Let’s plan the FHIR work and prioritise deliverables for the rest of this year.
We have 3 high level goals. Under each goal there’s a list of gaps in our current implementation that we are aware of or tasks to be executed to achieve each goal.
Compliance with HL7 testing scripts
2.1. Load HL7 terminologies and identify gaps.
THO (terminology.hl7.org) - HL7’s curated content that is released as a package as new versions of the HL7 spec are released
tx.fhir.org - an active FHIR Terminology Server loaded with terminologies useful for HL7 FHIR authors, including THO and special loaders for LOINC, SNOMED, etc.
3.1. Implement support for paging of lists within resources with e.g. CodeSystem-concepts using graphQL.
3.2. Implement validation of resources and xml support. Evaluate the use of fhir.resources library for this purpose.
3.3. Implement CapabilityStatement and TerminologyStatement resources.
3.4. Implement batch validation
3.5. Implement batch translation
3.6. Implement CodeSystem-$subsumes.
3.7. Implement ConceptMap-$closure.
In terms of prioritisation we will first focus on 1.1., 2.1. and 2.2. to discover any gaps. It will help us to identify bugs in our current implementation and missing features to be addressed. The ability to run compliance tests will give us confidence in our FHIR implementation.
Next we will focus on tasks from 3. in the order they are listed. $subsumes and $closure may be out of scope for this year.
@jon@jamlung let’s discuss if there are any adjustments needed.
To Do for WHO SMART Guidelines terminology content:
Link DAK IG resources where available
Also back the IG resources up
For non-IG resources, put them in an OCL repository
Queue up for further discussions:
Think about what the best model will be for these terminology resources - how can we find the differences in modeling, and how can we display them all consistently?
@rkorytkowski I took a stab at revising the FHIR roadmap that you posted. Let’s discuss during our architecture call this Friday
Full support for WHO SMART Guidelines and related projects
Turns out that unfortunately the “SMART on FHIR” project is unrelated – it’s a universal API for EHRs to fully support iPhone-like apps – doesn’t look like there is anything available out of the box to help us on this one
For us, support for WHO SMART Guidelines at this early stage means something like the following:
We maintain a compilation of all SMART Guideline terminology resources from the WHO and other authors like PEPFAR
Each resource should be tagged as to whether it should be published in “production”, or “test only”
Let’s discuss as a team how we want to do organize this so that it is as streamlined as possible
@jamlung is this something that you could take the lead in maintaining?
Implement a test script that we can run routinely to load and validate all of the compiled resources (both “production” and “test only”) in a test OCL environment – this should perform a diff (or something similar) so that we can easily understand gaps in OCL FHIR Core functionality; Script might need to drop resources afterwards?
Implement a script to load all “production” resources that we’ve compiled into a target OCL environment
Load HL7 terminologies and identify gaps – let’s focus only on THO (we’ll continue to talk with HL7 about tx.fhir.org, but anything like that will come later):
THO (terminology.hl7.org) - HL7’s curated content that is released as a package as new versions of the HL7 spec are released
3.1. Implement support for paging of lists within resources with e.g. CodeSystem-concepts using graphQL.
3.2. Implement validation of resources and xml support. Evaluate the use of fhir.resources library for this purpose.
3.3. Implement CapabilityStatement and TerminologyStatement resources.
3.4. Implement batch validation
3.5. Implement batch translation
3.6. Implement CodeSystem-$subsumes.
3.7. Implement ConceptMap-$closure.