ESTRO 37 Abstract book

ESTRO 37

S511

better than MLC tracking dosimetrically, but had a lower duty cycle and required several couch corrections to maintain the tumor exhale position inside the gating window. PO-0940 Porcine-lung-phantom based evaluation of proton dose calculations on 4DCBCT K. Niepel 1 , C. Kurz 1 , F. Kamp 2 , D. Hansen 3 , S. Rit 4 , S. Neppl 2 , J. Hofmaier 2 , D. Bondesson 5 , C. Thieke 2 , J. Dinkel 5 , C. Belka 2 , K. Parodi 1 , G. Landry 1 1 Ludwig-Maximilians-Universität München, Department of Medical Physics, Garching, Germany 2 LMU Munich, Department of Radiation Oncology, Munich, Germany 3 Aarhus University Hospital, Department of Oncology, Aarhus, Denmark 4 INSA-Lyon, CREATIS, Lyon, France 5 LMU Munich, Department of Radiology, Munich, Germany Purpose or Objective Daily time-resolved (4D) volumetric imaging may allow detecting the degradation of proton therapy dose distributions caused by fractional changes of the breathing pattern and anatomy. For this purpose, 4D images should enable accurate proton dose calculation. Cone beam computed tomography (CBCT) scanners are now installed at proton therapy gantries for patient positioning, and allow the reconstruction of 4DCBCT images. These are however not readily suitable for dose calculation. The objective of this study was to evaluate 4DCBCT image correction based on deformable image registration (DIR) of a diagnostic quality planning 4DCT scan. Material and Methods CBCT data-sets for a moving ex-vivo porcine lung phantom with 663 and 2350 projections were acquired with an Elekta XVI CBCT system. A reference time- resolved CT scan was additionally acquired. Data were binned into 10 motion phases using the projection images and the Amsterdam shroud approach. 4D images were reconstructed using the FDK reconstruction implemented in RTK (Reconstruction Toolkit) and by solving a 4D total variation problem using an accelerated ordered subset algorithm from the Gadgetron Toolkit. DIR of the diagnostic 4DCT to the single phases of the reconstructed 4DCBCTs was performed using Reggui (morphons) and plastimatch (b-splines with mean squared error), generating a 4D virtual-CT (4DvCT) exhibiting the geometry of the CBCT and the CT numbers of the diagnostic 4DCT. The water equivalent thickness (WET) from the phantom’s surface to the distal edge of a virtual target was calculated on each phase of the 4DvCT and compared to the 4DCT. Additionally, dose calculations from a robust treatment plan (3 mm positioning uncertainty and 3% range uncertainty) optimized on a single phase were performed on all phases and compared to the 4DCT reference dose distributions. A planning 4DCT and a CBCT scan of a lung cancer patient were additionally processed in a similar fashion. Results Based on visual inspection and WET analysis, the combination of morphons DIR and iterative reconstruction was identified as most accurate when considering the 663 projections datasets. The median WET difference to the CT was 1.4 mm. For most respiratory phases, the dose calculation on the 4DvCT approached the reference 4DCT, as shown in Figure 1. The plan was optimized on phase 3, where good agreement was observed, while the

phases most different from phase 3 (5 and 7) showed the worst agreement. In general the dose differences between 4DvCT and 4DCT were smaller than the differences observed from phase to phase. The method has additionally been tested on patient data, as demonstrated in Figure 2.

Conclusion Our results suggest that correction of iteratively reconstructed 4DCBCTs using a morphons DIR algorithm enables sufficiently accurate proton therapy dose calculations to detect dose distribution degradation. Acknowledgements: DFG-MAP, BFHZ PO-0941 3D motion validation with clinically used cine-MRI and an MR-LINAC phantom K. Dolde 1,2,3 , S. Schneider 4,5 , A. Pfaffenberger 1,3 , A. Hoffmann 4,5,6 1 German Cancer Research Center DKFZ, Medical Physics in Radiation Oncology, Heidelberg, Germany 2 University of Heidelberg, Department of Physics and Astronomy, Heidelberg, Germany 3 National Center for Radiation Research in Oncology,