Abstract
A Bayesian image reconstruction algorithm, One-Step-Late (OSL) algorithm which combines maximum likelihood expectation maximization (MLEM) with maximum a posteriori (MAP) estimation, was derived by Green in 1990. In Positron emission tomography (PET), image reconstruction using OSL combines the likelihood function with image prior. Use different way to reduce the effect of noise in the data with different image prior. However, image prior may over-smooth small objects and edges. Time-of-Flight PET system can provide time difference of annihilation photon pair and this TOF characteristic can improve image resolution and signal-to-noise ratio (SNR). In this study, we proposed a TOF iterative image reconstruction algorithm based on the Bayesian scheme. Using TOF technique can improve the image quality of image which reconstructed by Bayesian image reconstruction algorithm. The results of TOF-OSL and TOF-OSEM-OSL would be compared with the results of OSL and OSEM-OSL. TOF Bayesian image reconstruction algorithm can provide not only noise reduction but also can improve the image quality in PET.