基于扩散形状的DTI纤维跟踪算法研究

到目前为止，研究者们已经提出了许多种神经纤维束的三维可视化技术，其中，基于扩散跟踪的白质束成像技术在白质纤维束的可视化和分析中使用得最多。本文针对现有纤维跟踪算法存在的问题，提出了基于扩散形状的纤维跟踪算法，该算法结合了流线跟踪（streamline tracking，STT）法与张量弯曲（tensor deflection，TEND）法的优点，对不同的扩散形状采用不同的跟踪方向，尤其在平面扩散的情况提出了更接近纤维走行的跟踪方法，减少了跟踪方向与实际纤维走行的误差，能更完整、更准确地显示大脑白质的纤维走行。

A Study on Fiber Tracking Algorithm Based on the Diffusion Shape of DTI

Nowadays, various techniques of three-dimensional visualization with nerve fibers have been mentioned, and the technique of tractography is the most popular method in visualizing and analyzing with white matter fibers. According to the limitation of the traditional algorithms, this dissertation presents a new fiber tracking algorithm based on the diffusion shape. Absorbing the advantages of Streamline Tracking (STT) and Tensor Deflection (TEND), this algorithm uses different tracking strategies on corresponding diffusion shapes. It results in a better tracking pathway which is closer to real fiber in case of planar diffusion and visualizes the fibers of the brain white matter more accurately.