本論文之研究目的在於建立一個應用於乳房X光影像之腫塊的自動化分割與鑑別流程。影像分割的部分，首先透過二值化、形態學法與連通區域標記分析，移除影像中不屬於人體的部分。接著使用模糊C均值法做初步的影像分割，並使用洞填充演算法，將空心之區域補齊。最後藉由水平集法對區域輪廓進行修正。影像的鑑別的部分，對分割完的區域擷取5個幾何特徵與5個紋理特徵，然後使用支援向量機進行學習與分類。研究中特別比較循序浮動選擇法與個別選擇法在最佳特徵選取對腫塊判定能力的影響。本研究使用Mini-MIAS資料庫的乳房X光影像來做方法的驗證與分析。實驗結果顯示，在測試資料的分類表現，最佳模型為使用循序浮動選擇法保留8個特徵，此時假陽率為20.83%，真陽率為85.71%，準確率為79.52%，ROC曲線下面積(AUC分數)為0.85。

The purpose of this study is to construct a process to automatically segment and detect the breast masses on mammograms. For image segmentation, artifacts such as labels are removed first by following algorithms, thresholding, morphological closing, and connected-component labeling. Then, fuzzy C-means method is utilized to extract the suspicious mass regions, followed by flood-fill algorithm to fill up those regions. At last, the level set method is employed to refine the contour of the segmented regions. As for masses detection, with five geometric features and five texture features extracted from the segmented regions, support vector machine (SVM) is applied to classify the segmented region. In particular, sequential floating searching method (SFSM) and individual choosing method are compared to explore their effects on classifying the segmented regions. In this research, mammograms from mini-mammographic image analysis society (MIAS) are used to validate the proposed process. Experiment results show that the best classification is achieved by using SFSM with eight-selected features. The false positive rate is 20.83%, the true positive rate is 85.71%, the accuracy is 79.52%, and the AUC score is 0.85.

[1] 衛生福利部國民健康署癌症防治組。. 乳癌防治。. https://www.hpa.gov.tw/Pages/Detail.aspx?nodeid=614&pid=1124. Accessed 07,10,2020.
[2] 陳若瑀。. 認識乳房攝影。. https://epaper.ntuh.gov.tw/health/201501/special_1_2.htmlw. Accessed 07,10,2020.
[3] Jyoti Dabass, Shaveta Arora, Rekha Vig, and Madasu Hanmandlu. Segmentation techniques for breast cancer imaging modalities- A review. Proceedings of the 9th International Conference On Cloud Computing, Data Science and Engineering, Confluence 2019, pages 658-663, 2019.
[4] Z. Chen and R. Zwiggelaar. A combined method for automatic identi cation of the breast boundary in mammograms. In 2012 5th International Conference
on BioMedical Engineering and Informatics, pages 121-125, 2012.
[5] B. Gupta and M. Tiwari. A tool supported approach for brightness preserving contrast enhancement and mass segmentation of mammogram images using histogram modified grey relational analysis. Multidimensional Systems and
Signal Processing, 28(4):1549-1567, 2017.
[6] E. Kozegar, M. Soryani, H. Behnam, M. Salamati, and T. Tan. Mass segmentation in automated 3-d breast ultrasound using adaptive region growing and supervised edge-based deformable model. IEEE Transactions on Medical Imaging, 37(4):918-928, 2018.
[7] David Raba, Arnau Oliver, Joan Mart , Marta Peracaula, and Joan Espunya.Breast segmentation with pectoral muscle suppression on digital mammograms. In Pattern Recognition and Image Analysis, pages 471-478, Berlin,Heidelberg, 2005. Springer Berlin Heidelberg.
[8] D. Terzopoulos M. Kass, A. Witkin. Snakes: Active contour models. International Journal of Computer Vision, 1:321-331, 1987.
[9] Kanchan Lata Kashyap, Manish Kumar Bajpai, and Pritee Khanna. Globally supported radial basis function based collocation method for evolution of level set in mass segmentation using mammograms. Computers in Biology and
Medicine, 87:22-37, 2017.
