DMAE-EEG: A Pretraining Framework for EEG Spatiotemporal Representation Learning

• Published in: IEEE transaction on neural networks and learning systems Vol. PP; pp. 1 - 15
• Main Authors: Zhang, Yifan, Yu, Yang, Li, Hao, Wu, Anqi, Chen, Xin, Liu, Jinfang, Zeng, Ling-Li, Hu, Dewen
• Format: Journal Article
• Language: English
• Published: United States IEEE 02.07.2025
• Subjects: Electroencephalography (EEG); masked autoencoder; motion intention recognition; signal quality enhancement
• ISSN: 2162-237X; 2162-2388; 2162-2388
• DOI: 10.1109/TNNLS.2025.3581991

Electroencephalography (EEG) plays a crucial role in neuroscience research and clinical practice, but it remains limited by nonuniform data, noise, and difficulty in labeling. To address these challenges, we develop a pretraining framework named DMAE-EEG, a denoising masked autoencoder for mining generalizable spatiotemporal representation from massive unlabeled EEG. First, we propose a novel brain region topological heterogeneity (BRTH) division method to partition the nonuniform data into fixed patches based on neuroscientific priors. Second, we design a denoised pseudo-label generator (DPLG), which utilizes a denoising reconstruction pretext task to enable the learning of generalizable representations from massive unlabeled EEG, suppressing the influence of noise and artifacts. Furthermore, we utilize an asymmetric autoencoder with self-attention as the backbone in the proposed DMAE-EEG, which captures long-range spatiotemporal dependencies and interactions from unlabeled EEG data across 14 public datasets. The proposed DMAE-EEG is validated on both generative (signal quality enhancement) and discriminative tasks (motion intention recognition). In the quality enhancement, DMAE-EEG outperforms existing statistical methods with normalized mean squared error (nMSE) reduction of 27.78%-50.00% under corruption levels of 25%, 50%, and 75%, respectively. In motion intention recognition, DMAE-EEG achieves a relative improvement of 2.71%-6.14% in intrasession classification balanced accuracy across 2-6 class motor execution and imagery tasks, outperforming state-of-the-art methods. Overall, the results suggest that the pretraining framework DMAE-EEG can capture generalizable spatiotemporal representations from massive unlabeled EEG and enhance the knowledge transferability across sessions, subjects, and tasks in various downstream scenarios, advancing EEG-aided diagnosis and brain-computer communication and control, and other clinical practice.