Stream: genomics/committers
Topic: coded-annotation
Patrick Werner (Nov 27 2021 at 13:40):
I finished my branch:
http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/
I added an intro note the the CodedAnnotation data type, and added short descriptions to the places where codedAnnotation is used.
Patrick Werner (Nov 27 2021 at 13:40):
fyi @Mullai Murugan
Patrick Werner (Nov 27 2021 at 13:40):
This is the data type: http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/
Patrick Werner (Nov 27 2021 at 13:41):
Added the codedAnnotation to the Report (genomicReportNote): http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/StructureDefinition-genomics-report.html
Patrick Werner (Nov 27 2021 at 13:41):
and to the genomics base observation: http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/StructureDefinition-genomics-base.html
Kevin Power (Nov 27 2021 at 16:12):
Quick review tells me it looks good! We have been naming the extension slices with kebob case (genomics-report-note) but noticed you used camel case (genomicsReportNote).
Bret H (Nov 28 2021 at 03:46):
needs description here: http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/StructureDefinition-genomics-report-definitions.html#DiagnosticReport.extension:coded-note
Bret H (Nov 28 2021 at 03:48):
Will the extension only be in Diagnostic Report? That is, will the variant identifier component in Variant Finding Observation be kept or replaced in the observation with the coded-annotation?
Bret H (Nov 28 2021 at 03:50):
The coded-annotation is a very nice construct...very tempting to extend its use but that would be bad. let me know if you need help with any definition text.
Patrick Werner (Nov 28 2021 at 09:02):
Kevin Power said:
Quick review tells me it looks good! We have been naming the extension slices with kebob case (genomics-report-note) but noticed you used camel case (genomicsReportNote).
will change
Patrick Werner (Nov 28 2021 at 09:03):
Bret H said:
Will the extension only be in Diagnostic Report? That is, will the variant identifier component in Variant Finding Observation be kept or replaced in the observation with the coded-annotation?
Its an extension on DR, and a profile on GenomicsBase.note.
Patrick Werner (Nov 28 2021 at 09:03):
I didn't add too much text. Wasn't sure if more is needed
Bret H (Nov 28 2021 at 16:44):
yes, more description will be needed. The word 'annotation' has other connotations in genomics. I think something like 'This element is for recording at the Genomic report level, which aggregates the effects of all the observations found in the report...' Need to clarify 1) that this is about the report as a whole - not a specific observation, 2) the scope of the element.
Some would consider a clinvarID a variant annotation, and others would consider a Sequence ontology term applied to a variant to be an annotation. As the element is about the report as a whole, it needs to be clarified.
Thanks! @Patrick Werner
Bret H (Nov 28 2021 at 16:47):
not sure what you mean by 'profile on GenomicsBase.note'
Kevin Power (Nov 28 2021 at 17:03):
@Bret H - So in this case Annotation means the data type in FHIR but good point about the potential for confusion in this space. We probably should add some extra text.
By ‘profile on GenomicsBase.note’ he means in the GenomicsBase profile, we switched the data type of the ‘note’ attribute from Annotation to CodedAnnotation.
Kevin Power (Nov 28 2021 at 17:07):
And since it is on the base profile it would be available on all of our Observation profiles
Bret H (Nov 28 2021 at 17:08):
I know the FHIR annotation element. The need for clarification is confusion that will result within the genomics community at large. Also, begin in the Observations does not jive with 'Its an extension on DR...'
Kevin Power (Nov 28 2021 at 17:10):
Understood and agree
Bret H (Nov 28 2021 at 17:15):
Observation.note - current 'Definition A text note which also contains information about who made the statement and when.'
current 'Requirements Need to be able to provide free text additional information.'
maybe add something like 'The coded-annotation data type, while not allowing for or intending to make the content computable, does allow the author to indicate the type of note. This does not replace use of interpretation, or value, or components' to provide a little scoping in observation.note?
might even want to state an example type or two. It'll save us trouble later.
