Author: Michael J. Choi, Vindu Wathsala Kathriarachchi, Christopher L. Nelson, Andrew P. Soderstrom, Yawei Zhang 👨🔬
Affiliation: The University of Texas MD Anderson Cancer Center, MD Anderson, UF Health Proton Therapy Institute 🌍
Purpose: Cone Beam Computed Tomography (CBCT) is widely used in image-guided radiation therapy for patient positioning. While kV photons offer high image contrast, they are prone to artifacts caused by high-Z materials, which can be mitigated using MV photons. However, MV photons produce images with lower contrast. This study investigates the feasibility of using dual-energy CBCT (DE-CBCT), combining kV and MV photons, to enhance image quality by reducing artifacts from high-Z materials while preserving the contrast typically provided by kV photons in clinical practice.
Methods: A Winston-Lutz Cube phantom was imaged using 100 kVp and 6 MV photons on a Varian TrueBeam linear accelerator in Developer Mode. The projections were processed in MATLAB and reconstructed offline using the Reconstruction Toolkit, with kV only, MV only, and various combinations of both energy levels. Image quality was assessed by comparing the resulting images in terms of photon energy distribution, order of acquisition, and arc sector configuration. The performance of these combinations was evaluated against baseline kV images. Similar scans and evaluations were performed on other phantoms, including the CATPhan and an anthropomorphic pelvis phantom.
Results: DE-CBCT images with varying kV and MV photon compositions were successfully reconstructed. Qualitatively, MV projections significantly impacted CBCT image quality, particularly in highly heterogeneous regions, where image contrast decreased. However, MV projections still maintained acceptable contrast and image sharpness for clinical applications. The reduction of high-Z artifacts enhanced the visualization of the region of interest, leading to better resolution of fine details.
Conclusion: The development of DE-CBCT on clinically deployed systems shows promise. This approach offers an innovative solution for clinics, eliminating the need for a new machine or system retrofitting. The improved image quality has the potential to benefit patients by providing high-quality CBCT images throughout treatment.