTIC, the items on the left are more of the grouping (e.g., physio or device or usage context) …
- - Specialization / generalization also follow from left to right
- - …
- - Dain: Example of a RR from ECG (looks good) … but since they are totally obstructed, the RR (from vent … airflow) is zero! Need to be able to capture the full level of granularity + multiple instances of the same fundamental semantic
- Paul stated that the sooner we can agree and "concretize" the token order, the better off we will be. He proposed the following organization:
- - MDC_VENT_SET_PEEP ... (set value, set numeric limit high|low, etc.)
- - MDC_VENT_SETNUMLIMIT_PEEP_HIGH|LOW ... (set numeric limit high|low, etc.) (or MDC_VENT_SET_NUMLIMIT_PEEP ... )
- - MDC_VENT_NUM_PEEP ... (an numeric observation)
- - MDC_VENT_EVT_ APNEA ... (an event)
- One benefit of this approach is that it provides a consistent way to identify the VMD (device or organ system) as the first token after "MDC" and then the data type / modality as the second token after MDC. The third token would be the principal physiologic or device identifier, such as "PEEP" or "HEART_RATE" followed by zero or more discriminators. This allows you to easily and reliably find all instances of "NUM_HEART_RATE" or "NUM_PULSE_RATE" in cases where multiple technologies are used to acquire a numeric value. This is difficult to do in an automated manner using the existing nomenclature since different conventions are used in different places.
- This structure should also help focus the group's attention on specific data types / modalities first, like "numerics" and will facilitate systematic organization and queries based on the REFID.
Decisions/Issues:
Action(s):
- (Steven) forward a reference / copy of the standard(s) promoted by the IHTSDO expert
- (Todd) forward similar standards published by ISO TC215 WG3 (based on work from CEN TC251).
6 Planning discussion
- <tbd>Status/Discussion: - Target for the 2009.01.07 meeting:
- - Review the current groups
- - Complete a MindMap first pass exercise
- - Define the model (e.g., spreadsheets & columns) into which terms will be defined
- - Determine term definition work plan (focusing on numeric monitored paramters first) - what terms by which companies
Decisions/Issues:
Action(s):
- (Todd) Send out a general group e-mail to grigger the current term group definitions.
- (Todd) Define ACTION ITEMS table on the Vent TG wiki page
7 Next Meeting
- ChairStatus/Discussion: Decisions/Issues:
- Next meeting 2009.01.07
Action(s):
Next Meeting
(Reviewed & approved by PCD RTM Vent TG @ 2009.01.07 meeting)