tfmri.coils.compress_coils

compress_coils(kspace, coil_axis=- 1, out_coils=None, method='svd', **kwargs)[source]

Coil compression gateway.

This function estimates a coil compression matrix and uses it to compress kspace. If you would like to reuse a coil compression matrix or need to calibrate the compression using different data, use tfmri.coils.CoilCompressorSVD.

This function supports the following coil compression methods:

  • SVD: Based on direct singular-value decomposition (SVD) of k-space data 1. This coil compression method supports Cartesian and non-Cartesian data. This method is resilient to noise, but does not achieve optimal compression if there are fully-sampled dimensions.

Parameters

  • kspace – A Tensor. The multi-coil k-space data. Must have type complex64 or complex128. Must have shape [..., Cin], where ... are the encoding dimensions and Cin is the number of coils. Alternatively, the position of the coil axis may be different as long as the coil_axis argument is set accordingly. If method is "svd", kspace can be Cartesian or non-Cartesian. If method is "geometric" or "espirit", kspace must be Cartesian.

  • coil_axis – An int. Defaults to -1.

  • out_coils –

    An int. The desired number of virtual output coils.

  • method – A string. The coil compression algorithm. Must be "svd".

  • **kwargs – Additional method-specific keyword arguments to be passed to the coil compressor.

Returns

A Tensor containing the compressed k-space data. Has shape [..., Cout], where Cout is determined based on out_coils or other inputs and ... are the unmodified encoding dimensions.

References

1

Huang, F., Vijayakumar, S., Li, Y., Hertel, S. and Duensing, G.R. (2008). A software channel compression technique for faster reconstruction with many channels. Magn Reson Imaging, 26(1): 133-141.

2

Zhang, T., Pauly, J.M., Vasanawala, S.S. and Lustig, M. (2013), Coil compression for accelerated imaging with Cartesian sampling. Magn Reson Med, 69: 571-582. https://doi.org/10.1002/mrm.24267

3

Bahri, D., Uecker, M., & Lustig, M. (2013). ESPIRIT-based coil compression for cartesian sampling. In Proceedings of the 21st Annual Meeting of ISMRM, Salt Lake City, Utah, USA (Vol. 47).