磁振造影(Magnetic Resonance Imaging, MRI)透過非侵入的方式得到人體內器官的影像；氫質子磁振頻譜(Proton MR Spectroscopy)，利用共振的原理，收集氫質子的訊號，再轉成頻譜。近年來，醫生從磁振頻譜中得到病人腦內代謝物的資訊，以便觀察病理的變化，像是觀察腦膿瘍病人的代謝物，一直是臨床上診斷和治療的重要過程。在臨床上，不同回訊時間(Echo Time, TE)的頻譜所提供的資訊能夠增進醫生在診斷上的準確性。

本論文是使用獨立成分分析法(ICA)分析磁振頻譜，分析後得到的獨立成分代表組成這些輸入資料的成分，接著透過呂思穎學長使用主成分分析法(PCA)分析磁振頻譜的論文中提到的投影，用來協助觀察獨立成分和病人頻譜的關係。我們也討論ICA和PCA用在腦膿瘍病患頻譜的結果，以及實驗後發現的問題做討論並找出可能的原因，像是輸入資料前的比例正規化是否一定要執行、比例正規化的結果不如預期、有獨立成分的peak處在模糊位置混淆判斷等等。

Magnetic Resonance Imaging (MRI) can obtain the tissues of in vivo non-invasively. Proton MR Spectroscopy uses the resonance principle to collect the signals of proton and transforms them to spectrums. It provides information of metabolites in patient’s brain for doctors to observe the change of pathology. Observing the metabolites of brain abscess patients is most important process in clinical diagnosis and treatment. Then, doctors use different spectrums of echo time (TE) to enhance the accuracy in the diagnosis.

In our study, we use independent component analysis (ICA) to analyze MR spectroscopy. After analyzing, the independent components represent the elements which compose the input data. Then, we use the projection which is mentioned by Ssu-Ying Lu’s Thesis to help us observe the relationship between independent components and spectrums of patients. We also discuss the result of spectrums with using ICA and PCA and discover some questions (whether it need to do scale normalization before inputting data or not, the result of scale normalization doesn’t expect, and the peak in some independent components confuse us by locating in indistinct place) to discuss and to find possible reason after experiments.

致謝 i
中文摘要 ii
Abstract iii
目錄 iv
圖目錄 vi
表目錄 viii
第一章 緒論 1
1.1. 前言 1
1.2. 研究動機 4
1.3. 論文架構 5
第二章 研究方法與原理 6
2.1. 資料擷取 6
2.1.1. 正交線圈 6
2.1.2. PRESS (Point Resolved Spectroscopy)定位 6
2.1.3. MRS檔案格式 7
2.2. 渦電流校正和比例正規化 7
2.3. 獨立成分分析法 9
2.3.1. 定理 9
2.3.2. 非高斯性的量測 10
2.3.3. ICA的前處理 11
2.3.4. FastICA 演算法 12
2.3.5. ICA的模糊點 12
2.4. 投影 13
2.5. 資料收集 14
2.5.1. TE35 實驗資料 14
2.5.2. TE136 實驗資料 14
2.5.3. TE35+TE136合併 實驗資料 14
2.6. 實驗方法 15
第三章 結果與討論 23
3.1. 結果 23
3.1.1. TE35實驗結果 23
3.1.2. TE136實驗結果 25
3.1.3. TE35+TE136合併實驗結果 27
3.2. 討論 29
第四章 結論 66
第五章 參考文獻 67

[1] Magnetic Resonance - Technology Information Portal. Available: http://www.mr-tip.com
[2] B. R. Smith, "Magnetic resonance microscopy in cardiac development," Microsc Res Tech, vol. 52, pp. 323-30, Feb 1 2001.
[3] R. J. van Geuns, P. A. Wielopolski, H. G. de Bruin, B. J. Rensing, P. M. van Ooijen, M. Hulshoff, M. Oudkerk, and P. J. de Feyter, "Basic principles of magnetic resonance imaging," Prog Cardiovasc Dis, vol. 42, pp. 149-56, Sep-Oct 1999.
[4] H. Akutsu, A. Matsumura, T. Isobe, I. Anno, S. Takano, Y. Itai, and T. Nose, "Chronological change of brain abscess in (1)H magnetic resonance spectroscopy," Neuroradiology, vol. 44, pp. 574-8, Jul 2002.
[5] P. H. Lai, K. T. Li, S. S. Hsu, C. C. Hsiao, C. W. Yip, S. Ding, L. R. Yeh, and H. B. Pan, "Pyogenic brain abscess: findings from in vivo 1.5-T and 11.7-T in vitro proton MR spectroscopy," AJNR Am J Neuroradiol, vol. 26, pp. 279-88, Feb 2005.
[6] 黃佳民, "使用GAVA針對LCModel的Basis對活體氫質子磁振頻譜之定量研究," 2010.
[7] J. Hao, X. Zou, M. P. Wilson, N. P. Davies, Y. Sun, A. C. Peet, and T. N. Arvanitis, "A comparative study of feature extraction and blind source separation of independent component analysis (ICA) on childhood brain tumour 1H magnetic resonance spectra," NMR Biomed, vol. 22, pp. 809-18, Oct 2009.
[8] S. W. Provencher, "Automatic quantitation of localized in vivo 1H spectra with LCModel," NMR Biomed, vol. 14, pp. 260-4, Jun 2001.
[9] S. LI, "A tutorial on Principal Components Analysis," 2002.
[10] R. Stoyanova and T. R. Brown, "NMR spectral quantitation by principal component analysis," NMR Biomed, vol. 14, pp. 271-7, Jun 2001.
[11] S. Y. Lu, "The Classification of In Vivo Proton Magnetic Resonance Spectroscopy of Brain Abscesses Using Principal Component Analysis " 2011.
[12] J. Ma and Z. Sun, "MRS classification based on independent component analysis and support vector machines," Proceedings of the 5th International Conference on Hybrid Intelligent Systems, pp. 509–511, 2005.
[13] A. Hyvarinen, J. Karhunen, and E. Oja, "Independent Component Analysis," 2001.
[14] A. Hyvarinen and E. Oja, "Independent component analysis: algorithms and applications," Neural Netw, vol. 13, pp. 411-30, May-Jun 2000.
[15] U. Klose, "In vivo proton spectroscopy in presence of eddy currents," Magn Reson Med, vol. 14, pp. 26-30, Apr 1990.
[16] Y. W. Chen and D. Sugiki, "Segmentation of MR images using independent component analysis," Knowledge-Based Intelligent Information and Engineering Systems, Pt 2, Proceedings, vol. 4252, pp. 63-69, 2006.
[17] Z. J. Yuan and E. Oja, "A FastICA algorithm for non-negative independent component analysis," Independent Component Analysis and Blind Signal Separation, vol. 3195, pp. 1-8, 2004.
[18] A. Hyvarinen and E. Oja, "A fast fixed-point algorithm for independent component analysis," Neural Computation, vol. 9, pp. 1483-1492, Oct 1 1997.
[19] (2000). FastICA. Available: http://www.cis.hut.fi/projects/ica/fastica/