本論文提出一基於平均位移的方法能有效率地分割出腹主動脈瘤，使用者只需要輸入一組病人斷層掃描影像，在起始位置圈選動脈壁（管腔）內一小範圍（亦即血液），最後在終點位置點擊一下，系統將會根據初始圈選的範圍做腹主動脈的追踪。與商業軟體的腹主動脈瘤的分割需要逐張圈選想要的範圍相較，本論文可以節省許多人力避免因為疲勞而產生疏忽。與其它針對腹主動脈瘤分割的研究，如果使用者輸入的是腹部的斷層掃描的話，直接從第一張就可以開始進行本計劃的演算法；若是全身的斷層掃描也只需從橫隔模以下就可開始進行，不需要人為事先挑選有腹主動脈瘤的CT切片才能開始，使用上方便了許多。
實驗結果顯示，針對腹主動脈瘤的影像切割結果都令人滿意。

In this paper we propose a mean-shift based technique for a precise tracking and segmentation of abdominal aortic aneurysm (AAA) from computed tomography (CT) angiography images. Output data from the proposed method can be used for measurement of aortic shape and dimensions. Knowledge of aortic shape and size is very important for selection of appropriate stent graft device for treatment of AAA. Comparing to conventional approaches, our method is very efficient and can save a lot of manual labors.

目錄
第一章 緒論                       	1
第二章 相關文獻回顧                 	5
第三章 相關工作	                        9
第四章 腹主動脈影像追踪及腹主動脈瘤影像切割	13
  4.1 設置起始切片和終點切片	        15
  4.2 影像二值化	                        16
  4.2-1：二值化腔內影像Pb,s	        16
  4.2-2：二值化上皮組織影像Pt,s	        19
  4.3 腹主動脈追踪	                22
  4.4 腔內輪廓切割及上皮輪廓切割	        26
  4.4-1：射線邊緣偵測	                26
  4.4-2：射線長度分析	                28
  4.5 切片間上皮輪廓平滑化	                33
第五章 實驗結果與討論	                35
第六章 結論與未來研究方向	                42
參考文獻	                                44
附錄：英文論文	                        47

圖目錄
圖1.為主動脈圖，A所顯現的是正常主動脈；B則是胸主動脈瘤(TAA)；C為腹主動脈瘤(AAA)	2
圖2.為破裂的腹主動脈瘤，空心的箭頭所指處即為動脈瘤，而實心箭頭所指的則是血液流到腹部 (by James Heilman, MD)[3]	3
圖3.腹主動脈瘤與周圍器官；紅色箭頭為腔內；黃色箭頭為上皮組織；綠色箭頭為周圍器官組織	5
圖4.DICOM的Data element格式	10
圖5.CT儀器[13]	12
圖6.系統流程	14
圖7.設置起始切片；中間的紅色框為使用者圈選；外圍的藍色框為之後追踪用。	16
圖8.腔內區域初步二值化圖；α=3	17
圖9.形態學x與x1-x8	18
圖10.圖7經過二次擴張再二次侵蝕的腔內區域結果	19
圖11.各材質其Hounsfield Unit值域	20
圖12.導致支架影像原始值下降示意圖；黑色線為腔內輪廓；紅色虛線為支架；綠色線為上皮輪廓	20
圖13.左圖為原圖（理想的上皮組織區域為兩個綠色曲線間的部位）；右圖為本文上皮二值化的結果圖	22
圖14.Mean Shift演算法概念圖	23
圖15.高斯權重圖；σ=2	24
圖16.高斯Mean Shift執行結果	26
圖17.放射式邊緣偵測演算法概念圖	27
圖18.射線邊緣偵測(a)初步結果圖；(b)中間值濾波後的結果	28
圖19.圖17(b)射線長度展開圖；紅色是代表該射線的長度而綠色則為長度的一階導函數	29
圖20.紅色圈代表會執行線性內插	32
圖21.圖17經過一階導函數平滑後	32
圖22.圖20射線長度展開圖	33
圖23.某切片(b)與其上下各兩切片(c,d,e,f)及經過切片間上皮輪廓平滑化方法調整之結果顯示於(g)	34
圖24.(a)建議的起始切片；(b)為未放置支架建議的終點切片；(c)為出現有二個支架的切片；(d)為(c)的上張切片	36
圖25.範例一病人的實驗結果	39
圖26.範例六其起始切片執行結果(a)及上皮二值化圖(b)和下二張切片執行結果(c)、(d)；第19張切片(e)；第33張切片(f)；範例八最後一張切片(g)	41

表目錄
表1.DICOM標籤意義及資料長度	                        10
表2.Photometric Interpretation值及其所代表的意義	11
表3.Transfer Syntax UID值所代表的意義	        11

