If I understand correctly, one would need both magnitude and phase images in order to be able to reconstruct real and imaginary parts of the MR image, and this is the complex image on which one could apply the IFFT to work in the kspace.
This would imply there is no way to go back in the kspace if I only have a magnitude image.
Output of the inverse fourier transform on MR data

 Site Admin
 Posts: 523
 Joined: Fri Sep 02, 2005 2:21 pm
 Location: Leeds, UK
 Contact:
Re: Output of the inverse fourier transform on MR data
I agree with this. You would need the phase information too as well as the magnitude image.
However note that your program may still give you a "kspace" from a magnitude image if you do a iFFT, because it will assume that the imaginary parts are all zero, and all the info is in the real parts. It won't be the complex kspace as it was acquired.
However note that your program may still give you a "kspace" from a magnitude image if you do a iFFT, because it will assume that the imaginary parts are all zero, and all the info is in the real parts. It won't be the complex kspace as it was acquired.
Re: Output of the inverse fourier transform on MR data
Thank you xargon for your question, and special thanks to dave for this forum.
This question is not easily,intuitively, to understand. Not only in MRI data, but in general. What do the output complex data of the Inverse Fourier Transform represents or means? or more precisely for MRI, What is the phase for each pixel in image domain means?
I hope my reply revive this post to get a satisfying answer.
Regards
This question is not easily,intuitively, to understand. Not only in MRI data, but in general. What do the output complex data of the Inverse Fourier Transform represents or means? or more precisely for MRI, What is the phase for each pixel in image domain means?
I hope my reply revive this post to get a satisfying answer.
Regards