An Iterative Image Reconstruction for Differential X-ray Phase-contrast Computed Tomography

Abstract

The last decade saw the growth of the use of X-ray Phase-contrast Computed Tomography (PCT) imaging. Differential X-ray Phase-contrast CT (DPCT) has shown bright prospect for clinical use in recent years, but how to best reconstruct the phase-contrast tomographic image from incomplete or inconsistent projections remains challenging. We designed an iterative image reconstruction method for DPCT utilizing compressed sensing. This proposed method reconstructs the phase-contrast tomographic image of the refractive index decrement gradient with acceptable accuracy and image quality even when the projection data is undersampled or noisy. Real data experiments indicate this proposed method is accurate and practical and outperforms the analytic filtered back-projection reconstruction. This method shows potential for X-ray DPCT facilities in low dose tomography and fast acquisition imaging.