AuthorEskin, Joshua Daniel, 1960-
AdvisorBarrett, 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.
AbstractSemiconductor-based gamma-ray-imaging detectors are under development for use in high-resolution nuclear medicine imaging applications. These detectors, based on cadmium zinc telluride, hold great promise for delivering improved spatial resolution and detection efficiency over current methods. This dissertation presents work done on three fronts, all directed toward enhancing the practicality of these imaging devices. Electronic readout systems were built to produce gamma-ray images from the raw signals generated by the imagers. Mathematical models were developed to describe the detection process in detail. Finally, a method was developed for recovering the energy spectrum of the original source by using maximum-likelihood estimation techniques. Two electronics systems were built to read out signals from the imaging detectors. The first system takes signals from a 48 x 48-pixel array at 500 k samples per second. Pulse-height histograms are formed for each pixel in the detector, all in real time. A second system was built to read out four 64 x 64 arrays at 4 million pixels per second. This system is based on digital signal processors and flexible software, making it easily adaptable to new imaging tasks. A mathematical model of the detection process was developed as a tool for evaluating possible detector designs. One part of the model describes how the mobile charge carriers, which are released when a gamma ray is absorbed in a photoelectric interaction, induce signals in a readout circuit. Induced signals follow a "near-field effect," wherein only carriers moving close to a pixel electrode produce significant signal. Detector pixels having lateral dimensions that are small compared to the detector thickness will develop a signal primarily due to a single carrier type. This effect is confirmed experimentally in time-resolved measurements and with pulse-height spectra. The second part of the model is a simulation of scattering processes that take place when a gamma ray is absorbed within the detector volume. A separate simulation predicts the spreading of charge carriers due to diffusion and electrostatic forces. The models are used in a technique to improve the energy resolution of the detectors by estimation of the source spectrum using the Expectation-Maximization algorithm.
Degree ProgramGraduate College