Enhancing Privacy in Federated Learning: Secure Aggregation for Real-World Healthcare Applications
It addresses privacy concerns in real-world healthcare applications by making secure aggregation more practical, though it is incremental as it focuses on implementation within an existing framework.
This study implemented and benchmarked secure aggregation protocols in the Fed-BioMed framework for healthcare federated learning, showing they protect privacy with minimal impact: computational overhead is less than 1% on CPU and under 50% on GPU, and accuracy drops by no more than 2% compared to non-secure methods.
Deploying federated learning (FL) in real-world scenarios, particularly in healthcare, poses challenges in communication and security. In particular, with respect to the federated aggregation procedure, researchers have been focusing on the study of secure aggregation (SA) schemes to provide privacy guarantees over the model's parameters transmitted by the clients. Nevertheless, the practical availability of SA in currently available FL frameworks is currently limited, due to computational and communication bottlenecks. To fill this gap, this study explores the implementation of SA within the open-source Fed-BioMed framework. We implement and compare two SA protocols, Joye-Libert (JL) and Low Overhead Masking (LOM), by providing extensive benchmarks in a panel of healthcare data analysis problems. Our theoretical and experimental evaluations on four datasets demonstrate that SA protocols effectively protect privacy while maintaining task accuracy. Computational overhead during training is less than 1% on a CPU and less than 50% on a GPU for large models, with protection phases taking less than 10 seconds. Incorporating SA into Fed-BioMed impacts task accuracy by no more than 2% compared to non-SA scenarios. Overall this study demonstrates the feasibility of SA in real-world healthcare applications and contributes in reducing the gap towards the adoption of privacy-preserving technologies in sensitive applications.