Abstract Book

ESTRO 37

S547

magnetic resonance (MR) imaging. Hence, there is growing interest to investigate the technical feasibility of MR-integrated proton therapy (MRiPT). The aim was to operate an MRI system in the beam of a PT facility and to characterize the MR imaging performance during simultaneous irradiation. Material and Methods A 0.22 T open MR scanner (MrJ2200, Paramed Medical Systems) was installed in a compact Faraday cage at the fixed horizontal beamline of our PT facility. A beam guide in the wall of the cage allows beam transmission to the field-of-view (FOV) of the scanner. The scanner’s magnetic isocenter was aligned, such that a 10 mm diameter collimated proton beam of 125 MeV was stopped in the most distal image slice of the ACR Small Phantom, which was centrally positioned in the FOV inside a dedicated knee coil (Fig. 1). Prior to irradiation, the magnet was shimmed and the magnetic field homogeneity (MFH) was mapped over a 22 cm diameter spherical volume by a magnetic field camera (MFC3045, Metrolab). To assess the effect of magnetic fringe fields of the nearby beam line magnets, the MFH measurements were repeated while these magnets were energized for beam energies between 70-220 MeV. During irradiation, the phantom was imaged using T1 and T2-weighted spin echo (SE) sequences with parameter settings according to the phantom test guidance from the ACR. Additionally, two gradient echo (GRE S and GRE L ) scans were performed with a short repetition time (TR) and long echo time (TE): TR = 30 and 80 ms, and TE = of 8 and 30 ms, respectively, a flip angle of 20° and acquired voxel size of 0.63×0.79×5.00 mm 3 . A validated software tool (Matlab) was used to extract the ACR imaging parameters and to estimate a geometric transformation from image pairs with and without beam.

Conclusion This study demonstrates that in both cohorts, PET textural features derived using a fixed bin count are largely independent from SUV max , SUV mean , and MATV. Their independency decreases when using a fixed bin width. Clear guidelines for PET feature analysis, including SUV binning, and feature selection are necessary to avoid collinearity. The next step is to assess the prognostic value of PET textural features and its variability in multiple patient cohorts in both pre-treatment and response imaging. PO-0986 Characterization of in-beam MR imaging performance during proton beam irradiation A. Hoffmann 1,2,3 , S. Gantz 1,3 , P. Grossinger 1 , L. Karsch 1,3 , J. Pawelke 1,3 , A. Serra 4 , J. Smeets 5 , S. Schellhammer 1,3 1 OncoRay - National Center for Radiation Research in Oncology, Medical Radiation Physics, Dresden, Germany 2 Faculty of Medicine and University Hospital Carl Gustav Carus at the Technische Universität Dresden, Radiotherapy and Radiation Oncology, Dresden, Germany 3 Helmholtz-Zentrum Dresden-Rossendorf, Radiooncology- OncoRay, Dresden, Germany 4 Paramed Medical Systems, Research & Development, Genua, Italy 5 Ion Beam Applications SA, Research & Development, Louvain-la-Neuve, Belgium Purpose or Objective Given the sensitivity of proton therapy (PT) to anatomical changes, it could greatly benefit from integration with Poster: Physics track: Images and analyses

Results After shimming, the peak-to-peak MFH was 88 ppm, which is within the scanner’s operating specifications. The MFH measurements with and without energized beam line magnets showed no significant differences, but the baseline resonance frequency was increased by 70-110 Hz depending on beam energy. The SE and GRE image quality was sufficient for analysis. Differences in ACR parameters due to operating the beam line magnets or the beam were within measurement uncertainties. During empowering the beam line magnets, a sequence- dependent translation of 0.5-3 mm in frequency encoding direction was observed in the images, with GRE L being the most sensitive sequence (Fig. 2).