Program Information
Towards High Quality Low Dose MV-CBCT Through Detector Improvements
Y Hu1*, J Rottmann1 , M Myronakis1 , R Fueglistaller2 , A Wang3 , P Huber2 , D Morf2 , J Star-Lack3 , R Berbeco1 , (1) Department of Radiation Oncology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, and Harvard Medical School, Boston, MA, (2) Varian Medical Systems, CH-5405, Baden-Dattwil, Switzerland, (3) Varian Medical Systems, Palo Alto, CA,
Presentations
WE-F-605-12 (Wednesday, August 2, 2017) 1:45 PM - 3:45 PM Room: 605
Purpose: The present study explores the path by which electronic portal imaging devices (EPIDs) may be designed to improve soft-tissue contrast in megavoltage (MV) cone-beam computed tomography (CBCT), specifically by employing a multi-layer imager (MLI) architecture.
Methods: A validated Monte Carlo (MC) simulation is employed simulating indirect-conversion, single-layer (SLI) EPIDs with varying scintillation materials, thicknesses, and exposures. Changes in modulation transfer function (MTF) and noise power spectrum (NPS) for each SLI with scintillator thickness are predicted using cubic spline fitting of MC results. NPS vs. linac output is fit to a linear trend. MTF for an MLI is the weighted average of each layer (each modeled as SLI), where weight is determined by image contrast. MLI NPS is the sum of the NPS for each layer. Readout is modeled by modifying the transfer functions by a Sinc function (quantity-squared for NPS) according to pixel-width. Reconstruction is based on a filtered backprojection (FBP) algorithm and is simulated by invoking the central-slice theorem on the projection-domain MTF and NPS.
Results: The prototype MLI features a five-fold NPS improvement over the current clinical SLI. Each MLI layer improves low-dose performance due to increases in total scintillator thickness. Additional electronic noise incurred by the combined readout of all MLI layers diminishes improvement of low-dose image quality. Because of increased electronic noise, the MLI is not quantum-noise limited at a dramatically lower dose than SLI.
Conclusion: The MLI provides a platform for overall MV-CBCT image quality improvement over the current SLI by increasing signal gain. Additional electronic noise limits low-dose performance. To push quantum-noise limited performance to acceptable levels (i.e. per-projection doses of 1-2% of typical imaging doses), imager gain must be improved while limiting electronic noise. This may be achieved by improving detector design, use of novel scintillation materials, and new image processing techniques.
Funding Support, Disclosures, and Conflict of Interest: Grant from Varian Medical Systems R01CA188446-01 from National Cancer Institute
Contact Email: