ESTRO 2020 Abstract book

S203 ESTRO 2020

identify FM positions. In the second automatic step, the six voxels with the lowest grey value within the areas were determined and burned-in on the pCT. Slice-thickness of CT and mDIXON was 2.5 mm, in-plane resolution 1.7x1.7x2.5 mm. For this study, 226 CBCTs were retrospectively registered (Figure 1) on the pCT with burned-in FM. These registration results were compared to the 226 clinical CBCT to CT registrations to evaluate the burned-in FM position verification accuracy. For a proper comparison the different patient position on pCT and CT need to be taken into account. We corrected CBCT-pCT registrations with an FM based registration between CT and pCT. The residual error after this correction was caused by prostate deformation and inaccuracies in determining center-of-mass (COM) positions on the different modalities, and is quantified by calculating the root mean square (RMS) of the components of the difference vector of the COM positions of all markers. After correcting for the different patient position on pCT and CT, for each strategy an average and standard deviation in translations and rotations of the FM match on the available CBCTs was calculated. These were used to determine population mean, systematic, and random error and compared to each other. Wilcoxon Signed Ranks Test was used to indicate the significantly difference of the two strategies.

OC-0351 Deep learning for rectal spacer stratification in prostate boost radiotherapy C. Thomas 1,2 , I. Dregely 2 , I. Oksuz 2 , T. Guerrero- Urbano 2,3 , A. Greener 1 , A. King 2 , S. Barrington 2 1 Guy's and St. Thomas' NHS Foundation Trust, Medical Physics, London, United Kingdom ; 2 King's College London, School of Biomedical Engineering and Imaging Sciences, London, United Kingdom ; 3 Guy's and St. Thomas' NHS Foundation Trust, Clinical Oncology, London, United Kingdom Purpose or Objective For patients undergoing radiotherapy (RT) for prostate cancer, rectal toxicity is a potential serious complication that needs to be considered. Rectal spacing devices have been shown to reduce the incidence and severity of rectal toxicity, however the cost can be prohibitive for healthcare providers. A decision support system to stratify patients based on rectal dose and toxicity risk is therefore desirable. The aim of this retrospective, simulation study was to investigate the use of a deep learning (DL) network to predict rectal dose distributions, associated rectal toxicity risk and rectal spacer requirement in patients who may be suitable for simultaneous integrated boost RT using simulated intra-prostatic lesions. Material and Methods Twenty patients with histologically-proven prostate cancer were enrolled into a Health Research Authority approved clinical study [*****]. For this simulation study, targets and OARs were re-contoured using the methodology from a national dose escalation pilot study. Six simulated lesions were created within the prostate CTV for each patient, to represent dominant intra-prostatic lesions (DILs). As a reference standard, treatment plans were created in Eclipse v13.6 TPS for the 6 DILs for each patient. A Lyman- Kutcher-Burman NTCP model was then used to predict grade 2 (G2) rectal bleeding risk. A hierarchically dense 3D U-net was trained on binary structure maps (PTV60, PTV53, DIL, rectum and bladder) labelled with associated dose distribution, and then was used to predict rectal toxicity risk, using a leave-one-out cross-validation process, and compared to the reference standard. Results Within the cohort of simulated treatment plans, DVH- derived predicted risk of G2 rectal bleeding ranged from 3.7% to 10.4%, with median risk of 7.2%. Taking delineated structures as input, the network predicted rectal toxicity risk with upper and lower 95% confidence limits of 1.3% and -1.6% respectively. Using DL, 82% of patient plans were correctly stratified for rectal spacer insertion based on a proposed protocol of offering the intervention to those with toxicity risk above the population median.

Results The group mean, systematic and random errors did not differ significantly from each other (p> 0.05) (Table 1). The residual error of the FM registration between CT and pCT was below 1 mm, the largest component is in the slice direction.

Conclusion The accuracy of IGRT in our MR-only workflow using burned-in FM is comparable to our current CT-based workflow and therefore pCT with burned-in FM is suitable as a reference image for IGRT using CBCT in clinical use.