ESTRO 2023 - Abstract Book

S105

Saturday 13 May

ESTRO 2023

Results Submillimeter differences in positional breast reproducibility and stability were found between both arms. (p < 0.001 for non-inferiority). The left anterior descending artery near-max dose (14,6±12,0 Gy vs. 7,7±7,1 Gy, p=0,018) and mean dose (5,0±3,5 Gy vs. 3,0±2,0 Gy, p=0,009) were significantly improved in the MANIV-DIBH arm. The same applied for the V5Gy of the left ventricle (2,4±4,1 % vs. 0,8±1,6 %, p=0,001) as well as for the left lung V20Gy (11,4±2,8 % vs. 9,7±2,7 %, p=0,019) and V30Gy (8,0±2,6 % vs. 6,5±2,3 %, p=0,0018). Tolerance and treatment time were comparable. Conclusion MANIV provides better OARs sparing while reaching the irradiation accuracy of SGRT. MO-0143 Patch-based deep learning automatic organ segmentation for online adaptive prostate radiotherapy W. Mukaidani 1 , T. Shiinoki 2 , Y. Yuasa 1 , K. Fujimoto 2 , Y. Kawazoe 2 , Y. Ishihara 3 , A. Sawada 4 , Y. Manabe 2 , M. Kajima 2 , H. Tanaka 2 1 Yamaguchi University Hospital, Department of Radiological technology, Ube, Japan; 2 Yamaguchi University, Department of Radiation Oncology, Ube, Japan; 3 Japanese Red Cross Wakayama Medical Center, Department of Radiation Oncology, Wakayama, Japan; 4 Kyoto College of Medical Science, Faculty of Medical Science, Nantan, Japan Purpose or Objective Online adaptive radiotherapy (O-ART) is being introduced into clinical practice. For O-ART, treatment planning should be replanned using daily verification image because of the anatomical information change. Therefore, routine usage is limited due to the time-consuming task of manual segmentation and its accuracy variation among observers. This study proposes patch-wised (PW)-U-net-based automatic pelvic CT segmentation models for prostate radiotherapy planning and validate its performance compared with the conventional non-PW U-net model (C-U-net). Materials and Methods One-hundred patients who underwent radiotherapy for locally advanced prostate cancer were enrolled in this study. For all patients, CT scan was performed whole pelvis. The contours of organs including prostate, bladder and rectum delineated by the radiation oncologist for radiotherapy planning were registered as the ground truth segmentations. The dataset was split into training (n = 70) and test (n = 30) subsets. Figure 1 shows the workflow of this study. For the PW-U-net, the input CT and the corresponding bladder segmentation maps with 256 x 256 pixels were divided into 4 and 16 patches with 128 x 128 and 64 x 64 pixels, respectively. The prostate and rectum segmentation maps with 128 x 128 pixels were divided into 4 and 16 patches with 64 x 64 and 32 x 32 pixels, respectively. For each data, 2D-U-net model was constructed using the training data and, then verified with the test data. Finally, the divided images of predicted segmentation were concatenated to original image . To compare the segmentation performance between the 4-PW-U-net, 16-PW-U-net and C U-net, dice similarity coefficient (DSC) and Hausdorff distance (HD) between predicted and ground truth segmentation for prostate, bladder and rectum were evaluated. Mini-Oral: Trends in dosimetry planning