本研究的目的在於建立自動化系統，用於偵測B-Mode及M-Mode頸動脈超音波影像之動脈壁。在臨床的應用中，血管壁內層厚度(IMT: intima-media thickness)是一個重要指標，用於預測心血管疾病，如：心肌梗塞，腦中風等。當頸動脈血管壁內層厚度增加時，其身體其他部位的動脈血管壁內層厚度亦會增加。如果發生在心臟冠狀動脈，則心肌梗塞的風險即會增高。如果發生在腦部，則腦中風的可能會增加。因此，用超音波觀測頸動脈壁影像則成為一個最佳選擇。
在量測血管壁內層厚度的方面，主要是以B-Mode影像為主。傳統上是以受過專業訓練的醫生，以手工的方式描繪出。此舉不但耗時費力，並且其結果因人因時而異。因此，自動化量測血管壁內層厚度則突出其重要性。在本論文中提出兩種方法，用以量測血管壁內層厚度，分別是：1) Dual-Line Detection System（雙線偵測系統）及2) Snake-Based System。文中分別對兩種方法作詳細描述。前者是一新提出的架構，使用影像梯度及連續性作為特徵參數。針對此系統，使用誤差分析，比較自動化系統與人工描繪的結果，計算其誤差。至於後者，文中提出一方法，用以搜尋intima並提供給Snake作為初始曲線，Snake據此而能找出intima及adventitia。我們改進Cohen’s Snake，使之能應用於此研究目的。並提出一選擇時間常數的方法，解決曲線震盪及耗時的問題。針對Snake-Based System，我們比較自動化系統與人工描繪的結果，並計算其誤差。在實際使用 兩系統的經驗中，我們發現，Dual-Line Detection System的抗雜訊干擾的能力較Snake-Based System要強。
　　在量測血管壁隨時間的變化方面，主要是以M-Mode影像為主。傳統上仍是以受過專業訓練的醫生，以手工的方式描繪出。其缺點也是顯而易見。本論文提出一利用自我相似性來找出一與血管內層曲線相似的曲線，並進而找出血管壁隨時間變化的曲線。本系統可以處理大部分的M-Mode影像，但對於高雜訊的影像，仍有誤差產生。此誤差大多產生於當intima有缺損同時旁邊伴有高雜訊時。
此系統已建立完成並建制於德國茀萊堡大學醫院運動醫學科，當超音波檢查完成 後，即交由這些系統作分析。他們提供一個可靠的技術於量測頸動脈之血管壁內層厚度及血管壁彈性，這些參數在臨床上扮演很重要的意義。

The purpose of this study is to develop the automated image analysis system for identifying the intima and adventitia of common carotid artery (CCA) in B-mode sonographic images and the intima of both near and far walls of the common carotid artery in M-mode images. In clinical application, the intima-media thickness (IMT) is an important index in predicting or identifying the myocardial events such as myocardial infarctions or strokes. If the IMT in the CCA is increased, the IMT in another artery system might be also increased. If the increases take place in coronary arteries, the risk of a myocardial related disease is increased. If the increases take place in brain arteries, the possibility of resulting in strokes is arisen. Therefore, using the sonographic images to examine the CCA walls is a good choice.
Generally, B-mode images are used in measuring the IMT. The traditional method for tracing the intima and adventitia curves is manual measurement performed by experienced physicians. However, the results may differ among physicians and times, especially, with different professional training and experiences. Therefore, the development of an automated system for measuring the IMT becomes an important issue. In this study, two kinds of methods are proposed to measure the IMT. One is the Dual-Line Detection System, and the other is the Snake-Based System. The former uses the image gradient and continuity as features. In the accuracy analysis, comparison of manual measurement and automated analysis of the dual-line detection system is performed. The errors are calculated and listed in the tables. Another one is the snake-based system. A method is proposed to give the snake an initial curve in identifying the intima and adventitia. Cohen’s snake is modified for this application. In addition, we propose a method to choose the time-constant so that the problems of oscillation and computation-intensiveness of snakes are solved. As to the accuracy analysis, the results obtained from the automated system and from the manual measurement are compared. The errors are thereby calculated. Based on our clinical applications, it is found that the Dual-Line Detection System is more robust than the Snake-Based System in the images with large noises.
