Abstract:
X-ray Computed Tomography imaging (CT) is the standard modality for radiation treatment
planning for cancer patients in radiation oncology. Accurate treatment plan modeling depends upon an
undistorted mapping of the CT attenuation coefficient data into tissue density. In response to the reality of
larger body habitus among patients, manufacturers have developed CT scanner systems with larger bore
diameters, i.e., 80 cm. The larger bore sizes result in significant technical challenges for manufacturers with
respect to image uniformity, geometric distortion, and artifact suppression in the reconstructed images.
Furthermore, there is an increasing number of cancer patients present with implanted medical devices or
prostheses composed of relatively high-density metals. The resulting image artifacts from these sources
must be addressed in the treatment planning process as they tend to obscure the anatomy as well as result
in potentially inaccurate dose distribution representation. The manufacturers have recently introduced
software algorithms to suppress metal artifacts in the image data. Characterization of the image information
resulting from the use of such filters is important to radiation therapy modeling. This research seeks to
investigate the image quality throughout the transverse image plane (bore) of a newly installed General
Electric (GE) large-bore (80 cm) Discovery CT590 RT scanner at the IU Health Ball Memorial Hospital,
Department of Radiation Oncology. This research will investigate image contrast, spatial resolution,
Hounsfield Units consistency, geometric distortion, and image artifact reduction at several clinically
relevant radiographic techniques for the GE scanner using a standard imagery protocol as well as when
metal artifact filter is applied. These parameters will be assessed using standard CT radiographic test tools
using tissue equivalent phantoms both with and without the presence of metal artifacts.
