弥散频谱成像的研究进展

Research progress of diffusion spectrum imaging

研究表明人类大脑中的很多区域内存在着多方向的纤维束,这类区域的体素内水分子的平均弥散并不符合高斯分布。因此,基于单方向纤维束模型的弥散张量成像在揭示组织微观结构中存在先天缺陷。弥散频谱成像使用多b值多方向的扫描序列,采集水分子在整个q空间的弥散信息,通过一个具有高角分辨率的概率密度函数来描述水分子的弥散运动,能够可靠地观测单体素内多方向的纤维束,并据此进行纤维束追踪重建出真实而复杂的组织结构。作者全面介绍了近年来弥散频谱成像在基本原理、方法学和应用方面的研究进展,为推动国内弥散成像技术在科研和临床上的发展提供帮助。

Scientific studies showed that many areas in human brains contain fiber tracts running along multiple different directions, and the general average diffusion of water molecules within the image voxles of such areas does not statistically follow Gaussian distribution. Therefore, the model of diffusion tensor that hypothesizes a single fiber orientation is intrinsically defective in revealing the microstructures of brain tissues. Diffusion spectrum imaging（DSI） employs pulse sequences using multiple b-values along multiple gradient directions, sampling diffusion information of water molecules in the entire q-space, and quantitatively estimates water diffusion using a probability density function of high angular resolution diffusion imaging. DSI thus can reliably observe fibers running along multiple directions within any voxel in the image, which subsequently allows fiber tracking along different directions, leading to improved reconstruction reflecting true but complicated tissue structures in reality. This article aims to conduct a comprehensive survey of literatures published in recent years in the domain of DSI research, providing a summary on its fundamental principles, methodology, applications and important progresses, which may hopefully help promote the development of DSI in China in both research and clinical studies.