arXiv:2606.15449v1 Announce Type: new Abstract: Electronic prior authorization workflows require FHIR Questionnaire items to carry LOINC codes, yet most items in the HL7 Da Vinci CDS-Library lack these bindings. We treat this as a retrieval problem: given a Questionnaire item's text, find the correct LOINC code in a pool of 97,314 active codes. We compare six methods (TF-IDF, frozen MiniLM, BioBERT, BioLORD, contrastively fine-tuned MiniLM, and a TF-IDF+GPT reranker) on a 54-item evaluation set spanning three query styles (natural question, medium, and terse). No single method wins on every me
The increasing push for digitisation in healthcare and the maturation of AI models for natural language processing are converging to address long-standing interoperability challenges.
Improving the automated binding of medical terminology in digital health records is crucial for efficient prior authorization, data analysis, and the broader application of AI in healthcare workflows.
The development of effective transfer learning models for FHIR Questionnaire terminology binding promises to significantly reduce manual effort and improve the accuracy of medical data coding.
- · Healthcare Providers
- · Healthcare IT companies
- · AI/ML developers
- · Patients
- · Manual data coders
- · Legacy EHR systems
- · Healthcare administrative overhead
More accurate and efficient electronic prior authorization processes.
Accelerated development and deployment of AI agents that can interpret and act upon structured medical data.
Enhanced interoperability across disparate healthcare systems, potentially leading to more integrated and predictive healthcare models.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL