Scintillation imaging and predictive AI for real-time radiation protection in radiotherapy
Radiotherapy is one of the most effective treatments for cancer, capable of precisely targeting tumors while sparing healthy tissue. Yet, because it relies on invisible, high-energy radiation, even small deviations in beam delivery can have serious consequences for patient safety and public trust. As treatments become faster and more complex, from Intensity-Modulated Radiotherapy (IMRT) to Volumetric Arc Therapy (VMAT), Stereotactic Radiosurgery (SRS), and ultra-high dose-rate (UHDR-FLASH) therapy, the need for real-time control and verification becomes critical.
This PhD will further develop an existing system based on artificial intelligence and high-speed imaging. The idea is to bring the system to the clinic and to create a new generation of intelligent safety systems for radiotherapy. The goal is to move beyond traditional “check-after” approaches and instead predict and prevent errors while the radiotherapy treatment is being delivered. The candidate will build upon this existing prototype that monitors the radiation beam in real time, learns from data how the dose evolves, and autonomously reacts to protect patients from anomalies.
The work involves both experimental physics and AI modelling, using fast cameras, radiation detectors, and deep-learning architectures capable of operating within milliseconds. It will also include collaboration with medical physicists and clinicians to integrate the system into realistic treatment workflows.
This research offers a unique opportunity for a doctoral student interested in radiation physics, artificial intelligence, and medical technology to contribute to the future of safer, adaptive cancer treatment. The project combines cutting-edge science with a clear clinical mission: to make radiotherapy systems that can think, predict, and protect — in real time.