CRCC: Contrast-Based Robust Cross-Subject and Cross-Site Representation Learning for EEG

TL;DR

CRCC tackles the challenge of cross-site EEG generalization by decomposing domain shifts into three bias factors and introducing a two-stage training pipeline. The approach combines multi-dataset masked reconstruction pretraining with domain-discriminator signals, followed by fine-tuning that employs contrastive cross-subject/cross-site learning and site-adversarial optimization to distill domain-invariant neural representations. Across a self-constructed seven-site MDD/HC dataset, CRCC achieves consistent improvements over state-of-the-art baselines and delivers strong zero-shot generalization, including a 10.7 percentage-point gain in unseen environments. This work contributes a principled bias-aware framework and a robust multi-site benchmark that enhances the clinical reliability of EEG biomarkers for depression screening.

Abstract

EEG-based neural decoding models often fail to generalize across acquisition sites due to structured, site-dependent biases implicitly exploited during training. We reformulate cross-site clinical EEG learning as a bias-factorized generalization problem, in which domain shifts arise from multiple interacting sources. We identify three fundamental bias factors and propose a general training framework that mitigates their influence through data standardization and representation-level constraints. We construct a standardized multi-site EEG benchmark for Major Depressive Disorder and introduce CRCC, a two-stage training paradigm combining encoder-decoder pretraining with joint fine-tuning via cross-subject/site contrastive learning and site-adversarial optimization. CRCC consistently outperforms state-of-the-art baselines and achieves a 10.7 percentage-point improvement in balanced accuracy under strict zero-shot site transfer, demonstrating robust generalization to unseen environments.

Figures (7)

• Figure 1: The CRCC Framework: our model comprises two primary stages: the pre-training phase (right) and the fine-tuning phase (left). During pre-training, the model takes 10-second EEG Differential Entropy (DE) features as input. These features are processed by an encoder, with gradients backpropagated through a joint objective involving a noise-reducing decoder and a domain discriminator, the latter of which facilitates robust representation learning across multiple open-source datasets. In the subsequent fine-tuning stage, the decoder and domain discriminator are detached. The core fine-tuning objective integrates a standard cross-entropy loss for classification with two specialized components: a contrastive learning module designed for cross-subject feature extraction and an adversarial loss aimed at mitigating inter-site variability.
• Figure 2: Scales values of the subjectsThe bars are color-coded, with blue representing HC and green representing MDD patients. For each group, the bars from left to right indicate: (i) the mean score for that category, (ii) the mean score of correctly classified subjects, and (iii) the mean score of misclassified subjects. The y-axis denotes the respective scale scores across four diagnostic dimensions: HAMD and BDI, and HAMA and SAS (Anxiety).
• Figure 3: Zero-shot generalization performance across varying training site scales. The trend demonstrates a clear scaling law: the balanced accuracy on the unseen test set increases monotonically as the number of training sites grows from 2 to 6. The model achieves peak performance (69.75%) with 6 sites, underscoring the benefits of data diversity in enhancing cross-site robustness.
• Figure 4: Comparative Interpretability Analysis of EEGPT: Intuitively, this explores the basis of the model's classification decisions. Greater alignment between the focus of training data and generalization data indicates that their distributions are more similar from the model's perspective, thereby leading to superior generalization performance.
• Figure 5: Multi-level interpretability and consistency analysis of CRCC(ours) across various sites and generalization scenarios.