Zero-Shot EEG-to-Gait Decoding via Phase-Aware Representation Learning

2025-06-24

AI Summary

This study addresses the critical brain–computer interface (BCI) challenge of lower-limb movement intention decoding, aiming for zero-shot, calibration-free cross-subject gait reconstruction. To tackle key difficulties in EEG—namely weak phase consistency, limited causal interpretability, and substantial inter- and intra-subject variability—we propose a phase-aware multi-cycle gait reconstruction objective coupled with a relational domain modeling framework. Our approach integrates structured contrastive representation learning, relative contrastive loss, domain-dynamic decoding heads, and semantic-temporal joint alignment between EEG and motion embeddings. This enables biomechanically plausible and temporally coherent zero-shot generalization. Evaluated on benchmark datasets, our method achieves state-of-the-art cross-subject performance and demonstrates strong unsupervised gait-phase detection capability, significantly enhancing the deployability and generalizability of BCIs in real-world applications.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Humans and AI: Brain-Sensing and Analysis
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Accurate decoding of lower-limb motion from EEG signals is essential for advancing brain-computer interface (BCI) applications in movement intent recognition and control. However, challenges persist in achieving causal, phase-consistent predictions and in modeling both inter- and intra-subject variability. To address these issues, we propose NeuroDyGait, a domain-generalizable EEG-to-motion decoding framework that leverages structured contrastive representation learning and relational domain modeling. The proposed method employs relative contrastive learning to achieve semantic alignment between EEG and motion embeddings. Furthermore, a multi-cycle gait reconstruction objective is introduced to enforce temporal coherence and maintain biomechanical consistency. To promote inter-session generalization, during fine-tuning, a domain dynamic decoding mechanism adaptively assigns session-specific prediction heads and learns to mix their outputs based on inter-session relationships. NeuroDyGait enables zero-shot motion prediction for unseen individuals without requiring adaptation and achieves superior performance in cross-subject gait decoding on benchmark datasets. Additionally, it demonstrates strong phase-detection capabilities even without explicit phase supervision during training. These findings highlight the potential of relational domain learning in enabling scalable, target-free deployment of BCIs.

Problem

• Decoding lower-limb motion from EEG signals accurately
• Addressing inter- and intra-subject variability challenges
• Achieving zero-shot motion prediction for unseen individuals

Innovation

• Phase-aware EEG-motion semantic alignment
• Multi-cycle gait reconstruction for coherence
• Domain dynamic decoding for generalization