Security & Privacy

What are the main security features of FLAME?

FLAME implements multiple security layers to protect sensitive medical data:

1. Privacy-Preserving Analysis: Analyzes data without transferring sensitive information outside organizations.
2. Code Approval & Verification: An analysis algorithms is manually reviewed and approved before execution.
3. Encryption: Uses several methods of encryption (Paillier, RSA keys, envelope). Only previously registered participants can access results.
4. Analysis Execution Security: Analysis runs in isolated Docker containers.
5. Access Control:
   • Role-Based Access Control for different user types.
   • Identity provider integration (OIDC).
   • Keycloak authentication at Hub and Node level.
   • Permission-based authorization.
6. Data Security: Data remains on local nodes and is not transferred to central Hub.
7. Standards Compliance:
   • Alignment with Medical Informatics Initiative (MII) standards.
   • Integration with Data Integration Centers (DIC).
   • FHIR compliance.