基于凸多边形顶点检测的肝脏CT灌注图像肋骨和脊椎骨分割方法研究  

Rib and spine segmentation method based on convex polygon vertex detection in liver CT perfusion images

作　　者：陈兆学[1] 孟子渲 孟靖欣 CHEN Zhaoxue;MENG Zixuan;MENG Jingxin(School of Health Science and Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)

机构地区：[1]上海理工大学健康科学与工程学院,上海200093

出　　处：《上海理工大学学报》2022年第1期48-51,共4页Journal of University of Shanghai For Science and Technology

基　　金：上海理工大学“大学生创新创业训练计划”项目资助。

摘　　要：在参考肋骨和脊椎骨解剖学位置特征基础上,设计了高效的基于凸多边形顶点检测的肋骨和脊椎骨检测和分割方法。该方法首先针对肝脏CT灌注图像的基本特点,采用高斯滤波和最大相关准则法进行二值化;然后基于凸多边形检测的方法,对肋骨和脊椎骨二值化区域进行分割。实验结果表明,该算法有效地实现了肝脏CT灌注图像中肋骨和脊椎骨的分割。因计算量较小,并且在耗时比较大的连通区域求取和标记环节采用了快速实现方法,计算速度比较快,对于肝脏CT计算机辅助诊断系统中进一步进行肝脏分割和配准具有良好的临床应用与参考价值。An efficient method dedicated to detecting and segmenting of the rib and spine was introduced based on convex polygon vertex detection utilizing position and anatomic characteristics of the rib and spine. Firstly, according to the basic characteristics of liver CT perfusion images, the Gaussian blurring technique and maximum correlation criterion was utilized to obtain the binary image,and then the binarified regions of the rib and spine were segmented based on convex polygon vertex detection. Experimental results show that the algorithm can effectively segment the rib and spine in liver CT perfusion images. Due to relatively low computation cost together with adoption of fast implementation to the large time-consuming connected components labeling and analysis, the presented method runs with a most high speed. The method has great reference value in clinical application to further liver segmentation and registration in liver CT computer-aided diagnosis system.

关 键 词：肝脏CT灌注 凸多边形 图像分割 肋骨 脊椎骨 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]