磁振造影(Magnetic Resonance Imaging, MRI)是一個在臨床上常被拿來量測人體訊號的技術，它不僅擁有高解析度、成像快速、非侵入性等優點，由於它廣泛的應用範圍及發展性，以致於有更多的MRI應用技術被發研究人員所開發出來。肺部微灌流就是其中之一，根據指示劑稀釋定理可以從(AIF)及輸出資訊藉由解捲積的方式找到該組織的殘餘函數，最後從殘餘函數中可以定義出人體的生理參數，如血流、血量及傳遞時間。然而，解捲積的方法有很多種，其解出來的結果也因為方法的不同有不同的優缺點。本篇論文主要是探討使用奇異值分解(singular value decomposition, SVD)來解捲積，並透過加入雜訊分析的實驗，了解SVD在雜訊干擾的情況下是否一樣可以提供一個較穩定的結果。根據結果顯示，在六個正常人的生理參數分析加入不同雜訊的測試，每一個的結果都顯示出SVD在一個正常的條件及環境中，可以提供研究人員一個穩定的生理參數。

The technique of Magnetic Resonance Imaging (MRI) is used to estimate human signals in clinical. The MRI not only have advantages on high resolution, fast scanning and non-invasive, but also the huge range of application and potential progress. There are more application about MRI research and development by researcher, and one of the research is pulmonary perfusion. According to indicator dilution theory, the residue function of tissue can be estimated by deconvolution from artery input function (AIF) and output function. Then, the perfusion parameters can be defined from residue function such as blood flow, blood volume and Mean Transit Time (MTT). However, the methods of deconvolution are many different ways, and the results have different advantages and disadvantages in different methods. In this thesis, the main object is discussing with a method of deconvolution by singular value decomposition (SVD), and added an exam that about a Gaussian noise interfere with signals. The exam will test whether the SVD in an noise interference can prove a stable result or not. The results of analysis perfusion parameters which are added in different power of noise can show the SVD in a normal condition that provides a stable value of perfusion parameters for researcher.

1. Hiroto Hatabu, Eiji Tadamura, David L. Levin, Qun Chen, Wei Li, Ducksoo Kim, Pottumarthi V. Prasad, and Robert R. Edelman. Quantitative assessment of pulmonary perfusion with dynamic contrast-enhanced MRI. Magnetic Resonance in Medicine 1999; 42:1033-1038.

2. Anne M. Smith, Cécile B. Grandin, Thierry Duprez, Frédéric Mataigne, Guy Cosnard. Whole brain quantitative CBF and CBV measurements using MRI bolus tracking: Comparison of methodologies. Magnetic Resonance in Medicine 2000; 43:559-564.

3. Hidemasa Uematsu, Hiroki Yamada, Norihiro Sadato, Satoshi Muramoto, Hiroshi Inoue, Koji Hayashi, Yoshiharu Yonekura, Hirohiko Kimura, Hajime Sakuma, Tetsuya Matsuda, Nobushige Hayashi, Kazutaka Yamamoto, Yasushi Ishii. Assessment of hepatic portal perfusion using T2* measurements of Gd-DTPA. Journal of Magnetic Resonance Imaging e 1998; 8:650-654.

4. David L. Levin, Qun Chen, Ming Zhang, Robert R. Edelman, and Hiroto Hatabu. Evaluation of Regional pulmonary perfusion using ultrafast magnetic resonance imaging. Magnetic Resonance in Medicine 2001; 46:166-171.

5. Tiffany Ting-Fang Shin, Chih-Wei Yu, Chao-yu Hsu, Bang-Bin Chen. Dynamic Contrast Enhanced-Magnetic Resonance Imaging as a prognostic predictor of tumor response to anti-angiogenesis therapy for the treatment of hepatocellular carcinoma. J. Chinese Oncol. Soc. 24(5), 282-288.
6. Paul Meier and Kenneth L. Zierler. On the theory of the indicator-dilution method for measurement of blood flow and volume. J Appl Physiol 1954; 6:731-744.

7. Yoshiharu Ohno, Hiroto Hatabu, Kenya Murase, Takanori Higashino, Hideaki Kawamitsu, Hirokazu Watanabe, Daisuke Takenaka, Masahiko Fujii, Kazuro Sugimura. Quantitative assessment of regional pulmonary perfusion in the entire lung using three-dimensional ultrafast dynamic contrast-enhanced magnetic resonance imaging: Preliminary experience in 40 subjects. Journal of Magnetic Resonance Imaging 2004; 20:353-365.

8. Kenya Murase, Keiichi Kikuchi, Hitoshi Miki, Teruhiko Shimizu,, Junpei Ikezoe, Determination of arterial input function using fuzzy clustering for quantification of cerebral blood flow with dynamic susceptibility contrast-enhanced MR imaging. Journal of Magnetic Resonance Imaging 2001; 13:797-806.

9. Leon Axel. Cerebral blood flow determination by rapid-sequence computed tomography. Radiology 1980; 137:679-686.
10. Leif Qstergaard, Robert M. Weisskoff, David A. Chesler, Carsten Gyldensted, Bruce R. Rosen. High resolution measurement of cerebral blood flow using intravascular tracer bolus passages. Part I: Mathematical approach and statistical analysis. Magnetic Resonance in Medicine 1996; 36:715-725.

11. Thomas E. Conturo, Erbil Akbudak, Melanie S. Kotys, Maison L. Chen, Steve J. Chun, Raymond M. Hsu, Caitlin C. Sweeney, Joanne Markham. Arterial input functions for dynamic susceptibility contrast MRI: Requirements and signal options. Journal of Magnetic Resonance Imaging 2005; 22:697-703.

12. Jacek Ruminski. A Comparison of MTT Calculation Techniques in MRI Brain Perfusion Imaging. IFMBE Processding 2008; 22:509-512

13. Ona Wu, Leif Ostergaard, Robert M. Weisskoff, Thomas Benner, Bruce R. Rosen, and A. Gregory Sorensen. Tracer arrival timing-insensitive technique for estimating flow in MR perfusion-weighted imaging using singular value decomposition with a blook-circulant deconvolution matrix. Magnetic Resonance in Medicine 2003; 50:164-174.

14. Ho-Ling Liu, Yonglin Pu, Yijun Liu, Lisa Nickerson, Trevor Andrews, Peter T. Fox, Jia-Hong Gao. Cerebral blood flow measurement by dynamic contrast MRI using singular value decomposition with an adaptive threshold. Magnetic Resonance in Medicine 1999;42:167–172.

15. Frank Risse, Tristan A. Kuder, Hans-Ulrich Kauczor, Wolfhard Semmler, Christian Fink. Suppression of pulmonary vasculature in lung perfusion MRI using correlation analysis. European Society of Radiology 2009; DOI: 10.1007/s00330 -009-1464-9