基于QBI成像技术构建脑结构连接网络

Construction of Brain Anatomical Connectivity Network Based on Q-ball Imaging

目的为构建脑结构连接网络,本文提出一种基于Q空间球面成像(Q-ball imaging,QBI)技术的扩展多向流线跟踪算法,并根据纤维束路径走向构建脑结构连接矩阵。方法首先,根据磁共振图像进行脑组织分割,之后与标准模板对比完成灰质功能区域划分和标记;然后,采用球面插补计算取向分布函数(orientation diffusion function,ODF),进而根据扩展多向流线算法进行路径跟踪。最终根据跟踪出的纤维束数量及其路径所处功能区来构建结构连接矩阵。结果以真实脑数据为对象进行跟踪,生成连接矩阵,并进行连接通路分析。结论所得结果符合脑生理结构实际情况,适合于构建结构连接网络,为研究脑连接提供了一种新途径。

Objective To construct the brain anatomic connectivity, a new tracking method based on Q-ball im- aging （QBI） was proposed in this paper, and then the linkage matrices were generated according to the results. Methods First, the brain tissues were segmented out from the MRI images, and then the functional areas of the grey matter were labeled according to the standard template. Second, the orientation diffusion function （ODF） was calculated using spherical interpolation, and the paths were traced using the method of multiply directional streamline. Then the linkage matrices were constructed according to the number of nerve fiber paths and their start-end points. Results This algorithm was applied to the real brain dataset, and the connectivity matrices were generated. Conclusion The method could be used to track the nerve fibers with high angle resolution and to resolve crossing and branching paths, the tracked results are more reasonable and suitable for the construc- tion of the anatomic linkages. This paper paves a new way for brain connectivity research.