In 2DFT imaging, considerable disparity exists between sampling times for data collection in the frequency- and phase-encode directions. In the frequency-encode direction, all 256-512 samples of a signal are acquired in the space of a single echo (e.g., over 5-80 msec). Conversely, the time to obtain a single sample in the phase-encode direction is on the order of seconds to minutes, since essentially all lines of k-space must be collected to obtain the complete data set for Fourier reconstruction.
Most gross physiologic motions (respiration, swallowing, cardiac pulsation) occur over time frames of a hundred milliseconds to several seconds. Because these motions are slow in relation to the frequency-encode sampling interval, they typically produce only a small amount of spatial blurring locally in the frequency-encode direction. Conversely, since the phase sampling interval is generally equal or longer than the period of most physiologic motions, artifacts will be more prominent in this direction. Furthermore, these artifacts will be propagated in the phase-encode direction regardless of whether the physiologic motion has occurred in the frequency-encode, phase-encode, or slice-select directions.
Most gross physiologic motions (respiration, swallowing, cardiac pulsation) occur over time frames of a hundred milliseconds to several seconds. Because these motions are slow in relation to the frequency-encode sampling interval, they typically produce only a small amount of spatial blurring locally in the frequency-encode direction. Conversely, since the phase sampling interval is generally equal or longer than the period of most physiologic motions, artifacts will be more prominent in this direction. Furthermore, these artifacts will be propagated in the phase-encode direction regardless of whether the physiologic motion has occurred in the frequency-encode, phase-encode, or slice-select directions.
A simple and useful imaging strategy for all types of motion is simply to swap the phase- and frequency-encoding axes. Although this will not change the motion artifact, it will simply shift it into a different direction so that it will not spill over areas of interest. For example, in axial cranial imaging the PE direction is usually selected to be left-to-right so that orbital motion artifacts do not spill over onto the brain.
|
Advanced Discussion (show/hide)»
No supplementary material yet. Check back soon.
References
Axel L, Summers RM, Kressel HY, Charles C. Respiratory effects in two-dimensional Fourier transform MR imaging. Radiology 1986; 160:795-801.
Storey P, Chen Q, Li W, et al. Band artifacts due to bulk motion. Magn Reson Med 2002; 48:1028-1036.
Wood ML, Henkelman RM. MR image artifacts from periodic motion. Med Phys 1985;12:143-151.
Axel L, Summers RM, Kressel HY, Charles C. Respiratory effects in two-dimensional Fourier transform MR imaging. Radiology 1986; 160:795-801.
Storey P, Chen Q, Li W, et al. Band artifacts due to bulk motion. Magn Reson Med 2002; 48:1028-1036.
Wood ML, Henkelman RM. MR image artifacts from periodic motion. Med Phys 1985;12:143-151.
Related Questions
Why do motion artifacts often form into discrete ghosts?
Why do motion artifacts often form into discrete ghosts?