[10] Kanchan L. Kashyap, Manish K. Bajpai, Pritee Khanna, and George Giakos. Mesh-free based variational level set evolution for breast region segmentation and abnormality detection using mammograms. International Journal for Numerical Methods in Biomedical Engineering, 34(1):1-20, 2018.
[11] Sebastian Raschka and Vahid Mirjalili. Python 機器學習(Python Machine Learning)。劉立民、吳建華譯. 博碩文化, 2018.
[12] E. Malar, A. Kandaswamy, D. Chakravarthy, and A. Giri Dharan. A novel approach for detection and classification of mammographic microcalcifications using wavelet analysis and extreme learning machine. Computers in Biology and Medicine, 42(9):898-905, 2012.
[13] R.C. Gonzalez and R.E. Woods. Digital Image Processing. Upper Saddle River, 2008.
[14] Robert M. Haralick and Linda G. Shapiro. Computer and Robot Vision,Volume 1. Addison-Wesley, 1992.
[15] Bezdek James. Pattern Recognition With Fuzzy Objective Function Algorithms. Springer Science+Business Media, 1981.
[16] J. Anitha and J. Dinesh Peter. Mass segmentation in mammograms using a kernel-based fuzzy level set method. International Journal of Biomedical Engineering and Technology, 19(2):133-153, 2015.
[17] K. Somasundaram and T. Kalaiselvi. A method for filling holes in objects of medical images using region labeling and run length encoding schemes. 2010.
[18] Stanley Osher and James A. Sethian. Fronts propagating with curvature-dependent speed: Algorithms based on Hamilton-Jacobi formulations. Journal of Computational Physics, 79(1):12-49, 1988.
[19] J.A. Sethian. Level Set Methods and Fast Marching Methods: Evolving Interfaces in Computational Geometry, Fluid Mechanics, Computer Vision, and Materials Science. Cambridge University Press, 1999.
[20] Stanley Osher and Ronald Fedkiw. Level Set Methods and Dynamic Implicit Surfaces., volume 153. Springer-Verlag New York, Inc, 2003.
[21] Jos e Gomes and Olivier Faugeras. Reconciling distance functions and level sets. Journal of Visual Communication and Image Representation, 11(2):209-223, 2000.
[22] Chunming Li, Chenyang Xu, Changfeng Gui, and Martin D. Fox. Level set evolution without re-initialization: A new variational formulation. Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, I:430-436, 2005.
[23] Lawrence C. Evans. Partial Di erential Equation. American Mathematical Society, 1983.
[24] Zhiqiong Wang, Ge Yu, Yan Kang, Yingjie Zhao, and Qixun Qu. Breast tumor detection in digital mammography based on extreme learning machine. Neurocomputing, 128:175-184, 2014.
[25] Robert M. Haralick, K. Shanmugam, and Its'hak Dinstein. Textural Features for Image Clsassi cation. IEEE Transactions on Systems, Man and Cybernetics, SMC-3(6):610-621, 1973.
[26] P. Pudil, J. Novovi cov a, and J. Kittler. Floating search methods in feature selection. Pattern Recognition Letters, 15(11):1119-1125, 1994.
[27] Vladimir N. Vapnik. The Natural of Statistical Learning Theory. Springer Science+Business Media, 2000.
[28] Christopher M. Bishop. Pattern Recognition and Machine Learning. Springer Science+Business Media, 2006.
[29] Stephen Boyd and Lieven Vandenberghe. Convex Optimization. Cambridge University Press, 2004.
[30] John Platt. Sequential minimal optimization: A fast algorithm for training support vector machines. Technical Report MSR-TR-98-14, April 1998.
[31] Chih-Chung Chang and Chih-Jen Lin. Libsvm: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2:27:1-27:27, 2011.
[32] J.Suckling. The mammographic image analysis society digital mammogram database. International Congress Series, 1069:375-378, 1994.