颈总动脉超声图像中内中膜定位方法的研究

目的:在颈总动脉的B超图像中,由计算机自动定位颈总动脉正常内中膜组织位置。方法:编制Matlab程序,在采集的B超样本图像中进行实验。基本算法思想为在颈总动脉水平径向位图像中,搜索各像素列,探查像素自上而下的灰度值变化,当检测到长距离低像素(<5)而后出现高、低和更高的像素值变化时,认为找到了内中膜位置。进行多点探测,取最佳位置。结果:选取5幅正常颈总动脉的水平径向位B超图像,程序运行结果均能够准确定位内中膜位置。结论:利用计算机在颈总动脉超声图像中自动定位正常内中膜位置是可行的,为进一步实现内中膜厚度测量、动脉硬化斑块分析的自动化提供了基础。

Study on location method of intima-media membrane in B-mode ultrasound image of

Objective： To realize the automatic location of normal intima-media membrane of common carotid artery（CCA） in B-mode ultrasonic image by using computer. Methods： To conduct experiments on the ultrasonic image samples collected from clinics by using Matlab programs. The basic procedure is that we will search each pixel column in the longitudinal CCA image and detect the variation of gray value from top to bottom. When the long distance of pixels with low gray value（less than 5） and the gray value variation of big-small-bigger are detected, the program can make the decision that the intima-media membrane is located. Multipoint measurement was taken for the best location. Results： The program can locate the intima-media membrane in the all five samples containing longitudinal CCA image. Conclusion： It’s feasible that locate normal intima-media membrane in the ultrasonic CCA image automatically by computer software, which lays the foundation of further automatic measurement of intima-media thickness and analysis of plaque ingredients.