Program Information
Verification of the Simplified Monte Carlo Method for Simulation in An Inhomogeneous Phantom
S Mizutani1*, Y Takada1 , R Kohno2 , K Hotta2 , T Akimoto2 , (1) Graduate School of Pure and Applied Sciences, University of Tsukuba, Tsukuba, Ibaraki, (2) National Cancer Center Hospital East, Kashiwa, Chiba
Presentations
SU-E-T-805 (Sunday, July 12, 2015) 3:00 PM - 6:00 PM Room: Exhibit Hall
Purpose:
A simplified Monte Carlo (SMC) method has been developed to obtain fast and accurate dose calculation in heterogeneous tissues to improve the accuracy of dose calculation. We have applied the SMC method to calculation of dose kernels for the pencil beam scanning. While the SMC method tracks individual primary protons in medium, it simplifies the dose calculation by using a measured depth-dose distribution instead of considering nuclear reactions and tracking secondary particles. To verify the accuracy of SMC calculation, we compared a dose-calculation result using the SMC method with that using the Full Monte Carlo (FMC) method in an inhomogeneous phantom.
Methods:
As a model of the inhomogeneous media, we considered a phantom complied of a rectangular acrylic block of size 150*300*200 mm³ immersed in water within a virtual container with a size of 300*300*400 mm³.
Results:
We found excellent agreement of overall dose distributions between the SMC and FMC methods. As for the laterally integrated depth-dose distributions, slight difference was found in front of the second Bragg-Peak between the two algorithms. While it took 25717.0 s on 2.7 GHz Intel Core i7 for the FMC method to complete the calculation, it took 15.4 s for the SMC method to complete it. The SMC method can calculate the dose distribution approximately 1675 times faster than the FMC method.
Conclusion:
The dose distribution obtained by SMC method agreed well that obtained by the FMC method in a simple inhomogeneous phantom. In contrast, calculation time by the SMC method has been reduced by three orders in magnitude compared with the FMC method.
Contact Email: