No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
The full text of this article is not currently available.
A fast GPU-based Monte Carlo simulation of proton transport with detailed
modeling of nonelastic interactions
1.H. Paganetti, H. Jiang, K. Parodi, R. Slopsema, and M. Engelsman, “Clinical implementation of full Monte Carlo dose calculation in proton beam therapy,” Phys. Med. Biol. 53, 4825–4853 (2008).
2.K. Parodi, A. Ferrari, F. Sommerer, and H. Paganetti, “Clinical CT-based calculations of dose and positron emitter distributions in proton therapy using the fluka Monte Carlo code,” Phys. Med. Biol. 52, 3369–3387 (2007).
5.J. Perl, J. Shin, J. Schümann, B. Faddegon, and H. Paganetti, “TOPAS: An innovative proton Monte Carlo platform for research and clinical applications,” Med. Phys. 39, 6818–6837 (2012).
6.R. Kohno, K. Hotta, S. Nishioka, K. Matsubara, R. Tansho, and T. Suzuki, “Clinical implementation of a GPU-based simplified Monte Carlo method for a treatment planning system of proton beam therapy,” Phys. Med. Biol. 56, N287–N294 (2011).
8.D. Giantsoudi, J. Schuemann, X. Jia, S. Dowdell, S. Jiang, and H. Paganetti, “Validation of a GPU-based Monte Carlo code (gPMC) for proton radiation therapy: Clinical cases study,” Phys. Med. Biol. 60, 2257–2269 (2015).
9.L. Su, T. Liu, A. Ding, and X. G. Xu, “A GPU/cuda based Monte Carlo for proton transport: Preliminary results of proton depth dose in water,” Med. Phys. 39, 3945 (2012).
10.T. Osiecki, M. Tsai, A. Gattiker, D. Jamsek, S. Nassif, W. Speight, and C. Sze, “Hardware acceleration of an efficient and accurate proton therapy Monte Carlo,” Procedia Comput. Sci. 18, 2241–2250 (2013).
13. This is roughly the expected number of non-elastic events for 1 × 107 proton histories. The CPU calculations were performed using one i7-3820 3.6 GHz processor thread.
14.S. Leray, D. Mancusi, P. Kaitaneimi, J. C. David, A. Boudard, B. Braunn, and J. Cugnon, “Extension of the Liège intranuclear cascade model to light ion-induced collisions for medical and space applications,” J. Phys.: Conf. Ser. 420, 012065 (2013).
16. In this paper, “kernel” refers to a piece of code that runs on the GPU but called from the host CPU.
17.The definitions of the terms “elastic,” “non-elastic,” and “inelastic” can be found in ICRU Report No. 63, “Nuclear data for neutron and proton radiotherapy and for radiation protection” (ICRU Publications, Bethesda, MD, 2000).
18.The International Commission on Radiation Units and Measurements, “Nuclear data for neutron and proton radiotherapy and for radiation protection,” ICRU Report No. 63 (ICRU Publications, Bethesda, MD, 2000).
19.W. Schneider, T. Bortfeld, and W. Schlegel, “Correlation between CT numbers and tissue parameters needed for the Monte Carlo simulations of clinical dose distributions,” Phys. Med. Biol. 45, 459–478 (2000).
21.L. Urban, “Multiple scattering model in geant4,” CERN Report No. CERN-OPEN-2002-070 (CERN, Geneva, 2002).
22.O. Chibani, “New algorithms for the Vavilov distribution calculation and the corresponding energy loss sampling,” IEEE Trans. Nucl. Sci. 45, 2288–2292 (1998).
23.J. Baro, J. Sempau, J. M. Fernandez-Varia, and F. Salvat, “ penelope: An algorithm for Monte Carlo simulation of the penetration and energy loss of electrons and positrons in matter,” Nucl. Instrum. Methods Phys. Res. B 100, 31–46 (1995).
24.J. Cugnon, D. L’Hôte, and J. Vandermeulen, “Simple parameterization of cross-sections for nuclear transport studies up to the GeV range,” Nucl. Instrum. Methods Phys. Res. B 111, 215–220 (1996).
26.J. Ranft, “Estimation of radiation problems around high energy accelerators using calculations of the hadronic cascade in matter,” Part. Accel. 3, 129–161 (1972).
27.S. W. Peterson, J. Polf, M. Bues, G. Ciangaru, L. Archambault, S. Beddar, and A. Smith, “Experimental validation of a Monte Carlo proton therapy nozzle model incorporating magnetically steered protons,” Phys. Med. Biol. 54, 3217–3229 (2009).
28. For example, if the thickness of the range shifter is 4.34 cm, on average the beam at the extreme edge of the scanning region would “see” a thickness of 4.38 cm.
