ESTRO 2024 - Abstract Book

S4559

Physics - Machine learning models and clinical applications

ESTRO 2024

ML models can be integrated into clinical practice to provide PSQA outcome estimates for SRS treatments. Our results suggest that such models should undergo comprehensive verification to ensure their reliability and safety before clinical use. Although a regression model can provide an expected value for PSQA, the use of a classifier could offer a simpler and more intuitive indication of the risk of PSQA failure.

Keywords: machine learning, patient-specific QA, SRS

References:

[1] Lambri N, Hernandez V, Sáez J, et al. Multicentric evaluation of a machine learning model to streamline the radiotherapy patient specific quality assurance process. Phys Med. 2023;110:102593. doi:10.1016/j.ejmp.2023.102593