因應東方女性罹患乳癌的比例日漸攀升，為了提升早期的治癒率，影像診斷系統扮演不可或缺的角色，針對東方女性的乳房較小巧且緻密而言，比起西方女性更依賴於乳房超音波的診斷。目前雖有許多研究利用乳房超音波的電腦輔助診斷系統幫助醫師辨別乳房腫瘤，但此方法仍會受限於儀器和人為操作因素；另有一派學者提出分析乳房超音波的原始射頻訊號，已證實Nakagami 分佈可描述乳房超音波訊號特性，也能有效區分良性和惡性的腫瘤，但仍未有較完整的一套方法，且Nakagami 參數影像的解析度較差。
因此，我們希望結合灰階影像的輪廓特徵和原始射頻訊號的資訊做為新的診斷工具，由灰階影像可以獲取乳房腫瘤的外型，而Nakagami 參數影像則可以區分腫瘤內部的成分，由不同的物理量一起探討乳房腫瘤的分類。傳統B-mode 影像前置處理，我們提出新的參數MSR(Mean to Standard deviation Ratio)做為增強灰階影像對比的加權因子，且發現MSR 的分佈能夠區分乳房病灶和正常乳腺組織，便採用雙峰高斯分佈擬合其直方圖，以最佳整體臨界值法獲取二值化影像，再利用增強對比後的灰階影像和二值化影像進行自動化輪廓圈選，最後，萃取腫瘤輪廓並採用6 種輪廓特徵參數，驗證其惡性腫瘤的輪廓不規則度會高於良性腫瘤。另一方面，針對Nakagami 參數進行探討，採用先前研究提出的Nakagami參數成像方法，並觀測不同乳房病灶其Nakagami 參數影像上ROI 內的平均Nakagami 參數，可以區分出纖維囊腫、脂肪、及實質性腫瘤，且統計出良性腫瘤的平均Nakagami 參數會高於惡性腫瘤。
利用Fuzzy C-means 執行所有特徵參數的群聚分析，證實結合輪廓特徵參數和Nakagami 參數的方法分類乳房腫瘤，兩者之間可以相輔相成，表現最好的輪廓特徵參數和平均Nakagami 參數結合後，其診斷效率的準確度為81.7%、敏感度為80%、特異性為83%。

Breast cancer is the most common cancer in women worldwide. According to the lately statistics in Taiwan, the mortality from breast cancer has become the fourth of cancerous diseases among women. In addition, younger women tend to have dense breasts and Asian women tend to have denser ones. Not only ultrasound is capable of detecting masses even in dense breasts, but also more convenient, safer, and forming the images in real-time tool for patient in regularly physical examination. Computer-aided diagnosis (CAD) has been used to discriminate between benign and malignant tumor for ultrasonic B-mode scans, but this method makes the classification largely dependent on the skill of the operator. Several studies have shown that the Nakagami parameter estimated from the ultrasonic backscattered signals can be used to assist conventional B-mode scanning when classifying breast tumors.
Hence, we propose to combine ultrasonic B-mode scans with Nakagami parametric image for categorizing breast masses. We expect to acquire boundary feature and internal components of breast masses from B-mode scans and Nakagami parametric image. For boundary feature, irregular degree of contour in malignant tumor is higher than benign tumor. For internal components, the average Nakagami parameter of malignant tumor is lower than benign tumor.
We used Fuzzy C-means (FCM) to separate malignant cluster and benign cluster by all parameters. The average Nakagami parameter and contour features provide complementary characteristics in diagnosis ultrasound breast tumor image. The best efficiency is that accuracy is 81.7 %, sensitivity is 80 %, and specificity is 83 %.

[1] "Guidelines for management of breast cancer," World Health Organization, 2008.
[2] H. O. Adami, L. B. Signorello, and D. Trichopoulos, "Towards an understanding of breast cancer etiology," Semin. Cancer Biol., vol. 8, pp. 255-262, 1998.
[3] "Breast Cancer Facts & Figures 2007-2008," American Cancer Society, 2007.
