Bayesian Reconstruction Using A Novel Nonlocal MRF Prior for PET Transmission Tomography

In positron emission tomography, transmission scans can be performed to estimate attenuation correction factors (ACF) which are in turn used to correct the emission scans. Such an attenuation correction is crucial for quantitatively accurate PET reconstructions. Long scan times, although alleviating the noise effect of the count-limited scans, are blamed for patient uncomfortableness and movement. So transmission tomography often suffers from the noise effect because of the short scan time. Thus, reconstruction which is capable of overcoming the noise effect is highly needed. In this article, based on Markov random fields (MRF) theory, we put forward a novel nonlocal prior for PET transmission tomography. Comparing to the traditional local prior, the proposed prior can exploit the nonlocal information of the objective images and impose an effective regularization on the reconstruction problem for PET transmission tomography.