Title :
Noise and resolution properties of gamma penalized likelihood reconstruction
Author :
Wang, Wenli ; Goldstein, Charles ; Gindi, Gene
Author_Institution :
Dept. of Radiol. & Electr. Eng., State Univ. of New York, Stony Brook, NY, USA
Abstract :
In this paper, we (re)introduce the independent gamma prior for penalized likelihood reconstruction for both emission and transmission tomography. We give the fixed point expressions for bias and variance, and validate the expressions using Monte Carlo simulations. We also give a plug-in approach to get an approximate estimate of the variance image. We list the fixed point local point-spread function expression, investigate how the FWHM and volume of this local point-spread function vary with different α´s in both low and high intensity regions, and illustrate the noise-resolution trade offs
Keywords :
Bayes methods; Monte Carlo methods; conjugate gradient methods; covariance matrices; emission tomography; gamma distribution; image reconstruction; image resolution; maximum likelihood estimation; medical image processing; positron emission tomography; single photon emission computed tomography; Bayesian approach; FWHM variation; Monte Carlo simulations; PET; SPECT; bias; conjugate gradient algorithm; covariance matrix; emission tomography; fixed point expressions; gamma penalized likelihood reconstruction; high intensity regions; hyperparameters; independent gamma prior; local point-spread function; low intensity regions; noise properties; noise-resolution trade offs; plug-in approach; resolution properties; transmission tomography; variance; volume variation; Focusing; Image reconstruction; Laboratories; Poisson equations; Probability distribution; Proposals; Radiology; Smoothing methods; Tomography; Vocabulary;
Conference_Titel :
Nuclear Science Symposium, 1998. Conference Record. 1998 IEEE
Conference_Location :
Toronto, Ont.
Print_ISBN :
0-7803-5021-9
DOI :
10.1109/NSSMIC.1998.774362
