Using correlated sampson to accelerate CT-based Monte Carlo dose calculations for brachytherapy treatment planning

A. Sampson, Y. Le, D. Todor, J. Williamson

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

The aim of this study was to assess the impact on efficiency and accuracy of a correlated sampling Monte Carlo simulation code for evaluating clinical brachytherapy dose distributions, accounting for tissue-composition and applicator heterogeneities. This code was built upon an extensively benchmarked Monte Carlo code, PTRAN-CCG and a CT-like cross-section map derived from patient single-energy CT images. Differences between PTRAN-CCG and the correlated code, expressed relative to the statistical uncertainty were found to be normally distributed with a standard deviation of 1.00 and a mean of -0.109 indicating that the correlated and conventional Monte Carlo agree within statistical error. Correlated sampling increases efficiency by factors of 7.1, 9.1, and 10.4, for greater than 20%, 50% and 90% of D90, respectively, for a 2×2×2 mm voxel grid.

Original languageEnglish
Title of host publicationWorld Congress on Medical Physics and Biomedical Engineering
Subtitle of host publicationRadiation Oncology
PublisherSpringer Verlag
Pages311-314
Number of pages4
Edition1
ISBN (Print)9783642034725
DOIs
StatePublished - 2009
EventWorld Congress on Medical Physics and Biomedical Engineering: Radiation Oncology - Munich, Germany
Duration: Sep 7 2009Sep 12 2009

Publication series

NameIFMBE Proceedings
Number1
Volume25
ISSN (Print)1680-0737

Conference

ConferenceWorld Congress on Medical Physics and Biomedical Engineering: Radiation Oncology
Country/TerritoryGermany
CityMunich
Period09/7/0909/12/09

Keywords

  • Brachytherapy
  • Correlated sampling
  • Monte Carlo

Fingerprint

Dive into the research topics of 'Using correlated sampson to accelerate CT-based Monte Carlo dose calculations for brachytherapy treatment planning'. Together they form a unique fingerprint.

Cite this