Boundary reconstruction in limited-angle x-ray phase-contrast tomography

Abstract

ABSTRACT We investigate an algorithm for boundary-enhanced image reconstruction from limited-angle projection data inX-ray phase-contrast tomography. The algorithm exploi ts the fact that the imaging model of phase-contrasttomography establishes a sparse representation of the o bject that facilitates accurate image reconstruction formhighly incomplete measurement data. The developed algorithm may therefore bene“t a wide range of X-rayphase-contrast tomography applications by dramatically reducing data-acquisition times and limiting radiationdose.Keywords: X-ray phase-contrast tomography, limited-angle image reconstruction 1. INTRODUCTION Propagation-based phase-contrast tomography 1–6 (PCT) is a non-interferometric, in-line, imaging techniquethat exploits a contrast mechanism based on variations in an objects X-ray refractive index distribution. Atdiagnostic X-ray energies, variations in the real component of the refractive index of tissues are several ordersof magnitude larger than variations in the imaginary component, or equivalently, the X-ray attenuation coef-“cient. Accordingly, X-ray phase-contrast imaging may permit the visualization of tissues that have similarX-ray absorption properties. Additionally, unlike absorption-contrast, the phase-contrast mechanism persists atrelatively high energies, which permits low-dose imaging. Experimental studies of X-ray phase-contrast imaginghave revealed signi“cantly enhanced contrast over conventional radiographic methods in applications includingcancer detection.