SU-E-T-547: Influence of Phantom Voxel Size Setup On Monte Carlo Simulation in Gamma Knife Using EGSnrc.

PURPOSE: To investigate the influence of different phantom voxel size setup on the accuracy and efficiency when conducting Monte Carlo simulation of Gamma knife using EGSnrc.

METHODS: We simulated a 30mmκ60mmκ60mm size uniform water phantom and locate the phantom isocenter 70.6mm vertical down to the source in accord with measured condition for all different models. Dose calculation was executed within two different x direction voxel size setup schemes. One is completely according to measured intervals, which were 0.1mm big for all 300 voxels and the other is to divide voxels into 5 parts after picking two penumbra sections with 2mmκ4,0.5mmκ9,1mmκ5,0.5mmκ9,2mmκ4. After phantom voxel partition, Gamma Knief simulation and dose calculation, center cross section lateral profiles along X direction could be obtained for those two voxel size setup plans using EGSnrc. Those two plan profiles were plotted and compared with measured data respectively.

RESULTS: Results showed that great satisfaction can be achieved between profiles designed in 5 parts and measured data for all models. Differences can be reduced from as max as 8% for designed in measured intervals to less than 3% for designed in 5 parts compared with measured data respectively. It only takes 1/3 of designed in measured intervals time for phantom setup in 5 parts, with less than 15 mins for every simulation.

CONCLUSION: Deviding phantom voxels into 5 parts can make it more accurate and effective when conducting Monte Carlo simulation of Gamma knife using EGSnrc. Project 30970866 supported by National Natural Science Foundation of China.