On-device Learning of EEGNet-based Network For Wearable Motor Imagery Brain-Computer Interface

TL;DR

This paper introduces a lightweight on-device learning approach for EEG-based brain-computer interfaces, enabling real-time adaptation and improved accuracy on wearable devices using a low-power processor.

Contribution

It presents a novel on-device learning engine for EEGNet that adapts to individual users in real-time on low-power wearable hardware, addressing feature distribution drift.

Findings

Achieved up to 7.31% accuracy improvement over baseline.

Demonstrated real-time inference with 14.9 ms latency and low energy consumption.

Validated on Physionet EEG Motor Imagery dataset with a small memory footprint.

Abstract

Electroencephalogram (EEG)-based Brain-Computer Interfaces (BCIs) have garnered significant interest across various domains, including rehabilitation and robotics. Despite advancements in neural network-based EEG decoding, maintaining performance across diverse user populations remains challenging due to feature distribution drift. This paper presents an effective approach to address this challenge by implementing a lightweight and efficient on-device learning engine for wearable motor imagery recognition. The proposed approach, applied to the well-established EEGNet architecture, enables real-time and accurate adaptation to EEG signals from unregistered users. Leveraging the newly released low-power parallel RISC-V-based processor, GAP9 from Greeenwaves, and the Physionet EEG Motor Imagery dataset, we demonstrate a remarkable accuracy gain of up to 7.31\% with respect to the baseline with a memory footprint of 15.6 KByte. Furthermore, by optimizing the input stream, we achieve enhanced real-time performance without compromising inference accuracy. Our tailored approach exhibits inference time of 14.9 ms and 0.76 mJ per single inference and 20 us and 0.83 uJ per single update during online training. These findings highlight the feasibility of our method for edge EEG devices as well as other battery-powered wearable AI systems suffering from subject-dependant feature distribution drift.