Removing Neural Signal Artifacts with Autoencoder-Targeted Adversarial Transformers (AT-AT)

2025-02-07

AI Summary

To address severe EMG interference in EEG signals and the high computational cost and lengthy training time of existing deep learning denoising methods, this paper proposes the lightweight AT-AT model—a novel autoencoder-guided adversarial temporal Transformer architecture. The encoder first identifies critical timepoints of EMG contamination, enabling the adversarial Transformer to perform physiologically informed, precise denoising. By incorporating joint EEG/EMG modeling and neurophysiological constraints, the model achieves over 90% parameter reduction. Evaluated on a multi-source dataset comprising 67 subjects, AT-AT attains reconstruction correlation coefficients of 0.95 (at SNR = 2 dB) and 0.70 (at SNR = −7 dB), matching the performance of large-scale models while accelerating inference by 3.2×. The method thus achieves an unprecedented balance among reconstruction fidelity, physiological plausibility, and deployment efficiency.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Electromyogenic (EMG) noise is a major contamination source in EEG data that can impede accurate analysis of brain-specific neural activity. Recent literature on EMG artifact removal has moved beyond traditional linear algorithms in favor of machine learning-based systems. However, existing deep learning-based filtration methods often have large compute footprints and prohibitively long training times. In this study, we present a new machine learning-based system for filtering EMG interference from EEG data using an autoencoder-targeted adversarial transformer (AT-AT). By leveraging the lightweight expressivity of an autoencoder to determine optimal time-series transformer application sites, our AT-AT architecture achieves a>90% model size reduction compared to published artifact removal models. The addition of adversarial training ensures that filtered signals adhere to the fundamental characteristics of EEG data. We trained AT-AT using published neural data from 67 subjects and found that the system was able to achieve comparable test performance to larger models; AT-AT posted a mean reconstructive correlation coefficient above 0.95 at an initial signal-to-noise ratio (SNR) of 2 dB and 0.70 at -7 dB SNR. Further research generalizing these results to broader sample sizes beyond these isolated test cases will be crucial; while outside the scope of this study, we also include results from a real-world deployment of AT-AT in the Appendix.

Problem

• Removing EMG noise from EEG data
• Reducing model size and training time
• Ensuring filtered signals maintain EEG characteristics

Innovation

• Autoencoder-targeted adversarial transformers
• Reduced model size by 90%
• Achieved high correlation coefficients