Generally, the M-mode images are used for assessing the motion of CCA walls. The traditional way performed by experienced physicians is manual tracing. Therefore, it is tedious, time-consuming, unreliable, and subjective. In this study, an automated method that uses self-similarity is proposed to identify the intimal layers in motion of both near and far wall of CCA in M-mode images. This system works satisfactorily for most of the images. However, some errors might be occurred owing to large image noises. Most of errors come from the intima having discontinuity with large noises nearby.
These systems have been installed in the Department of Rehabilitation, Prevention and Sports Medicine, Freiburg University Hospital. After the sonographic examination, the images are processed by these systems off-line. They provide a reliable technique to measure the IMT and the wall motion of the CCA intima, which are important factors in clinic application.

[1] D. H. O’Leary et al., “Carotid-artery intima and media thickness as a risk factor for myocardial infarction and stroke in older adults,” The New England Journal of Medicine vol. 340, pp. 14-22, 1999.
[2] H. N. Hodis, W. J. Mack, et al., “The role of carotid artery intima-media thickness in predicating clinical coronary events,” Ann. Intern. Med., vol. 128, pp. 262-269, 1998.
[3] D. H. O’Leary et al., “Use of Sonography to evaluate carotid atherosclerosis in the elderly: The cardiovascular health study,” Stroke, vol. 22, no. 9, pp. 1155-1163, 1991.
[4] D. Baldassarre et al., “Common carotid intima-media thickness measurement: A method to improve accuracy and precision,” Stroke, vol. 25, pp. 1588-1592, 1994.
[5] M. A. Espeland, T. E. Craven, W. A. Riley, J. Corson, A. Romont, and C. D. Furberg, “Reliability of longitudinal ultrasonographic measurements of carotid intimal-medial thickness,” Asymptomatic Carotid Artery Progression Study Research Group,” Stroke, vol. 27, pp. 480-485, 1996.
[6] W. A. Riley, R. W. Barnes, W. B. Applegate, R. Dempsey, T. Hartwell, V. G. Davis, M. G. Bond, and C. D. Furberg, “Reproducibility of Noninvasive ultrasonic measurement of carotid artery plaque study,” Stroke, vol. 23, no. 8, pp. 1062-1068, 1992.
[7] L. E. Chambless, M. M. Zhong, D. Arnett, A. R. Folsom, W. A. Riley, and G. Heiss, “Variablity in B-mode ultrasound measurements in the atherosclerosis risk in communities (ARIC) study,” Ultrasound in Med. & Biol., vol. 22, no. 5, pp. 545-554, 1996.
[8] P. J. Touboul, P. Prati, P. Y. Scarabin, V. Adrai, E. Thibout, and P. Ducimetiere, “Use of monitoring software to improve the measurement of carotid wall thickness by B-mode imaging,” J. Hypertens. Suppl., vol. 10, pp. 37-41, 1992.
[9] R. H. Selzer, H. N. Hodis, H. Kwong-Fu et al., “Evaluation of computerized edge tracking for quantifying intima-media thickness of the common carotid artery from B-mode ultrasound images,” Atherosclerosis, vol. 111, pp. 1-11, 1994.
[10] T Gustavsson, Q Liang, I Wendelhag, J. Wikstrand, “A dynamic programming procedure for automated ultrasonic measurement of the carotid artery,” Proc. IEEE Computers Cardiology, pp. 297-300, 1994.
[11] T. Gustavsson, R. Abu-Gharbieh, G. Hamarneh, and Q Liang, “Implementation and comparison of four different boundary detection algorithms for quantitative ultrasonic measurements of the human carotid artery,” Proc. IEEE Computers Cardiology, pp. 69-72, 1997.
[12] Q. Liang, I. Wendelhag, J. Wikstrand, and T. Gustavsson, “A multiscale dynamic programming procedure for boundary detection in ultrasonic artery images,” IEEE Trans. Med. Imaging, vol. 19, no. 2, pp. 127-142, 2000.
[13] M. Kass, A. Witkin, and D. Terzopoulos, “Snake: Active contour models,” International Journal of Computer Vision, vol. 1, pp. 321-331, 1988.
[14] L. D. Cohen, “On active contour models and balloons,” Computer Vision, Graphics, and Image Processing: Image Understanding, vol. 53, no. 2, pp. 211-218, 1991.
