Statistical Models for Anisotropic X-Ray Dark-field Tomography

Anisotropic X-ray Dark-field Tomography (AXDT) is a recently introduced imaging modality to recover anisotropic scattering functions of volumetric objects from dark-field data acquired by an X-ray grating interferometer. Current reconstruction methods implicitly assume a Gaussian distribution of the noise in the negative logarithm of the data. In this paper we derive log-likelihood functions for use in statistical reconstruction methods, based on more accurate noise models of the measured X-ray grating interferometer data. We also discuss some properties of these functions relevant to optimization.     «

Anisotropic X-ray Dark-field Tomography (AXDT) is a recently introduced imaging modality to recover anisotropic scattering functions of volumetric objects from dark-field data acquired by an X-ray grating interferometer. Current reconstruction methods implicitly assume a Gaussian distribution of the noise in the negative logarithm of the data. In this paper we derive log-likelihood functions for use in statistical reconstruction methods, based on more accurate noise models of the measured X-ray...     »