Date of Award

Fall 2005

Document Type

Thesis - Restricted

Degree Name

Master of Science (MS)

Department

Electrical and Computer Engineering

First Advisor

Josse, Fabien

Second Advisor

Sabol, John M.

Third Advisor

Johnson, Michael T.

Abstract

Dual energy x-ray imaging is a technique that has been proven to enhance detectability of certain tissues by removing the undesired background clutter caused by contrasting tissues. This application takes advantage of the fact that different materials absorb different energy x-rays at rates proportional to their density and linear attenuation coefficient. Therefore, the make-up of the object of interest can be resolved by irradiating it with multiple x-ray spectrums and then digitizing the resultant detected values and applying mathematical algorithms. The possible combinations of energy spectrums, anatomic imaging tasks, and detector responses make parameter optimization difficult. While some optimization can be done experimentally, it becomes rapidly apparent that a mathematical model of dual-energy algorithms for use in the optimization of spectral input parameters for a given imaging task would be beneficial. This thesis describes the creation and validation of a spectral model for use in the simulation of a dual energy x-ray digital imaging system, with results that will be correlated to data obtained in a GE Healthcare Technologies Imaging Laboratory. The goal is to create a system, which will optimize the dual energy imaging technique with respect to dose, CNR, and energy separation. Furthermore, this paper describes the application of this model to the clinical task presented by Intravenous Urography, although the model will be applicable to other imaging applications.

Share

COinS

Restricted Access Item

Having trouble?