mri-q2

Computation of a matrix Q, representing the scanner configuration, used in a 3D magnetic resonance image reconstruction algorithm in non-Cartesian space. The k-space trajectory is the 3D Fourier transform of the scan trajectory used in an MRI. To reconstruct the image, an inverse Fourier transform needs to be performed on the data obtained during the k-space trajectory scan. This is straightforward if the k-space trajectory falls on a Cartesian grid, since a FFT can then be performed to obtain the original image. If the k-space trajectory does not follow a Cartesian coordinate system, it can be converted to Cartesian by gridding or an advanced reconstruction algorithm can be applied to the non-Cartesian grid. This kernel is meant to be used with non-Cartesian k-space trajectories and is part of an advanced image reconstruction algorithm, which is outlined in the paper referenced below.

Reference: http://impact.crhc.illinois.edu/shared/papers/cf08.stone.pdf

Code snippet

numK = 3072;
float *phiMag, phiR, phiI;

for (indexK = 0; indexK < numK; indexK++){
   float real = phiR[indexK];
   float imag = phiI[indexK];
   phiMag[indexK] = real*real + imag*imag;
}