首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Computerized quantification of carotid artery stenosis using MRA axial images
Authors:Kim Do-Yeon  Park Jong-Won
Institution:Department of Information and Communication Engineering, Chungnam National University, Taejon 305-764, Republic of Korea. dykim@ns.koepc.co.kr
Abstract:Currently, the North American symptomatic carotid endarterectomy trial, European carotid surgery trial, and common carotid method are used to measure the carotid stenosis for determining candidates for carotid endarterectomy using the projection angiography from different modalities such as digital subtraction angiography, rotational angiography, computed tomography angiography and magnetic resonance angiography. A new computerized carotid stenosis measuring system was developed using MR angiography axial image to overcome the drawbacks of conventional carotid stenosis measuring methods, to reduce the variability of inter-observer and intra-observer. The gray-level thresholding is one of the most popular and efficient methods for image segmentation. We segmented the carotid artery and lumen from three-dimensional time-of-flight MRA axial images using gray-level thresholding technique. Using the measured intima-media thickness value of common carotid artery for each case, we separated carotid artery wall from the segmented carotid artery region. After that, the regions of segmented carotid without artery wall were divided into region of blood flow and plaque. The calculation of carotid stenosis degree was performed as follows: carotid stenosis grading = (area measure of plaque/area measure of blood flow region and plaque) * 100%. No previous study has developed the carotid stenosis measuring method using MRA axial image. The new computerized stenosis measuring system has advantage over conventional caliper measuring methods; it will not only greatly increase the speed of stenosis measuring but also reduce the variability between readers. It should also reduce the variability between different institutions.
Keywords:
本文献已被 PubMed 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号