Positron Emission Tomography (PET) is an established functional imaging modality but its practical deployment is hampered by motion artefacts. This paper proposes a method for correcting reconstructed PET images through a list-mode rebinning process. Respiratory motion is compensated for by applying 2D affine transforms to the respiratory gated sequence of line-ofresponse events prior to image reconstruction. The affine transforms are derived from a non-rigid 3D/3D registration model derived from retrospectively gated MRI acquisitions of the same patient with free-breathing. The proposed motion correction scheme is evaluated with a simulation framework where the improvement in image quality is assessed by using data acquired from 10 normal subjects.
Adrian James Chung, Paolo G. Camici, Guang-Zhong Y