Secure and Privacy-Preserving Secondary Data Use A Framework for Cross-Domain Computation in the Encrypted Domain

Secondary use of data has become increasingly rel-evant over the past years, referring to the repurposing of data for cross-domain applications to extract multidisciplinary insights that would otherwise remain hidden. When it comes to secondary use of sensitive data, like health data, such efforts are frequently hampered by ethical, legal, and privacy constraints. As a result, recent efforts stress the need for infras-tructures that support privacy-aware data federation. Privacy Enhancing Technologies have emerged as a tool to preserve privacy throughout the entire data sharing and computation life-cycle, while ensuring adherence to regulations, such as the General Data Protection Regulation (GDPR). Contributing to this notion, this paper presents a privacy-aware data feder-ation framework leveraging Fully Homomorphic Encryption to enable secure cross-domain computations in the encrypted domain, avoiding centralized data storage and ensuring Data Providers retain total control over their data. Three real-world application scenarios are investigated in this work, leveraging health data to inform insights in the education domain while performance tests showcase the viability of the approach towards lawful, privacy-aware secondary use of sensitive data.