Federated learning cuts communication in brain-computer interface personalization
NEXUS-MI: Communication-Aware Federated Personalization for Gateway-Coordinated Motor-Imagery Brain-Computer Interfaces
Machine LearningNetworking and Internet Architecture
Summary
Brain-computer interfaces using brain signals to control devices need to be personalized for each person, but collecting enough data is hard and sharing raw data raises privacy concerns. The authors created a system called NEXUS-MI that lets multiple devices learn together without sharing raw brain data by coordinating updates through a local gateway. Their method smartly controls communication to reduce data sent without much loss in performance, though some users may experience less improvement. This work shows that managing how devices synchronize is important for making personalized brain-computer interfaces practical.
What this means in practice
- •For medical device developers: Reduce data transmission needs in personalized brain-computer interfaces for motor-impaired users by coordinating model updates locally.$Commercial implications: Enables more efficient and privacy-preserving adaptive BCI products for rehabilitation and assistive technologies.
- •For cloud infrastructure engineers: Implement communication-aware federated learning strategies to reduce server-to-client bandwidth in edge-coordinated decentralized systems.
Authors
Daniel Adu Worae, Aarthy Nagarajan
Abstract
Electroencephalography (EEG)-based motor-imagery brain-computer interfaces (MI-BCIs) vary across subjects and sessions, complicating personalization from limited calibration data. Federated learning can exploit shared representations without centralizing raw EEG, but existing federated MI studies largely assume regular synchronization. We introduce NEXUS-MI, a gateway-coordinated federated personalization framework that treats synchronization as a coupled learning-and-communication control problem. Raw EEG and classifier heads remain local, while an edge coordinator maintains the shared backbone. We evaluate NEXUS-MI through offline replay using BCI Competition IV Dataset 2a (BCICIV-2a; 9 subjects, 4 classes) and OpenBMI (54 subjects, 2 classes). Session 1 supports backbone learning, and Session 2 provides limited-calibration personalization and held-out testing. An ideal-link reference and six heterogeneous-link policies characterize gateway participation, buffering, stale-update admission, and backbone-download control. The principal comparison holds delayed-update handling fixed while contrasting non-adaptive and communication-aware synchronization. Paired subject-level comparisons use Holm adjustment, and robustness across five matched realizations is assessed by hierarchical bootstrap. Communication-aware coordination reduced server-to-client backbone traffic by approximately 42% on both datasets, while cohort-level accuracy differences were small and realization-dependent. Cohort averages also concealed subject-level vulnerability, with losses reaching approximately 12 percentage points on BCICIV-2a relative to the ideal-link reference. These findings establish gateway synchronization as an explicit design variable in federated MI personalization and motivate joint evaluation of personalized accuracy, communication cost, update freshness, and subject-level reliability.