Tailoring deep learning for real-time brain-computer interfaces: From offline models to calibration-free online decoding

TL;DR

This paper introduces RAP, a novel method that adapts offline deep learning models for real-time brain-computer interfaces, reducing calibration needs and computational complexity for online decoding.

Contribution

The paper presents RAP, a parameter-free, source-free domain adaptation technique that modifies existing offline DL models for efficient, calibration-free online BCI decoding.

Findings

RAP enables real-time, cross-subject decoding without calibration.

It reduces computational complexity during online decoding.

The method preserves user privacy and supports co-adaptive systems.

Abstract

Despite the growing success of deep learning (DL) in offline brain-computer interfaces (BCIs), its adoption in real-time applications remains limited due to three primary challenges. First, most DL solutions are designed for offline decoding, making the transition to online decoding unclear. Second, the use of sliding windows in online decoding substantially increases computational complexity. Third, DL models typically require large amounts of training data, which are often scarce in BCI applications. To address these challenges and enable real-time, cross-subject decoding without subject-specific calibration, we introduce realtime adaptive pooling (RAP), a novel parameter-free method. RAP seamlessly modifies the pooling layers of existing offline DL models to meet online decoding requirements. It also reduces computational complexity during training by jointly decoding consecutive sliding windows. To further alleviate data requirements, our method leverages source-free domain adaptation, enabling privacy-preserving adaptation across varying amounts of target data. Our results demonstrate that RAP provides a robust and efficient framework for real-time BCI applications. It preserves privacy, reduces calibration demands, and supports co-adaptive BCI systems, paving the way for broader adoption of DL in online BCIs. These findings lay a strong foundation for developing user-centered, high-performance BCIs that facilitate immediate feedback and user learning.