Ultrasound simulation on the cell broadband engine using the westervelt equation
ICA3PP'12 Proceedings of the 12th international conference on Algorithms and Architectures for Parallel Processing - Volume Part I
Hi-index | 0.00 |
Ultrasound imaging simulations provide an important tool in imaging research. However, realistic imaging simulations involve long execution times and require enormous computational resources that exceed the capabilities of serial computers. In this paper, we present a parallel three-dimensional ultrasound simulator that runs on computer clusters and uses a k-space method to compute acoustic propagations. The simulator achieved mean error values of 0.57%, which demonstrate its numerical accuracy. Parallel computing enhanced the simulator efficiency by enabling imaging simulations with short execution times. For instance, the parallel execution time of a simulated ultrasound image using 20 cluster nodes was 18.61 hours compared to a serial execution time of 357.50 hours.