Unstructured Text to FHIR Using Generative AI and openEHR | HL7 FHIR DevDays 2024, Minneapolis

[@medblocks]

If you're also a FHIR 🔥 nerd like me and want to learn how to work with FHIR, watch my exclusive webinar here: link.medblocks.com/fhir-dc207e

[@kaikewley2656]

The "3 Major Differences between FHIR and openEHR" slide (at 17:10) is excellent.

[@ROHITKUMAR-mu7ik]

All the best, Nice Idea and thought process

[@ROHITKUMAR-mu7ik]

Good one

[@adamv2628]

what struck me in this talk was your remark and comparison with a TV set : it tells me that FHIR has a limited capacity to stream all the openEHR clinical data in a proper way. Above that FHIR contains only a small set of mature (Level 5) resource-types. It looks to me as FHIR is on that part less suited to be used in combination with openEHR.

[@nedu47]

How did the Epic FHIR, and Aidbox FHIR build this (30 to 33) into their processes? I sent body max index data to Epic FHIR, which showed me a response saying the data was higher than expected.

[@RelaxWithUsNow]

Most likely additional business logic on top of the FHIR Facade or when the data got to the Epic EHR, validation also occurs at that level , thereby returning that validation error

[@dinesht6883]

Confused