Abstract
An SPECT image can be approximated as the convolution of the ground truth spatial radioactivity with the system point spread function (PSF). The PSF of an SPECT system is determined by the combined effect of several factors, including the gamma camera PSF, scattering, attenuation, and collimator response. It is hard to determine the SPECT system PSF analytically, although it may be measured experimentally. We formulated a blind deblurring reconstruction algorithm to estimate both the spatial radioactivity distribution and the system PSF from the set of blurred projection images. The algorithm imposes certain spatial-frequency domain constraints on the reconstruction volume and the PSF and does not otherwise assume knowledge of the PSF. The algorithm alternates between two iterative update sequences that correspond to the PSF and radioactivity estimations, respectively. In simulations and a small-animal study, the algorithm reduced image blurring and preserved the edges without introducing extra artifacts. The localized measurement shows that the reconstruction efficiency of SPECT images improved more than 50% compared to conventional expectation maximization (EM) reconstruction. In experimental studies, the contrast and quality of reconstruction was substantially improved with the blind deblurring reconstruction algorithm.
Highlights
Iterative methods are commonly used in single photon emission computed tomography (SPECT) reconstruction because of their ability to handle incomplete data and to incorporate a priori information in the process
The overall point-spread function (PSF) of the gamma camera depends on the source location inside the object and the shape of the object; this PSF can be incorporated into the iterative reconstruction algorithm [2]
Image restoration and deconvolution techniques can be performed on either the projection image [3, 4] or the reconstruction image [5, 6]. The results of these techniques are often improving image quality, the PSF functions used in these techniques are either assumed to be known or are estimated by neglecting the physics of the SPECT system, and some extra artifacts might be introduced
Summary
Iterative methods are commonly used in single photon emission computed tomography (SPECT) reconstruction because of their ability to handle incomplete data and to incorporate a priori information in the process. Because the sensitivity and resolution of SPECT are complex functions of many factors, such as scattering, medium attenuation, and collimator response, it is difficult to incorporate these factors analytically into the reconstruction process. The overall PSF of the gamma camera depends on the source location inside the object and the shape of the object; this PSF can be incorporated into the iterative reconstruction algorithm [2]. Because of the complexity and the object-dependent nature of the PSF model, it is impractical to apply the exact form of the PSF directly for pinhole imaging. The combined effect of scatter and detector PSF is approximately the same as low-pass filtering of the projection image. The results of these techniques are often improving image quality, the PSF functions used in these techniques are either assumed to be known or are estimated by neglecting the physics of the SPECT system, and some extra artifacts might be introduced
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
