**abstract:**
Phase-contrast imaging experimentally demonstrates greatly enhanced contrast over conventional attenuation-contrast imaging for biological soft tissue. Among all of the set-ups of phase-contrast imaging, the in-line holography is the instrumentally simplest method. In this talk, we discuss two reconstruction methods for holographic X-ray phase-contrast CT.
The first one is an extension of Bronnikovâ€™s work which is based on the assumption that X-ray beams travel along straight lines in the object, and Fresnel diffraction is used to describe the propagation of X-rays outgoing the object. We use the well-known Newton iterative formula to solve the nonlinear relationship between the photon intensity function and the phase coefficient function. This gives a quite straightforward nonlinear phase retrieval method to reconstruct the 3D distribution of the phase coefficient of the object.
The other method is an extension of the work by Jonas and Louis which is based on the Born approximations. We firstly establish a L^{p} error bound for Born series expansion of the scattered field, and then extend the one-order approximation reconstruction algorithms to high-order ones.
The results of computer simulations are discussed. The results show the validity of the proposed algorithms. The image quality and the applicability are enhanced to a certain extent.

Mon 15 Oct, 15:30 - 16:30, Aula Dini

<< Go back