ESTRO 2021 Abstract Book

S662

ESTRO 2021

Fonseca 1 , F. Verhaegen 1 1 Department of Radiation Oncology (MAASTRO), GROW – School for Oncology and Developmental Biology, Maastricht, The Netherlands; 2 medPhoton GmbH, radART Institute at the Paracelsus Medical University Salzburg, Salzburg, Austria; 3 MedAustron, Ion Therapy Centre, Wiener Neustadt, Austria Purpose or Objective Cone Beam Computed Tomography (CBCT) is a commonly used imaging method in adaptive radiotherapy to detect anatomical changes and setup errors before the treatment and to adapt the radiotherapy plan accordingly. However, CBCT imaging can lead to a large cumulative imaging dose for patients that undergo more hyperfractionated treatment schemes (≈2 Gy of total skin dose for 42 fractions1). Due to photon scatter, the CBCT image quality is still inferior compared to planning CT. In this study, we aim to address both drawbacks simultaneously with the medPhoton CBCT Imaging Ring System (IRS). The IRS is capable of performing non-isocentric image acquisitions and it can position each of its four collimation jaws independently during rotation, which enables target-specific CBCT acquisitions with optimized collimation settings. Target-optimized CBCT imaging has the potential to decrease the amount of photon scatter, while the imaging dose received by the patient is reduced. Materials and Methods Full field-of-view (FullFOV) and target-optimized limited field-of-view (LimFOV) CBCT acquisitions have been acquired for two different sites, the head and the thorax site, of an anthropomorphic phantom and for the pelvic site of a patient. The scan protocol that covers the target volume can either be defined by the lateral and anterior-posterior (AP) projections, or based on the planning CT (Figure 1 a). The selected target region is then used to optimize the collimator settings and the trajectories prior to image acquisition (Figure 1 b,c). A Monte Carlo (MC) model of the IRS was developed in Geant4 to score the imaging dose delivered by non- isocentric LimFOV and FullFOV trajectories around the anthropomorphic phantom. Extensive model validation was performed with CTDI and radiochromic film measurements. Results The acquired LimFOV scan of the patient's prostate showed improved image quality compared to the FullFOV scan (Figure 1 d,e). In the LimFOV patient scan, the scatter artifacts were reduced and both the contrast-to- noise ratio and the signal-to-noise ratio increased. We obtained similar image quality improvements for the LimFOV scans over the FullFOV for the phantom scans. Dose volume histograms displaying the dose received by the bones, lungs and the full body of the anthropomorphic thorax phantom are shown in Figure 2. The total dose received by the thorax site of the phantom is roughly 5 times higher for the FullFOV scan in comparison to the LimFOV scan. Conclusion We found that optimized target-specific CBCT acquisitions are a valuable addition to the clinical workflow as they increase image quality by decreasing scatter, while reducing the imaging dose. The increased imaging performance observed in the LimFOV acquisitions can be used clinically to focus the image acquisition on critical structures. MC simulations are a good complement for the CBCT acquisitions as they can be used to provide an independent estimation of patient imaging dose.