文献类型：期刊文献

英文题名：EDSF-Net: An enhanced dynamic spatiotemporal-frequency attention network for robust EEG decoding in motor imagery

作者：Chen, Weijie[1,4];Daly, Ian[2,5];Chen, Yixin[1,4];Li, Junxian[1,4];Wu, Xiao[1,4];Zhao, Ruiyu[1,4];Wang, Xingyu[1,4];Cichocki, Andrzej[3,6,7];Jin, Jing[1,4,8]

机构：[1]East China Univ Sci & Technol, Shanghai 200237, Peoples R China;[2]Univ Essex, Wivenhoe Pk, Colchester CO4 3SQ, Essex, England;[3]Nicolaus Copernicus Univ, PL-871001 Torun, Poland;[4]East China Univ Sci & Technol, Key Lab Smart Mfg Energy Chem Proc, Minist Educ, Shanghai 200237, Peoples R China;[5]Univ Essex, Sch Comp Sci & Elect Engn, Brain Comp Interfacing & Neural Engn Lab, Wivenhoe Pk, Colchester CO4 3SQ, Essex, England;[6]Polish Acad Sci, Syst Res Inst, 01-447b, Warsaw, Poland;[7]Nicolaus Copernicus Univ UMK, PL-87100 Torun, Poland;[8]East China Univ Sci & Technol, Sch Math, Shanghai 200237, Peoples R China

年份：2026

卷号：204

外文期刊名：NEURAL NETWORKS

收录：;WOS:【SCI-EXPANDED(收录号:WOS:001798589400001)】；

基金：This work was supported in part by the Brain Science and Brain-like Intelligence Technology-National Science and Technology Major Project under Grant 2022ZD0208900 and National Natural Science Foundation of China under Grant 62176090; in part by Shanghai Municipal Science and Technology Major Project under Grant 2021SHZDZX, This research is also supported by Project of Jiangsu Province Science and Technology Plan Special Fund in 2022 (Key research and development plan industry foresight, fundamental research fund for the central universities JKH01241605 and key core technologies) under Grant BE2022064-1, and in part by the Lingang Laboratory under Grant LGL8998.

语种：英文

外文关键词：Brain computer interface; Motor imagery; Electroencephalogram; Convolutional neural network

摘要：Motor imagery is a non-invasive process that operates independently of external stimuli, and can be used to establish a direct connection between the brain and external devices solely through the imagination of a specific movement. Nonetheless, the complexity and variability of neural patterns pose substantial challenges, as accurately decoding motor imagery from electroencephalography signals remains a significant obstacle. This paper introduces an enhanced dynamic spatiotemporal-frequency attention convolutional neural network (EDSF-Net) for the precise decoding of motor imagery. EDSF-Net employs a refined spatiotemporal attention mechanism, grounded in enhanced dynamic convolution (EDConv), to emphasize localized spatial features alongside high and low-frequency temporal characteristics. Subsequently, EDConv is utilized for global spatial feature extraction. Following this, group convolutions formed by EDConv are implemented to fuse the extracted features effectively. Ultimately, a synchronized channel-frequency attention mechanism is employed to capture critical channel and frequency domain information, facilitating the model's focus on features most pertinent to the task throughout the learning process. We conducted a comprehensive evaluation of the performance of EDSF-Net on two public datasets, BCI Competition IV 2a and OpenBMI. In the hold-out session experiments, EDSF-Net achieved decoding accuracies of 84.26% and 75.14%, respectively. In the leave-one-subject-out experiments, EDSF-Net attained decoding accuracies of 66.78% and 82.24%, respectively. These results show that EDSF-Net has robust generalization capabilities, affirming its efficacy in addressing complex pattern recognition tasks, with significant potential for diverse applications.