Leveraging Transfer Learning and User-Specific Updates for Rapid Training of BCI Decoders

TL;DR

This paper presents a transfer learning approach using CNNs that enables rapid personalization of EEG-based BCI decoders, significantly reducing calibration time and improving accuracy across subjects.

Contribution

It introduces a transfer learning pipeline with a two-layer CNN that quickly adapts to new subjects using minimal data, enhancing BCI usability outside labs.

Findings

Updated models improved accuracy by up to 22.1 percentage points.

Transfer learning reduced calibration data requirements.

CNN decoders can be personalized rapidly for practical BCI use.

Abstract

Lengthy subject- or session-specific data acquisition and calibration remain a key barrier to deploying electroencephalography (EEG)-based brain-computer interfaces (BCIs) outside the laboratory. Previous work has shown that cross subject, cross-session invariant features exist in EEG. We propose a transfer learning pipeline based on a two-layer convolutional neural network (CNN) that leverages these invariants to reduce the burden of data acquisition and calibration. A baseline model is trained on EEG data from five able-bodied individuals and then rapidly updated with a small amount of data from a sixth, holdout subject. The remaining holdout data were used to test the performance of both the baseline and updated models. We repeated this procedure via a leave-one-subject out (LOSO) validation framework. Averaged over six LOSO folds, the updated model improved classification accuracy upon the baseline by 10.0, 18.8, and 22.1 percentage points on two binary and one ternary classification tasks, respectively. These results demonstrate that decoding accuracy can be substantially improved with minimal subject-specific data. They also indicate that a CNN-based decoder can be personalized rapidly, enabling near plug-and-play BCI functionality for neurorehabilitation and other time-critical EEG applications.