ESTRO 2025 - Abstract Book

S3377

Physics - Machine learning models and clinical applications

ESTRO 2025

Conclusion: The MMDP model, trained through instance weighting anatomy-to-function training, can extract complementary features from multi-modality inputs, predicting dose distributions close to high-quality FLART plans. The MMDP based auto-planning algorithm can achieve lung dose painting, producing high-quality FLART plans by leveraging voxel-wise lung function information from various imaging techniques. It shows promises in promoting FLART planning consistency, quality, and efficiency.

Keywords: FLART, multi-modality learning, dose prediction

References: 1. Baschnagel AM, et al. A Phase 2 Randomized Clinical Trial Evaluating 4-Dimensional Computed Tomography Ventilation-Based Functional Lung Avoidance Radiation Therapy for Non-Small Cell Lung Cancer. International Journal of Radiation Oncology*Biology*Physics. 2024;119(5):1393-1402. 2. Yamamoto T, et al. Four-Dimensional Computed Tomography Ventilation Image-Guided Lung Functional Avoidance Radiation Therapy: A Single-Arm Prospective Pilot Clinical Trial. International Journal of Radiation Oncology*Biology*Physics. 2023;115(5):1144-1154. 3. Xiong T, et al. Automatic planning for functional lung avoidance radiotherapy based on function-guided beam angle selection and plan optimization. Physics in Medicine & Biology. 2024;69(15):155007.