ESTRO 2025 - Abstract Book

S3443

Physics - Machine learning models and clinical applications

ESTRO 2025

Conclusion: The cGAN framework generated realistic estimates of measured EPID fluence, enabling the creation of synthetic maps that support gamma analysis at various criteria and visual review of failed points. After appropriate reliability and safety verification, predictive models can serve as virtual QA tools to support PSQA of SRS treatments. Supported by: GR-2019-12370739.

Keywords: radiosurgery, PSQA, synthetic fluence