System Calibration and Image Reconstruction for a New Small-Animal SPECT System
AdvisorBarrett, Harrison H
Committee ChairBarrett, Harrison H
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractA novel small-animal SPECT imager, FastSPECT II, was recently developed at the Center for Gamma-Ray Imaging. FastSPECT II consists of two rings of eight modular scintillation cameras and list-mode data-acquisition electronics that enable stationary and dynamic imaging studies. The instrument is equipped with exchangeable aperture assemblies and adjustable camera positions for selections of magnifications, pinhole sizes, and fields of view (FOVs). The purpose of SPECT imaging is to recover the radiotracer distribution in the object from the measured image data. Accurate knowledge of the imaging system matrix (referred to as H) is essential for image reconstruction. To assure that all of the system physics is contained in the matrix, experimental calibration methods for the individual cameras and the whole imaging system were developed and carefully performed. The average spatial resolution over the FOV of FastSPECT II in its lowmagnification (2.4X) configuration is around 2.4 mm, computed from the Fourier crosstalk matrix. The system sensitivity measured with a ⁹⁹ᵐTc point source at the center of the FOV is about 267 cps/MBq. The system detectability was evaluated by computing the ideal-observer performance on SKE/BKE (signal-known-exactly/background-known-exactly) detection tasks. To reduce the system-calibration time and achieve finer reconstruction grids, two schemes for interpolating H were implemented and compared: these are centroid interpolation with Gaussian fitting and Fourier interpolation. Reconstructed phantom and mouse-cardiac images demonstrated the effectiveness of the H-matrix interpolation.
Degree ProgramOptical Sciences