Joint estimation of image and deformation parameters in motion-corrected PET

We consider an emission tomography reconstruction problem in which projection measurements from several time frames are available. Two strategies for doing motion-corrected image reconstruction are compared. In the first strategy, separate images are reconstructed from the measurements at each time frame. They are then consolidated by post-registration and averaging procedures. In the second strategy, we incorporate parameters to describe the effects of motion into the statistical model of the projections. Joint maximum likelihood estimation of image and motion parameters is then carried out. Each of these strategies involves the minimization of non-convex cost functions. Accordingly, we also propose some relevant optimization algorithm design options.