Toward Optimal Computation of Ultrasound Image Reconstruction Using CPU and GPU

Photoacoustic tomography (PAT) offers structural and functional imaging of living biological tissue with highly sensitive optical absorption contrast and excellent spatial resolution comparable to medical ultrasound (US) imaging. We report the development of a fully integrated PAT and US dual-modality imaging system, which performs signal scanning, image reconstruction and display for both photoacoustic (PA) and US imaging all in a truly real-time manner. The backprojection (BP) algorithm for PA image reconstruction is optimized to reduce the computational cost and facilitate parallel computation on a state of the art graphics processing unit (GPU) card. For the first time, PAT and US imaging of the same object can be conducted simultaneously and continuously, at a real time frame rate, presently limited by the laser repetition rate of 10 Hz. Noninvasive PAT and US imaging of human peripheral joints in vivo were achieved, demonstrating the satisfactory image quality realized with this system. Another experiment, simultaneous PAT and US imaging of contrast agent flowing through an artificial vessel was conducted to verify the performance of this system for imaging fast biological events. The GPU based image reconstruction software code for this dual-modality system is open source and available for download from http://sourceforge.net/projects/pat realtime .

As a rapidly maturing class of many-core parallel computing hardware, graphical processing unit (GPU) has enjoyed a surge of interest in the medical ultrasound community in the past decade. In particular, the software-level programmability of GPUs has significantly lowered the entry barrier for ultrasound imaging researchers (who might not be parallel computing specialists) to pursue fast realization of novel imaging algorithms that have known theoretical potential but have yet to demonstrate their real-time feasibility. This presentation shall highlight how GPUs have emerged as a new computing workhorse in realizing various ultrasound imaging innovations. Of particular note is the enabling role that GPUs have played in fostering practical pursuit of high-frame-rate ultrasound (HiFRUS) imaging innovations that are based on direct processing of pre-beamformed radiofrequency (RF) data acquired from individual array elements. Using the state-of-art GPU technology, it is readily possible to achieve >1,000 fps HiFRUS beamforming throughput. GPU can also play a pivotal role in facilitating fast realization of computationally intensive HiFRUS algorithms, such as adaptive beamforming, color-encoded flow speckle imaging, and eigen-processing. The real-time performance of these GPU computing kernels will be discussed, and their practical implementation on software-oriented open-platform research systems will be presented.

A method and apparatus for generating ultrasound pulses comprises producing a multi-level pulse sequence including a series of pulses. The series of pulses includes at least three pulses having three different amplitudes, respectively. The amplitudes are at least one of a positive non-zero voltage, a negative non-zero voltage, and an intermediate level voltage. The multi-level pulse sequence may be produced by a switch network having at least three different input voltage levels. The switch network outputs the multi-level pulse sequence at an output node to a transducer within a probe. Echo signals based on the series of pulses are received, and an ultrasound image is produced based on the received echo signals.

To understand the mechanism of information processing by a biological neural network, computer simulation of a large-scale spiking neural network is an important method. However, because of a high computation cost of the simulation of a large-scale spiking neural network, the simulation requires high performance computing implemented by a supercomputer or a computer cluster. Recently, hardware for parallel computing such as a multi-core CPU and a graphics card with a graphics processing unit (GPU) is built in a gaming computer and a workstation. Thus, parallel computing using this hardware is becoming widespread, allowing us to obtain powerful computing power for simulation of a large-scale spiking neural network. However, it is not clear how much increased performance the parallel computing method using a new GPU yields in the simulation of a large-scale spiking neural network. In this study, we compared computation time between the computing methods with CPUs and GPUs in a simulation of neuronal models. We developed computer programs of neuronal simulations for the computing systems that consist of a gaming graphics card with new architecture (the NVIDIA GTX 1080) and an accelerator board using a GPU (the NVIDIA Tesla K20C). Our results show that the computing systems can perform a simulation of a large number of neurons faster than CPU-based systems. Furthermore, we investigated the accuracy of a simulation using single precision floating point. We show that the simulation results of single precision were accurate enough compared with those of double precision, but chaotic neuronal response calculated by a GPU using single precision is prominently different from that calculated by a CPU using double precision. Furthermore, the difference in chaotic dynamics appeared even if we used double precision of a GPU. In conclusion, the GPU-based computing system exhibits a higher computing performance than the CPU-based system, even if the GPU system includes data transfer from a graphics card to host memory.

