Suppr超能文献

从单次试验信号中同时识别眼动注视相关电位和反应相关电位。

Simultaneous identification of eye fixation related potentials and reaction related potentials from single-trial signals.

作者信息

Moriyama Y, Tomita Y, Honda S, Matsuo N, Hitoshi U

机构信息

Faculty of Science & Technology, Keio University, Yokohama, Japan.

出版信息

Med Biol Eng Comput. 1997 Nov;35(6):671-6. doi: 10.1007/BF02510976.

Abstract

To obtain the effective components from event related potentials (ERPs), it has been necessary to average the waveforms of several trials. ERP data reflect the psychophysiological state of a subject. Time variation is an important feature in ERP analysis, and so the single-trial method is required. A method is proposed to identify simultaneously both eye fixation related potentials (EFRPs) and reaction related potentials (RRPs) using wavelength transforms. The EFRP is an ERP associated with saccadic eye movement. The RRP is defined as a component similar to P300 gained from the reaction signal. Six subjects participate in the oddball task. The task takes 30 min for each subject. Electroencephalograms (EEGs), electro-oculograms (EOGs) and electromyograms (EMGs) are simultaneously recorded. The EFRP is extracted for the offset of saccade from EOG, and the RRP is extracted for the onset of reaction from EMG. The results show that the estimated waveforms described well each of the components in the EFRP and RRP. Moreover, the simulation results show that the amplitudes of the lambda wave are estimated to within an error of 4%, and those of the latencies are within 0.4%, with an SNR of 4.5dB. Those of P300 were 11 and 4%, respectively. The reliability of the method is proved to be sufficient for estimating ERPs.

摘要

为了从事件相关电位(ERP)中获取有效成分,有必要对多次试验的波形进行平均。ERP数据反映了受试者的心理生理状态。时间变化是ERP分析中的一个重要特征,因此需要采用单次试验方法。本文提出了一种利用波长变换同时识别眼动相关电位(EFRP)和反应相关电位(RRP)的方法。EFRP是一种与眼球跳动相关的ERP。RRP被定义为从反应信号中获得的类似于P300的成分。六名受试者参与了Oddball任务。每个受试者的任务时长为30分钟。同时记录脑电图(EEG)、眼电图(EOG)和肌电图(EMG)。从EOG中提取眼跳偏移的EFRP,从EMG中提取反应起始的RRP。结果表明,估计波形能很好地描述EFRP和RRP中的每个成分。此外,模拟结果表明,λ波的幅度估计误差在4%以内,潜伏期误差在0.4%以内,信噪比为4.5dB。P300的幅度估计误差分别为11%和4%。该方法的可靠性被证明足以用于ERP的估计。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验