文献类型：期刊文献

英文题名：Discriminative Feature Extraction via Multivariate Linear Regression for SSVEP-Based BCI

作者：Wang, Haiqiang[1];Zhang, Yu[1];Waytowich, Nicholas R.[2];Krusienski, Dean J.[2];Zhou, Guoxu[3];Jin, Jing[1];Wang, Xingyu[1];Cichocki, Andrzej[3,4]

机构：[1]E China Univ Sci & Technol, Key Lab Adv Control & Optimizat Chem Proc, Minist Educ, Shanghai 200237, Peoples R China;[2]Old Dominion Univ, Dept Elect & Comp Engn, Norfolk, VA 23529 USA;[3]RIKEN, Brain Sci Inst, Lab Adv Brain Signal Proc, 2-1 Hirosawa, Wako, Saitama 3510198, Japan;[4]Skolkowo Inst Sci & Technol, Moscow, Russia

年份：2016

卷号：24

期号：5

起止页码：532

外文期刊名：IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING

收录：;EI(收录号：20162402479204);WOS:【SCI-EXPANDED(收录号:WOS:000376280900002)】；

基金：This work was supported in part by the National Natural Science Foundation of China under Grant 61305028, Grant 91420302, Grant 61573142, and Grant 61203127, by the Fundamental Research Funds for the Central Universities under Grant WH1314023, Grant WG1414005, WH1516018, and Grant WH1414022, by the Shanghai Chenguang Program under Grant 14CG31, and by the National Science Foundation (USA) under Grant 1421948 and Grant 1064912. Corresponding authors: Yu Zhang and Xingyu Wang (e-mail: yuzhang@ecust.edu.cn; xywang@ecust.edu.cn).

语种：英文

外文关键词：Brain-computer interface (BCI); canonical correlation analysis (CCA); electroencephalogram (EEG); multivariate linear regression (MLR); steady-state visual evoked potential (SSVEP)

摘要：Many of the most widely accepted methods for reliable detection of steady-state visual evoked potentials (SSVEPs) in the electroencephalogram (EEG) utilize canonical correlation analysis (CCA). CCA uses pure sine and cosine reference templates with frequencies corresponding to the visual stimulation frequencies. These generic reference templates may not optimally reflect the natural SSVEP features obscured by the background EEG. This paper introduces a new approach that utilizes spatio-temporal feature extraction with multivariate linear regression (MLR) to learn discriminative SSVEP features for improving the detection accuracy. MLR is implemented on dimensionality-reduced EEG training data and a constructed label matrix to find optimally discriminative subspaces. Experimental results show that the proposed MLR method significantly outperforms CCA as well as several other competing methods for SSVEP detection, especially for time windows shorter than 1 second. This demonstrates that the MLR method is a promising new approach for achieving improved real-time performance of SSVEP-BCIs.