Iterative SENSE reconstruction
In this clip,
we demonstrate how iterative image reconstruction methods
can be used to reconstruct images from undersampled MRI data
from multiple receive coils.
Ordinarily, such undersampling leads to gross image distortions
(see the images at the first iteration),
but our iterative reconstruction method,
used in conjunction with information
about receiver coil sensitivity,
converges to high quality images.
These methods can allow images
to be acquired more quickly
for functional brain imaging studies using MRI.
Click here to download the movie.
Joint work with Brad Sutton and Doug Noll.
Details.
The k-space trajectory has 2363 data points,
with 4us sampling,
roughly corresponding to a 32 x 32 image in a 11cm FOV.
The reconstructed images are 64 by 64, FOV 22 cm.
Back to Fessler home page