Discover and read the best of Twitter Threads about #smartonfhir
Most recents (2)
These #AI text generation algorithms are getting scarily good at general code, blogs and poems. But does it work on something so niche as @HL7 #FHIR?
My testing below 👇
My testing below 👇
Let's start simple, please generate me a #FHIR patient resource with a name and birth date. And yes ✅ it generates a flawless Patient instance with a name and birthDate value (that validates with no issues: simplifier.net/snippet/wardwe…)
It even includes some extra explanation...
It even includes some extra explanation...
Let's make it a bit harder.
This algorithm remembers our previous conversation, so we can just continue with asking it to add an extension for capturing the patient's 'Race' value, for which the US Core IG has a common extension: hl7.org/fhir/us/core/S… 👇
This algorithm remembers our previous conversation, so we can just continue with asking it to add an extension for capturing the patient's 'Race' value, for which the US Core IG has a common extension: hl7.org/fhir/us/core/S… 👇
I have found most non-technical explanations of FHIR (🔥) to be unsatisfactory. Key things to know are as follows… 1/🧵
FHIR (Fast Healthcare Interoperability Resources) defines a standard way to organize health data. By having a standard, a computer programmer can write an application (app) that requests data from a source location without having to ask anyone how the data are organized. 2/
This means that a computer application can be written once and work the same no matter which source (e.g., hospital) it requests data from. This is called interoperability. 3/
