Zero filling interpolation

Last revised by Andrew Murphy on 2 Apr 2020

Zero filling interpolation (ZIP) is the substitution of zeroes for unmeasured data points in order to increase the matrix size of the new data prior to Fourier transformation of MR data. This results in pixels smaller than the actual resolution of the image. The zero filling occurs in the periphery of k-space rather than the center where most of the signal comes from. 

For example

  • scan matrix 256 x 256
  • reconstruction matrix 512 x 512


Zero filling processes can be very practical in everyday clinical usage by reducing scanning times without much loss in resolution or SNR, since zero filled points contain neither signal nor noise. Hence SNR is unaffected 1

Zero filling is also used in MR spectroscopy, adding zeros to the end of the FID prior to Fourier transform. This results in increased frequency resolution in the resulting spectrum allowing chemical peaks to be more easily resolved 2

ADVERTISEMENT: Supporters see fewer/no ads