data2health / clinical-adaptor Goto Github PK

CDM to FHIR mappings
Note: Can we leverage CDM to BRIDG from Phase 1 and the existing BRIDG to FHIR mappings to do this? Or has Dragon done any of this already?
Note: CDM valid values are mapped to CDMH values (which are US Core)

Implement FHIR Server using HL7 Bulk FHIR to create a research data warehouse. The assumption is that EPIC EMR support FHIR Search API (http://www.hl7.org/fhir/search.html) and/or the FHIR Bulk Data API (https://github.com/smart-on-fhir/fhir-bulk-data-docs). Prototype limited to cancer use case and/or US Core.

This Repo appears to have been introduced to correspond to the FDA data submission, using FHIR as a canonical model. While related, I think this should be distinguished from the Clinical Harmonization (across CTSA hubs for federated query) problem. Some of the text for this Repo was borrowed from harmonization. Should we disambiguate this?

Profile selected should be US Core conformant
Note: that CDMH values were bound to US Core value set as part of Phase 1

FHIR Results should be visible via the Query Results Viewer this should be completed as part of phase I

export aggregated results from many institution to FHIR for investigators to import and use for analysis

Will need to set up a and store data results in a staging database from local institutions

Sandeep is moving forward by reversing the existing CDMH to FHIR mapping; will need to validate this is correct at some point

Adapter may need to work with FHIR Terminology Services
We maybe able to use CAMP FHIR by UNC for OMOP and PCORI
question: would CDM-based FHIR Servers be built and populated on regular basis like each institution's FHIR server?

Create a HIPPA / Safe Harbor compliant de-identification service that will consume search or bulk query results and will remove, encrypt, or obfuscate identifying attributes within resources and will recognize and remove small result groups that may provide identification by their uniqueness.

I am interested in helping with Step 4 Create Result Export Formats: CDISC SDTM, FHIR, CSV, part 2: "CSV/FHIR format export is a general export format more that can imported into common tools like excel or SAS or R for further analysis"

I can stand up a prototype web-app for testing/exploration pretty quickly using R and Shiny. I just have the following questions:

1. Are there any artefacts related to FHIR → CSV conversion that I should look at (unit tests, code in any language (any state of readiness is fine), or additional text descriptions) before I dive in?
2. Do we know what format the FHIR results on the FHIR Results Repository will have: XML, JSON, NDJSON, or something else?
3. Do we know what version of FHIR is likely to be used?
4. Is there already a recommended public test-server to simulate the behavior of the FHIR Results Repository?

What is the process of contributing prototype code to this project? If there isn't one, I can just create a new repository, instrument it for TravisCI, check in code, and post the URL here.

Alex F. Bokov
UT Health San Antonio
Institute for Integration of Medicine and Science
Clinical Informatics Research Division

transform query to FHIR via Adeptia using mappings

This should be completed from phase I

Each institution executes a bulk FHIR data request against the institution’s identified data warehouse pulling all data into a FHIR de-identified server (https://github.com/smart-on-fhir/fhir-bulk-data-docs/blob/master/export.md).  Sits behind their firewall. The bulk load or the FHIR data warehouse is automatically repeated on a regular basis (~ every night) in order to keep the information up-to-date.  the FHIR query will be executed against the institution's de-identified FHIR Sever.