Diffusion Tensor Imaging Enables Robust Mapping of the Deep Cerebellar Nuclei
Diffusion Tensor Imaging Enables Robust Mapping of the Deep Cerebellar Nuclei
复制标题
扩散张量成像能够实现小脑深部核团的稳健绘图
DOI：
--
复制
发表时间：
2007
期刊：
影响因子：
--
通讯作者：
S. Ying
中科院分区：
文献类型：
--
作者：
B. Landman;Annie Du;Wade D. Mayes;Jerry L Prince;S. Ying
人脑映射2007弥散张量成像实现了小脑深部核班尼特A的稳健映射。作者声明：Annie X. Du，Wade D.杰里·梅耶斯王子a，c，，莎拉H.应d，e a约翰霍普金斯大学医学院生物医学工程系，美国马里兰州巴尔的摩b约翰霍普金斯大学医学院神经病学系，美国马里兰州巴尔的摩c约翰霍普金斯大学电气和计算机工程系，美国马里兰州巴尔的摩d罗素H.摩根放射学和放射科学系，约翰霍普金斯大学医学院，巴尔的摩，MD，美国e眼科系，约翰霍普金斯大学医学院，巴尔的摩，MD，美国字数：496（500限制）介绍小脑深部核（DCN）与运动控制，平衡和空间处理密切相关，是神经退行性疾病的敏感靶点。结构和概率评估的最新进展使DCN识别成为可能[1，2]。然而，核边界的可靠定义甚至异质群体中的识别仍然是难以捉摸的，部分原因是高度可变的T2，特别是在齿状核中（例如，考虑到年龄和/或病理性铁积累）[3]。DCN的定量评估对小脑受累的神经退行性疾病的诊断、分期和预后有重要价值。白色物质完整性和连通性的扩散张量成像（DTI）研究已应用于临床和老龄人群[4，5]。然而，灰质核团的一般DTI分析受到低信号和对比度的阻碍。在这项研究中，我们使用DTI彩色图的基础上明确的对比度的白色物质束，包括核的DCN的特点。我们研究这些DTI衍生的观察结果的敏感性，在共济失调患者和对照组之间的齿状回灰质的差异。方法18例原发性孤立性小脑病变患者（11例男性/7例女性）和19例正常对照者（6例男性/13例女性）采用多层面单次激发EPI序列进行全脑扫描（标称分辨率2.2 mm）。每个序列使用32个扩散编码方向和5个平均最小加权（B 0）体积，使用3 T MR扫描仪（Intera，Philips Medical Systems，The Netherlands）。使用DTISudio（Susumu Mori，巴尔的摩，马里兰州）进行脚的纤维追踪。使用MIPAV（NIH，Bethesda，马里兰州）的色图和B 0（T2加权）体积对代表性受试者进行齿状核、顶核和球状/栓状核的描绘。球形和栓塞形细胞核不能单独区分，因此它们被合并为“插入”细胞核。由神经科专家参考组织学和结构MRI切片评估容量放置的准确性（图1）[6]。手动描绘所有受试者的双侧齿状核，同时对受试者组设盲。在控制年龄和性别的一般线性框架中比较各组的标准化B 0强度（T2 w）和齿状突体积。结果和讨论DTI彩色图能够识别对照组和患者的DCN，尽管有不同程度的萎缩和功能缺陷。彩色图体积对比度上级T2 w（B 0）和MP3（T1 w）（图1）。识别的区域在视觉上对应于在结构图像上观察到的细胞核（当细胞核在结构体积上可见时）。DCN的位置与已发表的参考文献在大小和形状上与小脑解剖结构一致（图2）。DCN与小脑脚的比较显示，DCN位于主要神经束之间，正如组织学所预期的那样（图3），因此各向异性降低的区域可以安全地解释为灰质而不是纤维交叉。共济失调患者表现出显著的齿状回体积减小（p<0.01）和T2 w升高（p<0.01）（图4）。齿状回体积减少与小脑疾病中典型的萎缩相一致，T2 w对比度增加提示铁沉积。可变的核体积降低概率方法的精度，而可变的T2降低基于T2 w对比度的描绘的可靠性。DTI提供了对比，使强大的识别DCN，并揭示临床相关的差异，齿状特征。参考文献[1]Dimitrova，A. et.al. 17（1）：240[2]Dimitrova，A.，et.al.（2006）Neuroimage.30（1）：12[ 3]Maschke，M.，et.al.（2004）J.Neurol.251（6）：740[4]Basser，P.J.，et.al.（2002）NMR. Biomed.15（7-8），456 [5]NagaePoetscher，L.M.，et.al.（2004）AJNR.25（8）：1325[6] 04 The Dog of the Woman（1995）
Human Brain Mapping 2007 Diffusion Tensor Imaging Enables Robust Mapping of the Deep Cerebellar Nuclei Bennett A. Landman , Annie X. Du , Wade D. Mayes , Jerry L. Prince a, c, , Sarah H. Ying d, e a Department of Biomedical Engineering, The Johns Hopkins University School of Medicine, Baltimore, MD, USA b Department of Neurology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA c Department of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, USA d The Russell H. Morgan Department of Radiology and Radiological Sciences, The Johns Hopkins University School of Medicine, Baltimore, MD, USA e Department of Ophthalmology, The Johns Hopkins University School of Medicine, Baltimore, MD, USA Word Count: 496 (500 Limit) Introduction Deep cerebellar nuclei (DCN) are intimately involved in motor control, balance, and spatial processing, and are sensitive targets of neurodegenerative disease. Recent progress in structural and probabilistic assessment has enabled DCN identification[1,2]. Yet, robust definition of nuclear boundaries and even identification in heterogeneous populations has remained elusive, in part due to highly variable T2, especially in the dentate (e.g., given age and/or pathological iron accumulation)[3]. Quantitative assessment of DCN could be invaluable to diagnosis, staging, and prognosis in neurodegenerative diseases with