基于非均匀螺旋线数据和布雷格曼迭代的快速磁共振成像方法  

A new fast magnetic resonance imaging method based on variabledensity spiral data acquisition and Bregman iterative reconstruction

作　　者：方晟[1] 吴文川[2] 应葵[3] 郭华[2] 

机构地区：[1]清华大学核能与新能源技术研究院,北京100084 [2]清华大学生物医学工程系生物医学影像研究中心,北京100084 [3]清华大学工程物理系,北京100084

出　　处：《物理学报》2013年第4期502-508,共7页Acta Physica Sinica

基　　金：国家自然科学基金(批准号:81101030)资助的课题~~

摘　　要：数据采集时间长是制约磁共振成像技术发展的重要瓶颈.为了解决这一问题,本文基于压缩感知成像理论,提出了一种结合非均匀螺旋线磁共振数据采集序列和布雷格曼迭代重建的快速磁共振成像方法,通过欠采样缩短数据采集时间.欠采样引起混迭伪影则通过非均匀螺旋线欠采样特性和布雷格曼迭代重建去除.水模磁共振成像实验和在体磁共振成像实验结果表明:欠采样情况下,所提出的方法能有效去除欠采样导致的混迭伪影,获得的图像结构信息完整的成像结果,在缩短采样时间的同时,具有较高的准确度.Data acquisition time is a bottle neck for increasing imaging speed of magnetic resonance imaging. To solve the problem, anew fast magnetic resonance imaging method based on variable-density spiral acquisition and Bregman iterative reconstruction is proposed in this paper, under the framework of compressed sensing. The proposed method increases the acquisition speed by data undersampling. The resulting undersampling aliasing artifact is removed by utilizing the intrinsic property of variable-density spiral and Bregman iterative recosntruction. The proposed method is validated by both phantom experiemnt and in vivo experiment. The experimental results demonstrate that the proposed method can effectively remove aliasing artifact from data undersampling, and achieve an image with well-preserved image structure information. Therefore this method can be used for reducing data acquisition time.

关 键 词：磁共振成像 非均匀螺旋线 全变分 布雷格曼迭代 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] O482.531[自动化与计算机技术—计算机科学与技术]