标题： AGTCNet：一种基于图-时间的合理运动想象脑电分类方法 显示英文标题

标题： AGTCNet: A Graph-Temporal Approach for Principled Motor Imagery EEG Classification

摘要： 脑机接口（BCI）技术利用脑电图（EEG）标志着一项变革性创新，使运动障碍个体能够平等地与环境互动。尽管其具有令人鼓舞的潜力，但由于个体和时间上神经活动的固有复杂性和变异性，以及EEG硬件限制，开发主体不变和会话不变的BCI系统仍然是一个重大挑战。虽然之前的研究试图开发稳健的BCI系统，但现有方法在捕捉多通道EEG信号中的复杂时空依赖性方面仍然无效。本研究通过引入注意力图时序卷积网络（AGTCNet），填补了这一空白，这是一种用于运动想象EEG（MI-EEG）分类的新颖图时序模型。具体而言，AGTCNet利用EEG电极的拓扑配置作为归纳偏差，并集成图卷积注意网络（GCAT）以联合学习表达性的时空EEG表示。所提出的模型显著优于现有的MI-EEG分类器，在使用紧凑架构的同时实现了最先进的性能，证明了其在BCI部署中的有效性和实用性。在BCI竞赛IV数据集2a上，AGTCNet在主体无关分类中实现了66.82%的移动平均准确率，当微调用于主体特定分类时进一步提高到82.88%。在EEG运动运动/想象数据集上，AGTCNet分别实现了4类和2类主体无关分类的64.14%和85.22%的移动平均准确率，对于主体特定分类进一步提高到72.13%和90.54%。 显示英文摘要

摘要： Brain-computer interface (BCI) technology utilizing electroencephalography (EEG) marks a transformative innovation, empowering motor-impaired individuals to engage with their environment on equal footing. Despite its promising potential, developing subject-invariant and session-invariant BCI systems remains a significant challenge due to the inherent complexity and variability of neural activity across individuals and over time, compounded by EEG hardware constraints. While prior studies have sought to develop robust BCI systems, existing approaches remain ineffective in capturing the intricate spatiotemporal dependencies within multichannel EEG signals. This study addresses this gap by introducing the attentive graph-temporal convolutional network (AGTCNet), a novel graph-temporal model for motor imagery EEG (MI-EEG) classification. Specifically, AGTCNet leverages the topographic configuration of EEG electrodes as an inductive bias and integrates graph convolutional attention network (GCAT) to jointly learn expressive spatiotemporal EEG representations. The proposed model significantly outperformed existing MI-EEG classifiers, achieving state-of-the-art performance while utilizing a compact architecture, underscoring its effectiveness and practicality for BCI deployment. With a 49.87% reduction in model size, 64.65% faster inference time, and shorter input EEG signal, AGTCNet achieved a moving average accuracy of 66.82% for subject-independent classification on the BCI Competition IV Dataset 2a, which further improved to 82.88% when fine-tuned for subject-specific classification. On the EEG Motor Movement/Imagery Dataset, AGTCNet achieved moving average accuracies of 64.14% and 85.22% for 4-class and 2-class subject-independent classifications, respectively, with further improvements to 72.13% and 90.54% for subject-specific classifications.