[4] 行政院衛生署, "The statistics of Department of Health"
[5] R. A. Kubik-Huch, "Imaging the young breast," The Breast, vol. 15, pp. S35-S40, 2006.
[6] A. T. Stavros, D. Thickman, C. L. Rapp, M. A. Dennis, S. H. Parker, and G. A. Sisney, "Solid breast nodules: use of sonography to distinguish between benign and malignant lesions," Radiology, pp. 123-34, 1995.
[7] P. M. Shankar, V. A. Dumane, J. M. Reid, V. Genis, F. Forsberg, C. W. Piccoli, and B. B. Goldberg, "Classification of ultrasonic B-mode images of breast masses using Nakagami distribution," IEEE Trans. Ultrason. Ferroelectr. Freq. Control, vol. 48, pp. 569-580, 2001.
[8] P. M. Shankar, V. A. Dumane, C. W. Piccoli, J. M. Reid, F. Forsberg, and B. B. Goldberg, "Classification of breast masses in ultrasonic B-MODE images using a compounding technique in the Nakagami distribution domain," Ultrasound Med. Biol., vol. 28, pp. 1295-1300, 2002.
[9] P. M. Shankar, V. A. Dumane, T. George, C. W. Piccoli, J. M. Reid, F. Forsberg, and B. B. Goldberg, "Classification of breast masses in ultrasonic B scans using Nakagami and K distributions," Phys. Med. Biol., vol. 48, pp. 2229-40, 2003.
[10] P. H. Tsui, C. K. Yeh, C. C. Chang, and Y. Y. Liao, "Classification of breast masses by ultrasonic Nakagami imaging: a feasibility study," Phys. Med. Biol., vol. 53, pp. 6027-6044, 2008.
[11] K. G. Kim, J. H. Kim, and B. G. Min, "Classification of malignant and benign tumors using boundary characteristics in breast ultrasonograms," J. Digit. Imaging, vol. 15, pp. 224-227, 2002.
[12] K. G. Kim, S. W. Cho, S. J. Min, J. H. Kim, B. G. Min, and K. T. Bae, "Computerized scheme for assessing ultrasonographic features of breast masses," Acad. Radiol, vol. 12, pp. 58-66, 2005.
[13] Y. H. Chou, C. M. Tiu, G. S. Hung, S. C. Wu, T. Y. Chang, and H. K. Chiang, "Stepwise logistic regression analysis of tumor contour features for breast ultrasound diagnosis," Ultrasound Med. Biol., vol. 27, pp. 1493-1498, 2001.
[14] C. M. Chen, Y. H. Chou, K. C. Han, G. S. Hung, C. M. Tiu, H. J. Chiou, and S. Y. Chiou, "Breast lesions on sonograms: Computer-aided diagnosis with nearly setting-independent features and artificial neural networks," Radiology, vol. 226, pp. 504-514, 2003.
[15] H. C. Yang, C. H. Chang, S. W. Huang, Y. H. Chou, and P. C. Li, "Breast Ultrasound Computer-Aided Diagnosis Using Both Acoustic and Image Features," IEEE Ultrason. Symp., pp. 2489-2492, 2007.
[16] W. C. Shen, R. F. Chang, and W. K. Moon, "Computer aided classification system for breast ultrasound based on breast imaging reporting and data system (BI-RADS)," Ultrasound Med. Biol., vol. 33, pp. 1688-1698, 2007.
[17] W. K. Moon, R. F. Chang, C. J. Chen, D. R. Chen, and W. L. Chen, "Solid breast masses: Classification with computer-aided analysis of continuous US images obtained with probe compression," Radiology, vol. 236, pp. 458-464, 2005.
[18] K. Horsch, M. L. Giger, L. A. Venta, and C. J. Vyborny, "Automatic segmentation of breast lesions on ultrasound," Med. Phys., vol. 28, pp. 1652-1659, 2001.
[19] J. Canny, "A computational approach to edge detection," IEEE Trans. Pattern Anal. Mach. Intell., vol. PAMI-8, pp. 679-698, 1986.
