Monte Carlo modeling of direct x-ray imaging systems

Show full item record

Redirect: RIT Scholars content from RIT Digital Media Library has moved from to RIT Scholar Works, please update your feeds & links!
Title: Monte Carlo modeling of direct x-ray imaging systems
Author: Matos, Norman
Abstract: We present a new model for the image-formation processes in a direct x-ray imaging system. The imaging system is composed of an x-ray source, a phantom (target) and a direct digital x-ray imager. Image-formation for this system is defined as a cascade of nine stages: (1) x-ray source photon generation, (2) photon-phantom interaction, (3) photon absorption in detector, (4) electron-hole pair generation, (5) charge collection, (6) charge spreading, (7) pixel creation, (8) addition of electronic noise, and (9) digitization. The output of the model is a raw x-ray image of the phantom with statistical properties representative of the properties of an x-ray image that would have been produced by a real direct x-ray imaging system. Signal-to-Noise Ratio (SNR) values at different exposure levels for two types of x-ray sources, a DN9 beam and a DN5 beam, were calculated for two scenarios: (1) a step wedge target and (2) no target (flat field images). These SNR values were compared to values calculated from images captured with a real digital x-ray imaging system employing an amorphous selenium detector. It was found that SNR values calculated from the model data compared favorably to SNR values calculated from the real data. For both x-ray beams, the SNR calculated from the model flat field images was consistently higher than the SNR calculated from the real system images throughout the exposure levels. The average difference between SNR values was 6.02% and 18.07% for the DN9 and DN5 x-ray sources, respectively. For the wedge target, the average difference in SNR values was found to be 12.97% for a DN5 x-ray source. In addition, we describe an image evaluation exercise in which synthetic images produced with the model were utilized to investigate the impact of fill factor and electronic noise at different pixel sizes on the ability to detect two targets common in mammography. The first target as an acrylic disc with a 6.25 mm diameter and 1.85 mm thickness. The second target was composed of six calcium discs each with a 0.4 mm diameter and a 0.6 mm thickness.
Record URI:
Date: 2008-02

Files in this item

Files Size Format View
NMatosThesis02-2008.pdf 1.628Mb PDF View/Open

The following license files are associated with this item:

This item appears in the following Collection(s)

Show full item record

Search RIT DML

Advanced Search