Ant colony method to control variance reduction techniques in the Monte Carlo simulation of clinical electron linear accelerators of use in cancer therapy

✍ Scribed by S. García-Pareja; M. Vilches; A.M. Lallena

• Publisher: Elsevier Science
• Year: 2010
• Tongue: English
• Weight: 570 KB
• Volume: 233
• Category: Article
• ISSN: 0377-0427
• DOI: 10.1016/j.cam.2008.03.052

✦ Synopsis

The Monte Carlo simulation of clinical electron linear accelerators requires large computation times to achieve the level of uncertainty required for radiotherapy. In this context, variance reduction techniques play a fundamental role in the reduction of this computational time. Here we describe the use of the ant colony method to control the application of two variance reduction techniques: Splitting and Russian roulette. The approach can be applied to any accelerator in a straightforward way and permits the increasing of the efficiency of the simulation by a factor larger than 50.