Nonlinear frequency modulated excitation signal and modified compressing filter for improved range resolution and side lobe level of ultrasound echoes

A fast GPU based bidiagonal solver for computational aeroacoustics

Synthetic aperture Radar (SAR) has been extensively used for space-borne Earth observations in recent times. In conventional SAR systems analog beam-steering techniques are capable of implementing multiple operational modes, such as the Stripmap, ScanSAR, and Spotlight, to fulfill the different requirements in terms of spatial resolution and coverage. Future RADAR satellites need to resolve the complex issues such as wide area coverage and resolution. Digital beam-forming (DBF) is a promising technique to overcome the problems mentioned above. In communication satellites DBF technique is already implemented. This paper discuses the relevance of DBF in space-borne RADAR satellites for enhancements to quality imaging. To implement DBF in SAR, processing of SAR data is an important step. This work focused on processing of Level 1.1 and 1.5 SAR image data. The SAR raw data is computationally intensive to process. To resolve the computation problem, high performance computing (HPC) is necessary. The relevance of HPC for SAR data processing using an off-the-shelf graphical processing unit (GPU) over CPU is discussed in this paper. Quantitative estimates on SAR image processing performance comparisons using both CPU and GPU are also provided as validation for the results.

Parallel multi-dimensional interpolation in the complex Fourier domain (also known as the non-uniform fast Fourier transform) encounters major challenges, due to computation issues such as increasing computation complexity and space complexity. For instance, the contemporary graphics processing unit (GPU) is limited by the relatively small memory size and the increasing size of the interpolator. This issue makes multidimensional Fourier domain interpolation problematic, while finding an optimized configuration remains an unsolved challenge in industrial applications, e.g. magnetic resonance imaging (MRI) or computerized tomography. To enhance the performance of multi-dimensional interpolation on GPU, a new parallel hierarchical tensor products tree approach is proposed. The method combines the composite 1D interpolators under the limitation imposed by the memory size of the device. The resultant run-time performance on the GPU varies with different configurations. The best-tuned method is 2.52-4.98× faster than the compressed sparse row (CSR) on the discrete GPU and 4.16-9.59× faster than CSR on the integrated GPU. The hierarchical tensor product interpolation is used to compute the multi-dimensional nonuniform fast Fourier transform. An acceleration of 30× was achieved in 3D MRI reconstruction.

Particle-mesh interpolations are fundamental operations for particle-in-cell codes, as implemented in vortex methods, plasma dynamics and electrostatics simulations. In these simulations, the mesh is used to solve the field equations and the gradients of the fields are used in order to advance the particles. The time integration of particle trajectories is performed through an extensive resampling of the flow field at the particle locations. The computational performance of this resampling turns out to be limited by the memory bandwidth of the underlying computer architecture. We investigate how mesh-particle interpolation can be efficiently performed on graphics processing units (GPUs) and multicore central processing units (CPUs), and we present two implementation techniques. The single-precision results for the multicore CPU implementation show an acceleration of 45-70×, depending on system size, and an acceleration of 85-155× for the GPU implementation over an efficient single-threaded C++ implementation. In double precision, we observe a performance improvement of 30-40× for the multicore CPU implementation and 20-45× for the GPU implementation. With respect to the 16-threaded standard C++ implementation, the present CPU technique leads to a performance increase of roughly 2.8-3.7× in single precision and 1.7-2.4× in double precision, whereas the GPU technique leads to an improvement of 9× in single precision and 2.2-2.8× in double precision.

PCB design aids

High-performance direct gravitational N-body simulations on graphics processing units

GPU Acceleration of CFD Algorithm: HSMAC and SIMPLE

The variable preconditioned (VP) Krylov subspace method with mixed precision is implemented on graphics processing unit (GPU) using compute unified device architecture (CUDA), and the linear system obtained from the edge element is solved by means of the method. The VPGCR method has the sufficient condition for the convergence. This sufficient condition leads us that the residual equation for the preconditioned procedure of VPGCR can be solved in the range of single precision. To stretch the sufficient condition, we propose the hybrid scheme of VP Krylov subspace method that uses single and double precision operations. The results of computations show that VPCG with mixed precision on GPU demonstrated significant achievement than that of CPU. Especially, VPCG-JOR on GPU with mixed precision is 41.853 times faster than that of VPCG-CG on CPU.

Hands-on Performance Tuning of 3D Finite Difference Earthquake Simulation on GPU Fermi Chipset

Traditional methods of target velocity search based on pulse compression using GPU (Graphics Processing Unit) with single precision floating point format are difficult to meet the requirements of real-time processing. In this paper, a novel method for accelerating search process with half precision floating point format, which is based on the latest NVIDIA GPU card Tesla P100, is proposed. The scaling factor is defined to avoid the overflow issues with half precision floating point numbers. And the expression of scaling factor for LFM (Linear frequency modulation) signal is also derived. Measurement results show that about 50% promotion of computation speed is attainable with half precision floating point format when compared to single precision floating point format, which indicates the method can accelerate the searching speed efficiently.