This wiki has undergone a migration to Confluence found Here
<meta name="googlebot" content="noindex">

Difference between revisions of "CG WG Call Notes leading to 2014 May WGM"

From HL7Wiki
Jump to navigation Jump to search
 
(5 intermediate revisions by the same user not shown)
Line 2: Line 2:
  
 
==Proposed Agenda==
 
==Proposed Agenda==
*  
+
* Continue general disucssion
  
  
 
==Attendees==
 
==Attendees==
*  
+
* Amnon Shabo, Scott Bolte, Siew Lam, Bob Milius, Grant Wood, Mollie Ullman-Cullere, Don Rule, Larry Babb, Chris Steiner (Elsevier)
  
  
 
==Draft Minutes==
 
==Draft Minutes==
* please fill in
+
* Discussed the FHIR effort presented last week and how it can be aligned with previous efforts of HL7 Clinical Genomics
 
+
* Discussed the aforementioned alignment on the various levels, from vocabularies used to draw codes assigned to attributes, on to the names of classes and attributes and up to the high level structure of the models
 +
* Amnon mentioned his comments to Don Rule's message, as follows:
 +
** Don: identifying  a data type that is explicitly the diplotype;
 +
*** Amnon: Indeed - that's a gap we haven't addressed, partly because I always thought it's more in scope for bioinformatics to standardize
 +
** Don: a predicted phenotype of “Normal or Rapid Metabolizer”;
 +
*** Amnon: That's an issue of resolution, similar to the problem in tissue typing, as nicely described by Bob Milius
 +
** Don: Phenotype might embed the specific facts that were used to derive it;
 +
*** Amnon: Not sure what you mean by "embed", but if it's leading to 'pre-coordination' of genotype and phenotype observations in a single encoding, that could be eventually challenging for CDS to  understand its constituents
 +
** Don: information about the patient includes a link to the S3 archive where all of the raw data is kept;
 +
*** Amnon: I believe that in addition to such links, we also need to be able to encapsulate key chunks of raw data within the HL7 structure (we had such a mechanism in the deprecated v3 specs and I hope to add it to the next release of the GTR). This is important for information exchange scenarios where often links get broken and having key raw data inline could help CDS doing a better job.
 +
** Don: It is very important to keep the data structure and the display completely independent;
 +
*** Amnon: That's actually provided by the GTR in a single structure, that is, instructions on rendering along with the respective structured data.
  
 
==April 15, 2014 -- Weekly call==
 
==April 15, 2014 -- Weekly call==

Latest revision as of 09:12, 23 April 2014

April 22, 2014 -- Weekly call

Proposed Agenda

  • Continue general disucssion


Attendees

  • Amnon Shabo, Scott Bolte, Siew Lam, Bob Milius, Grant Wood, Mollie Ullman-Cullere, Don Rule, Larry Babb, Chris Steiner (Elsevier)


Draft Minutes

  • Discussed the FHIR effort presented last week and how it can be aligned with previous efforts of HL7 Clinical Genomics
  • Discussed the aforementioned alignment on the various levels, from vocabularies used to draw codes assigned to attributes, on to the names of classes and attributes and up to the high level structure of the models
  • Amnon mentioned his comments to Don Rule's message, as follows:
    • Don: identifying a data type that is explicitly the diplotype;
      • Amnon: Indeed - that's a gap we haven't addressed, partly because I always thought it's more in scope for bioinformatics to standardize
    • Don: a predicted phenotype of “Normal or Rapid Metabolizer”;
      • Amnon: That's an issue of resolution, similar to the problem in tissue typing, as nicely described by Bob Milius
    • Don: Phenotype might embed the specific facts that were used to derive it;
      • Amnon: Not sure what you mean by "embed", but if it's leading to 'pre-coordination' of genotype and phenotype observations in a single encoding, that could be eventually challenging for CDS to understand its constituents
    • Don: information about the patient includes a link to the S3 archive where all of the raw data is kept;
      • Amnon: I believe that in addition to such links, we also need to be able to encapsulate key chunks of raw data within the HL7 structure (we had such a mechanism in the deprecated v3 specs and I hope to add it to the next release of the GTR). This is important for information exchange scenarios where often links get broken and having key raw data inline could help CDS doing a better job.
    • Don: It is very important to keep the data structure and the display completely independent;
      • Amnon: That's actually provided by the GTR in a single structure, that is, instructions on rendering along with the respective structured data.

April 15, 2014 -- Weekly call

Proposed Agenda

  • FHIR resources for core genomics information


Attendees

  • Siew Lam, Scott Bolte, Rubby Russel, Lloyd McKenzie, who else?


Draft Minutes

  • please fill in

April 8, 2014 -- Weekly call

Proposed Agenda

  • Updates
  • Phoenix WGM agenda


Attendees

  • Amnon Shabo, Grant Wood, Siew Lam, Scott Bolte, Rubby Russel, Larry Babb, Bob Milius, Lloyd McKenzie


Draft Minutes

  • Attendees gave updates on recent conf. calls
  • Discussed the Phoenix WGM agenda for CG and revised it
    • Added a discussion on how to align the FHIR Genomic core resources with the CG DIM and DAM
    • Family history on Wednesday Q1 will start with a general discussion followed by reviewing the various specs we now have for FH
    • On Tuesday Q3 there will be no solo meeting
    • Amnon will check the status of the the CG DIM PSS that was approved in the Domain Experts SD but not clear whether it passed the TSC
    • Lloyd confirmed a joint meeting with FHIR on Wed. Q4
  • Grant will send out the latest version

March 18, 2014 -- Weekly call

Proposed Agenda

  • Family health history
    • Alex Tsatsulin will present and demo the Power Lineage family history app (that will be lunched at ACMG next week)
    • Compare to the HL7 specs
    • Discussion


Attendees

  • Amnon Shabo, Grant Wood, Alex Tsatsulin, Siew Lam, Scott Bolte, Rubby Russel, Michael Bakhtin, Mollie Ullman-Cullere


Draft Minutes

  • The Power Lineage family history app was demoed by Alex
  • Discussion
    • pain points of getting EMR vendors and healthcare providers to implement family health history structures that can be good input to risk assessment algorithms
    • Comparing to HL7 v3 and FHIR
    • Pros and cons of flat vs. hierarchical representation of the pedigree
  • Decided to continue this discussion in the weekly call next week
  • FHIR genomics resources are likely to be discussed in the week after