Mercier Manuel R, Bickel Stephan, Megevand Pierre, Groppe David M, Schroeder Charles E, Mehta Ashesh D, Lado Fred A
Department of Neurology, Montefiore Medical Center, 111 East 210th Street, Bronx, NY 10467, USA; Department of Neuroscience, Albert Einstein College of Medicine, 1410 Pelham Parkway South, Bronx, NY 10461, USA; Centre de Recherche Cerveau et Cognition (CerCo), CNRS, UMR5549, Pavillon Baudot CHU Purpan, BP 25202, 31052 Toulouse Cedex, France.
Department of Neurology, Montefiore Medical Center, 111 East 210th Street, Bronx, NY 10467, USA.
Neuroimage. 2017 Feb 15;147:219-232. doi: 10.1016/j.neuroimage.2016.08.037. Epub 2016 Aug 21.
While there is a strong interest in meso-scale field potential recording using intracranial electroencephalography with penetrating depth electrodes (i.e. stereotactic EEG or S-EEG) in humans, the signal recorded in the white matter remains ignored. White matter is generally considered electrically neutral and often included in the reference montage. Moreover, re-referencing electrophysiological data is a critical preprocessing choice that could drastically impact signal content and consequently the results of any given analysis. In the present stereotactic electroencephalography study, we first illustrate empirically the consequences of commonly used references (subdermal, white matter, global average, local montage) on inter-electrode signal correlation. Since most of these reference montages incorporate white matter signal, we next consider the difference between signals recorded in cortical gray matter and white matter. Our results reveal that electrode contacts located in the white matter record a mixture of activity, with part arising from the volume conduction (zero time delay) of activity from nearby gray matter. Furthermore, our analysis shows that white matter signal may be correlated with distant gray matter signal. While residual passive electrical spread from nearby matter may account for this relationship, our results suggest the possibility that this long distance correlation arises from the white matter fiber tracts themselves (i.e. activity from distant gray matter traveling along axonal fibers with time lag larger than zero); yet definitive conclusions about the origin of the white matter signal would require further experimental substantiation. By characterizing the properties of signals recorded in white matter and in gray matter, this study illustrates the importance of including anatomical prior knowledge when analyzing S-EEG data.
虽然人们对使用穿透深度电极的颅内脑电图（即立体定向脑电图或S-EEG）进行中尺度场电位记录有着浓厚兴趣，但白质中记录的信号却一直被忽视。白质通常被认为是电中性的，并且在参考导联中常常被包含在内。此外，重新设置电生理数据的参考导联是一个关键的预处理选择，它可能会极大地影响信号内容，进而影响任何给定分析的结果。在本立体定向脑电图研究中，我们首先通过实验说明了常用参考导联（皮下、白质、全局平均、局部导联）对电极间信号相关性的影响。由于这些参考导联中的大多数都包含白质信号，接下来我们考虑皮质灰质和白质中记录的信号之间的差异。我们的结果表明，位于白质中的电极触点记录到的是混合活动，其中一部分来自附近灰质活动的容积传导（零时间延迟）。此外，我们的分析表明，白质信号可能与远处的灰质信号相关。虽然附近组织的残余被动电扩散可能解释这种关系，但我们的结果表明，这种长距离相关性有可能源于白质纤维束本身（即远处灰质的活动沿轴突纤维传播，时间延迟大于零）；然而，关于白质信号起源的确切结论还需要进一步的实验证实。通过描述白质和灰质中记录的信号特性，本研究说明了在分析S-EEG数据时纳入解剖学先验知识的重要性。