Patrick Werner (Nov 29 2021 at 11:03):
i already added this text on obs.note.short:
Notes SHALL not contain information which can be captured in a structured way
Patrick Werner (Nov 29 2021 at 11:03):
Happy to also add Brets test to the definition
Kevin Power (Nov 29 2021 at 15:48):
Hey all - Between meetings today I made some updates to some of the wording. Tried keeping the new data type and even extension pretty generic, but then moved some of the more guidance into our GenomicsReport and GenomicsBase profiles. And just tweaked some of the wording in some cases. If you want to see what I changed, please take a look at the commit:
https://github.com/HL7/genomics-reporting/commit/1ecb1f1c7e3b2950d98caf78aa84be3fc3976ce8
Welcome any/all feedback.
Kevin Power (Nov 29 2021 at 15:49):
I do wonder if we need to update one of our example reports and example observations to use CodeAnnotation? Anyone up for that?
Jamie Jones (Nov 29 2021 at 15:54):
I agree, and feel this should tie into the eMERGE example report, since they were the impetus for the update @Mullai Murugan any thoughts on adapting how the eMERGE "canned" text went in, using our new structure? (other input also welcome)
Patrick Werner (Nov 29 2021 at 15:56):
thanks @Kevin Power for your contribution. I like your changes.
Patrick Werner (Nov 29 2021 at 15:56):
:+1: for adding the emerge disclaimer text to the emerge example report
Bret H (Nov 29 2021 at 16:07):
@Kevin Power small addition to "Description: "Extension adding a CodeableConcept for type to the Annotation data-type"
What do you think? Trying to clarify that the codeableconcept is not the value of the Annotation - I know it would be obvious to anyone that bothered to look at the data-type but working for the lower common denominator.
Kevin Power (Nov 29 2021 at 16:12):
@Bret H - Did you review my last changes?
Bret H (Nov 29 2021 at 16:18):
@Kevin Power Yes. The ones in https://github.com/HL7/genomics-reporting/commit/1ecb1f1c7e3b2950d98caf78aa84be3fc3976ce8 , much appreciated. Especially, the comments on the structure def page for the extension.
Was wondering if adding 'COdeableConcept for type' (as above) would be useful to make it clear that the Annotation data-type content is not a codeableConcept for the idle reader who might not have looked at the FHIR Annotation data-type? Or is that a step too far?
Kevin Power (Nov 29 2021 at 16:20):
Adding that text where exactly?
Kevin Power (Nov 29 2021 at 16:20):
On the Data Type profile? http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/StructureDefinition-coded-annotation.html
Bret H (Nov 29 2021 at 16:34):
In http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/artifacts.html#extensions
The additional text in the structure def html page for the extension makes the point clear but in the artifacts page the sentence is by itself.
...let me know if that's not clear.
Kevin Power (Nov 29 2021 at 16:36):
Ahhh dang it, sorry I missed that. Will change.
Bret H (Nov 29 2021 at 16:41):
BTW I use notepad++ find all in files and point to the folder of the local git repo when editing. not perfect but helped me.
Kevin Power (Nov 29 2021 at 16:43):
Yup, I use Visual Studio Code in the same way, but I messed up and only searched the input/fsh folder, not entire input folder
https://media4.giphy.com/media/xT5LMzIK1AdZJ4cYW4/giphy.gif?cid=c623cb35cg7hg47onlhf6miv9pz7l5z6dehl9i93hdtdcj8n&rid=giphy.gif&ct=g
Kevin Power (Nov 29 2021 at 16:46):
Take a look now @Bret H
Mullai Murugan (Nov 30 2021 at 14:48):
Patrick Werner said:
Added the codedAnnotation to the Report (genomicReportNote): http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/StructureDefinition-genomics-report.html
@Patrick Werner @Kevin Power is it ok to update github directly? I would like to add "report summary text, interpretation text" to the examples on this page
Patrick Werner (Nov 30 2021 at 14:50):
Yes of course, feel free to commit to the branch directly
Mullai Murugan (Nov 30 2021 at 15:58):
Jamie Jones said:
I agree, and feel this should tie into the eMERGE example report, since they were the impetus for the update Mullai Murugan any thoughts on adapting how the eMERGE "canned" text went in, using our new structure? (other input also welcome)
I think we can use test disclaimer and report comments as emerge examples for GenomicReportNote and summary text for the note field on genomicsbase. Any suggestions on where we should include this information?
