近年來，由於現代人的生活品質不斷地提昇，過度精緻的飲食，加上體能運動的缺乏，往往會造成身體上的疾病。心血管疾病也因此而成為人類死亡的主要病因，並且巳經大幅地增加人類的死亡率。而心臟病和中風則是其中較為知名的心血管疾病，這類疾病通常在發病前沒有任何徵兆。因此，早期發現病人心血管中的可疑疾病因子，是用於預防治療的主要目的。冠狀動脈粥狀斑及冠狀動脈鈣化也巳經被廣泛地證實是形成心血管疾病的主要成因，這也是臨床上醫生在對病人進行診斷時，所需要觀察的部分。因此，我們的研究想要藉著對心血管冠狀動脈電腦斷層掃瞄影像的分析，了解血管所包含的成份，並且更進一步地來協肋醫生進行診斷。
在本篇論文中，我們發展了一套新的電腦輔助系統，本系統可以自動化地分割出冠狀動脈，並且對心血管疾病進行量化分析。首先，我們提出了一個新的變形模組，稱為雙邊變形模組。這個新的方法可以自動化並準確地同時擷取出血管內腔及血管外壁的邊界。而在量化分析的部分，我們採用了區域線性映對分類模組，做為血管組織分類架構的主要核心。配合重新定義的分類特徵轉換函式，區域線性映對的分類機制能夠有效率地區別出血管內腔、冠狀動脈粥狀斑以及冠狀動脈鈣化，並且能準確地獲得血管組織的分類結果。由實驗的結果中，我們所提出的血管冠狀動脈電腦斷層掃瞄影像分析系統，其效能及準確性巳經得到了驗證。

In recent years, the changes in human life styles, e.g. elegant diet and lack of exercise have caused drastic increase in many diseases. Especially, coronary artery disease has become one of the major causes of human death, which has greatly increased the mortality of human being. Heart attack and stroke are well-known coronary artery diseases, which usually occur without prior symptoms. Therefore, early detection in patients with suspicious coronary risk factors is the major goal of the preventive therapeutic strategies. Coronary atherosclerotic plaque and coronary calcification are well-recognized as the major factors in the development of coronary artery disease. By analyzing the coronary artery in the computed tomography angiogram, physicians can realize the composition within the artery and give the accurate treatment or preventive screening for patients.
In this thesis, we have developed a computer-aimed system for automatic segmentation and quantitative analysis of coronary artery disease. First we presented a novel deformable modal, called dual-snake, which can automatically and precisely extract boundaries of lumen and adventitia at the same time. In the analysis procedure, the local linear map (LLM) classifier is employed for tissue classification of coronary artery. With designed feature transformation function, LLM classification mechanism can effectively discriminate lumen and plaques and obtain accurate results of classification. The experimental results have demonstrated the efficiency and accuracy of the proposed image analysis system.

[1] Schroeder S, Kuettner A, Kopp AF, Heuschmidt M, Burgstahler C, Herdeg C, and Claussen CD, “Noninvasive evaluation of the prevalence of noncalcified atherosclerotic plaques by multi-slice detector computed tomography: results of a pilot study,” Int J Cardiol., Dec;92(2-3):151-5, 2003.
[2] M. Sonka, X. Zhang, and M. Siebes et al., ”Segmentation of intravascular ultrasound images: A knowledge based approach,” IEEE Trans. on Medical Imaging.14: 719-732. 1995.
[3] S. Aaro and M. Dahlborn, “The longitudinal axis rotation of the apical vertebra, the vertebral, spinal, and rib cage, deformity in idiopathic scoliosis studied by computer tomography,” Spine 6: 567-721981.
[4] D. Rueckert, P. Burger etc., “Automatic Tracking of the Aorta in Cardiovascular MR Images Using Deformable Models,” IEEE Transactions on Medical Imaging, 1997.
[5] L. D. Cohen and I. Cohen, “Finite-element methods for active contour models and balloons for 2-D and 3-D images,” IEEE Trans. Pattern Anal. Machine Intell., vol. 15, pp. 1131-1147, Nov. 1993.
[6] S. R. Wang and Y. N. Sun, “ 3D Segmentation of Aorta Arch on MR Images by Skeleton-based Shape Model,” Proc. The 6th World Multi-Conference on Systemics, Cybernetics and Informatics, 2002.
[7] C. H. Su, “Computer Visual System for 3D Flow Analysis of Aortic MR Images”, master thesis of CSIE, NCKU, Tainan, Taiwan, R.O.C., 2001.
[8] C.-M. Chen, H. H.-S. Lu, and A.-T. Hsiao, "A Dual Snake Model of High Penetrability for Ultrasound Image Boundary Extraction," Ultrasound in Medicine and Biology, 27, 12, 1651-1665, 2001.
[9] T. W. Nattkemper, H. Ritter, W. Schubert, “A neural classifier enabling high-throughput topological analysis of lymphocytes in tissue sections,” IEEE Transactions on Information Technology in Biomedicine 5(2): 138-149, 2001.
[10] M. Kass, A. Witkin, and D. Terzoulos, “Snake: Actour contour models,” Int J Comput Vision, 1:321–331, 1987.
[11] X. Papademetris, A. Sinuas, D. Dione, and J. Duncan, “Estimation of 3-D Left Ventricular Deformation From Medical Images Using Biomechanical Models,” IEEE Trans. on Medical Imaging, Vol. 21, No. 7, pp.786-800, July 2002.
[12] C. von Birgelen, C. D. Mario, W. Li, J. C. Schuurbiers, C. J. Slager, P. J. de Feyter, J. R. Roelandt, and P. W. Serruys, “Morphometric analysis in three-dimensional intracoronary ultrasound: An in vitro and in vivo study performed with a novel system for the contour detection of lumen and plaque,” Amer. Heart J., vol. 132, pp. 516–527, 1996.
[13] W. Li, E. J. Gussenhoven, Y. Zhong, S. H. K. The, C. D. Mario, S. Madretsma, F. van Egmond, P. J. de Feyter, H. Pieterman, H. van Urk, H. Rijsterborgh, and N. Bom, “Validation of quantitative analysis of intravascular ultrasound images,” Int. J. Cardiac Imag., vol. 6, pp. 247–254, 1991.
[14] C. von Birgelen, G. S. Mintz, A. Nicosia, D. P. Foley, W. J. van der Giessen, N. Bruining, S. G. Airiian, J. R. T. C. Roelandt, P. J. de Feyter, and P. W. Serruys, “Electrocardiogram-gated intravascular ultrasound image acquisition after coronary stent deployment facilitates on-line three-dimensional reconstruction and automated lumen quantification,” J. Amer. Coll. Cardiol., vol. 30, pp. 436–443, 1997.
[15] J. W. Sammon, Jr., “A nonlinear mapping for data structure analysis,” IEEE Trans. Comput., vol. C-18, pp. 401–409, 1969.
[16] J. Mao and K. Jain, “Artificial neural networks for feature extraction and multivariate data projection,” IEEE Trans. Neural Networks, vol. 6, pp. 296–317, Feb. 1995.
[17] E. Oja and S. Kaski, Kohonen Maps. Amsterdam, The Netherlands: Elsevier, 1999.