- 學位論文
利用相似度決定白質神經纖維束成像之最佳擴散張量梯度數目
Using Coherence Index to Determine the Optimum Diffusion Tensor Encoding Steps for White Matter Tractography
摘要
近年來磁振造影(Diffusion MRI)技術的成熟發展，帶領人類揭開在神經科學研究中相當古老且深遠的議題，而人類大腦的神秘面紗終得以慢慢地被解開。其中擴散張量磁振造影(DTI)，因以非侵入方式之醫學臨床診斷造影，探測大腦內白質神經纖維結構之神經傳遞，利用此技術所提供之擴散主要方向，能夠連續追蹤水份子傳遞方向，而實際預測可能神經纖維方向，再利用Tractography技術重建神經纖維路徑，此話題是目前受到相當重視之熱門課題之一。 目前國際上著名研究腦功能科學發展單位，陸續發表Tractography演算法相關論文及發展應用工具，無非希望帶給人類較不同以往對大腦複雜神經結構一種理性客觀認識與判定。於是，亦提出Tractography區域性顯微結構演算法，為Selective Distance-Weighted Interpolation(SDWI)，主要概念基於高相似度、距離為權重內插方法而來，以選擇周圍最佳能量場，當做其前進之最有效方向，成功找到大腦內主要之白質神經纖維束圖像，更將以(2D phantom)成束管狀假體，驗証其顯微結構追踨演算法之可信度。 在廣泛Tractography技術應用在腦神經科學發展與臨床疾病診斷治療上，佔有與日俱增之極俱重要角色，所以受到相當大的關注與相關應用議題之深入解析資訊，於是我們提出一個嶄新量化公式，Tractography相似度指標(coherence index)，說明最佳(MR-DTI)擴散脈衝梯度數目與Tractography追踨纖維結構走向正確度之相關性，對於現今傳統Tractography技術未被證實發展研究中，提供一項舉足輕重之臨床最佳實驗參考數據。
並列摘要
Recently, the technique of Diffusion Magnetic Resonance Imaging (Diffusion MRI) has developed to make progress in explaining perpetual controversy about neuroscience in human brain. Diffusion Tensor magnetic resonance Imaging (DTI) has been recognized as an important tool to reveal the axonal fiber tracts in cerebral white matter noninvasively in the clinical diagnosis image. By probing the translational displacement of water molecules, it provides the primary direction of water molecular diffusion which is correlated to the main pathway of fiber bundles. The eigenvector of the diffusion tensor at each location in the cerebral white matter represents the fiber orientation at the same location. Based on this information, 3D reconstruction and visualization of white matter fiber pathways can be produced by tractography. The last, MR-DTI has achieved a reasonable cognition and judgement, making a difference from the past acquisition strategies for complicated structure of white matter bundles. International eminent functional brain science research center has investigated related lectures for tractography algorithm and available tools, and we also proposed a novel fiber tracking algorithm. It extracted salient tensor feature using a local regularization theory that represented SDWI method. By using a phantom image made up of PE fibers with known fiber pathways to validate stability of this algorithm. Extensive attention and profound analysis has been given to recover identifiable anatomical structures that correspond to fiber tracjectories, and the position of brain lesions in vast Tractography innovation has played more important role. We has derived from Tractography quantitative index a formula that determines the correlation between optimum diffusion tensor encoding steps and white matter tractography. Its dominant strategy has not been proven by the traditional Tractography research and related application in the clinical experiment parameter.
參考文獻
延伸閱讀
- Charoenkwan, P. (2014). 設計最佳化演算法預測蛋白質功能和辨認神經細胞影像 [doctoral dissertation, National Chiao Tung University]. Airiti Library. https://doi.org/10.6842/NCTU.2014.00450
- Chen, N. M. (2008). 以疊代演算法設計整型高斯光束為強度及相位均勻分佈之純相位繞射元件 [master's thesis, National Taiwan University]. Airiti Library. https://doi.org/10.6342/NTU.2008.03211
- Hsu, C. M., & Cheng, J. C. (2009). 利用類神經網路、望想函數及粒子群最佳化以達成印刷電路版設計最適化. 明新學報, 35(2), 95-114. https://doi.org/10.29688/MHJ.200908.0010
- Wang, S. (2009). Diffusion tensor MR imaging as a biomarker for the evaluation of whitematter injury in rodent models [doctoral dissertation, The University of Hong Kong]. Airiti Library. https://www.airitilibrary.com/Article/Detail?DocID=U0029-1812201200016276
- Su, H. T. (2017). Coordination of Cerebral White Matter Tracts in the Long-term Dance and Piano Training [master's thesis, National Taiwan University]. Airiti Library. https://doi.org/10.6342/NTU201702436