> FHIR builds on previous data format standards from HL7, like HL7 version 2.x and HL7 version 3.x. But it is easier to implement because it uses a modern web-based suite of API technology, including a HTTP-based RESTful protocol, and a choice of JSON [JSON-LD], XML or RDF for data representation.[1] One of its goals is to facilitate interoperability between legacy health care systems, to make it easy to provide health care information to health care providers and individuals on a wide variety of devices from computers to tablets to cell phones, and to allow third-party application developers to provide medical applications which can be easily integrated into existing systems. [2]
> FHIR provides an alternative to document-centric approaches by directly exposing discrete data elements as services. For example, basic elements of healthcare like patients, admissions, diagnostic reports and medications can each be retrieved and manipulated via their own resource URLs.
> Perhaps dictation to unstructured text to chart form fields is the clinical workflow that needs optimization.
Medical annotation, Terminology Server, NER: Named Entity Recognition, WSD: Word-sense Disambiguation, AI Summarization, look up and pre-fill the correct forms and form fields in the chart (according to the unstructured notes), Medical Coding
From the article: https://allthingsopen.org/articles/the-case-for-open-source-... :
> currently keeping client journal entries with Apple Pages
Those are "unstructured notes" in medical informatics or health informatics.
Health informatics: https://en.wikipedia.org/wiki/Health_informatics
For the medical dictation to transcription to unstructured notes workflow,
OpenAI Whisper is free but not open source, and it hallucinates; https://www.wired.com/story/hospitals-ai-transcription-tools...
Perhaps dictation to unstructured text to chart form fields is the clinical workflow that needs optimization.
Which chart form fields work with other physicians' tools too?
FHIR is an open spec: https://en.wikipedia.org/wiki/Fast_Healthcare_Interoperabili... :
> FHIR builds on previous data format standards from HL7, like HL7 version 2.x and HL7 version 3.x. But it is easier to implement because it uses a modern web-based suite of API technology, including a HTTP-based RESTful protocol, and a choice of JSON [JSON-LD], XML or RDF for data representation.[1] One of its goals is to facilitate interoperability between legacy health care systems, to make it easy to provide health care information to health care providers and individuals on a wide variety of devices from computers to tablets to cell phones, and to allow third-party application developers to provide medical applications which can be easily integrated into existing systems. [2]
> FHIR provides an alternative to document-centric approaches by directly exposing discrete data elements as services. For example, basic elements of healthcare like patients, admissions, diagnostic reports and medications can each be retrieved and manipulated via their own resource URLs.
FHIR spec: https://build.fhir.org/
Which EHRs / EMRs support FHIR?
FHIR / Open Source Implementations: https://confluence.hl7.org/display/FHIR/Open+Source+Implemen...
FHIR / Public Test Servers: https://confluence.hl7.org/display/FHIR/Public+Test+Servers
/? awesome clinical open source: https://www.google.com/search?q=awesome+clinical+open+source
> Perhaps dictation to unstructured text to chart form fields is the clinical workflow that needs optimization.
Medical annotation, Terminology Server, NER: Named Entity Recognition, WSD: Word-sense Disambiguation, AI Summarization, look up and pre-fill the correct forms and form fields in the chart (according to the unstructured notes), Medical Coding
Clinical coding or Medical coding: https://en.wikipedia.org/wiki/Clinical_coder
From https://news.ycombinator.com/item?id=38739890#38744177 re: Linked Data in heath informatics:
> Yeah, so it's at that workflow where they're not getting linked data edges out of their own unstructured data input.