[20] J. Shan, H. D. Cheng, and Y. Wang, "A completely automatic segmentation method for breast ultrasound images using region growing," JCIS-2008 Proceedings, 2008.
[21] Y. L. Huang and D. R. Chen, "Watershed segmentation for breast tumor in 2-D sonography," Ultrasound Med. Biol., vol. 30, pp. 625-632, 2004.
[22] D. R. Chen, R. F. Chang, W. J. Wu, W. K. Moon, and W. L. Wu, "3-D breast ultrasound segmentation using active contour model," Ultrasound Med. Biol., vol. 29, pp. 1017-1026, 2003.
[23] C. Davatzikos and R. N. Bryan, "Using a deformable surface model to obtain a shape representation of the cortex," IEEE Trans. Med. Imaging, vol. 15, pp. 785-795, 1996.
[24] C. Y. Xu and J. L. Prince, "Snakes, shapes, and gradient vector flow," IEEE Trans. Image Process., vol. 7, pp. 359-369, 1998.
[25] J. R. Cheng and S. W. Foo, "Dynamic directional gradient vector flow for snakes," IEEE Trans. Image Process., vol. 15, pp. 1563-1571, 2006.
[26] Y. S. Chen and T. D. Lin, "An iterative approach to removing the “Closure Noise” using disk expansion method," IEEE Signal Process. Lett., vol. 2, pp. 105-107, 1995.

[27] C. K. Yeh, Y. S. Chen, W. C. Fan, and Y. Y. Liao, "A disk expansion segmentation method for ultrasonic breast lesions," Pattern Recognit., vol. 42, pp. 596-606, 2009.
[28] W. C. Hoffman, "Statistical methods in radio-wave propagation," 1958.
[29] P. M. Shankar, "A general statistical model for ultrasonic backscattering from tissues," IEEE Trans. Ultrason. Ferroelectr. Freq. Control, vol. 47, pp. 727-736, 2000.
[30] P. H. Tsui and S. H. Wang, "The effect of transducer characteristics on the estimation of Nakagami paramater as a function of scatterer concentration," Ultrasound Med. Biol., vol. 30, pp. 1345-1353, 2004.
[31] P. M. Shankar, V. A. Dumane, C. W. Piccoli, J. M. Reid, F. Forsberg, and B. B. Goldberg, "Computer-aided classification of breast masses in ultrasonic B-scans using a multiparameter approach," IEEE Trans. Ultrason. Ferroelectr. Freq. Control, vol. 50, pp. 1002-1009, 2003.
[32] P. M. Shankar, C. W. Piccoli, J. M. Reid, J. Forsberg, and B. B. Goldberg, "Application of the compound probability density function for characterization of breast masses in ultrasound B scans," Phys. Med. Biol., vol. 50, pp. 2241-2248, 2005.
[33] R. KOLÁŘ, R. JIŘÍK, and J. JAN, "Estimator Comparison of the Nakagami-m Parameter and Its Application in Echocardiography," Radioengineering, vol. 13, pp. 8-12, 2004.
[34] F. Davignon, J. F. Deprez, and O. Basset, "A parametric imaging approach for the segmentation of ultrasound data," Ultrasonics, vol. 43, pp. 789-801, 2005.
[35] P. H. Tsui, C. C. Huang, C. C. Chang, S. H. Wang, and K. K. Shung, "Feasibility study of using high-frequency ultrasonic Nakagami imaging for characterizing the cataract lens in vitro," Phys. Med. Biol., vol. 52, pp. 6413-6425, 2007.
[36] A. Madabhushi and D. N. Metaxas, "Combining low-, high-level and empirical domain knowledge for automated segmentation of ultrasonic breast lesions," IEEE Trans. Med. Imaging, vol. 22, pp. 155-169, 2003.
[37] 李維寧, "Efficient Speckle Tracking Technique and Its Applications in Ultrasonic Breast Imaging," 碩士論文, 國立台灣大學電機工程研究所, 2003.
[38] 樂穎杰, "Ultrasound Images in Carpal Tunnel Syndrome Diagnosis," 碩士論文, 國立清華大學生醫工程與環境科學研究所, 2007.