Publication Details

Full-wave nonlinear ultrasound simulation on distributed clusters with applications in high-intensity focused ultrasound

JAROŠ Jiří, RENDELL Alistair P. and TREEBY Bradley E. Full-wave nonlinear ultrasound simulation on distributed clusters with applications in high-intensity focused ultrasound. International Journal of High Performance Computing Applications, vol. 30, no. 2, 2016, pp. 137-155. ISSN 1741-2846. Available from: http://hpc.sagepub.com/content/early/2015/04/28/1094342015581024.full.pdf
Czech title
Celovlnové nelineární ultrazvukov simulace prováděné na distribuovaných clusterech aplikované v ultrazuvku o vysoké intenzitě
Type
journal article
Language
english
Authors
Jaroš Jiří, doc. Ing., Ph.D. (DCSY FIT BUT)
Rendell Alistair P. (ANU)
Treeby Bradley E. (UCL)
URL
Keywords

High Intensity Focused Ultrasound, Fourier Pseudospectral Methods, Westervelt Equation, FFTW, Large-Scale Problems, Distributed Computing

Abstract

Model-based treatment planning and exposimetry for high-intensity focused ultrasound (HIFU) requires the numerical simulation of nonlinear ultrasound propagation through heterogeneous and absorbing media. This is a computationally demanding problem due to the large distances travelled by the ultrasound waves relative to the wavelength of the highest frequency harmonic. Here, the k-space pseudospectral method is used to solve a set of coupled partial differential equations equivalent to a generalised Westervelt equation. The model is implemented in C++ and parallelised using the message passing interface (MPI) for solving large-scale problems on distributed clusters. The domain is partitioned using a 1D slab decomposition, and global communication is performed using a sparse communication pattern. Operations in the spatial frequency domain are performed in transposed space to reduce the communication burden imposed by the 3D fast Fourier transform. The performance of the model is evaluated using grid sizes up to 4096 x 2048 x 2048 grid points distributed over a cluster using up to 1024 compute cores. Given the global nature of the gradient calculation, the model shows good strong scaling behaviour, with a speed-up of 1.7x whenever the number of cores is doubled. This means large-scale simulations can be distributed across high numbers of cores on a cluster to minimise execution times with a relatively small overhead. The effiacy of the model is demonstrated by simulating the ultrasound beam pattern for a HIFU sonication of the kidney.

Published
2016
Pages
137-155
Journal
International Journal of High Performance Computing Applications, vol. 30, no. 2, ISSN 1741-2846
Book
International Journal of High Performance Computing Applications
Publisher
SAGE Publications
DOI
UT WoS
000374772700001
EID Scopus
BibTeX
@ARTICLE{FITPUB10833,
   author = "Ji\v{r}\'{i} Jaro\v{s} and P. Alistair Rendell and E. Bradley Treeby",
   title = "Full-wave nonlinear ultrasound simulation on distributed clusters with applications in high-intensity focused ultrasound",
   pages = "137--155",
   booktitle = "International Journal of High Performance Computing Applications",
   journal = "International Journal of High Performance Computing Applications",
   volume = 30,
   number = 2,
   year = 2016,
   publisher = "SAGE Publications",
   ISSN = "1741-2846",
   doi = "10.1177/1094342015581024",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/10833"
}
Back to top