Multi-View Contrastive Network (MCNet) for Motor Imagery Classification

2025-02-18

AI Summary

To address the limited deep feature learning capability caused by small-sample electroencephalography (EEG) data in motor imagery (MI)-based brain–computer interfaces (BCIs), this paper proposes a knowledge-driven time–space–frequency multi-view contrastive network. Methodologically, it introduces, for the first time, a tri-domain collaborative knowledge-guided data augmentation strategy and a dual-path contrastive learning framework, integrating time–frequency decomposition, spatial filtering, and neurophysiological priors to construct cross-view contrastive modules and cross-model consistency constraints; supervised contrastive learning is further incorporated to enhance discriminative representation learning. Evaluated on four benchmark MI EEG datasets, the proposed method consistently outperforms ten state-of-the-art approaches, achieving an average classification accuracy improvement of 3.2–5.7%. It demonstrates strong generalizability, architectural compatibility with diverse neural backbones, and effectively alleviates the small-sample bottleneck inherent in EEG-based MI decoding.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• 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
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Objective: An electroencephalography (EEG)-based brain-computer interface (BCI) serves as a direct communication pathway between the human brain and an external device. While supervised learning has been extensively explored for motor imagery (MI) EEG classification, small data quantity has been a key factor limiting the performance of deep feature learning. Methods: This paper proposes a knowledge-driven time-space-frequency based multi-view contrastive network (MCNet) for MI EEG decoding in BCIs. MCNet integrates knowledge from the time, space, and frequency domains into the training process through data augmentations from multiple views, fostering more discriminative feature learning of the characteristics of EEG data. We introduce a cross-view contrasting module to learn from different augmented views and a cross-model contrasting module to enhance the consistency of features extracted between knowledge-guided and data-driven models. Results: The combination of EEG data augmentation strategies was systematically investigated for more informative supervised contrastive learning. Experiments on four public MI datasets and three different architectures demonstrated that MCNet outperformed 10 existing approaches. Significance: Our approach can significantly boost EEG classification performance beyond designated networks, showcasing the potential to enhance the feature learning process for better EEG decoding.

Problem

• Improves motor imagery EEG classification
• Addresses small data quantity issue
• Enhances deep feature learning process

Innovation

• Multi-View Contrastive Network
• Time-space-frequency integration
• Cross-view contrasting module