Scaling Neuroscience Research using Federated Learning

TL;DR

This paper presents a federated learning framework for biomedical data, enabling privacy-preserving brain age prediction across multiple MRI data sites with improved convergence in heterogeneous environments.

Contribution

It introduces a federated learning architecture and training policies tailored for biomedical data, demonstrating effective brain age prediction without data sharing.

Findings

Semi-Synchronous protocol accelerates convergence in heterogeneous settings

Federated learning achieves accurate brain age prediction across multiple sites

Privacy is maintained by sharing only aggregated, encrypted parameters

Abstract

The amount of biomedical data continues to grow rapidly. However, the ability to analyze these data is limited due to privacy and regulatory concerns. Machine learning approaches that require data to be copied to a single location are hampered by the challenges of data sharing. Federated Learning is a promising approach to learn a joint model over data silos. This architecture does not share any subject data across sites, only aggregated parameters, often in encrypted environments, thus satisfying privacy and regulatory requirements. Here, we describe our Federated Learning architecture and training policies. We demonstrate our approach on a brain age prediction model on structural MRI scans distributed across multiple sites with diverse amounts of data and subject (age) distributions. In these heterogeneous environments, our Semi-Synchronous protocol provides faster convergence.