Duru Adil Deniz, Duru Dilek Göksel, Yumerhodzha Sami, Bebek Nerses
School of Physical Education and Sports, Neuroscience in Sports Research Lab., Marmara University, Istanbul, 34800, Turkey.
Department of Biomedical Engineering, Istanbul Arel University, Istanbul, 34537, Turkey.
Brain Imaging Behav. 2016 Jun;10(2):424-36. doi: 10.1007/s11682-015-9397-1.
Diffusion tensor imaging (DTI) allows in vivo structural brain mapping and detection of microstructural disruption of white matter (WM). One of the commonly used parameters for grading the anisotropic diffusivity in WM is fractional anisotropy (FA). FA value helps to quantify the directionality of the local tract bundle. Therefore, FA images are being used in voxelwise statistical analyses (VSA). The present study used Tract-Based Spatial Statistics (TBSS) of FA images across subjects, and computes the mean skeleton map to detect voxelwise knowledge of the tracts yielding to groupwise comparison. The skeleton image illustrates WM structure and shows any changes caused by brain damage. The microstructure of WM in thalamic stroke is investigated, and the VSA results of healthy control and thalamic stroke patients are reported. It has been shown that several skeleton regions were affected subject to the presence of thalamic stroke (FWE, p < 0.05). Furthermore the correlation of quantitative EEG (qEEG) scores and neurophysiological tests with the FA skeleton for the entire test group is also investigated. We compared measurements that are related to the same fibers across subjects, and discussed implications for VSA of WM in thalamic stroke cases, for the relationship between behavioral tests and FA skeletons, and for the correlation between the FA maps and qEEG scores.Results obtained through the regression analyses did not exceed the corrected statistical threshold values for multiple comparisons (uncorrected, p < 0.05). However, in the regression analysis of FA values and the theta band activity of EEG, cingulum bundle and corpus callosum were found to be related. These areas are parts of the Default Mode Network (DMN) where DMN is known to be involved in resting state EEG theta activity. The relation between the EEG alpha band power values and FA values of the skeleton was found to support the cortico-thalamocortical cycles for both subject groups. Further, the neurophysiological tests including Benton Face Recognition (BFR), Digit Span test (DST), Warrington Topographic Memory test (WTMT), California Verbal Learning test (CVLT) has been regressed with the FA skeleton maps for both subject groups. Our results corresponding to DST task were found to be similar with previously reported findings for working memory and episodic memory tasks. For the WTMT, FA values of the cingulum (right) that plays a role in memory process was found to be related with the behavioral responses. Splenium of corpus callosum was found to be correlated for both subject groups for the BFR.
扩散张量成像（DTI）可实现活体脑结构映射以及检测白质（WM）的微观结构破坏。用于对白质各向异性扩散进行分级的常用参数之一是分数各向异性（FA）。FA值有助于量化局部纤维束的方向性。因此，FA图像正被用于体素级统计分析（VSA）。本研究对不同受试者的FA图像采用基于纤维束的空间统计学（TBSS）方法，并计算平均骨架图以检测纤维束的体素级信息，从而进行组间比较。骨架图像展示了白质结构，并显示出由脑损伤引起的任何变化。本研究对丘脑卒中患者白质的微观结构进行了研究，并报告了健康对照组和丘脑卒中患者的VSA结果。结果表明，存在丘脑卒中时，几个骨架区域受到了影响（FWE，p < 0.05）。此外，还研究了整个测试组定量脑电图（qEEG）评分和神经生理学测试与FA骨架之间的相关性。我们比较了不同受试者中与相同纤维相关的测量结果，并讨论了丘脑卒中病例中白质VSA的意义、行为测试与FA骨架之间的关系以及FA图与qEEG评分之间的相关性。通过回归分析获得的结果未超过多重比较的校正统计阈值（未校正，p < 0.05）。然而，在FA值与脑电图θ频段活动的回归分析中，发现扣带束和胼胝体存在相关性。这些区域是默认模式网络（DMN）的一部分，已知DMN参与静息态脑电图θ活动。发现脑电图α频段功率值与骨架FA值之间的关系支持两个受试者组的皮质 - 丘脑 - 皮质循环。此外，对两个受试者组的神经生理学测试，包括本顿面部识别（BFR）、数字广度测试（DST）、沃林顿地形记忆测试（WTMT）、加利福尼亚言语学习测试（CVLT）与FA骨架图进行了回归分析。我们发现与DST任务对应的结果与先前报道的工作记忆和情景记忆任务的结果相似。对于WTMT，发现参与记忆过程的扣带（右侧）的FA值与行为反应相关。胼胝体压部在两个受试者组的BFR测试中均被发现存在相关性。
