Skip to main content
Skip to "About this site"
Skip to section menu
Canada.ca
Services
Departments
Language selection
Français
Search and menus
Search and menus
Library and Archives Canada
Search
Topics menu
Discover the Collection
Search the Collection
Services for the Public
Services and programs
You are here:
Home
Services and Programs
Theses Canada
Search - Theses Canada
Renewing our web presence
Search - Theses Canada
Page Content
OCLC number
46553891
Author
Wang, Hui,1964-
Title
Development of a portal imager and of tools for radiation treatment verification.
Degree
Ph. D. -- McGill University, 1996
Publisher
Ottawa :National Library of Canada = Bibliothè€que nationale du Canada,1997.
Description
3 microfiches.
Notes
Includes bibliographical references.
Abstract
A prototype electrostatic imager has been developed for megavoltage portal imaging in radiation therapy. The imager utilizes amorphous selenium (a-Se) with a front metal plate as the imager receptor and a high resolution voltmeter probe for image readout. Imaging characteristics of a-Se have been investigated theoretically through Monte Carlo simulations, and experimentally by measuring radiation discharging curves and phantom tests. The results of this study have shown that the prototype imager has high sensitivity, good spatial resolution and low noise level. Our study also reveals the potential of electrostatic imaging with metal/a-Se in megavoltage imaging. Two computer algorithms have also been developed for automatic segmentation and contrast-enhancement of digital portal images, and for radiation field shape verification. Based on a priori knowledge of the properties of portal images, the segmentation and contrast-enhancement algorithm employs multiple criteria and dynamic reasoning to achieve optimal segmentations of individual images, and has been proved to be accurate, robust and fast. The algorithm for radiation field shape verification is an adaptation of the chamfer matching technique to a specific application: matching closed contours. By incorporating geometric features of the radiation field and using a simple minimization method which is more specific to this task, the algorithm appears to be able to improve the matching results of the standard method.
ISBN
0612197832
9780612197831
Date modified:
2020-01-02
Primary navigation (left column)
Theses Canada
Search Theses Canada
Search help for Theses Canada