[15] L. D. Cohen and I. Cohen, “Finite element methods for active contour methods and balloons for 2D and 3D images,” IEEE Trans. Patt. Anal. Machine Intell., vol. 15, no. 11, pp. 1131-1147, 1993.
[16] W. M. Neuenschwande, P. Fua, L. Iverson, G. Szekely, and O. Kubler, “Ziplock snake,” Int. J. Computer Vision, vol. 25, no. 3, pp. 191-201, 1997.
[17] K. F. Lai and R. T. Chin, “Deformable contours: Modeling and extraction,” IEEE Trans. Patt. Anal. Machine Intell, vol. 17, no. 11, pp. 1084-1090, 1995.
[18] Mathematical Society of Japan, “Encyclopedic Dictionary of Mathematics,” 2nd Edition, The MIT Press, Cambridge Massachusetts, and London, England, 1993.
[19] A. P. G. Hoeks, P. J. Brands, F. A. M. Smeets, and R. S. Reneman, “Assessment of the distensibility of superficial arteries,” Ultrasound in Med. & Biol., vol. 16, no. 2, pp. 121-128, 1990.
[20] R. W. Stadler, J. A. Taylor, and R. S. Lees, “Comparison of B-mode, M-mode and echo-tracking methods for measurement of the arterial distension waveform,” Ultrasound in Med. Biol., vol. 23, no. 6, pp. 879-887, 1997.
[21] W. M. Siebert, “Circuits, signals, and systems,” Camberage, MA, The MIT Press, 1986.
[22] V. Caselles, R. Kimmel, and G. Sapiro, “Geodesic active contours,” International Journal of Computer Vision,” vol. 22, no. 1, pp. 61-79, 1997.
[23] I.D.G. Macleod, “Comments on techniques for edge detection,” Proc. of the IEEE, vol. 60, p. 344, 1972.
[24] P. A. Couvignou, N. P. Papanikolopoulos, M. Sullivan, and P. K. Khosla, “The use of active deformable models in model-based robotic visual servoing,” J. Intell. Robotic Syst., vol. 17, pp. 195-221, 1996.
[25] W. Neuenschwander, “Elastic deformable contour and surface models for 2-D and 3-D image segmentation,” Ph.D. dissertation, Swiss Federal Institute of Technology, Zurich, Switzerland.
[26] A. P. G. Hoeks, C. Willekes, P. Boutouyrie, P. J. Brands, J. M. Willigers, and R. S. Reneman, “Automated detection of local artery wall thickness based on M-line signal processing,” Ultrasound in Med. & Biol., vol. 23, no. 7, pp. 1017-1023, 1997.
[27] Y. Tardy, D. Hayoz, J.-P. Mignot, P. Richard, H. R. Brunner, and J.-J. Meister, “Dynamic non-invasive measurements of arterial diameter and wall thickness,” Journal of Hypertension, vol. 10, pp. 105-109, 1992.
[28] Da-chuan Cheng, Arno Schmidt-Trucksäss, Markus Sandock, Qin Pu, and Hans Burkhardt, “Automatic detection of the intimal and the adventitial layers of the common carotid artery wall in ultrasound B-mode images using snakes,” Proceedings of the 10th International Conference on Image Analysis and Processing (ICIAP’99), pp. 452-457, 1999.
[29] Da-chuan Cheng, Arno Schmidt-Trucksäss, Kuo-sheng Cheng, and Hans Burkhardt, “Using snakes to detect the intimal and adventitial layers of the common carotid artery wall in sonographic images,” Journal of Computer Methods and Programs in Biomedicine, vol. 67, pp. 27-37, 2002.
[30] Arno Schmidt-Trucksäss, Da-chuan Cheng, Markus Sandrock, Martin Huonker, Joseph Keul†, Rainer Rauramaa, Jürgen Schulte-Mönting, and Hans Burkhardt, “Computerized analysing system using the active contour in ultrasound measurement of carotid artery intima-media thickness,” Clinical Physiology, vol. 21, no. 5, pp. 561-569, 2001.
[31] Arno Schmidt-Trucksäss, Markus Sandrock, Da-chuan Cheng, Hans-Michael Müller, Manfred W. Baumstark, Rainer Rauramaa, Aloys Berg, and Martin Huonker, “Quantitative Measurement of Carotid Intima-Media Roughness – Effect of Age and Manifest Coronary Artery Disease,” accepted to publish in Atherosclerosis.
