A survey of GPU-based acceleration techniques in MRI reconstructions

Abstract

Image reconstruction in magnetic resonance imaging (MRI) clinical applications has become increasingly more complicated. However, diagnostic and treatment require very fast computational procedure. Modern competitive platforms of graphics processing unit (GPU) have been used to make high-performance parallel computations available, and attractive to common consumers for computing massively parallel reconstruction problems at commodity price. GPUs have also become more and more important for reconstruction computations, especially when deep learning starts to be applied into MRI reconstruction. The motivation of this survey is to review the image reconstruction schemes of GPU computing for MRI applications and provide a summary reference for researchers in MRI community.

Simultaneous multi slice (SMS) balanced steady state free precession first-pass myocardial perfusion cardiovascular magnetic resonance with iterative reconstruction at 1.5 T

“…While iterative reconstruction can significantly improve CMR image quality, such an approach is computationally intensive compared to standard reconstruction [ 45 ]. Techniques to reduce reconstruction time such as by use of a graphics processing unit are feasible, have been applied to CMR data with iterative reconstruction [ 46 ], with substantial increase of reconstruction speed [ 47 ]. Hence, rapid iterative reconstruction of SMS data may be feasible with dedicated hardware and thereby facilitate implementation into routine clinical care.…”

Tolerogenic Iron Oxide Nanoparticles in Type 1 Diabetes: Biodistribution and Pharmacokinetics Studies in Nonobese Diabetic Mice

“…T1D development is invariably associated with changes in vascular permeability in pancreatic blood vessels. [40] This increased microvascular permeability is observed already before high blood glucose diagnosis in prediabetic mice. We hypothesized that nanoparticulate contrast agents might accumulate in inflamed pancreatic islets via the EPR effect, inducing changes in tissue contrast.…”

Positive‐contrast susceptibility imaging based on first‐order primal‐dual optimization

“…Parallel computing is a type of classic computation to speed up the computer speed and graphics processing unit (GPU)‐based platforms have been used to make high performance parallel computing available. Recently, GPU‐based acceleration techniques have attracted much attention and are successfully applied in MRI reconstruction . The GPU‐based technique can also be applied in positive‐contrast susceptibility imaging to improve computational efficiency.…”