Penalized-likelihood image reconstruction for x-ray fluorescence computed tomography

摘要: X-ray fluorescence computed tomography (XFCT) allows for the reconstruction of distribution elements within a sample from measurements x rays produced by irradiation with monochromatic synchrotron radiation. XFCT is not transmission modality, but rather stimulated emission modality; thus correction attenuation incident and photons essential if accurate images are to be obtained. This challenging because map is, in general, known only at stimulating beam energy various energies interest. We make use empirically fitted analytic expressions x-ray coefficients express unknown maps as linear combinations quantities elemental concentrations then develop an iterative image algorithm based on penalized-likelihood methods that have been developed medical tomography. Studies numerical phantoms indicate approach able produce qualitatively quantitatively reconstructed even face severe attenuation. also apply method real synchrotron-acquired data demonstrate marked improvement quality relative filtered backprojection reconstruction.