隶属度区域水平集方法提取B超图像病灶  

Lesion Extraction from B-type Ultrasound Image Using Subordinate Degree Region Level Set Method

作　　者：杨谊[1] 喻德旷[1] 申洪[2] 

机构地区：[1]南方医科大学生物医学工程学院,广州510515 [2]南方医科大学基础医学院,广州510515

出　　处：《生物医学工程学杂志》2015年第4期779-787,共9页Journal of Biomedical Engineering

基　　金：广东省科技计划项目资助(2010B060300001);广东省产学研项目资助(2011B090400037);广东省科技产业技术研究与开发专项资助(C1040471)

摘　　要：B超图像在医学临床诊断中有着重要应用,但广泛存在的灰度分布不均匀、对比度低、伪影和噪声干扰以及目标边界模糊等问题,给自动分割带来了困难。本文在区域水平集模型的基础上定义反映轮廓线像素点对目标/背景两个区域隶属度的因子,通过概率分布估计模型计算和比较各像素点的隶属度,以此为依据对像素点进行区域归属判别,由区域水平集迭代获得连续光滑的曲线。本文将B超图像目标分割看作对感兴趣目标区域的局部分割,将水平集的计算求解约束到局部范围,从而减少计算量。实验结果表明与几种水平集模型相比,本文方法对所测试的B超图像的分割在精度和速度上均有一定的改进。B-type ultrasound images have important applications in medical diagnosis. However, the widely spread intensity inhomogeneity, low-scale contrast, constructed defect, noise and blurred edges all make it difficult to implement automatic segmentation of lesion in the images. Based on region level set method, a subordinate degree region level set model was proposed, in which subordinate degree probabi!.ity of each pixel was defined to ref!.ect the pixet subjection grade to target and background respectively. Pixels were classified to either target or background by calcu- lation of their subordinate degree probabilities, and edge contour was obtained by region level set iterations. In this paper, lesion segmentation is regarded as local segmentation of specific area, and the calculation is restrained to the local sphere abide by the contour, which greatly reduce the calculation complexity. Experiments on B-type ultrasound images showed improved results of the proposed method compared to those of some popular level set methods.

关 键 词：水平集方法 隶属度区域判别 B超图像 病灶边界提取 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]