32.X. Du, T. Lui, W. Ji, and G. Xu, “Evalulation of vectorized Monte Carlo algorithm on GPUs for a neutron eigenvalue problem,” in Proceedings of International Conference on Mathematics and Computational Methods Applied to Nuclear Science and Engineering (M&C 2013), Sun Valley, Idaho, USA, May 5–9, 2013, on CD-ROM (American Nuclear Society, LaGrange Park, IL, 2013), pp. 2513–2522.
33.NVIDIA Corporation, CURAND library, PG-05328-032_V01, 2010.
and N. Bell
, “Thrust: A parallel template library
,” version 1.7.0, 2010, available at http://thrust.github.io
35.J. Sempau, A. Sánchez-Reyes, F. Salvat, H. Oulad ben Tahar, S. B. Jiang, and J. M. Fernández-Varea, “Monte Carlo simulation of electron beams from an accelerator using penelope,” Phys. Med. Biol. 46, 1163–1186 (2001).
36. On Fermi cards, Jia et al. Ref. 7 report the calculation speed to be considerably extended by atomic operations. They use multiple dose tallies to mitigate this effect.
37.G. A. P. Cirrone et al., “Hadrontherapy: A geant4-based tool for proton/ion therapy studies,” Prog. Nucl. Sci. Technol. 2, 207–212 (2011).
38.D. A. Low, W. B. Harms, S. Mutic, and J. A. Purdy, “A technique for the quantitative evaluation of dose distributions,” Med. Phys. 25, 656–661 (1998).
39.S. V. Förtsch, A. A. Cowley, J. V. Pilcher, D. M. Whittal, J. J. Lawrie, J. C. Van Standen, and E. Friedland, “Continuum yields from 12C (p,p′) at incident proton energies of 90 and 200 MeV,” Nucl. Phys. A 485, 258–270 (1988).
40.M. M. Meier, D. A. Clark, C. A. Goulding, J. B. McClelland, G. L. Morgan, C. E. Moss, and W. B. Amian, “Differential neutron production cross sections and neutron yields from stopping-length targets for 113 MeV protons,” Nucl. Sci. Eng. 102, 310–321 (1989).
41.A. A. Cowley, ICRU Report No. 63 (ICRU Publications, Bethesda, MD, 2000).
42. Compiling with the -use_fast_math flag enables the use of faster, but less accurate functions for standard mathematical operations on NVIDIA GPUs.
43. This is our institutional research cluster, which consists of ∼300 nodes, with 16 to 32 CPU cores per node. The CPU core processor speed varied between 2.4 and 3 GHz.
44.J. Ma, H. Wan Chan Tseung, and C. Beltran, “A GPU-accelerated and Monte Carlo-based intensity modulated proton therapy optimization system,” Med. Phys. 41, 121707 (9pp.) (2014).
Article metrics loading...
Very fast Monte
simulations of proton transport have been implemented recently on graphics
processing units (GPUs). However, these MCs usually use simplified models for
nonelastic proton–nucleus interactions. Our primary goal is to build a GPU-based
transport MC with detailed modeling of elastic and nonelastic
Using the cuda framework, the authors implemented GPU kernels for the
following tasks: (1) simulation of beam spots from our possible scanning nozzle
configurations, (2) proton propagation through CT geometry, taking into account
scattering, and energy loss straggling, (3) modeling of the
intranuclear cascade stage of nonelastic interactions when they
occur, (4) simulation of nuclear evaporation, and (5) statistical error estimates
on the dose. To validate our MC, the authors
performed (1) secondary particle yield calculations in proton
collisions with therapeutically relevant nuclei, (2)
calculations in homogeneous phantoms, (3) recalculations of complex head and neck
treatment plans from a commercially available treatment planning system, and
compared with geant 4.9.6p2/TOPAS.
Yields, energy, and angular distributions of secondaries from nonelastic
collisions on various nuclei are in good agreement with the
geant 4.9.6p2 Bertini and Binary cascade models. The 3D-gamma pass
rate at 2%-2 mm for treatment plan simulations is typically 98%. The net
computational time on a NVIDIA GTX680 card, including all CPU–GPU data transfers,
is ∼20 s for 1 × 107
Our GPU-based MC is the first of its kind to include a detailed nuclear
model to handle nonelastic interactions of protons with any nucleus. Dosimetric
calculations are in very good agreement with geant 4.9.6p2/TOPAS. Our
being integrated into a framework to perform fast routine clinical QA of
pencil-beam based treatment plans, and is being used as the dose calculation
engine in a clinically applicable MC-based IMPT treatment planning system. The
detailed nuclear modeling will allow us to perform very fast linear energy
transfer and neutron
estimates on the GPU.
Full text loading...
Most read this month