计算机工程

• 图形图像处理 • 上一篇    下一篇

基于折射角的稀疏投影角度相衬CT 图像重建

司 凯1,孙丰荣1,宋尚玲2,秦 峰1   

  1. (1. 山东大学信息科学与工程学院,济南250100; 2. 山东大学第二医院设备部,济南250033)
  • 收稿日期:2014-03-03 出版日期:2015-03-15 发布日期:2015-03-13
  • 作者简介:司 凯(1989 - ),男,硕士研究生,主研方向:医学图像处理;孙丰荣,教授;宋尚玲,博士;秦 峰,硕士研究生。
  • 基金项目:
    国家自然科学基金资助项目(61071053);山东省自然科学基金资助项目(ZR2014FM006)。

Sparse Projection Angle Phase Contrast CT Image Reconstruction Based on Refractive Angle

SI Kai  1,SUN Fengrong  1,SONG Shangling  2,QIN Feng  1   

  1. (1. School of Information Science and Engineering,Shandong University,Jinan 250100,China; 2. Department of Equipment,The Second Hospital of Shandong University,Jinan 250033,China)
  • Received:2014-03-03 Online:2015-03-15 Published:2015-03-13

摘要: 已有基于X 射线吸收衬度机制的计算机断层成像(CT)技术很难对由轻元素构成的弱吸收物质进行高质量成像。X 射线相位衬度CT 成像是对弱吸收物质具有超高分辨率的一种CT 技术,但该技术成像时间长、所需X射线辐射剂量大,不利于临床推广应用,因此,研究稀疏投影角度条件下的X 射线相位衬度CT 图像重建问题,基于压缩感知图像重建理论,使用折射角信息减少X 射线辐射剂量,提出一种X 射线相位衬度CT 图像重建算法。 实验结果表明,与滤波反投影算法相比,该算法在稀疏投影角度下可以得到较高质量的重建图像,在实际数据实验中能获得较高的峰值信噪比和数值准确性。

关键词: X 射线相衬, 计算机断层成像, 图像重建, 压缩感知, 折射角

Abstract: It is not easy for conventional X-ray absorption contrast based Computed Tomography ( CT ) imaging technique to get a high quality image from the weak absorption materials comprising of light elements. X-ray phase contrast CT imaging is a CT imaging technique aiming at weak absorption materials and possessing ultrahigh resolution. However,X-ray phase contrast CT imaging takes too long time and requires too high X-ray radiation dose,which can set back the promotion of clinical applications. So it has significant meaning to study the image reconstruction for phase contrast CT using few projection views. This paper is under the theory of Compressed Sensing(CS),uses refractive angle information,aims to reduce the X-ray radiation dose,proposes a X-ray phase contrast CT image reconstruction algorithm. Experimental results show that compared with Filtered Back-projection (FBP) algorithm,this algorithm can use few projection views to obtain reconstructed images of higher quality,and can get higher Peak Signal to Noise Ratio(PSNR) and numerical accuracy in the actual data experiment.

Key words: X-ray phase contrast, Computed Tomography (CT ), image reconstruction, Compressed Sensing (CS ), refractive angle

中图分类号: