Centre for Molecular Biology and Neuroscience, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway.
PLoS One. 2010 Jan 7;5(1):e8595. doi: 10.1371/journal.pone.0008595.
Diffusion MRI (dMRI) is widely used to measure microstructural features of brain white matter, but commonly used dMRI measures have limited capacity to resolve the orientation structure of complex fiber architectures. While several promising new approaches have been proposed, direct quantitative validation of these methods against relevant histological architectures remains missing. In this study, we quantitatively compare neuronal fiber orientation distributions (FODs) derived from ex vivo dMRI data against histological measurements of rat brain myeloarchitecture using manual recordings of individual myelin stained fiber orientations. We show that accurate FOD estimates can be obtained from dMRI data, even in regions with complex architectures of crossing fibers with an intrinsic orientation error of approximately 5-6 degrees in these regions. The reported findings have implications for both clinical and research studies based on dMRI FOD measures, and provide an important biological benchmark for improved FOD reconstruction and fiber tracking methods.
弥散磁共振成像(dMRI)广泛用于测量脑白质的微观结构特征,但常用的 dMRI 测量方法在解析复杂纤维结构的方向结构方面能力有限。虽然已经提出了几种有前途的新方法,但这些方法与相关组织学结构的直接定量验证仍然缺失。在这项研究中,我们使用个体髓鞘染色纤维方向的手动记录,定量比较了离体 dMRI 数据得出的神经元纤维方向分布(FOD)与大鼠脑髓鞘结构的组织学测量结果。我们表明,即使在具有交叉纤维复杂结构的区域,也可以从 dMRI 数据中获得准确的 FOD 估计,这些区域的固有方向误差约为 5-6 度。研究结果对基于 dMRI FOD 测量的临床和研究都具有重要意义,并为改进的 FOD 重建和纤维追踪方法提供了重要的生物学基准。