Accelerating image reconstruction in three-dimensional optoacoustic tomography on graphics processing units

Abstract

Purpose: Optoacoustic tomography (OAT) is inherently a three-dimensional (3D) inverse problem. However, most studies of OAT image reconstruction still employ two-dimensional imaging models. One important reason is because 3D image reconstruction is computationally burdensome. The aim of this work is to accelerate existing image reconstruction algorithms for 3D OAT by use of parallel programming techniques. Methods: Parallelization strategies are proposed to accelerate a filtered backprojection (FBP) algorithm and two different pairs of projection/backprojection operations that correspond to two different numerical imaging models. The algorithms are designed to fully exploit the parallel computing power of graphics processing units (GPUs). In order to evaluate the parallelization strategies for the projection/backprojection pairs, an iterative image reconstruction algorithm is implemented. Computer simulation and experimental studies are conducted to investigate the computational efficiency and numerical accuracy of the developed algorithms. Results: The GPU implementations improve the computational efficiency by factors of 1000, 125, and 250 for the FBP algorithm and the two pairs of projection/backprojection operators, respectively. Accurate images are reconstructed by use of the FBP and iterative image reconstruction algorithms from both computer-simulated and experimental data. Conclusions: Parallelization strategies for 3D OAT image reconstruction are proposed for the first time. These GPU-based implementations significantly reduce the computational time for 3D image reconstruction, complementing our earlier work on 3D OAT iterative image reconstruction.

Optical and Optoacoustic Model-Based Tomography: Theory and current challenges for deep tissue imaging of optical contrast

“…For many of the algebraic inversion algorithms, such matrix sizes can be computationally prohibitive. This limitation can be mitigated via on-the-fly calculation of matrix elements [31] and utilization of model-separability in wavelet-packet domain [32].…”

Near-Infrared Fluorescence Imaging and Spectroscopy in Random Media and Tissues

“…Data digitalization was performed with two or four in-house-designed 32-channel boards containing analog-to-digital converters with a sampling rate of up to 40 MHz, field-programmable gate array (FPGA, Stratix, Altera) for signal processing, and data transfer to personal computer through the Ethernet protocol. A standard dual-core personal computer running LOIS software is used to capture the data and rapidly reconstruct optoacoustic images (<30 s/10 million voxels) using multicore GPU in a CUDA-enabled NVIDIA video card [99].…”

Optoacoustic Imaging and Tomography: Reconstruction Approaches and Outstanding Challenges in Image Performance and Quantification

“…In addition to the algorithmic complexity, additional implementation aspects include the different memory consumption requirements between algorithms that only involve standard linear algebra operations for inversion, such as BP, versus the extensive memory size needed to store matrices for some MB schemes. The graphic processing units (GPUs), which are increasingly used in medical imaging in the recent years due to their vast computational power [87,88], are expected to mitigate or significantly reduce complications concerned with computational complexities in optoacoustic imaging.…”