MAXIMUM A POSTERIORI RECONSTRUCTION OF PATLAK PARAMETRIC IMAGE FROM SINOGRAMS IN DYNAMIC PET

Parametric imaging using Patlak graphical method has been widely used to analyze dynamic PET data. The conventional way to generate Patlak parametric image is to reconstruct dynamic images first and then perform Patlak graphical analysis on the time activity curves pixel-by-pixel. In this paper we present a Bayesian method for reconstructing Patlak parametric images directly from raw sinogram data by combining the Patlak plot model with image reconstruction. A preconditioned conjugate gradient algorithm is used to find the maximum a posteriori solution. We conduct computer simulations to validate the proposed method. The comparison with conventional indirect approaches shows that the proposed method results in more accurate estimate of the parametric image