CAMEL-CLIP: Channel-aware Multimodal Electroencephalography-text Alignment for Generalizable Brain Foundation Models

TL;DR

CAMEL-CLIP is a novel EEG-text alignment model that enhances robustness to channel heterogeneity, enabling better generalization across diverse EEG tasks through innovative encoding and contrastive learning techniques.

Contribution

It introduces a channel-aware multimodal pretraining framework with semantic positional encoding, dynamic channel projection, and dual-level contrastive learning for EEG-text alignment.

Findings

Achieves state-of-the-art performance in linear-probing tasks.

Outperforms existing models with full-finetuning.

Demonstrates robustness to channel heterogeneity.

Abstract

Electroencephalography (EEG) foundation models have shown promise for learning generalizable representations, yet they remain sensitive to channel heterogeneity, such as changes in channel composition or ordering. We propose channel-aware multimodal EEG-text alignment contrastive language-image pretraining (CAMEL-CLIP), a contrastive EEG-text multimodal foundation model designed to be robust to heterogeneous channel configurations and widely applicable to diverse downstream tasks. CAMEL-CLIP introduces three key components: (1) channel attribute-based positional encoding, which identifies channels through semantic information; (2) dynamic channel projection, which generates variable-length embeddings by independently projecting each channel without feature compression; and (3) dual-level contrastive learning, which jointly performs channel-level and sample-level contrastive learning to capture both channel-specific and global signal characteristics. Experimental results demonstrate that CAMEL-CLIP achieves state-of-the-art performance under linear-probing and outperforms existing foundation models that rely on full-finetuning.