Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/91814
標題: MR 影像之乳房組織分類使用限制能量最小化方法
Classification of Breast MR Imaging Using Constrained Energy Minimization
作者: Ci-You Zeng
曾啟侑
關鍵字: MRI
Constrained Energy Minimization
Kernel-based CEM
Iterative CEM
MRI
約束能量最小化 (Constrained Energy Minimization, CEM)
KCEM
ICEM
引用: [1] C.-I Chang, Hyperspectral Imaging: Techniques for Spectral Detection and Classification, Kluwer, 2003. [2] C.-I Chang, Hyperspectral Data Processing: Signal Processing Algorithm Design and Analysis Wiley, 2013. [3] Y. C. Ouyang, H.M. Chen, J.W. Chai, C.C.C. Chen, S.K. Poon, C.W. Yang, S.K. Lee and C.-I Chang, “Band expansion-based over-complete independent component analysis for magnetic resonance image analysis,” IEEE Trans. Biomedical Engineering, vol. 55, no. 6, pp. 1666-1677, June 2008 [4] J.-W. Chai, H.M. Chen, C.C.C. Chen, Y.C. Ouyang, S.K. Lee and C.-I Chang “Quantitative analysis of brain magnetic resonance images using support vector machine in conjunction with independent component analysis,” J. Magnetic Resonance Imaging, vol. 32, pp. 24-34, 2010. [DOI 10.1002/jmri.22210]. [5] BI-RADS Breast Imaging Reporting and Data System – Magnetic Resonance Imaging, First Edition, American College of Radiology, 2003. [6] C. K. Kuhl, R. K. Schmutzler, C.C. Leutner, A. Kempe, E. Wardelmann, A. Hocke, M. Maringa, U. Pfeifer, D. Krebs and H. H. Schild, 'Breast MR Imaging Screening in 192 Women Proved or Suspected to Be Carriers of a Breast Cancer Susceptibility Gene: Preliminary Results I' Radiology 2000; 215:267-279. [7] C. M. Wang, Sheng-Chi Yang, P.C. Chung, Y.N. Chung, C.C. Chen, Ching-Wen Yang, and Chia-Hsien Wen, “Orthogonal Subspace Projection-Base Approach to Classification of MR Image Sequences,” Computerized Medical Imaging and Graphics, December, 2001, vo1. 25, no. 6, pp. 465-476. [8] B. Schölkopf and A. J. Smola, Learning with Kernels. Cambridge, MA. MIT press, 2002. [9] D. D. Stark, and William G. Bradley., Magnetic Resonance Imaging, vol. 1, 3rd, Mosby Inc.,1999. [10] G. A. Wright, “Magnetic resonance image,” IEEE Signal Processing Mag., pp. 56– 66, Jan. 1997. [11] G. Sebastiani and P. Barone, “Mathematical principles of basic magnetic resonance image in medicine,” Signal Process, vol. 25, pp. 227–250, 1991. [12] A. H. Andersen, Z Zhang, MJ Avison, DM Gash, “Automated segmentation of multispectral brain MR images,” J. Neurosci Methods 2002; 122:13-23. [13] L. P. Clarke, R. P. Velthuizen, M. A. Camacho, et al. MRI segmentation: methods and applications. Magn. Reson. Imaging 1995; 13: 343-368. [14] H. Soltanian-Zadeh, D. J. Peck, J. P. Windham, T. Mikkelsen, “Brain tumor segmentation and characterization by pattern analysis of multispectral NMR images,” NMR Biomed 1998; 11: 201-208. [15] F. B. Mohamed, S. Vinitski, S. H. Faro, C. F. Gonzalez, J. Mack, T. Iwanaga. “Optimization of tissue segmentation of brain MR images based on multispectral 3D feature maps,” Magn. Reson. Imaging 1999; 17: 403-409. [16] C.-I Chang, Hyperspectral Imaging: Techniques for Spectral Detection and Classification, Kluwer, 2003. [17] C.-I Chang, Hyperspectral Data Processing: Signal Processing Algorithm Design and Analysis Wiley, 2013. [18] J. W. Chai, H.M. Chen, C.C.C. Chen, Y.C. Ouyang, S.K. Lee and C.-I Chang “Quantitative analysis of brain magnetic resonance images using support vector machine in conjunction with independent component analysis,” J. Magnetic Resonance Imaging, vol. 32, pp. 24-34, 2010. [DOI10.1002/jmri.22210]. [19] N. Otsu, “A threshold selection method from gray-level histogram,” IEEE Transactions on Systems Man Cybernet, SMC-8 pp. 62-66, 1978. [20] P. C. Chung, Senior Member, IEEE, Chuin-Mu Wang, Sheng-Chih Yang and Hsian-He Hsu “Tissues Classification for Breast MRI Contrast Enhancement Using Spectral Signature Detection Approach,” IEEE International Conference, Vol. 5, pp. 3917 - 3921, Oct. 2006. [21] J. C. Harsanyi, W. Farrand, and C.