Evaluation of Image Quality From KV Cone-Beam CT with Couch Offset for Two Image Protocols
C Kim*, R Ambrose, E Furhang, Beth Israel Medical Center, New York, NYSU-E-J-16 Sunday 3:00PM - 6:00PM Room: Exhibit Hall
Purpose: To evaluate kV cone-beam CT image quality for two clinical reconstruction algorithms when the volume of interest deviates from isocenter
Methods: The CatPhan phantom (The Phantom Lab.) was scanned using CBCT on a TrueBeam LINAC (Varian Medical Systems). The phantom contains various sections designed to assess certain image quality parameters. Images were acquired with the phantom centered about the imaging isocenter (as typically done for routine CBCT QA), and then also acquired with phantom offset 5 cm laterally and longitudinally. At each position, scans were taken using two techniques typically used in the clinic: Head & Pelvis. The centered-phantom CBCT was used as a baseline, and image quality for this scan was compared against that for each offset scan. The end points of the image quality comparison were four parameters: MTF, low contrast detectability, uniformity and noise. Analysis of all CT sets were performed by the PIPSpro QA software (Standard Imaging, Inc.).
Results: For the pelvis protocol, the critical frequency (f50, lp/cm) in lateral and longitudinal offset scans was worse than that of the reference position, meaning lowered small object detectability with high contrast in offset scans. There is no significant difference found in low contrast detectability and noise level for offset scans in both Pelvis and Head protocols. The uniformity index for the lateral shift cases better than the reference image sets.
Conclusions: The results show that even though off-centered CBCT is a useful tool to handle when the volume deviates from isocenter, a precaution is needed because of a decrease in high contrast detectability. A further study of image quality comparison for couch offset will consider different acquisition parameter settings.