Effect of Clinical Acquisition Parameters to Dose Distribution Adaptive Radiotherapy in Two Institutions
Abstract
Introduction
This work aims to examine image quality and dose based on Cone Beam Computed Tomography using phantoms in the two institutions
Materials and methods
The Rando and Catphan Phantoms were scanned using identical image acquisition and reconstruction settings with Cone Beam Computed Tomography (CBCT). The calibration curve was obtained based on the Hounsfiled Unit (HU) values on the Gammex 467 Phantom.. This study was performed at two institutions equipped with the same CBCT and CT Simulator. The acquisition with CBCT HU values that closely matched CT HU values employed a protocol of 120 kVp, Bowtie Filter (F1), and 80 mA for 10 ms within a Small Field of View (FOV). For the Medium FOV, the closest protocol also used 120 kVp, a Bowtie Filter, and 64 mA for 32 ms, while for the Large FOV, it was set at 120 kVp, F1, and 50 mA for 40 ms. The parameters chosen from each reconstructed FOV served as calibration curves for planning head and neck and pelvic treatments. Planning followed the Task Group 119 recommendations incorporating IMRT techniques.
Result
The best acquisition protocol for accurate dose calculations in head and neck cases is Field of View Small with 20 collimation using bowtie filter with 120 kVp and 80 mA for 10 ms/frame, while for Pelvic cases, it is FOVMedium with 20 collimation using bowtie filter with 120 kVp and 64 mA for 32 ms/frame. In image quality assessments, the lowest contrast visibility and uniformity scores were observed with FOV Small with 20 collimation using bowtie filter with 120 kVp and 80 mA for 10 ms/frame, using the reconstruction parameters Quality Image (QI) of 0.65% and 1%.
Conclusion
Selecting the appropriate CBCT acquisition protocol can lead to favorable outcomes in both dose calculation and image quality.
© 2026 J. Vica Intan Priliana, Sayid Mubarok, Levina Almira, Dwi Seno K Sihono, Supriyanto Ardjo Pawiro, published by Polish Society of Medical Physics
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.