ESTRO 2021 Abstract Book

S1453

ESTRO 2021

nozzle position: 17 – 24 – 31 – 38 - 41.3 cm, couch rotation 0° - 180° - 270°). The effect of partial surface loss on the registration accuracy (TRE) was investigated using an Iterative Closest Point algorithm.

Results The five-camera system improved the visibility and symmetry compared to the installed configuration for all gantry, couch and nozzle configurations, and patients combined (p < 0.001 for both). Specifically for the closest nozzle position the five-camera system outperforms the installed system (p < 0.001) as shown in figure 2.

Conclusion An SGRT system specifically designed for a compact gantry proton therapy system can improve the visibility of the patients’ surface during treatment and should be further explored to improve the SGRT use in proton therapy. PO-1729 Classification 、 GPR and dose difference prediction for patient-specific QA based Unet++ Y. Huang 1,1 1 Shanghai Chest Hospital, Shanghai Jiao Tong University, Department of Radiation Oncology, Shanghai, China Purpose or Objective To investigate the Unet++ to classify fail or pass plans, predict gamma passing rates (GPR) for different gamma criteria, and predict dose difference for virtual patient specific quality assurance. Materials and Methods 109 IMRT plans were selected including 568 fields. All plans used Portal Dosimetry for dose verification pre- treatment. Planar dose distribution of each field was used as input for Unet++, QA classification results, gamma passing rates of different gamma criteria, and dose difference were used as output.pre-treatment. Planar dose distribution of each field was used as input for Unet++, QA classification results, gamma passing rates of different gamma criteria, and dose difference were used as output.