Use of Weiner deconvolution to improve the quality of MR images

A computer software technique has been developed to insert an iterative modified Wiener deconvolution filter (an optimal filter) following the quadrature detector of an MR imager. This is done to suppress as well as possible the noise found in the receiving antenna. This technique convolves the complex signals sampled directly from the antenna quadrature detector with a transfer function that is the power spectrum of the estimated signal without noise divided by the power spectrum of noise alone. The image produced by this process is used to calculate the power spectrum of the signal without noise