Grinberg Farida, Maximov Ivan I, Farrher Ezequiel, Shah N Jon
Institute of Neuroscience and Medicine - 4, Forschungszentrum Juelich GmbH, Juelich, Germany,; Department of Neurology, Faculty of Medicine, RWTH Aachen University, JARA, Aachen, Germany.
Institute of Neuroscience and Medicine - 4, Forschungszentrum Juelich GmbH, Juelich, Germany.
Magn Reson Imaging. 2018 Jan;45:7-17. doi: 10.1016/j.mri.2017.08.007. Epub 2017 Sep 1.
Conventional fibre tractography methods based on diffusion tensor imaging exploit diffusion anisotropy and directionality in the range of low diffusion weightings (b-values). High b-value Biexponential Diffusion Tensor Analysis reported previously has demonstrated that fractional anisotropy of the slow diffusion component is essentially higher than that of conventional diffusion tensor imaging whereas popular compartment models associate this slow diffusion component with axonal water fraction. One of the primary aims of this study is to elucidate the feasibility and potential benefits of "microstructure-informed" whole-brain slow-diffusion fibre tracking (SDIFT) in humans. In vivo diffusion-weighted images in humans were acquired in the extended range of diffusion weightings≤6000smm at 3T. Fast and slow diffusion tensors were reconstructed using the bi-exponential tensor decomposition, and a detailed statistical analysis of the relevant whole-brain tensor metrics was performed. We visualised three-dimensional fibre tracts in in vivo human brains using deterministic streamlining via the major eigenvector of the slow diffusion tensor. In particular, we demonstrated that slow-diffusion fibre tracking provided considerably higher fibre counts of long association fibres and allowed one to reconstruct more short association fibres than conventional diffusion tensor imaging. SDIFT is suggested to be useful as a complimentary method capable to enhance reliability and visualisation of the evaluated fibre pathways. It is especially informative in precortical areas where the uncertainty of the mono-exponential tensor evaluation becomes too high due to decreased anisotropy of low b-value diffusion in these areas. Benefits can be expected in assessment of the residual axonal integrity in tissues affected by various pathological conditions, in surgical planning, and in evaluation of cortical connectivity, in particular, between Brodmann's areas.
基于扩散张量成像的传统纤维束成像方法利用低扩散权重（b值）范围内的扩散各向异性和方向性。先前报道的高b值双指数扩散张量分析表明，慢扩散成分的分数各向异性本质上高于传统扩散张量成像，而流行的 compartment 模型将这种慢扩散成分与轴突水分数联系起来。本研究的主要目的之一是阐明“微观结构信息”全脑慢扩散纤维追踪（SDIFT）在人体中的可行性和潜在益处。在3T条件下，采集了扩散权重≤6000s/mm²扩展范围内的人体活体扩散加权图像。使用双指数张量分解重建快速和慢速扩散张量，并对相关全脑张量指标进行了详细的统计分析。我们通过慢扩散张量的主要特征向量，利用确定性流线法在人体活体大脑中可视化三维纤维束。特别是，我们证明了慢扩散纤维追踪提供的长联合纤维数量明显更多，并且与传统扩散张量成像相比，能够重建更多的短联合纤维。SDIFT被认为是一种有用的补充方法，能够提高所评估纤维通路的可靠性和可视化。在皮质前区域，由于这些区域低b值扩散的各向异性降低，单指数张量评估的不确定性变得过高，SDIFT在该区域尤其具有信息量。在评估受各种病理状况影响的组织中的残余轴突完整性、手术规划以及评估皮质连接性，特别是布罗德曼区域之间的连接性方面，有望带来益处。
