基于快速Hessian矩阵的MSCT图像血管增强（英文）  

Enhancement of MSCT Vascular Image Based on Fast Hessian Matrix

作　　者：王丽瑶 汪毅[1] 陈晓冬[1] 赵聪[1] 张佳琛[1] 贾忠伟[2] 陈向志[3] 郁道银[1] Wang Liyao;Wang Yi;Chen Xiaodong;Zhao Cong;Zhang Jiachen;Jia Zhongwei;Chen Xiangzhi;Yu Daoyin(Key Laboratory on Opto-Electronics Information Technology of Ministry of Education, School of Precision Instrument and Opto-Electronics Engineering, Tianjin University, Tianjin 3 0 0 0 7 2 ,China;Department of Cardiovascular Medicine, Pe o pled Liberation Army 254 Hospital, Tianjin 300142 , China;Department of Radiology, PeopleJ s Liberation Army 254 Hospital, Tianjin 300142 ,China)

机构地区：[1]天津大学精密仪器与光电子工程学院光电信息技术教育部重点实验室,天津300072 [2]中国人民解放军第二五四医院心血管内科,天津300142 [3]中国人民解放军第二五四医院放射科,天津300142

出　　处：《纳米技术与精密工程》2018年第1期41-47,共7页Nanotechnology and Precision Engineering

摘　　要：针对传统的Frangi管状结构滤波方法存在的计算量大、耗时长的问题,本文使用一种基于快速Hessian矩阵的滤波算法对MSCT图像进行冠状动脉增强滤波.通过对Hessian矩阵多项式系数的条件分析,预识别出不满足条件的元素为非血管元素,可省略对该部分元素的Hessian矩阵特征值求解,从而降低计算量,减少滤波耗时.实验结果表明,在保证有效评价图像中管状结构相似度的情况下,该方法可将滤波耗时平均降低至传统Frangi算法的30.63%.Conventional filtering methods of the Frangi tubular structure is restricted by large computa-tional complexity as well as time-consuming issue. In this paper, a filtering algorithm based on fast-Hes- sian matrix is applied to solve the above problems, which can perform enhancement filtering on the coro-nary artery in the MSCT images. After analyzing the conditions of the Hessian matrix polynomial coeffi-cients ,the elements that fail to meet the predefined conditions are picked out and classified as non-vascu- lar elements, which will be neglected during the process of calculating the Hessian matrix eigenvalue. Thus, both the computational complexity and the filtering time wi ll be reduced. Experimental results show that this algorithm can reduce the average filtering computational time to 30. 63% of the traditional Frangi algorithm, with the similarity of the tubular structure in the effective evaluation image well pre-served.

关 键 词：Frangi滤波 血管增强 冠状动脉 三维分割 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]