扩散张量MR成像研究

Research on diffusion tensor imaging of MR

本文一方面综述了8年来国际上扩散张量磁共振成像(DTI)的研究进展情况。包括总结了到目前为止所使用的4种提高DTI成像精度的方法,并指出精度的提高依赖于成像中脉冲序列的优化、实验方法的完善和后处理算法两个方面。文中比较了目前研究扩散张量的两种基本模型:扩散张量模型和q-空间模型。指出这两种模型侧重应用于不同的场合。前者侧重于研究器官或宏观组织中的扩散,而后者侧重于研究小到细胞尺度(μm量级)的扩散行为。两者在应用研究方面是互补的,所要求的实验条件是有差别的。另一方面结合文献对扩散张量模型的实验条件进行了理论分析,认为b因子取得过高并不合理。并用DTI实验数据和结果进行了初步验证。为了改进扩散张量模型本文探讨了考虑多指数衰减的设想。文中综述了DTI的导出量和一些实验结果,在此基础上分析了设计和优化各向异性指标(DAI)的原则。对于DTI的重要导出结果神经纤维束成像(fibertractography)重点分析了造成其成像精度不高的主要因素,指出了改进纤维束传导方向甄选算法和寻求纤维束方向场的几何性质是两种可能的解决办法。

A review of the research on diffusion tensor imaging of MRI was undertaken in this paper. The four main schemes to improve the accuracy of the outcome was reviewed. A comparison was taken between the two methods of diffusion research: q-space-DWI and DTI. The two methods were applied to complementary research aspects?Analysis and experiments of the DTI model showed that high b value was not proper for DTI experiments- then the limitation and probable improvement schemes of DT model was discussed. We reviewed some of the rotationally invariant scalar quantities derived from diffusion tensor- and gave some FA and MD results derived from our own experiment. We also discussed fiber tractography and fiber tract following technique- two obstacles and possible resolutions were presented.