Automated metadata transformation in a medical data integration center: Implementation of an algorithm and standardized quality analysis

This study presents a novel approach to metadata management, while focusing on the development of an automated, microservice-based infrastructure at the University Medical Center Göttingen’s Medical Data Integration Center (UMG-MeDIC). Given the critical role of high-quality metadata in supporting reliable clinical research and data interoperability, this research addresses the challenges of metadata extraction, storage, and quality assurance across multiple data formats used in healthcare. Through a mixed-methods, single-case design, the metadata framework was developed, adopting a convergence format that integrates metadata from standards such as CDISC, OMOP, openEHR, and FHIR. This format enables consistent metadata representation and accommodates missing values, thus preserving data integrity and supporting FAIR principles. Key quality metrics, including completeness, consistency, and relevance, were defined and operationalized to assess metadata reliability systematically. The microservice architecture used enhances scalability and adaptability, demonstrating a replicable model for other data integration centers (DICs). This work contributes a scalable framework for metadata management, with potential for further application.