陳中明2006-07-262018-06-292006-07-262018-06-292003-07-31http://ntur.lib.ntu.edu.tw//handle/246246/22368神經纖維路徑是探索大腦結構與功能 之間的關係以及臨床診斷的重要資訊。傳 統上，此一資訊往往需藉由侵入性的方式 而達到。而擴散張量磁振造影(DTI)是目前 以非侵入式的方法觀察活體內大腦白質神 經纖維方向的最主要技術。然而，由於雜 訊與DTI 無法分辨交叉神經纖維的特性， 過去基於區域性最佳化的神經纖維追蹤演 算法，皆無法有效的在DTI 影像中找出合 理的神經纖維路徑。在此兩年期計畫的第 一年中，我們提出一個全域性的神經纖維 追蹤的演算法，此法不但有很好的雜訊免 疫力，也能進一步融入細胞大小的資訊。The nerve fiber trace may serve as crucial information for exploiting the relationship between the brain structure and the brain function and for the clinical diagnosis. Conventionally, acquiring this information usually needs to rely on invasive means. As contrast, the diffusion tensor magnetic resonance imaging (DTI) is nowadays the most prevailing approach to observe in vivo nerve fibers in the white matter noninvasively. Nevertheless, because of the noise and the inherent characteristics of the DTI that cannot differentiate crossover nerve fibers, those previous fiber tracking algorithms based on the ideas of local optimization all fail to find the reasonable nerve fiber traces from the DTI images. In the first year of the two-year project, we have developed a global fiber tracking algorithm, which not only has an excellent noise-immune ability, but also can further include the cell-size information. Keywords: Fiber Tracking, Diffusion Tensor Magnetic Resonance Imaging, Q-space, Cell Size of Nerve Fiber, Global Optimizationapplication/pdf93651 bytesapplication/pdfzh-TW國立臺灣大學醫學工程學研究所神經纖維追蹤擴散張量磁振造 影Q-空間神經纖維細胞大小全域性 最佳化reporthttp://ntur.lib.ntu.edu.tw/bitstream/246246/22368/1/912213E002081.pdf