ESTRO 2021 Abstract Book

S1484

ESTRO 2021

Conclusion The quality of the palliative treatments of 120 patients treated at the dedicated outpatient clinic was evaluated in terms of motion. 19% of the patients moved considerably during the treatment with large variations between treatment sites. No relation with pain was observed. These findings stimulate individualized treatment planning margins for patients treated with palliative intent based on treatment site and expected motion. PO-1758 A bone suppression method to improve tumour motion monitoring in intensity-based 2D/3D registration I. Gulyas 1 , P. Trnkova 1 , D. Georg 1 , A. Renner 1 1 Medical University of Vienna, Department of Radiation Oncology, Vienna, Austria Purpose or Objective Real time tumour motion monitoring (TMM) is a crucial process for respiration management in lung cancer radiotherapy. Motion monitoring can be difficult as the visibility of the target is impaired when located behind bony structures. The aim of this work is a proof of concept of a novel bone suppression method for a 2D/3D marker-less based TMM algorithm. Materials and Methods A real time bone suppression method was implemented in the in-house developed software for ultra-fast intensity-based 2D/3D tumour registration (FIRE). The method operates on both, the digitally reconstructed radiograph (DRR) and X-ray images. In the first step, bony structures are obtained from CT data by thresholding. The resulting bone-DRR is subtracted from intra-fractional X-rays using ray casting to get a soft- tissue-only image, which represents the bone-suppressed input image for subsequent tumour registration. The accuracy of TMM with bone suppression was tested on a 3D printed tumour placed in an in-house developed breathing phantom (ARDOS) . A 5 cm cranio-caudal tumour trajectory was simulated with a step size of 2 mm and therefore 26 anterior-posterior X-ray images were acquired (120 kVp, 25 mA and 20 ms) on the MedPhoton imaging ring to evaluate the impact of the bone suppression. The mean intensity values of six regions of interest (ROIs) within bone segments, six ROIs in lung segments and one ROI in the tumour were used to evaluate contrast. To obtain a ground truth tumour trajectory, image acquisitions were repeated