Secure Neuroimaging Analysis using Federated Learning with Homomorphic Encryption

Federated learning (FL) enables distributed computation of machine learning models over various disparate and remote data sources, without requiring to transfer any individual sample to a centralized location. This results in improved model generalization and efficient scaling of computation as more sources and larger datasets are added to the federation. Nevertheless, recent membership inference attacks show that private or sensitive personal data can sometimes be leaked or inferred when model parameters or summary statistics are shared with a central site, requiring improved security solutions. In this work, we propose a framework for secure FL using fully-homomorphic encryption (FHE). Specifically, we use the CKKS construction, an approximate, floating point compatible scheme that benefits from ciphertext packing and rescaling. In our evaluation on a large-scale brain MRI dataset, we use our proposed secure FL framework to train a deep learning model to predict a person's age from distributed MRI scans, a common benchmarking task, and demonstrate that there is no degradation in the learning performance between the encrypted and non-encrypted federated models.