A user-friendly SSVEP-based brain-computer interface using a time-domain classifier
A user-friendly SSVEP-based brain-computer interface using a time-domain classifier
复制标题
DOI：
10.1088/1741-2560/7/2/026010
复制
发表时间：
2010-04-01
影响因子：
4
通讯作者：
Sullivan, Thomas J.
中科院分区：
文献类型：
--
作者：
Luo, An;Sullivan, Thomas J.
介绍了一种基于稳态视觉诱发电位(SSVEP)的人机界面系统。使用低噪声干电极记录单通道脑电。与传统的基于凝胶的多传感器脑电系统相比，干式传感器更方便、更舒适、更具成本效益。搭建了一个硬件系统，显示四个LED灯盘以不同的频率闪烁，并与脑电信号采集同步。视觉刺激经过精心设计，使光敏者的潜在风险降至最低。我们描述了一种新的刺激锁定道间相关(SLIC)方法用于SSVEP分类，该方法利用脑电信号对刺激起始集进行时间锁定。研究了参数的不同选取对算法性能的影响。使用SLIC方法，平均光检测率为75.8%，错误率非常低(假阳性率为8.4%，误识率为1.3%)。与传统的基于频域的方法相比，SLIC方法更健壮(对用户的烦扰更少)，也适用于不规则的刺激模式。
We introduce a user-friendly steady-state visual evoked potential (SSVEP)-based brain-computer interface (BCI) system. Single-channel EEG is recorded using a low-noise dry electrode. Compared to traditional gel-based multi-sensor EEG systems, a dry sensor proves to be more convenient, comfortable and cost effective. A hardware system was built that displays four LED light panels flashing at different frequencies and synchronizes with EEG acquisition. The visual stimuli have been carefully designed such that potential risk to photosensitive people is minimized. We describe a novel stimulus-locked inter-trace correlation (SLIC) method for SSVEP classification using EEG time-locked to stimulus onsets. We studied how the performance of the algorithm is affected by different selection of parameters. Using the SLIC method, the average light detection rate is 75.8% with very low error rates (an 8.4% false positive rate and a 1.3% misclassification rate). Compared to a traditional frequency-domain-based method, the SLIC method is more robust (resulting in less annoyance to the users) and is also suitable for irregular stimulus patterns.