Secure federated optimization protects medical data by modality
Secure Multimodal Federated Optimization via Hierarchical DNA-Based Key Management
Cryptography and Security
Summary
When different institutions work together to train AI models from varied medical data types like images, DNA, and health records, it’s important to keep each type of data safe and private. The authors propose a system that uses special keys to lock and unlock each data type separately, so only authorized parts can access it. They also make sure that when updates are shared for model training, the individual data contributions remain hidden, protecting privacy. This method combines careful key management, secure data locking, and safe sharing of updates to keep diverse medical data secure during collaboration.
What this means in practice
- •For hospital data teams: Enable collaborative medical model training while keeping each data type securely encrypted and only locally accessible as authorized.
- •For biotech data engineers: Create secure pipelines that manage different genomic and clinical data keys separately to maintain strict privacy controls during optimization.
Authors
Yuping Yan, Yaochu Jin
Abstract
Federated optimization over heterogeneous multimodal data enables distributed institutions to collaboratively optimize a shared model using medical images, genomic profiles, clinical records, and physiological signals without centralizing raw data. However, protecting modality-specific data throughout the optimization lifecycle remains insufficiently explored, particularly when heterogeneous modalities require separate encryption keys and selective local access. This paper proposes a secure multimodal federated optimization framework that separates local data protection from federated-update protection. A trusted key-management authority securely provisions one client key through Hybrid Public Key Encryption (HPKE), from which each client locally derives modality-specific symmetric keys using the domain-separated HMAC-based Extract-and-Expand Key Derivation Function (HKDF). Heterogeneous modalities are mapped into a unified DNA-domain representation and processed through a DNA-origami-inspired structural transformation before authenticated encryption with AES-GCM. During local optimization, only the required modalities are decrypted within an authorized client environment. Individual model updates are independently protected through secure aggregation, allowing the coordinator to observe only aggregated updates rather than each client's contribution. Together, these components establish a structured approach to modality-specific protection and secure collaborative optimization, integrating hierarchical key management, authenticated local encryption, and aggregation-level update protection within a unified framework.