一种基于管状特征的冠脉中心线自动提取算法 - journal of .pdf

第32 卷第1期 东 北 大 学 学 报 ( 自 然 科 学 版 ) Vol32, No. 1 2 0 1 1 年 1 月 Journal of Northeastern U ni ersity( Natural Science) Jan. 2 0 1 1 1 2 2 1, 2 王胜军 , 付 玲 , 康 雁 , 刘积仁 ( 1. , 110819; 2. , 110819) : , ; , , ; , , CTA , , : ; ; ; ; : TP 391. 72 : A : 20 11) Automatic Centerline Extraction Based on Tubular Feature for Coronary Arteries 1 2 2 1, 2 WA N G She ngj un , F U L ing , K A N G Yan , L I U Jiren ( 1. School of Information Science Engineering , Northeastern Uni ersity, Shenyang 110819, China; 2. School of SinoDutch Biomedical and Information Engineering, Northeastern Uni ersity, Shenyang 110819, China. Corresponding author : WA NG Shengjun, Email: wangshengjun @ neusoft. com) Abstract: To sol e the problem of improper centerline extraction due to the improper segmentation of coronary arteries and ob iously changed essel cur ature, an algorithm based on tubular structure w as proposed for the automatic centerline extraction of coronary arteries. The coronary artery centerline w as extracted roughly on the basis of the configuration of the skeleton nodes, and then the criteria w ere proposed for the three kinds of blood essel structures according to the tubular structure of coronary so as to detect the actualities of the improper segmentation of coronary arteries. As a result, the disconnected centerline w as regulated and joined on the basis of local ascular structure, t