Investigation of variance reduction techniques parameters to enhance the efficiency for a 12 MV photon beam

Monte Carlo (MC) calculations used to require long CPU times, which made it prohibitive for routine clinical use. Several variance reduction techniques (VRTs) are included in the BEAMnrc MCcode to significantly improve the efficiency of photon beams for which the computation time constitutes a problematic parameter. This work focused on the optimization of VRTs parameters to improve the efficiency by BEAMnrc code. This work deals with the use of different VRTs parameters alone or combined, such as directional Bremsstrahlung splitting (DBS), uniform Bremsstrahlung splitting (UBS), photon and electron splitting, Russian roulette (RR), range rejection, and Bremsstrahlung cross section enhancement (BCSE) included in BEAMnrc code. We found that the relative improvement in efficiency due to UBS without RR, ICM_SPLIT (e-/e+), BCSE, and DBS is over 85, 130, 3489, and 4652 times higher in comparison to analog simulation for total photon fluence efficiency in 12 MV photon beam, respectively. VRTs combined in parallel simulation lead to enhance the dose efficiency by 19321 times compared to analog simulation. It is demonstrated that it is possible to further improve the efficiency by optimizing VRTs parameters combined and at the same time preserving its accuracy in parallel simulation of a 12 MV photon beam.It is demonstrated that it is possible to further improve the efficiency by optimizing VRT sparameters combined and at the same time preserving its accuracy in parallel simulation of a 12 MV photon beam