Abstract
A novel positron emission tomography (PET) scanner design based on a room-temperature pixelated CdTe solid-state detector is being developed within the framework of the Voxel Imaging PET (VIP) Pathfinder project [1]. The simulation results show a great potential of the VIP to produce high-resolution images even in extremely challenging conditions such as the screening of a human head [2]. With unprecedented high channel density (450 channels/cm3) image reconstruction is a challenge. Therefore optimization is needed to find the best algorithm in order to exploit correctly the promising detector potential. The following reconstruction algorithms are evaluated: 2-D Filtered Backprojection (FBP), Ordered Subset Expectation Maximization (OSEM), List-Mode OSEM (LM-OSEM), and the Origin Ensemble (OE) algorithm. The evaluation is based on the comparison of a true image phantom with a set of reconstructed images obtained by each algorithm. This is achieved by calculation of image quality merit parameters such as the bias, the variance and the mean square error (MSE). A systematic optimization of each algorithm is performed by varying thereconstruction parameters, such as the cutoff frequency of the noise filters and the number of iterations. The region of interest (ROI) analysis of the reconstructed phantom is also performed for each algorithm and the results are compared. Additionally, the performance of the image reconstruction methods is compared by calculating the modulation transfer function (MTF). The reconstruction time is also taken into account to choose the optimal algorithm. The analysis is based on GAMOS [3] simulation including the expected CdTe and electronic specifics.
Highlights
Details about the design, geometry, electronics and the simulation of expected performance of the Voxel Imaging positron emission tomography (PET) (VIP) scanner based on pixelated CdTe detectors is presented in [2]
The modulation transfer function (MTF) was calculated by taking the 1-D Fourier transform (FT) of a line profile recorded through the center of the point spread function (PSF) in the direction of the profile
region of interest (ROI) analysis Once the best reconstruction parameters are chosen based on the results shown in the subsection 4.1, the ROI analysis is performed
Summary
Geometry, electronics and the simulation of expected performance of the VIP scanner based on pixelated CdTe detectors is presented in [2]. A very low scatter fraction of 3.95% (according to the NEMA NU 2-2001 standard [5]) is achieved due to the good energy resolution provided by the CdTe detectors. The choice of reconstruction technique is one of the crucial factors in getting a good final image. For this reason the designers of PET systems often develop their own algorithm (or modify existing algorithms) to adapt it to the specifics of the scanner. In case of the VIP PET scanner, we test and evaluate a few standard methods to pick the optimal one in terms of image quality and reconstruction time
Sign-in/Register to view highlights & summary 
Published Version
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