arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.12910eess.SP
收藏

基于有效连接的无监督脑电通道选择方法

Effective Connectivity-Based Unsupervised Channel Selection Method for EEG

New York Institute of Technology(纽约理工学院);University of Essex(埃塞克斯大学);University of Isfahan(伊斯法罕大学);Isfahan University of Medical Sciences(伊斯法罕医科大学)

Neda Abdollahpour, N. Sertac Artan, Ian Daly, Mohammadreza Yazdchi, Zahra Baharlouei

AI总结:

提出基于有效连接的无监督EEG通道选择方法,通过ICEC准则量化通道因果影响,在三个数据集上以更少通道实现最高准确率。

摘要:

分析神经数据(如脑电图(EEG)数据)通常需要处理高维数据集,其中并非所有通道都提供同等有意义的信息。选择最相关的通道对于提高计算效率和确保对神经动态的稳健洞察至关重要。本研究引入了基于有效连接(ICEC)的通道重要性准则,用于量化每个通道中的有效连接(EC)。有效连接指的是一个神经区域对另一个神经区域施加的因果影响,提供了信息方向性流动的洞察。利用该准则,我们提出了一种考虑通道间交互强度的无监督通道选择方法。为了评估所提出的通道选择方法,我们将其应用于四个类别的三个知名EEG数据集。评估涉及使用五种有效连接度量计算ICEC准则:部分有向相干(PDC)、广义PDC(GPDC)、重整化PDC(RPDC)、有向传递函数(DTF)和直接DTF(dDTF)。为了聚焦通道选择的效果,我们采用公共空间模式(CSP)算法进行特征提取,并使用支持向量机(SVM)对所有参与者进行分类。结果与其他基于CSP的方法进行了比较。评估包括比较在五种有效连接度量下,使用和不使用所提出方法的参与者特定准确率。结果显示,所有参与者的性能均持续提升,且所选电极数量显著减少。与最先进的方法相比,我们的方法在三个数据集上取得了最高准确率:82%(22个通道中的13个)、86.01%(59个通道中的29个)和87.56%(118个通道中的48个)。

英文摘要:

Analyzing neural data such as Electroencephalography (EEG) data often involves dealing with high-dimensional datasets, where not all channels provide equally meaningful informa- tion. Selecting the most relevant channels is crucial for improving computational efficiency and ensuring robust insights into neural dynamics. This study introduces the Importance of Channels based on Effective Connectivity (ICEC) criterion for quantifying effective connectivity (EC) in each channel. Effective connectivity refers to the causal influence one neural region exerts over another, providing insights into the directional flow of information. Using this criterion, we propose an unsupervised channel selection method that accounts for the intensity of interactions among channels. To evaluate the proposed channel selection method, we applied it to three well-known EEG datasets across four categories. The assessment involved calculating the ICEC criterion using five effective connectivity metrics: partial directed coherence (PDC), generalized PDC (GPDC), renormalized PDC (RPDC), directed transfer function (DTF), and direct DTF (dDTF). To focus on the effect of channel selection, we employed the Common Spatial Pattern (CSP) algorithm for feature extraction and a Support Vector Machine (SVM) for classification across all participants. Results were compared with other CSP-based methods. The evaluation included comparing participant- specific accuracies with and without the proposed method across five effective connectivity metrics. The results showed consistent performance improvements and a significant reduction in the number of selected electrodes for all participants. Compared to state-of-the-art methods, our approach achieved the highest accuracies: 82% (13 out of 22 channels), 86.01% (29 out of 59 channels), and 87.56% (48 out of 118 channels) across three datasets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