Assessing and Optimizing Pinhole SPECT Imaging Systems for Detection Tasks

Persistent Link:
http://hdl.handle.net/10150/195931
Title:
Assessing and Optimizing Pinhole SPECT Imaging Systems for Detection Tasks
Author:
Gross, Kevin Anthony
Issue Date:
2006
Publisher:
The University of Arizona.
Rights:
Copyright © 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.
Abstract:
The subject of this dissertation is the assessment and optimization of image quality for multiple-pinhole, multiple-camera SPECT systems. These systems collect gamma-ray photons emitted from an object using pinhole apertures. Conventional measures of image quality, such as the signal-to-noise ratio or the modulation transfer function, do not predict how well a system's images can be used to perform a relevant task. This dissertation takes the stance that the ultimate measure of image quality is to measure how well images produced from a system can be used to perform a task. Furthermore, we recognize that image quality is inherently a statistical concept that must be assessed for the average task performance across a large ensemble of images.The tasks considered in this dissertation are detection tasks. Namely we consider detecting a known three-dimensional signal embedded in a three-dimensional stochastic object using the Bayesian ideal observer. Out of all possible observers (human or otherwise) the ideal observer sets the absolute upper bound for detection task performance by using all possible information in the image data. By employing a stochastic object model we can account for the effects of object variability, which has a large effect on observer performance.An imaging system whose hardware has been optimized for ideal observer detection task performance is an imaging system that maximally transfers detection task relevant information to the image data.The theory and simulation of image quality, detection tasks, and gamma-ray imaging are presented. Assessments of ideal observer detection task performance are used to optimize imaging hardware for SPECT systems as well as to rank different imaging system designs.
Type:
text; Electronic Dissertation
Keywords:
Bayesian Ideal Observer; Detection Task; Optimization; Image Quality; SPECT
Degree Name:
PhD
Degree Level:
doctoral
Degree Program:
Optical Sciences; Graduate College
Degree Grantor:
University of Arizona
Advisor:
Kupinski, Matthew A.
Committee Chair:
Kupinski, Matthew A.

Full metadata record

DC FieldValue Language
dc.language.isoENen_US
dc.titleAssessing and Optimizing Pinhole SPECT Imaging Systems for Detection Tasksen_US
dc.creatorGross, Kevin Anthonyen_US
dc.contributor.authorGross, Kevin Anthonyen_US
dc.date.issued2006en_US
dc.publisherThe University of Arizona.en_US
dc.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.en_US
dc.description.abstractThe subject of this dissertation is the assessment and optimization of image quality for multiple-pinhole, multiple-camera SPECT systems. These systems collect gamma-ray photons emitted from an object using pinhole apertures. Conventional measures of image quality, such as the signal-to-noise ratio or the modulation transfer function, do not predict how well a system's images can be used to perform a relevant task. This dissertation takes the stance that the ultimate measure of image quality is to measure how well images produced from a system can be used to perform a task. Furthermore, we recognize that image quality is inherently a statistical concept that must be assessed for the average task performance across a large ensemble of images.The tasks considered in this dissertation are detection tasks. Namely we consider detecting a known three-dimensional signal embedded in a three-dimensional stochastic object using the Bayesian ideal observer. Out of all possible observers (human or otherwise) the ideal observer sets the absolute upper bound for detection task performance by using all possible information in the image data. By employing a stochastic object model we can account for the effects of object variability, which has a large effect on observer performance.An imaging system whose hardware has been optimized for ideal observer detection task performance is an imaging system that maximally transfers detection task relevant information to the image data.The theory and simulation of image quality, detection tasks, and gamma-ray imaging are presented. Assessments of ideal observer detection task performance are used to optimize imaging hardware for SPECT systems as well as to rank different imaging system designs.en_US
dc.typetexten_US
dc.typeElectronic Dissertationen_US
dc.subjectBayesian Ideal Observeren_US
dc.subjectDetection Tasken_US
dc.subjectOptimizationen_US
dc.subjectImage Qualityen_US
dc.subjectSPECTen_US
thesis.degree.namePhDen_US
thesis.degree.leveldoctoralen_US
thesis.degree.disciplineOptical Sciencesen_US
thesis.degree.disciplineGraduate Collegeen_US
thesis.degree.grantorUniversity of Arizonaen_US
dc.contributor.advisorKupinski, Matthew A.en_US
dc.contributor.chairKupinski, Matthew A.en_US
dc.contributor.committeememberBarrett, Harrison H.en_US
dc.contributor.committeememberClarkson, Eric W.en_US
dc.identifier.proquest1822en_US
dc.identifier.oclc659746372en_US
All Items in UA Campus Repository are protected by copyright, with all rights reserved, unless otherwise indicated.