cerebellar involvement. Diffusion tensor imaging (DTI) investigations of white matter integrity and connectivity have been applied across clinical and aging populations[4,5]. Yet, general DTI analyses of gray matter nuclei have been hindered by low signal and contrast. In this study, we characterize the DCN using DTI colormaps based on well defined contrast of white matter tracts that encompass the nuclei. We investigate the sensitivity of these DTI derived observations to differences in dentate gray matter between ataxia patients and controls. Methods A multi-slice, single-shot EPI sequence achieved whole brain coverage (2.2 mm isotropic nominal resolution) in 18 patients (11M/7F) with idiopathic isolated cerebellar disease and 19 controls (6M/13F). Each sequence utilized 32 diffusion encoding directions and five, averaged minimally weighted (B0) volumes with a 3T MR scanner (Intera, Philips Medical Systems, The Netherlands). Fiber tracking of peduncles was performed with DTIStudio (Susumu Mori, Baltimore, Maryland). Delineations of the dentate, fastigial, and globose/emboliform nuclei were performed on representative subjects with colormap and B0 (T2-weighted) volumes with MIPAV (NIH, Bethesda, Maryland). Globose and emboliform nuclei could not be individually distinguished, so they were combined into “interposed” nuclei. Accuracy of volume placement was assessed by an expert neurologist with reference to histological and structural MRI sections (Fig.1)[6]. Bilateral dentate nuclei were manually delineated for all subjects while blind to subject group. Normalized B0 intensities (T2w) and dentate volumes were compared across groups in a general linear framework controlling for age and sex. Results and Discussion DTI colormaps enabled identification of DCN on controls and patients despite varying degrees of atrophy and functional deficit. Contrast on colormap volumes was superior to T2w(B0) and MPRAGE(T1w)(Fig.1). Identified regions visually corresponded to nuclei observed on structural images (when the nuclei were visible on structural volumes). Placement of the DCN agrees with published references both in terms of size and shape relative to the cerebellar anatomy(Fig.2). Comparison of the DCN with the cerebellar peduncles reveals that the DCN lie between major tracts, as expected from histology(Fig.3), so the areas of reduced anisotropy may be safely interpreted as gray matter rather than fiber crossings. Ataxia patients exhibited significant decreased dentate volume(p<0.01) and elevated T2w(p<.01)(Fig.4). Decreased dentate volume is agreement with atrophy typically presenting in cerebellar disease, and increased in T2w contrast suggests iron deposition. Variable nuclei volumes reduce the precision of probabilistic methods, while variable T2’s reduce the reliability of delineation based on T2w contrasts. DTI provides contrasts that enable robust identification of DCN and reveal clinically relevant differences in dentate characteristics. References[1]Dimitrova,A.,et.al.(2002)Neuroimage.17(1):240[2]Dimitrova,A.,et.al.(2006)Neuroimage.30(1):12[ 3]Maschke,M.,et.al.(2004)J.Neurol.251(6):740[4]Basser,P.J.,et.al.(2002)NMR.Biomed.15(7-8),456[5]NagaePoetscher,L.M.,et.al.(2004)AJNR.25(8):1325[6]Duvernoy,H.M.(1995)Springer Abstract: Human Brain Mapping 2007 Human Brain Mapping 2007