Chemical Shift: Phase Effects
Doesn't the chemical shift between water and fat protons also result in a phase shift between them? If this is so, then why aren't chemical shift artifacts seen in the phase-encode direction also?
|
|
It is indeed true that the chemical shift between water and fat protons produces both a frequency shift and a phase shift between the two species. In routine spin-warp imaging, however, the resulting extra phase of fat protons relative to water protons does not accumulate from one phase-encoding step to the next. The differential phase shift between water and fat is therefore constant at a given location between successive phase-encoding steps. Since the Fourier transform takes into account the phase difference in assigning spatial location to a signal in this direction, no water-fat misregistration will occur along the phase-encoding axis in routine spin-echo or gradient-echo imaging.
EPI image with chemical shiftartifact in PE direction (↑)
In echo-planar imaging (EPI) with Cartesian sampling, however, the opposite phenomenon occurs. Here, chemical-shift artifacts are minimal in the frequency-encode direction but may be extremely large in the phase-encode direction! The reason for this difference between spin-warp and echo-planar techniques lies in the way data is collected in the latter.
In EPI all the lines of raw data are collected immediately following a single RF excitation. The bandwidth per pixel in the frequency-encode direction is relatively large (i.e., on the order of 1500 Hz), so that chemical-shift artifacts are not noticeable. In the phase-encode direction, however, the bandwidth per pixel is extremely small (i.e., on the order of 10 Hz), since all the phase-encode lines are acquired in a time frame on the order of 100 msec. At 1.5T the fat/water chemical shift is about 220 Hz, and this low bandwidth in the phase-encode direction is translated into a noticeable artifact that can be up to 1-3 cm in width (depending on parameters chosen). Special corrections and modifications to the echo-planar technique must be made to minimize the fat signal and eliminate this artifact.
In EPI all the lines of raw data are collected immediately following a single RF excitation. The bandwidth per pixel in the frequency-encode direction is relatively large (i.e., on the order of 1500 Hz), so that chemical-shift artifacts are not noticeable. In the phase-encode direction, however, the bandwidth per pixel is extremely small (i.e., on the order of 10 Hz), since all the phase-encode lines are acquired in a time frame on the order of 100 msec. At 1.5T the fat/water chemical shift is about 220 Hz, and this low bandwidth in the phase-encode direction is translated into a noticeable artifact that can be up to 1-3 cm in width (depending on parameters chosen). Special corrections and modifications to the echo-planar technique must be made to minimize the fat signal and eliminate this artifact.
References
Jezzard P, Balaban RS. Correction for geometric distortion in echo planar images from Bo field variations. Magn Reson Med 1995;34:65-73.
Jezzard P, Balaban RS. Correction for geometric distortion in echo planar images from Bo field variations. Magn Reson Med 1995;34:65-73.
Related Questions
What is a chemical-shift artifact?
What is a chemical-shift artifact?