Self-supervised Learning for Electroencephalogram: A Systematic Survey

TL;DR

This paper surveys self-supervised learning for EEG to address label scarcity and inter-subject variability. It develops a taxonomy organized into predictive, generative, contrastive, and hybrid SSL methods, detailing representative pretext tasks, architectures, and losses. It maps SSL EEG approaches to downstream tasks such as emotion recognition, motor imagery, sleep staging, and seizure detection, evaluating relevant datasets and practical considerations. It concludes with identified challenges and actionable directions, including EEG-oriented pretext tasks, graph-based SSL, heterogeneous/seamless multimodal SSL, and knowledge-driven frameworks to improve generalization and interpretability.

Abstract

Electroencephalogram (EEG) is a non-invasive technique to record bioelectrical signals. Integrating supervised deep learning techniques with EEG signals has recently facilitated automatic analysis across diverse EEG-based tasks. However, the label issues of EEG signals have constrained the development of EEG-based deep models. Obtaining EEG annotations is difficult that requires domain experts to guide collection and labeling, and the variability of EEG signals among different subjects causes significant label shifts. To solve the above challenges, self-supervised learning (SSL) has been proposed to extract representations from unlabeled samples through well-designed pretext tasks. This paper concentrates on integrating SSL frameworks with temporal EEG signals to achieve efficient representation and proposes a systematic review of the SSL for EEG signals. In this paper, 1) we introduce the concept and theory of self-supervised learning and typical SSL frameworks. 2) We provide a comprehensive review of SSL for EEG analysis, including taxonomy, methodology, and technique details of the existing EEG-based SSL frameworks, and discuss the difference between these methods. 3) We investigate the adaptation of the SSL approach to various downstream tasks, including the task description and related benchmark datasets. 4) Finally, we discuss the potential directions for future SSL-EEG research.

Figures (12)

• Figure 1: The general process of SSL-integrated EEG analysis. The black arrows represent forward propagation, the green and blue arrows denote backpropagation based on pretext task loss and downstream task loss, respectively.
• Figure 2: The taxonomy of the typical self-supervised learning methods and self-supervised EEG analysis methods
• Figure 3: Categories of Self-supervised learning for EEG analysis
• Figure 4: The comparison of spatial predictive and temporal predictive SSL EEG analysis methods
• Figure 5: The general process of transformation predictive method for EEG analysis. Different signal transformation techniques are applied to EEG signals to generate augmented samples and pseudo labels. The model can capture critical signal-level features for downstream tasks by correctly predicting the transformation method.