Abstract #0103
Hierarchically Semiseparable Generalized Encoding Matrix Compression for Fast Distortion Corrected Inverse Imaging
Stephen F Cauley 1,2 , Kawin Setsompop 1,2 , Dan Ma 3 , Yun Jiang 3 , Elfar Adalsteinsson 4 , Lawrence Wald 1,2 , and Mark Griswold 3,5
1
Athinoula A. Martinos Center for Biomedical
Imaging, MGH/HST, Charlestown, MA, United States,
2
Dept.
of Radiology, Harvard Medical School, Boston, MA, United
States,
3
Dept.
of Biomedical Engineering, Case Western Reserve
University, Cleveland, Ohio, United States,
4
Harvard-MIT
Div. of Health Sci. and Tech., Dept. of Electrical
Engineering and Computer Science, Cambridge, MA, United
States,
5
Dept.
of Radiology, Case Western Reserve University and
University Hospitals of Cleveland, Cleveland, Ohio,
United States
Reconstruction of non-Cartesian data can be a
computationally demanding problem. Iterative numerical
solutions often involve repeated evaluation of Discrete
Fourier or NUFT operators, coil sensitivity profiles,
and other physical MR parameters. Alternatively,
Hierarchically Semiseparable (HSS) modeling can be used
to compute an approximate inverse of the generalized
encoding matrix. The HSS model can be computed prior to
data collection and is ideal for time series
reconstruction, e.g. fMRI, cardiac imaging, and MR
fingerprinting. We demonstrate a 40x speed-up when
compared to state-of-the-art iterative solvers for the
reconstruction of distortion corrected spiral data.
This abstract and the presentation materials are available to members only;
a login is required.
Join Here