[32] Markus Sandrock, Da-chuan Cheng, Daniel Schmitz, Arno Schmidt-Trucksäss, “Quantification of the wall inhomogeneity in B-mod sonographic images of the carotid artery,” Journal of Ultrasound Med., vol. 21, pp. 1395-1404, 2002.
[33] Rainer Rauramaa, Sari B. Väisänen, Le-Anh Luong, Arno Schmidt-Trucksäss, Ilkka M.Penttilä, Claude Bouchard, Jari Töyry, Steve E. Humphries, “Stromelysin-1 and interleukin-6 gene promoter polymorphisms are determinants of asymptomatic carotid artery atherosclerosis,” Arterioscler Thromb Vasc. Biol., vol. 20, no. 12, pp. 2657-2662, 2000.
[34] J. Sethian, “Level set methods,” Cambridge Univ. Press, 1996.
[35] J. Sethian, “Level set methods and fast marching methods: Evolving interfaces in computational geometry, fluid mechanics, computer vision, and materials sciences,” Cambridge Monograph on Applied and Computational Mathematics, Cambridge University Press 1999.
[36] R. Ronfard, “Region-based strategies for active contour models,” International Journal of Computer Vision, vol. 13, no. 2, pp. 229-251, 1994.
[37] A. Chakraborty, L. Staib, and J. Duncan, “Deformable boundary finding in medical images by integrating gradient and region information,” IEEE Trans. Med. Imaging, vol. 15, pp. 859-870, 1996.
[38] N. Paragios and R. Deriche, “Geodesic active regions for motion estimation and tracking,” International Conference on Computer Vision, Corfu, Greece, 1999.
[39] C. Chesnaud, P. Refregier, and V. Boulet, “Statistical region snake-based segmentation adapted to different physical noise models,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 21, no. 11, pp. 1145-1156, 1999.
[40] S. Jehan-Besson, M. Barlaud, and G. Aubert, “Video object segmentation using eulerian region-based active contours,” International Conference on Computer Vision, Vancouver, Canada, 2001.
[41] N. Paragios and R. Deriche, “Coupled geodesic active regions for image segmentation: A level set approach,” European Conference in Computer Vision, Dublin, Ireland, June, 2000.
[42] N. Paragios and R. Deriche, “Geodesic active regions: A new paradigm to deal with frame partition problems in computer vision,” Journal of Visual Communication and Image Representation, vol. 13, pp. 249-268, 2002.
[43] N. Paragios and R. Deriche, “Geodesic active regions and level set methods for supervised texture segmentation,” International Journal of Computer Vision, vol. 46, no. 3, pp. 223-247, 2002.
[44] C. Chesnaud, P. Refregier, and V. Boulet, Statistical Region Snake-Based Segmentation Adapted to Different Physical Noise Models, IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 21, no. 11, pp. 1145-1157, 1999.
[45] P. Radeva and J. Vitrià, “Region based approach for discriminant snakes,” Proceedings IEEE Int. Conf. on Image Processing, pp. 801-804, 2001.
[46] D. Gil, P. Radeva, J. Saludes, “Segmentation of artery wall in coronary IVUS images: A probabilistic approach,” 15th International Conference on Pattern Recognition, Barcelona, vol. 4, pp. 352-355, 2000.
[47] Yusuf Akgul, Chandra Kambhamettu and Maureen Stone, “Automatic Motion Analysis of Tongue Surface from Ultrasound Images,” IEEE Workshop on Biomedical Image Analysis, Santa Barbara CA, June 26-27, 1998.
[48] M. Mignotte and J. Meunier, “An unsupervised multiscale approach for the dynamic contour-based boundary detection issue in ultrasound imagery,” International Conference on Computer Vision, Pattern Recognition and Image Processing, CVPRIP'00, vol. 2, pp. 366-369, Atlantic City, New Jersey, USA, March 2000.
[49] A. Bosnjak, V. Burdin, V. Torrealba, G. Montilla, B. Solaiman, C. Roux, “Dynamical segmentation of the left ventricle in echocardiographic image sequences,” Proceedings of the 23rd Annual EMBS International Conference, Istanbul Turkey, pp. 2634-2637, Oct. 2001.
[50] Frank H. Netter, “Atlas of human anatomy,” Ciba-Geigy Corporation, plate 64, 1989.