Artifacts in Spiral Imaging & Correction Methods
Craig H. Meyer1
1Department of Biomedical Engineering, University of Virginia, Charlottesville, VA, United States

Synopsis

This educational talk will cover the sources of spiral image artifacts and methods for correcting for these artifacts. Spiral imaging is a promising technique with short scan times, high SNR efficiency, and robustness to flow and motion. A number of technical advances have enabled high-quality spiral scans on modern imaging hardware. However, spiral scans have different artifacts than Cartesian scans. Sources of spiral artifacts include aliasing of objects outside the field of view, eddy currents, and main field inhomogeneity. Eddy current effects can be eliminated through a one-time calibration and image reconstruction techniques can remove image blur due to inhomogeneity.

Overview of Presentation

Spiral k-space scanning is a promising alternative to conventional Cartesian k-space scanning. Spiral scanning can reduce scan time by a factor of 5-10, has high SNR efficiency, and is relatively robust to flow and motion. Variable-density spiral trajectories can enable accelerated image reconstruction methods such as compressed sensing and deep learning. A number of technical advances have enabled high-quality spiral scans on modern imaging hardware, so that spiral methods are becoming clinically availably for applications such as MR fingerprinting and show promise for use at low field strength. However, spiral scans can have artifacts that are different from those of Cartesian scans. This talk will cover spiral artifacts and methods that have been developed to correct for them.

Field of View and Aliasing

For conventional Cartesian scans, if the field of view (FOV) is set too small in the phase encoding direction, then the familiar aliasing or “wrap” artifact appears in that direction, which takes the form of a duplicate of the object that may overlap the desired image. Fortunately, no aliasing occurs in the Cartesian readout direction, because the temporal anti-aliasing filter prevents it. For spiral scans, there is no dedicated readout direction, because both the x and y gradients are active throughout the readout, so that the effective readout direction rotates throughout the scan. Thus, aliasing is not prevented by the temporal filter, and setting the FOV too small can result in swirly artifacts. These artifacts can be reduced or eliminated by designing spirals for a larger FOV, by using spiral trajectories that vary in density as a function of k-space radius, or by using image reconstruction methods to compensate for under-sampling.

k-Space Trajectory Estimation

For high-quality images, it is essential for the image reconstruction program to have an accurate estimate of the spiral k-space trajectory. While modern MRI gradient systems produce very good spiral trajectories, residual gradient eddy current effects and system timing delays can result in space-varying image blur and geometric distortion. To compensate for these effects, the most important step is to measure the time delay between the prescribed spiral gradient waveforms and the actual gradient waveforms. To further reduce the k-space trajectory error, one solution is to measure the trajectory axis using MR methods [1] or a field camera [2]. Using this data, it is possible to perform a one-time calibration for gradient delays and eddy currents along each physical axis [3-5]. A properly calibrated spiral scan has minimal image quality degradation with modern hardware.

Main Field Inhomogeneity

The primary effect of main field inhomogeneity on a Cartesian scan is relatively minor geometric distortion. The effects on a spiral scan can be significant image blur and geometric distortion, if not corrected. These effects are larger for spiral scans with longer readouts and for single-shot scans. An important step in avoiding these effects is to demodulate at the correct center frequency for each slice. For inhomogeneity that varies slowly in space, various image reconstruction methods based on conjugate phase reconstruction can remove image blur. These include methods that require a separate field map and those that automatically or semi-automatically derive the field map [6]. For inhomogeneity with rapid spatial variations, such as near the sinuses, model-based reconstruction methods can be used to correct for geometric distortion [7]. Concomitant gradient field effects can also lead to spiral image blur [8], which can also be corrected in image reconstruction [6,8]. These effects are more significant for MRI scanners at lower fields and farther away from isocenter. To avoid blurring due to chemical shift, fat suppression or water-fat imaging is often used for spiral scans with longer readouts.

Acknowledgements

The speaker would like to acknowledge funding support from NIH R01 EB028773 and contributions by colleagues and by University of Virginia graduate students.

References

  1. Duyn JH, Yang Y, Frank FA, van der Veen Simple JW. Simple Correction Method for k-Space Trajectory Deviations in MRI. JMR 1998;132: 150-153.
  2. Kang D, Yarach U, In MH, Gray EM, Trzasko JD, Jo HJ, Shu Y, Huston J 3rd, Bernstein MA. The effect of spiral trajectory correction on pseudo-continuous arterial spin labeling with high-performance gradients on a compact 3T scanner. Magn Reson Med. 2020 Jul;84(1):192-205.
  3. Tan H, Meyer CH. Estimation of k-space trajectories in spiral MRI. Magn Reson Med. 2009 Jun;61(6):1396-404.
  4. Addy NO, Wu HH, Nishimura DG. Simple method for MR gradient system characterization and k-space trajectory estimation. Magn Reson Med. 2012 Jul;68(1):120-9.5.
  5. Campbell-Washburn AE, Xue H, Lederman RJ, Faranesh AZ, Hansen MS. Real-time distortion correction of spiral and echo planar images using the gradient system impulse response function. Magn Reson Med. 2016 Jun;75(6):2278-85.
  6. Chen W, Sica CT, Meyer CH. Fast conjugate phase image reconstruction based on a Chebyshev approximation to correct for B0 field inhomogeneity and concomitant gradients. Magn Reson Med. 2008 Nov;60(5):1104-11.
  7. Sutton BP, Noll DC, Fessler JA. Fast, iterative image reconstruction for MRI in the presence of field inhomogeneities. IEEE Trans Med Imaging. 2003 Feb;22(2):178-88.
  8. King KF, Ganin A, Zhou XJ, Bernstein MA. Concomitant gradient field effects in spiral scans. Magn Reson Med. 1999 Jan;41(1):103-12.
Proc. Intl. Soc. Mag. Reson. Med. 30 (2022)