Wu Ye, Lin Weili, Shen Dinggang, Yap Pew-Thian
Department of Radiology and BRIC, University of North Carolina, Chapel Hill, USA.
Inf Process Med Imaging. 2019 Jun;11492:319-331. doi: 10.1007/978-3-030-20351-1_24. Epub 2019 May 22.
Fiber tractography in baby diffusion MRI is challenging due to the low and spatially-varying diffusion anisotropy, causing most tractography algorithms to yield streamlines that fall short of reaching the cortex. In this paper, we introduce a method called asymmetry spectrum imaging (ASI) to improve the estimation of white matter pathways in the baby brain by (i) incorporating an asymmetric fiber orientation model to resolve subvoxel fiber configurations such as fanning and bending, and (ii) explicitly modeling the range (or ) of typical diffusion length scales in the developing brain. We validated ASI using in-vivo baby diffusion MRI data from the Baby Connectome Project (BCP), demonstrating that ASI can characterize complex subvoxel fiber configurations and accurately estimate the fiber orientation distribution function in spite of changes in diffusion patterns. This, in turn, results in significantly better diffusion tractography in the baby brain.
由于婴儿扩散磁共振成像（MRI）中扩散各向异性较低且空间变化，纤维束成像具有挑战性，这导致大多数纤维束成像算法生成的流线无法到达皮层。在本文中，我们引入了一种称为不对称频谱成像（ASI）的方法，以通过以下方式改进婴儿大脑中白质通路的估计：（i）纳入不对称纤维取向模型以解析亚体素纤维构型，如扇形和弯曲；（ii）明确建模发育中大脑典型扩散长度尺度的范围（或 ）。我们使用来自婴儿连接组项目（BCP）的体内婴儿扩散MRI数据对ASI进行了验证，表明尽管扩散模式发生变化，ASI仍可表征复杂的亚体素纤维构型并准确估计纤维取向分布函数。这进而在婴儿大脑中产生明显更好的扩散纤维束成像。
