ESTRO 2025 - Abstract Book

S4323

RTT - Treatment planning, OAR and target definitions

ESTRO 2025

Fig.2: Heatmaps of dosimetric differences between ATP and manual planning. Conclusion:

The current work has brought the automation of radiotherapy treatment planning to a real-world practice. The multi-institutional assessments show that the proposed ATP solution is noninferior to human-driven planning across multiple disease sites. It is demonstrated that the single-institutional DL models can be adapted to meet the needs of external institutions by adjusting the tolerance of predicted clinical goals. More efforts will be devoted to further improve the acceptability and applicability of the auto-plans in various contexts. The clinically accessible and instantly deliverable ATP solution will lead to higher efficiency and potentially more homogenous plans, and provide a foundation for facilitating online replanning in personalized adaptive radiotherapy. References: 1. Meyer P, Biston MC, Khamphan C, et al. Automation in radiotherapy treatment planning: Examples of use in clinical practice and future trends for a complete automated workflow. Cancer Radiother . 2021;25(6-7):617-622. 2. Liu R , Bai J , Zhao K ,et al.A New Deep-Learning-based Model for Predicting 3D Radiotherapy Dose Distribution In Various Scenarios[C]//2020 13th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI). 2020, pp. 748-753 3. Cornell M, Kaderka R, Hild SJ, et al. Noninferiority Study of Automated Knowledge-Based Planning Versus Human Driven Optimization Across Multiple Disease Sites. Int J Radiat Oncol Biol Phys . 2020;106(2):430-439. Keywords: online auto-planning, muti-center validation