Seeing Is Believing - FAIR Metadata for Medical Imaging Data in the SPHN Semantic Interoperability Framework.

Area of Science:

• Health Informatics
• Medical Imaging
• Bioinformatics

Background:

• Medical imaging is crucial for healthcare diagnostics, treatment, and research.
• Integrating diverse imaging data into national health networks presents significant interoperability challenges.
• Existing data structures often lack the semantic richness required for advanced research applications.

Purpose of the Study:

• To develop a standardized metadata framework for medical imaging data within the Swiss Personalized Health Network (SPHN).
• To enhance the FAIR (Findable, Accessible, Interoperable, Reusable) principles for medical imaging data.
• To facilitate the integration of imaging data with other clinical information for research and precision medicine.

Main Methods:

• Developed an RDF Schema extension specifically for SPHN imaging metadata.
• Leveraged established standards such as DICOM (Digital Imaging and Communications in Medicine) and SNOMED CT (Systematized Nomenclature of Medicine -- Clinical Terms).
• Modeled essential imaging metadata and established links to clinical data (diagnoses, patient visits, biosamples).

Main Results:

• Successfully created a standardized and structured metadata schema for medical imaging.
• The extension enhances semantic coherence by linking imaging data to relevant clinical information.
• Addressed key interoperability gaps in medical imaging data management.

Conclusions:

• The developed SPHN RDF Schema extension provides a robust solution for standardizing medical imaging metadata.
• This standardization is essential for unlocking the full potential of medical imaging data in clinical research and precision medicine.
• The approach promotes data reusability and facilitates advanced analytical applications within the SPHN.