'''Python implementation of the Segmented GRAPPA algorithm.'''
import numpy as np
from pygrappa import cgrappa
[docs]def seggrappa(kspace, calibs, *args, **kwargs):
'''Segmented GRAPPA.
See pygrappa.grappa() for full list of arguments.
Parameters
----------
calibs : list of array_like
List of calibration regions.
Notes
-----
A generalized implementation of the method described in [1]_.
Multiple ACS regions can be supplied to function. GRAPPA is run
for each ACS region and then averaged to produce the final
reconstruction.
References
----------
.. [1] Park, Jaeseok, et al. "Artifact and noise suppression in
GRAPPA imaging using improved k‐space coil calibration and
variable density sampling." Magnetic Resonance in
Medicine: An Official Journal of the International Society
for Magnetic Resonance in Medicine 53.1 (2005): 186-193.
'''
# Do the reconstruction for each of the calibration regions
recons = [cgrappa(kspace, c, *args, **kwargs) for c in calibs]
# Average all the reconstructions
return np.mean(recons, axis=0)
if __name__ == '__main__':
pass