Mantini D, Perrucci M G, Cugini S, Ferretti A, Romani G L, Del Gratta C
Institute of Advanced Biomedical Technologies, G. D'Annunzio University Foundation, Department of Clinical Sciences and Bio-imaging, G. D'Annunzio University, Chieti, Italy.
Neuroimage. 2007 Jan 15;34(2):598-607. doi: 10.1016/j.neuroimage.2006.09.037. Epub 2006 Nov 16.
The simultaneous recording of EEG and fMRI is a promising method for combining the electrophysiological and hemodynamic information on cerebral dynamics. However, EEG recordings performed in the MRI scanner are contaminated by imaging, ballistocardiographic (BCG) and ocular artifacts. A number of processing techniques for the cancellation of fMRI environment disturbances exist: the most popular is averaged artifact subtraction (AAS), which performs well for the imaging artifact, but has some limitations in removing the BCG artifact, due to the variability in cardiac wave duration and shape; furthermore, no processing method to attenuate ocular artifact is currently used in EEG/fMRI, and contaminated epochs are simply rejected before signal analysis. In this work, we present a comprehensive method based on independent component analysis (ICA) for simultaneously removing BCG and ocular artifacts from the EEG recordings, as well as residual MRI contamination left by AAS. The ICA method has been tested on event-related potentials (ERPs) obtained from a visual oddball paradigm: it is very effective in attenuating artifacts in order to reconstruct clear brain signals from EEG acquired in the MRI scanner. It performs significantly better than the AAS method in removing the BCG artifact. Furthermore, since ocular artifacts can be completely suppressed, a larger number of trials is available for analysis. A comparison of ERPs inside the magnetic environment with those obtained out of the MRI scanner confirms that no systematic bias in the ERP waveform is produced by the ICA method.
同时记录脑电图（EEG）和功能磁共振成像（fMRI）是一种很有前景的方法，可将有关脑动力学的电生理信息和血液动力学信息结合起来。然而，在MRI扫描仪中进行的EEG记录会受到成像、心冲击图（BCG）和眼动伪迹的干扰。目前存在多种用于消除fMRI环境干扰的处理技术：最常用的是平均伪迹减法（AAS），它对成像伪迹效果良好，但在去除BCG伪迹方面存在一些局限性，这是由于心电波持续时间和形状的变异性所致；此外，目前在EEG/fMRI中尚未使用任何衰减眼动伪迹的处理方法，并且在信号分析之前只是简单地剔除受污染的时段。在这项工作中，我们提出了一种基于独立成分分析（ICA）的综合方法，用于同时从EEG记录中去除BCG和眼动伪迹，以及AAS留下的残余MRI污染。ICA方法已在从视觉Oddball范式获得的事件相关电位（ERP）上进行了测试：它在衰减伪迹方面非常有效，以便从在MRI扫描仪中采集的EEG重建清晰的脑信号。在去除BCG伪迹方面，它比AAS方法表现得明显更好。此外，由于眼动伪迹可以被完全抑制，因此有更多的试验可用于分析。对磁环境中的ERP与从MRI扫描仪外部获得的ERP进行比较，证实ICA方法不会在ERP波形中产生系统偏差。
