A New Entropic Algorithm to Measure the Impact of Magnetic Field on Dose Distribution: Application to MRI-Guided Radiation Therapy

Complete text of publication follows: Purpose/Objective(s): The integration of magnetic resonance imaging (MRI), providing an efficient soft-tissue contrast, with either a linear accelerator (MR-linac) or internal radiation sources opens real perspective for a better radiotherapeutic treatment. However, the magnetic fields will modify the localization of radiation dose always induced by charged particles. The fast development of these new facilities is one of the recent big challenges for the next generation of treatment planning software. Furthermore, the demand in terms of speed and accuracy of these new modelling is more current than ever, the long term objective being to reach the real-time computation. Materials/Methods: Our theoretical physics group proposes a direct resolution of the 3D linear Boltzmann transport equation (LBTE) thanks to an angular momentum closure based on the principle of entropy maximization. This algorithm, which rests on moments method, originally developed for the energetic particles transport in magnetized plasmas is perfectly suited for modeling dose deposition for MRI-guided radiotherapy. In addition, all particles are treated simultaneously by this method. We are working on the validation of this original method through a protocol including comparisons with referent Monte-Carlo codes and with experiment campaigns on a large number of heterogeneous shapes. A significant reduction of the simulation time is observed. Results: We confirm the ability of our entropic closure to take efficiently into account magnetic effects on dose deposition for complex geometries. The robustness of our approach lying on a strong physical argument allows addition and control of complex physical effects. We also present our investigations of the changes induced by the presence of a magnetic field for heterogeneous media. We precisely describe simulations on various elementary and complex geometrical configurations providing a three dimensional analysis of these effects and the direct consequences. We confirm a significant modification (lack or increase) of the dose deposition at interfaces between different volumetric mass density materials and we are able to apply it to DICOM images extracted from real cases. Conclusion: This study illustrates that the development of this new generation of algorithms close from the original LBTE is more justified than ever. The current lack of treatment planning software for this new generation of facilities of MRI-guided radiotherapy is now obvious. Our model confirms its ability to study and model systematic differences in the dose exposure of the tumor and organs-at-risk caused by the presence of a magnetic field. If MRI-guided radiotherapy, as a perspective for real-time tumor tracking, is a real advance for modern oncology, the control of dose exposure of healthy tissue requires efficient treatment planning algorithms. Preliminary results indicate that the proposed method could be an useful tool for complex treatment plans using MRI-guided radiotherapy