ESTRO 2020 Abstract Book

S887 ESTRO 2020

Results 3D errors based on 15 CBCT’s per immobilization type are shown in Fig. 2a, while 1D group mean (M) & group systematic (S) errors in the sagittal plane are shown in Fig. 2b. The setup accuracy achieved with 3DPrIm was at least comparable with TM for the occiput & mandible. 3DPrIm setup accuracy was reduced for the more caudal sub-ROI’s (C6 & larynx), due to the current absence of lower neck immobilization. All patients and RTTs evaluated the 3DPrIm as usable and patient-friendly.

Purpose or Objective In MR-guided radiotherapy, contours of the target are propagated for each fraction to perform online daily adaptations. Conventional methods use deformable image registration which is often too slow for online use during treatment sessions. In this study, we investigate fast organ contour propagation from pre-treatment to fraction images in MR-guided prostate radiotherapy with convolutional neural networks (CNNs) that are trained for combined image registration and contour propagation. Material and Methods Five prostate cancer patients underwent twenty fractions of image-guided external beam radiotherapy on a 1.5T MR- Linac system (Unity, Elekta AB). For each patient, a pre- treatment 3D T2-weighted TSE MRI was used to delineate the CTV. The same scan was repeated during each fraction, with CTV contours being manually adapted if necessary. A 3D-CNN was trained in a supervised manner for combined image registration and contour propagation on a synthetically generated ground truth of randomly deformed images and prostate contours. The network estimated the propagated contour and a deformation field between the two input images (Fig1). The training set consisted of a synthetically generated ground truth of 15,000 randomly deformed images and prostate contours. We performed a leave-one-out cross-validation on the five patients and propagated the contours from the pre- treatment to the fraction scans. Three variants of the CNN, aimed at investigating supervision based on maximizing contour overlap, minimizing the error in the deformation vector fields over the field-of-view and, and a combination of the two were compared to a previously validated deformable registration approach (Elastix), by computing contour overlap, Hausdorff distances, and the prostate centroid distance for each of the propagated contours.

Conclusion 3DPrIm is feasible in terms of workflow, is tolerable by HNC patients and can achieve clinically acceptable setup accuracy for the head region. 3DPrIm therefore is a valid option to pursue enhancements in RT of HNC, such as spot- size reducing bolus in IMPT. Further optimization of the initial design, however, with added lower neck immobilization, is required towards clinical use of 3DPrIm in RT of HNC. PO-1624 Fast contour propagation for MR-guided prostate radiotherapy using convolutional networks K.A. Eppenhof 1,2 , M. Maspero 2 , M.H. Savenije 2 , H.C. De Boer 2 , J.R. Van der Voort van Zyp 2 , B.B. Raaymakers 2 , A.J. Raaijmakers 1,2 , M. Veta 1 , J.P. Pluim 1 , C.A. Van den Berg 2 1 Eindhoven University of Technology, Biomedical Engineering, Eindhoven, The Netherlands ; 2 UMC Utrecht, Radiation Oncology, Utrecht, The Netherlands

Results The neural networks trained on contour overlap or the combined objective achieved significantly better Hausdorff distances between predicted and ground truth contours than Elastix, with an additional much shorter time duration of 0.5 seconds ( Tab 1, Fig2).