Event-related modulation of steady-state visual evoked potentials for eyes-closed brain computer interface
Event-related modulation of steady-state visual evoked potentials for eyes-closed brain computer interface
复制标题
DOI：
10.1109/embc.2015.7318758
复制
发表时间：
2015-11
期刊：
影响因子：
--
通讯作者：
S. Nishifuji;Yuya Sugita;H. Hirano
中科院分区：
文献类型：
--
作者：
S. Nishifuji;Yuya Sugita;H. Hirano
脑机接口（Brain Computer Interface，BCI），又称脑机接口，是将脑电信号（EEG）的调制转换为用户的意图，从而在没有语音和身体运动的情况下与他人进行交流。脑机接口已被研究和开发为一种重要的手段，用于严重运动残疾的残疾人，如肌萎缩性侧索硬化症和肌营养不良症患者与他们的照顾者之间的沟通援助。最先进的脑机接口在信息传输速率和分类精度方面取得了优异的性能。然而，由于视觉模态的要求，大多数传统的脑机接口仍然无法用于眼控制受损的患者。本研究的目的是开发一种新的2级脑机接口，这是独立于眼控制，包括睁眼使用事件相关调制的稳态视觉诱发电位（SSVEP）在闭眼条件下的心理任务。11名年龄在21-24岁之间的健康受试者被招募，并被要求在闭眼条件下执行两项心理任务中的每一项：分别对闪烁刺激的心理关注和对他们最喜欢的动物的图像回忆。在闪烁频率为10 Hz，刺激强度为3-5 lx的条件下，几乎所有被试的后区的SSVEP的幅度都被认为是通过执行心理任务来调制的，这被用来表达用户的二元意图，即执行心理任务之一或不执行（休息）。在所有被试中，心理焦点的分类性能（80%）大于图像回忆的分类性能（75%）。缩短用于分类的数据长度将提高所提出的BCI的信息传输速率。
Brain computer interfaces (BCIs), also be referred to be as brain machine interfaces, transform modulations of electroencephalogram (EEG) into user's intents to communicate with others without voice and physical movement. BCIs have been studied and developed as one of the important means for communication-aid between disabled with severe motor disabilities such as amyotrophic lateral sclerosis and muscular dystrophy patients and their caregivers. State-of-art BCIs have achieved the outstanding performance in information transfer rate and classification accuracy. However, most of conventional BCIs are still unavailable for patients with impaired oculomotor control due to requirement of visual modality. The present study aimed at developing a novel 2-class BCI which was independent of oculomotor control including eye-opening using event-related modulation of steady state visual evoked potential (SSVEP) associated with mental tasks under eyes-closed condition. Eleven healthy subjects aged 21-24 years old were recruited and directed to perform each of two mental tasks under an eyes-closed condition; mental focus on flicker stimuli and image recall of their favorite animals, respectively. The magnitudes of SSVEP in the posterior regions of almost all the subjects were seen to be modulated by performing the mental tasks under the conditions of the flickering frequency of 10 Hz and stimulus intensity of 3-5 lx, which was used to express a user's binary intent, namely, performing one of the mental tasks or not (rest). The classification performance on the mental focus, 80 %, was larger than that on the image recall, 75 %, in average across all the subjects. Shortening of the data length used for classification would improve the information transfer rate of the proposed BCI.