胸部高分辨率CT片中肺实质的自动分割  

Automatic Segmentation of Pulmonary Parenchymain Thoracic High Resolution CT

作　　者：陈旭[1] 庄天戈[1] 

机构地区：[1]上海交通大学生物医学工程系,上海200030

出　　处：《上海交通大学学报》2002年第7期946-948,共3页Journal of Shanghai Jiaotong University

基　　金：上海市科技发展基金资助项目 ( 0 0 41190 11)

摘　　要：以早期弥漫性肺疾病的高分辨率 CT图片为研究对象 ,针对图片本身的特点 ,提出了一种新的肺部全自动分割法 .该方法结合了阈值技术和区域生长技术 .它首先将待处理源图像二值化 ,并根据用 Sobel算子抽取出的躯干边缘 ,得到纯粹的躯干部二值图像 .在中值滤波去噪后 ,对其中黑白区域分别进行连通域标记 ,并将面积较小区域 (如气管等 )的像素值取反 .最后再经形态滤波得到肺部精确的模板 .该算法能自动去除气管 /支气管对应的图像区域 。A novel segmentation method was proposed to extract the pulmonary parenchyma automatically from the HRCT of early diffuse lung disease (DLD). This method took the characteristics of the HRCT images into consideration and combined thresholding, region growing as well as morphology techniques together. It first selected a CT number threshold to make the original image a binary one and dealt it with a torso mask that derived from the contours extracted with a Sobel operator. After denoising with a median filter, it then labeled the white/black connectivity components in that binary image respectively and negated the value of pixels inside the regions with smaller areas. It subsequently filterse the image with a morphologically close operator and got a perfect lung mask finally. This fast and easy implementing method can eliminate the regions of the trachea/bronchi effectively and automatically. What's more, both the sensitivity to the selection of threshold value and necessitated interaction are reduced to the least extent.

关 键 词：胸部高分辨率CT片 肺实质 自动分割 图像分割 阈值法 区域生长 连通域标记 中值滤波 

分 类 号：R814.42[医药卫生—影像医学与核医学] TP391.41[医药卫生—放射医学]