Definitions to data flow: Operationalizing MIABIS in HL7 FHIR

Area of Science:

• Biomedical Informatics
• Biobanking
• Data Standards

Background:

• Biobanks are crucial for data-driven biomedical research, but lack standardized formats for sample data, hindering interoperability.
• Existing conceptual models like MIABIS face adoption challenges due to abstractness and varied implementations.
• Integrating diverse data sources (healthcare, registries, research) into biobanks requires a machine-readable solution.

Purpose of the Study:

• To develop a pragmatic, operational implementation of MIABIS for real-world data exchange.
• To create a standardized, machine-readable format for granular, sample-related biobank data.
• To enhance the interoperability and reusability of biobank data in research and clinical settings.

Main Methods:

• Systematic evaluation of data exchange standards, including HL7 FHIR and OMOP CDM.
• Development of a FHIR-based representation of MIABIS, supporting complex biobank structures.
• Creation of supporting tools: a Python library and an implementation guide for usability.

Main Results:

• Nine interoperable FHIR profiles for core MIABIS entities were created, adhering to FHIR standards.
• An open-source Python library was developed for schema validation and FHIR interaction abstraction.
• The FHIR-based MIABIS implementation was validated using real-world data at BBMRI-ERIC and demonstrated compatibility with federated infrastructures.

Conclusions:

• A machine-readable, interoperable MIABIS implementation is delivered, facilitating data exchange across biobanks and health systems.
• Integration with HL7 FHIR provides reusable tools and a framework for future data model evolution.
• This approach supports integration into clinical and research workflows, improving data discoverability, reuse, and collaboration.