Effects of variance reduction techniques on the efficiency and accuracy of Monte Carlo simulation for the single source channel of Gamma Knife

• DOI: 10.3760/cma.j.issn.1004-4221.2016.08.014
• Publication Date: 2016-08-15
• Authors: Yuan Tian; Yingjie Xu; Xinxin Ren; Pan Ma; Kuo Men
• Keywords: Variance reduction

Objective To evaluate the effects of four types variance reduction techniques (ring counter grid, high electron cutoff energy, termination tracking in some structures, and emission direction-biased sampling source) on efficiency accuracy Monte Carlo simulation for single source channel Leksell Gamma Knife.

Methods The Knife was modeled using software MCNP. Four were used to simulate dose distribution water-like phantom. The computation result compared between techniques.

Results All substantially improved had little effect (relative error less than 2.5%). However, if energy above 50 keV, became quite inaccurate due neglect scattering high-energy electrons their dosimetric contribution penumbra. When penumbra ignored, Profile platform overestimated underestimated.

Conclusions Rational use can improve Knife. impact should be carefully evaluated.

Key words: Gamma knife; Monte Carlo; Computation efficiency; Variance