基于K-means聚类与改进随机游走算法的冠脉光学相干断层图像斑块分割

发布时间：2018-06-04 02:51

  本文选题：K-means聚类 + 随机游走算法　； 参考：《生物医学工程学杂志》2017年06期

【摘要】：光学相干断层成像技术(OCT)现已发展成为国内外较热门的冠状动脉内影像技术,其中冠脉OCT图像的斑块区域分割对易损斑块的识别和研究有着重大意义。本文提出了一种基于K-means聚类与改进随机游走的新算法,实现了对冠脉钙化、纤维化斑块和脂质池的半自动化分割。本文主要创新点为改进了随机游走算法的权函数,将图像中像素间的边与种子点之间的距离加入到了权函数定义中,增加了弱边界的权值,防止了过分割现象的发生。本文基于以上方法对9名冠状动脉粥样硬化患者的OCT图像进行了斑块区域分割。通过对比医生手动分割结果,证明了本文方法具有良好的精度和鲁棒性,以期本文方法可对冠心病的临床诊断起到一定的辅助作用。
[Abstract]:Optical coherence Tomography (Oct) has developed into a popular intracoronary image technology at home and abroad. The segmentation of plaque region in coronary OCT image is of great significance to the identification and study of vulnerable plaque. In this paper, a new algorithm based on K-means clustering and improved random walk is proposed to realize semi-automatic segmentation of coronary artery calcification, fibrosis plaque and lipid pool. The main innovation of this paper is to improve the weight function of the random walk algorithm. The distance between the edges of pixels and the seed points in the image is added to the definition of the weight function, which increases the weight value of the weak boundary and prevents the phenomenon of over-segmentation. Based on the above methods, OCT images of 9 patients with coronary atherosclerosis were segmented into plaque regions. By comparing the results of manual segmentation by doctors, it is proved that this method has good accuracy and robustness, and it is expected that this method can play an auxiliary role in the clinical diagnosis of coronary heart disease.
【作者单位】： 河北大学电子信息工程学院;中国医学科学院北京协和医院心内科;
【基金】：国家自然科学基金项目(61473112) 河北省自然科学基金项目(F2015201196) 教育厅科学技术研究计划(QN2015135);教育厅科学技术研究计划(QN2014166)
【分类号】：R541.4;TP391.41
，

本文编号：1975533