Joint Image and Motion Reconstruction for Positron Emission Tomography

Abstract

Patient motion is one of the main contributing factors to image degradation in PET imaging. If ignored in the reconstruction algorithm, motion can lead to blurring in the reconstructed image and thus be a severe limiting factor for correct diagnosis or quantification. Motion compensation methods currently used for cardiac PET imaging involve gating, followed by static image reconstruction of individual gated data. The main disadvantage of such techniques is that they trade motion blurring for image noise: the less motion blurring present within a gate, the more gates are needed, reducing the data in each gate. As the data present in each gate is reduced, a corresponding increase in noise in the reconstructed image is observed.