基于图像特征配准的数字血管减影算法  

Registration Based on Image Features in DSA

作　　者：周宗恒[1] 李炜[1] 曹津升[1] 倪重匡[1] 

机构地区：[1]上海复旦大学计算机系,上海200437

出　　处：《计算机工程与应用》2001年第23期169-171,共3页Computer Engineering and Applications

摘　　要：数字减影血管成象(DSA)是一种在医学诊断以及介入治疗中起重要作用的技术。在临床应用中,由于能源能量的波动、设备噪声、病人移动、器官蠕动等,通常的数字减影结果常常出现伪影。伪影消除问题成为数字减影技术中的难点,也是引起关注的热点。文章提出了消除伪影的一套新的算法———基于图像局部特征消除伪影。包括建立排除模板、控制点判定准则与方法、位移向量优化、配准方法等。该方法已用VC编制成软件,并且装备数字化X-射线装置,在计算机试验及临床实验中都取得了满意的效果。Digital subtraction angiography is an important technology in both medical diagnoses and interposal therapy.In practice,due to undulation of energy,pigeons of devices,movement of patient,squirm of organisms,artifacts often appear in the results of usual methods of digital subtraction.Elimination of artifacts has become the difficulty and hotspot of digital subtraction. In this article,we raise a new method of eliminating artifacts,which is based on the local characteristic of image,including the establishment of elimination template,the rules of judging controlling points,the optimizing of movement vector,adjusting methods,etc.This method has been realized in VC program which has been installed in x-ray medical devices with satisfying results in both experiments and practice.

关 键 词：数字血管减影算法 图像特征配准 医学诊断 计算机 

分 类 号：R445[医药卫生—影像医学与核医学] TP391.41[医药卫生—诊断学]