deepmr.linops.NUFFTOp

deepmr.linops.NUFFTOp#

class deepmr.linops.NUFFTOp(*args: Any, **kwargs: Any)[source]#

Non-Uniform Fast Fourier Transform operator.

K-space sampling trajectory are expected to be shaped (ncontrasts, nviews, nsamples, ndims).

Input images are expected to have the following dimensions:

  • 2D MRI: (nslices, nsets, ncoils, ncontrasts, ny, nx)

  • 3D MRI: (nsets, ncoils, ncontrasts, nz, ny, nx)

where nsets represents multiple sets of coil sensitivity estimation for soft-SENSE implementations (e.g., ESPIRIT), equal to 1 for conventional SENSE and ncoils represents the number of receiver channels in the coil array.

Similarly, output k-space data are expected to be shaped (nslices, nsets, ncoils, ncontrasts, nviews, nsamples).

__init__(coord=None, shape=None, basis_adjoint=None, weight=None, device='cpu', threadsperblock=128, width=4, oversamp=1.25)[source]#

Initiate the linear operator.

Methods

__init__([coord, shape, basis_adjoint, ...])

Initiate the linear operator.

forward(x)

Apply Non-Uniform Fast Fourier Transform.

to(*args, **kwargs)

Copy to different devices

Attributes

H

Return adjoint linear operator.