Attenuation correction of polychromatic L-shell X-ray fluorescence computed tomography imaging

Polychromatic L-shell X-ray fluorescence computed tomography is a promising imaging technique to explore the element distribution of the sample. However, the image quality of large samples would be decreased dramatically due to the present of the self-absorption. In this investigation, an attenuation correction algorithm based on theories related to X-ray fluorescence was proposed. With this method, the attenuation coefficients at the incident energies were expressed as a function of known X-ray energies and unknown platinum concentrations. Then the attenuation coefficients were calculated based on the theories and added in the contribution value of the pixel in Maximum Likelihood Expectation Maximization reconstruction method. The element distributions can be got through continuous iterations. Finally, the feasibility of this method was tested by Monte Carlo simulation. The results indicated that, for a 4-mm diameter phantom containing an object of 2 mm in diameter with 0.10 wt% platinum solutions, the CNR of reconstructed images increased from 53.5 to 203.8 after attenuation correction while the relative error decreases from 97% to 2%.     

C-library raft: Reconstruction algorithms for tomography. Applications to X-ray fluorescence tomography

There are many reconstruction algorithms for tomography, raft for short, and some of them are considered “classic” by researchers. The so-called raft library, provide a set of useful and basic tools, usually needed in many inverse problems that are related to medical imaging. The subroutines in raft are free software and written in C language; portable to any system with a working C compiler. This paper presents source codes written according to raft routines, applied to a new imaging modality called X-ray fluorescence tomography.

Program summary

Program title: raftCatalogue identifier: AEJY_v1_0Program summary URL:http://cpc.cs.qub.ac.uk/summaries/AEJY_v1_0.htmlProgram obtainable from: CPC Program Library, Queen?s University, Belfast, N. IrelandLicensing provisions: GNU General Public Licence, version 2No. of lines in distributed program, including test data, etc.: 218 844No. of bytes in distributed program, including test data, etc.: 3 562 902Distribution format: tar.gzProgramming language: Standard C.Computer: Any with a standard C compilerOperating system: Linux and WindowsClassification: 2.4, 2.9, 3, 4.3, 4.7External routines:

•  

  raft:   autoconf 2.60 or later – http://www.gnu.org/software/autoconf/   GSL scientific library – http://www.gnu.org/software/gsl/   Confuse parser library – http://www.nongnu.org/confuse/

raft-fun: gengetopt – http://www.gnu.org/software/gengetopt/gengetopt.htmlNature of problem: Reconstruction algorithms for tomography, specially in X-ray fluorescence tomography.Solution method: As a library, raft covers the standard reconstruction algorithms like filtered backprojection, Novikov?s inversion, Hogan?s formula, among others. The input data set is represented by a complete sinogram covering a determined angular range. Users are allowed to set solid angle range for fluorescence emission at each algorithm.Running time: 1 second to 15 minutes, depending on the data size.     

具有加速因子的OSEM重建算法用于X射线荧光CT研究

有序子集最大期望值算法(Ordered Subsets Expectation Maximization,OSEM)具有较高的图像重建质量和较短的计算时间,已经被应用于内源CT(如SPECT、PET、同步辐射X射线荧光CT)的图像重建中。本文提出了一种具有加速因子的OSEM算法应用于X射线荧光CT的图像重建,通过引入加速因子h来调制校正因子的步长加快OSEM算法的收敛速度,研究了不同加速因子和不同子集数的AOSEM算法对重建图像质量的影响。计算机模拟及实验结果表明,在获得同等质量重建图像的同时,具有加速因子的OSEM算法的重建速度是常规OSEM的两倍。