Mt. Chan et al., A BAYESIAN-APPROACH TO PET RECONSTRUCTION USING IMAGE-MODELING GIBBS PRIORS - IMPLEMENTATION AND COMPARISON, IEEE transactions on nuclear science, 44(3), 1997, pp. 1347-1354
We demonstrate that (i) classical methods of image reconstruction from
projections can be improved upon by considering the output of such a
method as a distorted version of the original image and applying a Bay
esian approach to estimate from it the original image (based on a mode
l of distortion and on a Gibbs distribution as the prior) and (ii) by
selecting an ''image-modeling'' prior distribution (i.e., one which is
such that it is likely that a random sample from it shares important
characteristics of the images of the application area) one can improve
over another Gibbs prior formulated using only pairwise interactions.
We illustrate our approach using simulated Positron Emission Tomograp
hy (PET) data from realistic brain phantoms. Since algorithm performan
ce ultimately depends on the diagnostic task being performed. we exami
ne a number of different medically relevant figures of merit to give a
fair comparison. Based on a training-and-testing evaluation strategy,
we demonstrate that statistically significant improvements can be obt
ained using the proposed approach. We also present a statistical verif
ication of the normality condition required for the above statistical
claim.