CEA-List developed Phoebe, a powerful computing code, to simulate the way photons, electrons, and positrons interact with matter or human tissue. Built around a Monte Carlo algorithm, the code is modular and can be implemented on most commercially available IT systems.
Phoebe calculates the amount of radiation received by a material—such as human tissue in medical use cases—by simulating the interaction of ionizing radiation with matter. Phoebe can be used in many areas of applied nuclear physics, from medical physics to radiation protection, nuclear instrumentation, and ionizing radiation metrology.
For example, the code can be used to calculate and validate the mathematical correction to apply to experimental measurements, or to calculate the dose of radiation received by a patient during a radiotherapy session. Phoebe can also be used to calculate the dimensions of biological protection systems and radiation detectors.
Phoebe leverages the Monte Carlo method, widely recognized as the most precise way of modeling interactions between radiation and matter. It uses validated physics models from Penelope, an existing Monte Carlo code widely used in the medical field. Phoebe can be adapted and extended by adding modules, for variance reduction or to handle additional geometries, for instance.
Phoebe is also designed to be deployed on most commercially available IT systems, a notable advantage in terms of flexibility.
Phoebe offers several major advantages, including:
Monte Carlo simulations are widely used in radiotherapy to calculate the dose of radiation received by a patient in the area around the target tumor. However, this approach cannot provide a timely estimation of the dose received by healthy tissue outside of the target area, due to strong attenuation of the beam in the collimation system.
Built on an existing method, CEA-List researchers developed an approach in which particles are artificially introduced into out-of-field zones, reducing the calculation variance and dramatically reducing simulation times. The result obtained in this way is then corrected mathematically to establish an estimated dose.
The new method was implemented in Phoebe and experimentally tested. In clinical settings, our approach may be used to study the long-term effects of out-of-field radiation doses, with the end goal of improving treatment practices.
Ines a portable and parallel Monte-Carlo simulation code of class-II for electromagnetic showers, J. Garcia Hernandez, A. Croc. Présentation conférence MCMA 2019.
Calcul accéléré de la dose périphérique en radiothérapie, thèse de Valentin Champciaux, 2021.
A breakdown of the pseudo-deterministic transport variance reduction method: Formalization and usage considerations, Valentin Champciaux, Juan Carlos Garcia Hernandez, Mathieu Agelou. Computer Physics Communications, Elsevier, 2021, 264, pp.107979.