Compressive sensing (CS) has been shown to enable dramatic acceleration of MRI acquisition in some applications. Being\r\nan iterative reconstruction technique, CS MRI reconstructions can be more time-consuming than traditional inverse Fourier\r\nreconstruction. We have accelerated our CS MRI reconstruction by factors of up to 27 by using a split Bregman solver combined\r\nwith a graphics processing unit (GPU) computing platform. The increases in speed we find are similar to those we measure for\r\nmatrix multiplication on this platform, suggesting that the split Bregman methods parallelize efficiently. We demonstrate that the\r\ncombination of the rapid convergence of the split Bregman algorithm and the massively parallel strategy of GPU computing can\r\nenable real-time CS reconstruction of even acquisition data matrices of dimension 40962 or more, depending on available GPU\r\nVRAM. Reconstruction of two-dimensional data matrices of dimension 10242 and smaller took ~0.3 s or less, showing that this\r\nplatform also provides very fast iterative reconstruction for small-to-moderate size images.
Loading....