Model based patient pre-selection for intensity-modulated proton therapy (IMPT) using automated treatment planning and machine learning

Highlights: • A novel tool for pre-selection of head&neck patients for proton therapy was developed. • The overall pre-selection prediction accuracy using machine learning was 87%. • This tool can be used to screen larger patient populations for proton therapy. • This tool can help to avoid delays in the start of (proton) radiotherapy. Patient selection for intensity modulated proton therapy (IMPT), using comparative photon therapy planning, is workload-intensive and time-consuming. Pre-selection aims at avoidance of manual IMPT planning for patients that are in the end ineligible. We investigated the use of machine learning together with automated IMPT treatment planning for pre-selection of head and neck cancer patients, and validated the methodology for the Dutch model based selection (MBS) approach.