3D乳房彈性超音波之腫瘤診斷

Abstract

乳癌是女性癌症中的主要死因。然而，如果乳癌能夠在早期就被檢測到以及給予適當的治療，死亡率就能夠被有效降低。在最近幾年裡，許多研究都指出利用乳房彈性超音波來診斷會擁有優於傳統超音波的效能。在這篇論文裡的主要目的是使用我們所提議的切割方法來切割出腫瘤的輪廓並且利用從三維乳房彈性超音波的影像中所擷取出的特徵來診斷腫瘤。首先，先利用所提議的切割方法得到腫瘤在三維中的輪廓。接著，利用切割出來的輪廓以及B-mode影像來分別擷取腫瘤在影像中的紋理特徵、形狀特徵，以及建立與腫瘤有最小距離相合的橢球模型後，腫瘤與此橢球的異同點特徵，並利用同樣的腫瘤輪廓以及彈性影像得到腫瘤在影像中的彈性特徵。這些特徵被用來診斷良性與惡性的腫瘤。在此實驗中我們總共使用了40個經過病理驗證的腫瘤，其中包含了20個良性病例以及20個惡性病例。從實驗結果來看，結合形狀特徵、腫瘤與橢球的異同點特徵以及彈性特徵來診斷會擁有最好的效果，可達到準確率92.50% (37/40)，敏感性90.00% (18/20)，特異性95.00% (19/20)以及ROC曲線面積0.987。因此，利用乳房彈性超音波能夠更準確的診斷腫瘤。

Breast cancer is the major cause of cancer-related mortality in women. However, the death rate can be effectively decreased if the breast cancer can be detected early and treated appropriately. In recent years, many studies have indicated that the elastography has the better diagnosis performance than conventional US. In this paper, the main purposes are extracting the tumor contour by using the proposed segmentation methods and diagnosing the tumor by using the features extracted from 3-D elastography images. At first, the 3-D tumor contour is obtained by using the proposed segmentation methods. Then, the features containing texture information, shape information, ellipsoid fitting information are extracted respectively by using the segmented 3-D tumor contour and B-mode images, and the features containing elasticity information are calculated using the same contour and elastographic images. These features are used to diagnose the benign and malignant lesions. In this experiment, totally 40 biopsy-proved lesions containing 20 benign tumors and 20 malignant tumors are used. From the experimental results, the combination of shape, ellipsoid fitting and elastographic features has the best performance with accuracy 92.50% (37/40), sensitivity 90.00% (18/20), specificity 95.00% (19/20), and the area under the ROC curve Az 0.987. Therefore, the tumors can be diagnosed more precisely by using the elastography images.