| Unlike computed tomography(CT) and magnetic resonance imaging(MRI) that provide great details in anatomical structures, PET provides metabolic and functional information about a particular region of interest(ROI). Similar to other nuclear imaging techniques, the image quality of PET is often limited by the spatial resolution of a system, the counting statistics of a study and the effectiveness of a radiopharmaceutical for targeting the function of interest. Therefore, the image quality of PET can be improved by the combination of improved PET instrumentation, reconstruction algorithms and choice of appropriate radiopharmaceutical.The rapid growth in genetics and molecular biology combined with the development of techniques for genetic engineering has led to an increased interest in in vivo small animal imaging. The imaging environment for small animal PET studies is generally very favorable. Scatter and attenuation are reduced, count rates are modest, and fewer detectors are required to cover a given solid angle. This allows more flexibility in the detector design which is reflected in the varied approaches to animal PET systems currently being considered. In 2003, Tai et al. proposed and tested a novel geometry for PET named PET-Insert system in which one or more high-resolution detector modules are integrated into a conventional PET scanner with lower resolution detectors. Later on, this geometry activated the design of small animal PET as well. In this paper, we simulated and validated the geometry of PET-Insert system via GATE, a generic Monte Carlo simulation platform based on Geant4.Virtual-pinhole PET(VP-PET) is the earliest PET-Insert system. It can locally enhance the spatial resolution and contrast recovery near the add-on detectors, and depending on the configuration, may also increase the sensitivity of the system. This novel scanner geometry makes the reconstruction problem more challenging compared to the reconstruction of data from a stand-alone PET scanner, as new techniques are needed to model and account for the non-standard acquisition. In this dissertation, we also simulated a prototype small animal PET designed as a PET-Insert system for animal imaging and conducted the reconstruction with two algorithms: FBP(Filtered Back-Projection) and PML-EM(Penalized Maximum Likelihood-Expectation Maximization). A quantitative comparison of PML-EM and FBP was performed as well on data acquired from insert–scanner coincidences using Derenzo phantom and Shepp-Logan phantom.Monte Carlo simulations studies are nowadays an essential tool in a number of applications in the evolving area of emission tomography: modeling of imaging systems, developing and assessing of tomographic reconstruction algorithms and evaluating correction methods for improved image quantification. Geant4 Application for Tomographic Emission(GATE) is a generic Monte Carlo simulation platform designed to meet the specific needs of PET and SPECT applications, based on the Geant4 libraries, a well-established code for radiation transport. Besides the features provided by Geant4, GATE also involves specific modules necessary to perform realistic simulations, such as complex source distributions and the management of time-dependent processes(detector and source movements, radioactive decay, dynamic acquisition).In the simulation, demonstrating the coincidence simultaneously and quickly between two kinds of detector rings, PET scanner and insert detector, is an unexpected breakthrough owing to the recent update of GATE by OpenGATE collaboration. The results showed well that images of dot sources have good quality reconstructed with FBP, and phantom study suggested the improved iterative algorithm can deal with noise effectively, resulting in no artifact in the image. These are kept in a good agreement with the theoretical and experimental results. Therefore, the PET-Insert system is a promising design approach for small animal PET. |
PDF Full Text Request