LLM-Integrated Normalization and Knowledge for FHIR (LINK-FHIR)
LLM-Integrated Normalization and Knowledge for FHIR (LINK-FHIR) - PubMed PubMed
A system called LINK-FHIR was developed to provide automated machine-to-machine integration using LLM-integrated normalization and knowledge for FHIR.
Summary
This study proposes LINK-FHIR, a novel system designed to convert diverse electronic health record data into standardized Fast Healthcare Interoperability Resources (FHIR). The system leverages fine-tuned Large Language Models (LLMs) to efficiently process various data formats, including unstructured clinical notes and semi-structured lab reports.
Details
LINK-FHIR aims to solve the current challenge of converting diverse electronic health record data—which often lack standardized formats—into compliant FHIR resources. The core technology utilizes fine-tuned Large Language Models (LLMs) within a unified pipeline. This allows the system to process and convert various types of data, such as unstructured clinical notes, semi-structured lab reports, and structured tables, into FHIR-compliant resources. LINK-FHIR features dual interfaces: an automated machine-to-machine integration interface and an intuitive user interface for data visualization and management. Furthermore, it offers flexible deployment options to ensure compliance with healthcare security and privacy regulations. Comprehensive evaluation demonstrated the system's robust performance across diverse data formats, suggesting significant potential to enhance Health Information Exchange interoperability and improve operational efficiency across healthcare institutions.
Technology Note
FHIR（Fast Healthcare Interoperability Resources）は医療データ交換の国際標準。このエントリの関連技術: FHIR
Original content copyright by respective publishers