pygrappa.cgsense

Python implementation of iterative and CG-SENSE.

pygrappa.cgsense.cgsense(kspace, sens, coil_axis=-1)[source]

Conjugate Gradient SENSE for arbitrary Cartesian acquisitions.

Parameters:
  • kspace (array_like) – Undersampled kspace data with exactly 0 in place of missing samples.
  • sens (array_like) – Coil sensitivity maps.
  • coil_axis (int, optional) – Dimension of kspace and sens holding the coil data.
Returns:

res – Single coil unaliased estimate (imspace).
Return type:

array_like

Notes

Implements a Cartesian version of the iterative algorithm described in [1]. It can handle arbitrary undersampling of Cartesian acquisitions and arbitrarily-dimensional datasets. All dimensions except coil_axis will be used for reconstruction.

This implementation uses the scipy.sparse.linalg.cg() conjugate gradient algorithm to solve A^H A x = A^H b.

References

[1]Pruessmann, Klaas P., et al. “Advances in sensitivity encoding with arbitrary k‐space trajectories.” Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine 46.4 (2001): 638-651.