2nd ESTRO Forum 2013 S447 Conclusions: PBC calculates hot spot dose-volume less accurate than AAA in HT tissues. Therefore hot spot information is not precise in PBC plans than AAA plans and max dose constrains could be above than intended. Additionally, PBC could calculate OAR doses less than actual, which might cause over doses and morbidity in some OARs, such as spinal cord. Consequently AAA should be preferred to PBC for target including HT such as lung. Purpose/Objective: More accurate and

more »

... complex radiation techniques are required in order to improve the dose distribution over the target volume while sparing organs at risk. The development of new dose calculation algorithms which are both more accurate and faster than those used in clinical routine is required. Along with statistical methods like reduced (fast) Monte Carlo, we propose an alternative method based on a deterministic approach which provides advantage in reduced cost and improved precision. Materials and Methods: Our deterministic algorithm is based on the solution of Fokker Planck equation by using the multi-group in energy method combined with a specific angular momentum closure. For each energy-group, the equations for 2 angular moments are closed with the algebraic relation deduced from the principle of entropy minimization. (1) This method is already implemented in the plasma physics community for description of the energetic electron and photon transport combining good efficiency and precision. CELIA laboratory has developed a computing platform dedicated for validation of this algorithm for medical applications validation. Two aspects are considered : dose deposition calculation and optimization of the treatment plans. (1) Dubroca, Feugeas, Frank, Angular moment model for the Fokker-Planck equation, Eur. Phys. J. D 60 (2010) 301-307 Results: This deterministic code is compared to direct Monte-Carlo simulations using Geant4 and Penelope in the case of homogeneous and heterogeneous phantoms (water and bone media). It shows good precision in dose deposition and very short calculation time. We also present the results of an experimental campaign of sources characterizations on linear medical accelerators (energy spectrum). These data were implemented in the code for dose calculations. Conclusions: This work is the result of a multidisciplinary and transversal collaboration involving laboratories in fundamental physics, applied mathematics and cancer centers in the framework of a regional project with European financial grants started in 2011. The promising results obtained for electrons transport will be further extended to photons. The algorithm in dose optimization process is now under development.