Bob Dolin (Nov 30 2021 at 19:35):
@Mullai Murugan I wonder if test disclaimer should be part of the proposal to address https://jira.hl7.org/browse/FHIR-32084, where we can also consider eMerge's use of PlanDefinition?
Kevin Power (Nov 30 2021 at 19:36):
What 'information' to you mean @Mullai Murugan ? If you mean where to put the test disclaimer / report comments, I think the idea would be to take any examples where the eMERGE specific extensions are used, and switch them over to either the extension on GenomicsReport or one of the Observation.note attributes? Or did I misunderstand your question?
Kevin Power (Dec 14 2021 at 22:00):
After the call today, I pushed up some changes for this. I introduced a new CodeSystem and ValueSet, and setup an example binding.
http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/StructureDefinition-coded-annotation.html
Kevin Power (Dec 30 2021 at 21:29):
I merged latest QA changes from master into this branch, so if there is more to do, you should pull down changes first before making additional change.
Kevin Power (Jan 06 2022 at 17:01):
BTW - builds are back, and you can see the new example as XML for the coded note extension here:
http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/DiagnosticReport-PGxGenomicsReportEMERGE.xml.html
Wanted to note, the extensions are not in the rendered form, but you can review directly in the XML. It looks like this:
<extension
url="http://hl7.org/fhir/uv/genomics-reporting/StructureDefinition/genomic-report-note">
<valueAnnotation>
<extension
url="http://hl7.org/fhir/uv/genomics-reporting/StructureDefinition/annotation-code">
<valueCodeableConcept>
<coding>
<system
value="http://hl7.org/fhir/uv/genomics-reporting/CodeSystem/coded-annotation-types-cs"/>
<code value="test-disclaimer"/>
</coding>
</valueCodeableConcept>
</extension>
<text
value="This test was developed and its performance determined by this laboratory. It has not been cleared or approved by U.S. Food and Drug Administration.
Since FDA Approval is not required for clinical use of this test, this laboratory has established and validated the test's accuracy and precision,
pursuant to the requirement of CLIA '88. This laboratory is licensed and/or accredited under CLIA and CAP (CAP# xxxxxxx / CLIA# xxxxxxxxxx)."/>
</valueAnnotation>
</extension>
Patrick Werner (Jan 11 2022 at 14:23):
i'm currently merging the master into the coded-annotation branch.
Kevin Power (Jan 11 2022 at 22:57):
I pushed the backing out the attempt at backporting the R5 DiagnosticReport.note and kept our extension. I also added some new examples on Variant observations (simple statement that they were Sanger confirmed):
http://build.fhir.org/ig/HL7/genomics-reporting/branches/pw-codedNotes/Observation-Pgx-var-1019.json.html (search for 'sanger' to see how it looks)
Kevin Power (Jan 11 2022 at 22:59):
I suppose this is probably enough examples if others thing we are ready to finalize the JIRAs, vote/approve, then merge?
Patrick Werner (Jan 18 2022 at 14:31):
- merged master into the branch
- resolved all conflicts
- removed all .DS_Store files which were already commited and causing trouble (also removed from GH Repo)
Patrick Werner (Jan 18 2022 at 14:31):
will create proposed resolutions for the corresponding Jiras next
Patrick Werner (Jan 18 2022 at 19:57):
proposed resolutions provided.
Last updated: Apr 12 2022 at 19:14 UTC
