S1758
Physics - Dose prediction/calculation, optimisation and applications for particle therapy planning
ESTRO 2026
Oncology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China
contrast-enhanced images.
Purpose/Objective:
Online adaptive radiotherapy allows treatment plans to be adapted to anatomical variations between fractions. Nevertheless, its clinical implementation remains highly challenging, as the entire workflow must be accomplished within a constrained timeframe. This study sought to establish and evaluate a deep learning–based workflow for online adaptive proton therapy. Material/Methods: This study incorporated three deep learning tasks: synthetic CT (sCT) generation, robust dose prediction, and plan weight prediction. Both the weight prediction and robust dose prediction were based on a 3D Attention U-Net architecture. The input data for these two models can be generated simultaneously, thereby avoiding additional preprocessing and ensuring the feasibility of clinical implementation in the adaptive workflow. The sCT generation model adopted a conditional generative adversarial network, taking CBCT as the input. The proposed adaptive workflow was implemented and evaluated for both prostate and brain cases, with comparative analysis of the contribution of each step.
Conclusion: This study demonstrates that DirectSPR on photon- counting CT accurately predicts SPR values in iodine contrast-enhanced regions for both phantom and patient scenarios, achieving dosimetric accuracy comparable to plain CT planning. This advancement streamlines clinical workflows by eliminating the need for separate non-contrast CT acquisitions or VNC conversion, reducing patient imaging dose and improving efficiency while maintaining plan quality. Keywords: PCCT, DirectSPR, iodine contrast
Proffered Paper 3803
a deep learning-driven workflow for online adaptive proton therapy: implementation and initial evaluation Bo Pang 1 , Zhiyong Yang 1 , Kunyu Yang 1 , Yu Chang 1 , Mei Chen 2 1 Cancer Center, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China. 2 Department of Radiation
Made with FlippingBook - Share PDF online