為了改善雙重磁化準備快速梯度迴訊(MP2RAGE)成像技術受到反向隔熱脈衝激發效率的影響導致T1弛緩參數量測誤差的問題，三重磁化準備快速梯度迴訊(MP3RAGE)在2012年被提出。此技術利用一次成像過程中收取三組不同反轉時間的高解析度原始影像，並藉由對於非線性方程組求解，實際算出T1弛緩參數與反向脈衝效率。

本研究裡，對MP3RAGE序列提出新的影像合併方法，並推導出其雜訊模型，以針對人腦之對比雜訊比(CNR)進行最大化，找出脈衝序列的最佳反轉時間及偏折角等掃描參數。此外，也針對不同腦脊髓液T1弛緩參數、反向脈衝激發效率的設定、切面數(切面厚度)、相位編碼方式進行比較與討論，也將最佳化的結果與MP2RAGE在1.5T磁場進行比較。

實驗結果得知，相較於單一的反向脈衝激發效率，不同組織分別使用各自的反向脈衝激發效率模擬所得的參數，能在人體實驗中獲得最佳的CNR；此外，腦脊髓液T1弛緩參數對於影像CNR的影響有限；當切面數目越多時，不管在模擬或實驗的雜訊都隨之提高，CNR便顯得較差但擁有較佳的空間解析度；MP3RAGE中央相位編碼的CNR會優於線性相位編碼；最後，使用中央相位編碼的MP2RAGE之CNR會略優於MP3RAGE，但差距並不顯著。

A novel sequence, magnetization prepared 3 rapid acquisition gradient echoes (MP3RAGE), were proposed by our group earlier in 2012. Three high-resolution volumes with distinct inversion times were collected by this sequence to perform accurate T1 mapping as well as estimation of inversion efficiency.

In this study, a novel reconstruction of MP3RAGE data was presented to obtain unbiased T1WI without interference of inhomogeneity, spin density, and T2* effect. Error propagation was conducted to estimate noise level. Therefore, optimization of the contrast-to-noise ratio (CNR) of whole brain can be achieved. In addition, investigation on inversion efficiency, the T1 relaxation time of CSF, the slice number (or slice thickness), and the order of phase encoding was made through numerical simulation and in vivo experiments at 1.5 Tesla.

In conclusion, a tissue-dependant inversion efficiency can obtain better CNR than universal efficiency from 0.75 to 0.95. For the T1 relaxation time of CSF widely ranging from 2300 to 4500 ms, only minor difference was found. In addition, a larger slice number could lead to lower CNR mainly due to higher noise level with higher spatial resolution. It was also found that centric phase encoding can offer better contrast than linear phase encoding for MP3RAGE, which is in agreement with those reported by MPRAGE and MP2RAGE. At last, compared with MP3RAGE, MP2RAGE with centric phase encoding has slightly better CNR, which can not be visually identified.

審定書 i
致謝 ii
摘要 iii
Abstract iv
目錄 v
圖目錄 vii
表目錄 viii
第一章 簡介 1
第一節 背景 1
第二節 研究動機與目的 4
第二章 多重磁化準備快速梯度迴訊:脈衝序列與影像重建 5
第一節 雙重磁化準備快速梯度迴訊:MP2RAGE 5
第二節 三重磁化準備快速梯度迴訊:MP3RAGE 9
第三節 雜訊模型推導 13
第四節 CNR 定義 15
第三章 CNR 最佳化模擬 17
第一節 CNR 最佳化模擬參數定義與驗證 17
第二節 實驗參數模擬與討論 26
第四章 人體實驗 34
第一節 實驗設計34
第二節 人體實驗35
第三節 實驗討論50
第五章 結論 53
參考文獻 55

[1] P. F. Van de Moortele, E. J. Auerbach, C. Olman, E. Yacoub, K. Ugurbil, and S. Moeller, "T1 weighted brain images at 7 Tesla unbiased for Proton Density, T2* contrast and RF coil receive B1 sensitivity with simultaneous vessel visualization," Neuroimage, vol. 46, pp. 432-46, Jun 2009.
[2] J. P. Marques, T. Kober, G. Krueger, W. van der Zwaag, P. F. Van de Moortele, and R. Gruetter, "MP2RAGE, a self bias-field corrected sequence for improved segmentation and T1-mapping at high field," Neuroimage, vol. 49, pp. 1271-81, Jan 15 2010.
[3] M. Tanner, G. Gambarota, T. Kober, G. Krueger, D. Erritzoe, J. P. Marques, et al., "Fluid and white matter suppression with the MP2RAGE sequence," J Magn Reson Imaging, vol. 35, pp. 1063-70, May 2012.
[4] T. Kober, C. Granziera, D. Ribes, P. Browaeys, M. Schluep, R. Meuli, et al., "MP2RAGE multiple sclerosis magnetic resonance imaging at 3 T," Invest Radiol, vol. 47, pp. 346-52, Jun 2012.
[5] P.-T. Chen, "3D High Resolution T1 Mapping of Human Brain," September 2012
[6] D. G. Norris, H. Ludemann, and D. Leibfritz, "An Analysis of the Effects of Short T2 Values on the Hyperbolic-Secant Pulse," Journal of Magnetic Resonance, vol. 92, pp. 94-101, Mar 1991.
[7] D. C. Zhu and R. D. Penn, "Full-brain T1 mapping through inversion recovery fast spin echo imaging with time-efficient slice ordering," Magn Reson Med, vol. 54, pp. 725-31, Sep 2005.
[8] R. Deichmann, C. D. Good, O. Josephs, J. Ashburner, and R. Turner, "Optimization of 3-D MP-RAGE sequences for structural brain imaging," Neuroimage, vol. 12, pp. 112-27, Jul 2000.
[9] H. H. Ku, "Notes on the use of propagation of error formulas," Journal of Research of the National Bureau of Standards. Section C: Engineering and Instrumentation, vol. 70C, p. 263, Oct 1966
[10] W.-F. Chen, "Voxel-based Cortical Thickness Measurement of Human Brain Using Magnetic Resonance Imaging," January 2008
[11] R.-H. Chung, "Stability analysis of voxel-based cortical thickness measurement of human brain," September 2012.
[12] S.-J. Wu, "Optimization of contrast and signal homogeneity for high resolution 3D MRI of human brain at 1.5 Tesla," Aug 2011.
[13] P.-T. C. Wei-Feng Hung, Tzu-Chao Chuang, Hing-Chiu Chang,Ming-Ting Wu, "High Resolution Volumetric T1 Mapping Using a Novel MP3RAGE Method," ISMRM, April 2013.
[14] W. D. Rooney, G. Johnson, X. Li, E. R. Cohen, S. G. Kim, K. Ugurbil, et al., "Magnetic field and tissue dependencies of human brain longitudinal 1H2O relaxation in vivo," Magn Reson Med, vol. 57, pp. 308-18, Feb 2007.