基于超声图像边缘的乳腺肿瘤良恶性判别

目的：提取乳腺肿瘤超声图像的边缘，判别乳腺肿瘤的良恶性。方法：提出了一种改进的各向异性扩散滤波，即根据图像不同的梯度选择不同的扩散系数的扩散方程，滤除噪声的同时很好地保留了图像的边缘信息。然后在Level Set方法中提出一种新的变分公式，完全避免了重初始化步骤，且水平集函数的初始化灵活，采用手工勾画粗略边界的半自动分割流程，不仅提高了分割准确性，同时也进一步提高了分割效率。结果：采用综合灰度共生矩阵计算乳腺肿瘤的纹理特征和模糊C均值的方法进行良恶性判别，正确率为72.64%。结论：实验证明，本文算法能高效准确地提取出肿瘤边界，为肿瘤良恶性的判别提供可靠的依据。

Edge based differentiation between the malignant and the benign breast tumor in ultrasound images

Objective:Extracting the edge of breast tumor in ultrasonic images and distinguishing malignant from benign lesions.Method:This paper proposes an improved anisotropic diffusion filtering method which chooses diffusion equation with different diffusion coefficient according to the gradient of the image,and this method can smooth the image as well as finely preserve the boundary.Then presenting a new variational formulation in Level Set method which can avoid the step of re-initialization and make the formati...