EEG-X: Device-Agnostic and Noise-Robust Foundation Model for EEG

2025-11-12

AI Summary

To address cross-device data heterogeneity and low signal-to-noise ratio (SNR) in EEG analysis, this paper proposes a generalizable representation learning framework. Methodologically, it introduces position-aware channel embeddings and a dictionary-inspired convolutional transform (DiCT) layer to enable adaptive modeling of arbitrary electrode configurations; further, it integrates noise-aware masking, dual-space (raw/latent) joint denoising reconstruction, and DiCT-based feature projection to jointly optimize robust representation learning and signal recovery. Extensive evaluations on multi-device, multi-task, and cross-domain benchmarks demonstrate significant improvements over state-of-the-art methods—particularly under unseen device conditions and high-noise regimes—highlighting strong generalization and transferability. The code and pretrained models are publicly released to ensure reproducibility and facilitate real-world deployment.

Technology Category

• Machine Learning: Representation Learning
• Computer Vision: Representation Learning for Vision
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Foundation models for EEG analysis are still in their infancy, limited by two key challenges: (1) variability across datasets caused by differences in recording devices and configurations, and (2) the low signal-to-noise ratio (SNR) of EEG, where brain signals are often buried under artifacts and non-brain sources. To address these challenges, we present EEG-X, a device-agnostic and noise-robust foundation model for EEG representation learning. EEG-X introduces a novel location-based channel embedding that encodes spatial information and improves generalization across domains and tasks by allowing the model to handle varying channel numbers, combinations, and recording lengths. To enhance robustness against noise, EEG-X employs a noise-aware masking and reconstruction strategy in both raw and latent spaces. Unlike previous models that mask and reconstruct raw noisy EEG signals, EEG-X is trained to reconstruct denoised signals obtained through an artifact removal process, ensuring that the learned representations focus on neural activity rather than noise. To further enhance reconstruction-based pretraining, EEG-X introduces a dictionary-inspired convolutional transformation (DiCT) layer that projects signals into a structured feature space before computing reconstruction (MSE) loss, reducing noise sensitivity and capturing frequency- and shape-aware similarities. Experiments on datasets collected from diverse devices show that EEG-X outperforms state-of-the-art methods across multiple downstream EEG tasks and excels in cross-domain settings where pre-trained and downstream datasets differ in electrode layouts. The models and code are available at: https://github.com/Emotiv/EEG-X

Problem

• Addressing EEG dataset variability from different recording devices
• Improving EEG signal robustness against low signal-to-noise ratio
• Enhancing cross-domain generalization for EEG representation learning

Innovation

• Location-based channel embedding for cross-device generalization
• Noise-aware masking reconstructs denoised neural signals
• Dictionary-inspired convolutional transformation reduces noise sensitivity