-I. Chang, 'Detection of subpixel spectral signatures in hyperspectral image sequences,' in Proc. American Society of Photogrammetry & Remote Sensing Annu. Meeting, Reno, 1994, pp. 236-247. [22] J. C. Harsanyi, “Detection and classification of subpixel spectral signatures in hyperspectral image sequences,” Ph.D. dissertation, Dept, Elect. Eng., Univ. Maryland Baltimore County, Baltimore, MD, 1993. [23] BI-RADS Breast Imaging Reporting and Data System – Magnetic Resonance Imaging, First Edition, American College of Radiology, 2003. [24] C. K. Kuhl, R.K. Schmutzler, C.C. Leutner, A. Kempe, E. Wardelmann, A. Hocke, M. Maringa, U. Pfeifer, D. Krebs and H. H. Schild, 'Breast MR Imaging Screening in 192 Women Proved or Suspected to Be Carriers of a Breast Cancer Susceptibility Gene: Preliminary Results I' Radiology 2000; 215:267-279. [25] S. Heywang-Kobrunner and R. Beck, 'Contrast-enhanced MRI of the breast', 2nd edition, Springer-Verlag, 1996. [26] C.-I Chang, Hyperspectral Data Processing: Algorithm Design and Analysis. New York: Wiley, 2012. [27] G. Camps-Valls and L. Bruzzone, “Kernel-based methods for hyperspcetral image classifucation,” IEEE Trans. Geoscience and Remote Sensing, vol. 43, no. 6, pp.1351-1362, June 2005. [28] H. Kwon and N.M. Nasrabadi, “Kernel orthogonal subspace projection for hyperspectral signal classification,” IEEE Trans. on Geoscience and Remote Sensing, vol. 43, no. 12, pp. 2952-2962, Dec. 2005. [29] M. Girolami, 'Mercer Kernel-based clustering in feature space,' IEEE Trans.Neural Networks, vol. 13, no.3, pp. 780-784, 2002. [30] M. Huang, W. Yu, D. Zhu, “An Improved Image Segmentation Algorithm Based on the Otsu Method,” IEEE 2012 13th ACIS International Conference on, pp. 135 - 139 [31] M. W. Vannier, R. L. Butterfield, D. Jordan, W. A. Murphy, R. G. Levitt, and M.Gado, “Multispectral analysis of magnetic resonance images,” Radiology, vol. 154, no. 1, pp. 221–224, 1985. [32] M. W. Vannier, T. K. Pilgram, C. M. Speidel, L. R. Neumann, D. L. Rickman, and L. D. Schertz, “Validation of magnetic resonance imaging (MRI) multispectral tissue classification,” Computerized Medical Imaging and Graphics, vol. 15, no. 4, pp. 217–223, 1991. [33] T. Taxt and A. Lundervold, “Multispectral analysis of the brain using magnetic resonance imaging,” IEEE Transactions on Medical Imaging, vol. 13, no. 3, pp. 470–481, 1994.
摘要: 本文的目標是藉由使用磁共振成像技術來制定了一套自動的方式,它可 以計算乳房密度和自動檢測乳腺腫瘤。由於對某些高風險人群對來說,乳房 X 光檢查在高乳腺密度案例上,其用途有限,所以我們的目的是利用磁共振 影像來測量乳腺密度,進而提升患乳腺癌風險評估的準確性。基於內核的分 類一般採用非線性內核來解決線性不可分的問題。我們提出了一種光譜特徵 檢測技術,約束能量最小化(CEM),它可以成功地顯示高對比度圖像的分 類結果。我們增強的圖像後,使用閾值技術則乳房 MRI 可成功地分為三大 組織(脂肪組織,腺體組織和腫瘤),的目的是為了計算乳房密度和腫瘤密 度。 基於內核的方法最近已吸入相當的興趣在高光譜圖像分析,因為它在透過一 個非線性映射函數來擴展特徵至更高維空間中的能力。我們研究了一種基於 內核的 CEM,稱為內核 CEM(K-CEM),它採用了不同的內核來將原來的 數據特徵擴大到高的維度空間上,使 CEM 可以被運算。實驗來進行 CEM 和 K-CEM 之間的比較分析和研究。 我們提出了一個新的類型 CEM,稱為迭代 CEM(ICEM)。將前一個數據樣 本藉由分類器分類之結果做為下一次迭代的訓練樣本,並送入同一分類器分 類。我們的想法是在這樣一種方式,樣本量不一定要大,而隨機抽樣的問題 也可以在反覆實施 CEM 中被解決。我們希望這項研究將針對不同分類的乳 腺組織和體積測量提供一個準確而有效的方法。最後,我們將在各種實驗中 比較各種方法的評價對於疾病的嚴重程度和預測。
This thesis target is to develop a set of automatic way by using MRI technique which can calculate breast density and automatically detect breast tumor. Our goal is to measure MR breast density to recover breast cancer risk assessment for certain high-risk populations for whom mammography is of limited usefulness due to high breast density. Kernel-based classifiers generally use non-linear kernels to solve linear non-separable problems. We present a spectral signature detection technology, constrained energy minimization (CEM), which could successfully show the classified results in high contrast images. After the images were enhanced, we then applied thresholding techniques hence can successfully classify breast MRIs into three major tissues (fatty tissue, glandular tissue and tumor). The breast density and tumor density can be calculated correctly. We have investigated a kernel-based CEM, called Kernel CEM (K-CEM) which employs various kernels to expand the original data space into a higher dimensional feature space that CEM can be operated on. Finally, we present an iterative CEM (ICEM) where data samples classified by a classifier in a previous iteration are fed into the same classifier as training samples for classification in the next iteration. The idea is to implement a CEM algorithm iteratively in such a way that the sample size is not necessarily to be large while the random sampling issue can be also resolved. This study presents an efficient method for classification and volume measurement of different breast tissue.
URI: http://hdl.handle.net/11455/91814
文章公開時間: 2015-07-16
Appears in Collections:通訊工程研究所

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.