乳癌是現今婦女最常見罹患的癌症，其乳癌致死人數往往在各國十大癌症死因排行中佔居前三名，但隨著醫療技術的不斷進步，乳癌致死率可以因患者早期發現早期治療而降低致死率，醫學診斷乳房腫瘤良惡性的方法中又以超音波影像具有非侵入式、即時造影及費用便宜等優點，因此為目前最普遍使用的診斷方式，目前現行的醫學診斷方式需依賴醫師豐富的臨床經驗，透過超音波影像中腫瘤的輪廓、大小、形狀分析腫瘤良惡性，但超音波影像中經常包含大量的雜訊及紋理特徵，容易影響醫師正確判讀結果，在診斷上可能會因為醫師的經驗不足而造成誤判影響到後續的病情控制，因此如果自動描繪的乳房腫瘤輪廓能夠與醫師手動描繪的腫瘤輪廓極為相似，將有助於提升醫師診斷腫瘤良惡性的正確率，在本研究中將提出乳房超音波影像腫瘤輪廓描繪方法，首先需先透過影像前處理去除超音波影像雜訊並強化影像對比度，使後續做腫瘤輪廓切割具有較高的準確率，腫瘤輪廓切割方法採用3D region growing方法描繪出腫瘤輪廓，最後再透過腫瘤輪廓後處理階段分析陰影在超音波影像中具有的特性，將3D region growing所描繪的腫瘤輪廓中包含腫瘤陰影的區域移除，在本研究中使用20個case進行實驗，分別為10個良性case，10個惡性case，經過實驗結果顯示利用所提出的腫瘤輪廓切割方法，能夠與醫師手動描繪的腫瘤輪廓極為相似 Breast malignant and benign tumor exist discrepancy in shape and size on sonography, information about shape and size provided by tumor’s contour is important in clinical diagnosis. Due to ultrasound image includes noises and tissue texture, clinical diagnosis must be highly depends on expertise experience. However, manually sketch the three-dimensional (3D) breast tumor contour is a time-consuming and complicated task. Automatic contouring which provided the similar contour with manual sketch of the breast tumor in the ultrasonic images might assist physicians in making a correct diagnosis. This study presents an efficient method for automatically detecting 3D contours of breast tumors in sonography. The proposed method performs voxel nearest neighbor (VNN), Wiener filter and unsharp filter to enhance contrast and reduce noise. The 3D region growing algorithm is utilized to obtain the contour of breast tumor and then contour post-processing analyzes the extracted contour to diminish the shadow region of tumor. This study evaluates total of 20 tumor cases, comprising 10 benign and 10 malignant cases. The results of computer simulation reveal that the proposed 3D segmentation method provides a robust contouring for breast US image. This approach always identifies similar contours as that obtained by manual contouring of the breast tumor and could save much of the time required to sketch precise contours
