Program Information
Comparison of Varian Advanced CBCT Reconstruction and Standard Clinical TrueBeam CBCT Reconstruction
A Katsis*, E Abel , Varian Medical Systems, Palo Alto, California
Presentations
SU-H4-GePD-I-4 (Sunday, July 30, 2017) 4:30 PM - 5:00 PM Room: Imaging ePoster Lounge
Purpose: The ability to perform treatment planning using an on-board cone beam CT (CBCT) scanner is a necessity if adaptive planning is to be considered in the future. Currently, treatment planning is performed with diagnostic CT scans, while on-board CBCT is used solely for patient positioning during treatment. This study explores the feasibility of and provides a quantitative evaluation of a novel advanced reconstruction algorithm, as it compares to a standard clinical TrueBeam CBCT, and diagnostic CT reconstruction.
Methods: Scans of a Catphan 504 phantom, were taken on a bench top CBCT scanner with TrueBeam imaging geometry and reconstructed using both the standard and advanced CBCT methods. To determine the true CT number for each material insert, the spatially dependent x-ray spectrum after filtration was determined through a Monte Carlo simulation of the TrueBeam imaging chain up to the detector in TOPAS. The CT number accuracy and standard deviation for each material were then compared for both reconstructions. Finally, a scan of the same calibration phantom taken on a diagnostic CT scanner, the gold standard, was compared to the CBCT scans.
Results: The Varian advanced CBCT reconstruction method was found to reduce the CT number uncertainty, as compared to the standard TrueBeam CBCT reconstruction, from greater than ±30 HU to less than ±5 HU in some cases. The diagnostic CT reconstruction method was found to produce CT number uncertainties of approximately ±10 HU.
Conclusion: We have demonstrated a quantitative improvement in the CT number accuracy and precision of the Varian advanced CBCT algorithm. The improved accuracy is heading in the direction for use of setup CBCT scans to help inform time-of-treatment planning decisions.
Funding Support, Disclosures, and Conflict of Interest: Alexander Katsis and Eric Abel are employees of Varian Medical Systems.
Contact Email: