EEG2GAIT: A Hierarchical Graph Convolutional Network for EEG-based Gait Decoding

2025-04-02

AI Summary

Decoding lower-limb gait dynamics from EEG faces three key challenges: difficulty in modeling spatial dependencies across brain regions, inaccurate extraction of time-frequency features, and scarcity of high-quality synchronized gait-EEG data. To address these, we propose a Hierarchical Graph Convolutional Network (GCN Pyramid) that explicitly captures multi-scale topological relationships among cortical regions; design a Hybrid Time–Frequency Supervised Reward (HTSR) loss function to jointly optimize temporal continuity and spectral consistency in joint-angle prediction; and introduce GED—the first large-scale synchronized Gait-EEG Dataset—comprising 50 healthy subjects. Evaluated on both GED and the publicly available MoBI dataset, our method achieves statistically significant improvements over state-of-the-art approaches in joint-angle reconstruction accuracy. Ablation studies confirm the efficacy of each component. Furthermore, saliency mapping identifies neuroanatomically plausible motor-related regions—including the supplementary motor area and sensorimotor cortex—as critical for gait decoding.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Humans and AI: Brain-Sensing and Analysis
• Machine Learning: Multimodal Learning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Decoding gait dynamics from EEG signals presents significant challenges due to the complex spatial dependencies of motor processes, the need for accurate temporal and spectral feature extraction, and the scarcity of high-quality gait EEG datasets. To address these issues, we propose EEG2GAIT, a novel hierarchical graph-based model that captures multi-level spatial embeddings of EEG channels using a Hierarchical Graph Convolutional Network (GCN) Pyramid. To further improve decoding accuracy, we introduce a Hybrid Temporal-Spectral Reward (HTSR) loss function, which combines time-domain, frequency-domain, and reward-based loss components. Moreover, we contribute a new Gait-EEG Dataset (GED), consisting of synchronized EEG and lower-limb joint angle data collected from 50 participants over two lab visits. Validation experiments on both the GED and the publicly available Mobile Brain-body imaging (MoBI) dataset demonstrate that EEG2GAIT outperforms state-of-the-art methods and achieves the best joint angle prediction. Ablation studies validate the contributions of the hierarchical GCN modules and HTSR Loss, while saliency maps reveal the significance of motor-related brain regions in decoding tasks. These findings underscore EEG2GAIT's potential for advancing brain-computer interface applications, particularly in lower-limb rehabilitation and assistive technologies.

Problem

• Decoding gait dynamics from EEG signals with spatial dependencies
• Improving accuracy via hybrid temporal-spectral loss function
• Addressing scarcity of high-quality gait EEG datasets

Innovation

• Hierarchical GCN Pyramid for EEG spatial embeddings
• Hybrid Temporal-Spectral Reward loss function
• New Gait-EEG Dataset with synchronized data