Abstract Book

S960

ESTRO 37

adipose tissue. MC predicted images show that anatomical details of real patients can be obtained using an image acquired with the photons emitted from a HDR 192 Ir source. A clinical trial for prostate and gynecological patients is in preparation. EP-1789 Sensitivity analysis of EPID-based 3D dose reconstruction for VMAT QA A. Alhazmi 1 , C. Gianoli 1 , S. Neppl 2 , J. Martins 1 , S. Veloza 1 , M. Podesta 3 , F. Verhaegen 3 , M. Reiner 2 , C. Belka 2 , K. Parodi 1 1 Ludwig-Maximilians-Universität München, Department of Medical Physics- Faculty of Physics, Garching, Germany 2 LMU Munich, Department of Radiation Oncology, Munich, Germany 3 Maastricht University Medical Center MUMC, Department of Radiation Oncology MAASTRO, Maastricht, The Netherlands Purpose or Objective The goal of this study is to investigate the sensitivity of an EPID-based 3D dose reconstruction algorithm to detect geometric and dosimetric errors of VMAT plans for A VMAT plan for head & neck case was generated using the TPS Monaco®. The plan was delivered using ELEKTA synergy® linac equipped with PerkinElmer® aSi EPID. The acquired frame images were used to reconstruct 3D dose distribution with 1 mm isotropic resolution using an in- house developed algorithm. The algorithm converted the acquired frame images into planar dose distribution and back-projected them upstream into a numerically modeled cylindrical water phantom. The planar dose distributions of the water phantom were deconvolved by depth-specific scatter and attenuation kernels, thus reconstructing a 3D dose distribution. The kernels were obtained by making use of scatter and attenuation models to iteratively estimate the parameters from a set of reference measurements. The obtained parameters served as a look-up table for reconstruction of arbitrary measurements. For each frame image, the reconstructed 3D dose distribution was rotated in accordance with the correlated projection angle of the gantry. A sum of the reconstructed 3D dose distributions resulted in the integrated 3D dose distribution of the VMAT delivery. For geometric sensitivity testing, first, two selected neighboring leaves were shifted by 2 mm in all control points of the original plan file. Second, a controlled rotational shift of the gantry angles in the plan files of 1 degree was applied. For dosimetric sensitivity testing, the total monitor units of the original plans were increased by 4%. The resulting 3D dose distributions were compared to the original plan using the well-known gamma evaluation with (3%, 3 mm) as acceptance criteria. Results Figure (1) shows different views of the original 3D dose distributions calculated by the TPS and the reconstructed 3D dose distribution. The gamma evaluation between the plan and the unmodified acquisition was fulfilled by exceeding the 95% of the acceptance criteria. The modified acquisitions resulted in decreased passing rates of 93%, 91% and 83% of the acceptance criteria for the shifted leaves, gantry angle and added dose, respectively. Figure (2) (a, c and d) show the gamma evaluations of the unmodified and modified acquisitions at the iso-center plane. In Figure (2) (b) the effects of the leaves shift at the corresponding slice is illustrated. Quality Assurance (QA). Material and Methods

EP-1788 A complete solution for HDR brachytherapy measurements G.P. Fonseca 1 , M. Podesta 1 , R. Voncken 1 , M.R. Van den Bosch 1 , B.G.L. Vanneste 1 , L. Lutgens 1 , F. Verhaegen 1 1 Department of Radiation Oncology MAASTRO- GROW- School for Oncology and Developmental Biology- Maastricht University Medical Center, Physics, Maastricht, The Netherlands Purpose or Objective Dose measurements in brachytherapy require high positioning accuracy due to the very steep dose gradients. Daily QA and applicator commissioning are a heavy burden in the clinic and an accurate in vivo dose verification method is not available. This work proposes the use of an Imaging Panel (IP) as a complete solution for brachytherapy measurements ranging from daily QA, applicator commissioning up to in vivo dosimetry. Material and Methods Two different IPs (Varex and Perkin Elmer) were calibrated using a HDR 192 Ir source with a robotic arm for accurate positioning. The IPs were used to acquire 192 Ir gamma-ray images of ring applicators, register the source movement inside the applicator and to fully verify a treatment plan with multiple catheters (3D Cartesians coordinates, dwell time and air kerma strength). Additionally, a Monte Carlo (MC) model capable of predicting IP responses during a brachytherapy treatment (Figure 1a) was created and verified using heterogeneous phantoms (Figure 1b).

Results The time-resolved applicator commissioning method has 0.2 mm accuracy (2D coordinates). Dwell positons can be visualized on top of a projection of the applicator and compared against the treatment planning system. Measurements with the applicator within a water phantom (Figure 1c) can detect positioning errors larger than 1 mm (3D coordinates), swapped catheters and dwell time errors (≥0.1 s). Measurements with heterogeneous phantom showed a difference ranging from 1.6% between adipose tissue and muscle up to 11.2% between bone and muscle. Figure 1d shows a predicted IP image (MC) for a head and neck patient (base of tongue tumor). The highlighted region (rectangle) shows a high intensity region due to the dwell position and anatomical information mostly related to the spine and air cavities. Conclusion The proposed system can be used for commissioning applicators (imaging and dosimetry) with an accurate, efficient and time-resolved method. Moreover, the same system can be used for in vivo dosimetry. Experimental results shows that the IP can detect differences in tissue composition even for soft tissues such as muscle and