[1] 	施俊哲, “胸、腹主動脈瘤微創治療新趨勢--支架血管治療主動脈瘤,” 臨床醫學 60:4=358 2007.10[民96.10], pp. 271-282, 10 2007. 
[2] 	L. Ying-Siu, "heart disease and health," [Online]. Available: http://home.educities.edu.tw/leeyingsiu/slide-pci-heartdisease.pdf.
[3] 	James Heilman, MD, "A ruptured AAA as seen on CT," 2 June 2011. [Online]. Available: http://commons.wikimedia.org/wiki/File:RupturedAAA.png.
[4] 	T. D. Pham and J. Golledge, "Geo-statistically Constrained Fuzzy Segmentation of Abdominal Aortic Aneurysm CT Images," IEEE International Conference on Fuzzy Systems, 2008. FUZZ-IEEE 2008. (IEEE World Congress on Computational Intelligence)., pp. 1446 - 1451, 1-6 June 2008. 
[5] 	E.M, Majd, U.U. Sheikh, and S.A.R Abu-Bakar, "Automatic Segmentation of Abdominal Aortic Aneurysm in Computed Tomography Images Using Spatial Fuzzy C-Means," Signal-Image Technology and Internet-Based Systems (SITIS), 2010 Sixth International Conference on, pp. 170 - 175, 15-18 December 2010. 
[6] 	Marko Subasic, Sven Loncaric, Erich Sorantin, "3D image analysis of abdominal aortic aneurysm," SPIE Medical Imaging, p. 388, 17 February 2001. 
[7] 	T.F. Cootes, C.J. Taylor, D.H. Cooper, J. Graham, "Active Shape Models-Their Training and Application," Computer Vision and Image Understanding, vol. 61, no. 1, pp. 38-59, January 1995. 
[8] 	Marleen de Bruijne, Bram van Ginneken, Max A Viergever, Wiro J Niessen, "Interactive segmentation of abdominal aortic aneurysms in CTA images," Medical Image Analysis, vol. 8, no. 2, pp. 127-138, June 2004. 
[9] 	Silvia D. Olabarriaga, Jean-Michel Rouet, Maxim Fradkin, Marcel Breeuwer, and Wiro J. Niessen, "Segmentation of Thrombus in Abdominal Aortic Aneurysms From CTA With Nonparametric Statistical Grey Level Appearance Modeling," Medical Imaging, IEEE Transactions, vol. 24, no. 4, pp. 477-485, April 2005. 
[10] 	Onno Wink, Wiro J. Niessen, Max A. Viergever, "Minimum Cost Path Determination Using a Simple Heuristic Function," Pattern Recognition, 2000. Proceedings. 15th International Conference, vol. 3, pp. 988-1001, 2000. 
[11] 	S. Arya, D. M. Mount, N. S. Netanyahu, R. Silverman, and A. Y. Wu, "An optimal algorithm for approximate nearest neighbor searching fixed dimensions," Journal of the ACM, vol. 45, no. 6, pp. 891-923, November 1998. 
[12] 	Christos Zohios, Georgios Kossioris, and Yannis Papaharilaou, "Geometrical methods for level set based abdominal aortic aneurysm thrombus and outer wall 2D image segmentation," Computer Methods and Programs in Biomedicine, vol. 107, pp. 202-217, August 2012. 
[13] 	"Taipei Veterans General Hospital - the department of Radiology," [Online]. Available: http://wd.vghtpe.gov.tw/RAD/site.jsp?id=2951.
[14] 	Stefan W‥orz, Hendrik von Tengg-Kobligk, Verena Henninger, Fabian Rengier, Hardy Schumacher, Dittmar B‥ockler, Hans-Ulrich Kauczor, and Karl Rohr, "3-D Quantification of the Aortic Arch Morphology in 3-D CTA Data for Endovascular Aortic Repair," IEEE Transactions on Biomedical Engineering, vol. 57, pp. 2359-2368, October 2010. 
[15] 	Edward R. Dougherty, An introduction to morphological image processing, 1992. 
[16] 	Thorsten Behrens, Karl Rohr,H.S. Stiehl, "Robust segmentation of tubular structures in 3-D medical images by parametric object detection and tracking," IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 33, no. 4, pp. 554-561, August 2003. 
[17] 	Roc’ıo Cabrera Lozoya, Olivier Bouchot, Tadeusz Sliwa, Eric Steinmetz, Yvon Voisin and Alain Lalande, "A 4D Patient-Specific Modelling of the Thoracic Aorta from Cine-MR Images," 2011 Seventh International Conference on Signal-Image Technology and Internet-Based Systems (SITIS), pp. 269-276, 2011. 
[18] 	Yefeng Zheng, Matthias John, Rui Liao, Alois N‥ottling, Jan Boese, J‥org Kempfert, Thomas Walther, Gernot Brockmann, and Dorin Comaniciu, "Automatic Aorta Segmentation and Valve Landmark Detection in C-Arm CT for Transcatheter Aortic Valve Implantation," IEEE Transactions on Medical Imaging, vol. 31, no. 12, pp. 2307-2321, December 2012. 
[19] 	Fukunaga, K, Hostetler, L, "The estimation of the gradient of a density function, with applications in pattern recognition," IEEE Transactions on Information Theory, vol. 21, pp. 32-40, January 1975. 
[20] 	Yizong Cheng, "Mean shift, mode seeking, and clustering," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 17, pp. 790-799, August 1995. 
[21] 	O. Wink, W. J. Niessen, M. A. Viergever, "Fast quantification of abdominal aortic aneurysms from CTA volumes," Medical Image Computing and Computer-Assisted Interventation — MICCAI' 98 Lecture Notes in Computer Science, vol. 1496, pp. 138-145, 1998. 
[22] 	[Online]. Available: http://163.13.127.10/cht/experiments/medical_image/movie.rar.