Title: Acceleration of Monte Carlo Based Treatment Planning - Criteria when Adjoint Calculations are Faster
Authors: NIEVAART Victor AlexanderMOSS Raymond
Citation: Proceedings of the 13th International Congress on Neutron Capture Therapy p. 575-578
Publisher: ENEA
Publication Year: 2008
JRC Publication N°: JRC48351
URI: http://www.enea.it/com/ingl/New_ingl/publications/neutron_capture.html
http://publications.jrc.ec.europa.eu/repository/handle/JRC48351
Type: Contributions to Conferences
Abstract: Up until now, treatment planning in Boron Neutron Capture Therapy is only performed using Monte Carlo-based techniques. The conventional radiotherapy community has become more interested in such techniques, as they are impressed by the accuracy that can be achieved with Monte Carlo calculations. However, a disadvantage of the method is still the long times needed to obtain results with reliable statistics. Although computer power has become faster and cheaper over the years, it is still impossible to calculate a hundred or more different beam positions within a few days, which is the time a treatment planner in BNCT normally needs to produce an acceptable plan. With more calculated beam positions, a better treatment plan can be composed which can maximise the dose in the tumours whilst sparing the organs at risk. In normal (forward) Monte Carlo calculations, the particles start at the beam opening and travel into the tissue where they may or may not hit a target, e.g. tumour, organ at risk. In adjoint Monte Carlo calculations, the particles start at the target and travel out of the tissue to where the information is recorded. This information can be translated as if the particle started at the place of recording. With this method, the same information is gathered as with forward Monte Carlo but instantly all around the irradiated tissue. In a realistic head phantom with 10 organs at risk and 10 tumours, the adjoint techniques are 1.8 to 3.3 times faster than the forward MC calculations when 1020 different orientations of a gamma beam with a diameter larger than 5 cm are simulated. In the case of a neutron beam, the adjoint technique is faster by 6.6 to 20 times, than the forward MC. In general, in the case of small diameter beams, adjoint MC calculations are only preferable for a large number of beams and a small number of regions of interest. For larger beam sizes, the adjoint method is more favourable than the forward calculations when there are fewer beams and/or many regions of interest.
JRC Institute:Institute for Energy and Transport

Files in This Item:
There are no files associated with this item.


Items in repository are protected by copyright, with all rights reserved, unless otherwise indicated.