17:15 - 1° Speaker: Riccardo Mirabelli
Titolo: "Beta Radio Guided Surgery: physical principles and medical application"
Abstract: "Radio Guided Surgery is a technique helping the surgeon in the resection of tumors: a radiolabeled tracer is administered to the patient before surgery and then the surgeon evaluates the completeness of the resection with a handheld detector sensitive to emitted radiation. In recent years, an innovative approach to radio guided surgery has been proposed, exploiting the intraoperative localization of beta-particle-emitting tracers (e.g., 18F-FDG, 68Ga-PSMA or 90Y-DOTATOC). The beta probe designed for this application is based on a scintillating crystal of p-terphenyl read by a silicon photomultiplier.
After laboratory and Monte Carlo tests, this technique, based on the detection of short-penetration beta particles, has been recently validated for the first time in vivo with patient with prostatic cancer and GEP-NET with encouraging results."
17:40 - 2° Speaker: Lorenzo Arsini
Titolo: “Graph Neural Networks for Radiotherapy: from CT scan to dose distribution"
Abstract: “For every Radiotherapy treatment, the essential step that precedes the delivery of radiation is the so called “treatment plan optimisation”, where beams’ direction, energy and fluence are chosen in order to fit the dose medical prescriptions. In this phase, estimating the delivered dose both quickly and accurately is crucial, but, unfortunately, current methods hardly meet both needs.
In this context, I will show how a Deep Learning model allows us to compute the dose as a smooth function of the beam's entrance point, energy and fluence, maintaining high precision, but with negligible computation times. I will also give you a taste of how such a Deep-Learning-based dose computation could also open the path for next generation differentiable optimizers."
Ambra Mariani, Antonio Iovino, Elena Pompa Pacchi, Valentina Dompè, Victor Miralles