Effects of the Phantom Shape on the Gradient Artefact of Electroencephalography (EEG) Data in Simultaneous EEG-fMRI

Effects of the Phantom Shape on the Gradient Artefact of Electroencephalography (EEG) Data in Simultaneous EEG-fMRI
复制标题

DOI:
10.3390/app8101969
复制
发表时间:
2018-10-01
影响因子:
2.7
通讯作者:
Alzoubi, Khawla
Alzoubi, Khawla
中科院分区:
综合性期刊4区
文献类型:
--
作者:
Chowdhury, Muhammad E. H.;Khandakar, Amith;Alzoubi, Khawla

文献摘要

被引文献

相似文献

在同时采集脑电图和功能磁共振成像 (fMRI) 数据期间,由于磁共振 (MR) 扫描仪中的时变场梯度,脑电图 (EEG) 信号会受到梯度伪影 (GA) 的严重影响。在 MR 扫描仪内记录 EEG 时,GA 是伪影的主要贡献者,其中大多数来自 EEG 帽和受试者头部的相互作用。许多研究人员一直在联合 EEG-fMRI 研究中使用球形模型来表征 EEG 数据中的 GA。在这项研究中,我们研究了模型形状如何影响 GA 的表征。在执行定制和标准回波平面成像 (EPI) 序列期间,分别使用球形模型、头形模型和六名人类受试者记录脑电图数据。在使用头部形状模型进行试验时,GA 的均方根 (RMS) 电压在 EEG 通道上的空间电位图与人类头部相关的空间电位图非常相似,而不是球形模型获得的空间电位图。这通过测量平均相似性指数 (0.85/0.68) 得到证实。此外,配对t检验显示,头部模型和球形模型的数据与受试者的数据显着不同(p < 0.005),而头部模型和球形模型的数据之间的差异不显着(p = 0.07)。这项研究的结果强烈表明,头形模型应该用于并发 EEG-fMRI 中的 GA 表征研究。
Electroencephalography (EEG) signals greatly suffer from gradient artefacts (GAs) due to the time-varying field gradients in the magnetic resonance (MR) scanner during the simultaneous acquisition of EEG and functional magnetic resonance imaging (fMRI) data. The GAs are the principal contributors of artefacts while recording EEG inside an MR scanner, and most of them come from the interaction of the EEG cap and the subject's head. Many researchers have been using a spherical phantom to characterize the GA in EEG data in combined EEG-fMRI studies. In this study, we investigated how the phantom shape could affect the characterization of the GA. EEG data were recorded with a spherical phantom, a head-shaped phantom, and six human subjects, individually, during the execution of customized and standard echo-planar imaging (EPI) sequences. The spatial potential maps of the root-mean-square (RMS) voltage of the GA over EEG channels for the trials with a head-shaped phantom closely mimicked those related to the human head rather than those obtained for the spherical phantom. This was confirmed by measuring the average similarity index (0.85/0.68). Moreover, a paired t-test showed that the head-shaped phantom's and the spherical phantom's data were significantly different (p < 0.005) from the subjects' data, whereas the difference between the head-shaped phantom's and the spherical phantom's data was not significant (p = 0.07). The results of this study strongly suggest that a head-shaped phantom should be used for GA characterization studies in concurrent EEG-fMRI.