CEA DRF/ISVFJ/Neurospin/UNIRS, Gif-sur-Yvette, France; Université Paris-Saclay, Orsay, France.
Research Centre Juelich, Institute of Neuroscience and Medicine, Juelich, Germany.
Neuroimage. 2019 Jun;193:10-24. doi: 10.1016/j.neuroimage.2019.02.055. Epub 2019 Mar 5.
A GPU-based tool to generate realistic phantoms of the brain microstructure is presented. Using a spherical meshing technique which decomposes each microstructural item into a set of overlapping spheres, the phantom construction is made very fast while reliably avoiding the collisions between items in the scene. This novel method is applied to the construction of human brain white matter microstructural components, namely axonal fibers, oligodendrocytes and astrocytes. The algorithm reaches high values of packing density and angular dispersion for the axonal fibers, even in the case of multiple white matter fiber populations and enables the construction of complex biomimicking geometries including myelinated axons, beaded axons, and glial cells. The method can be readily adapted to model gray matter microstructure.
本文提出了一种基于 GPU 的工具,用于生成逼真的脑微观结构模型。该工具使用球形网格技术,将每个微观结构项分解为一组重叠的球体,从而快速构建模型,同时可靠地避免了场景中各物体之间的碰撞。该新颖方法应用于构建人类大脑白质微观结构组件,即轴突纤维、少突胶质细胞和星形胶质细胞。该算法实现了高的轴突纤维堆积密度和角向分散度,即使在存在多种白质纤维群体的情况下,也能够构建包括髓鞘轴突、珠状轴突和神经胶质细胞在内的复杂仿生几何形状。该方法可以很容易地适应于模拟灰质微观结构。