Metal Related Artifacts: Imaging Techniques and Challenges
Alissa Burge

Synopsis

MRI is an excellent imaging modality for evaluation of a wide variety of pathologies, due to its combination of spatial resolution, tissue contrast, and sensitivity for edema. The utilization of MRI in the presence of metal, however, is complicated by a variety of associated artifacts, largely related to metal susceptibility. Though sometimes technically demanding, appropriate modification of pulse sequence acquisition parameters can mitigate the effects of metal related artifacts, and the utilization of more advanced metal reduction sequences, where necessary, can result in further reductions in artifact, yielding high quality diagnostic scans in patients with metallic implants.

Introduction

MRI is an excellent imaging modality for evaluation of a wide variety of pathologies, due to its combination of spatial resolution, tissue contrast, and sensitivity for edema. The utilization of MRI in the presence of metal, however, is complicated by a variety of associated artifacts, largely related to metal susceptibility. Though sometimes technically demanding, appropriate modification of pulse sequence acquisition parameters can mitigate the effects of metal related artifacts, and the utilization of more advanced metal reduction sequences, where necessary, can result in further reductions in artifact, yielding high quality diagnostic scans in patients with metallic implants.

Conventional Sequences in the Presence of Metal

Many variables contribute to the degree and extent of artifact observed about a given implant upon images generated by a particular pulse sequence. Artifacts related to B0 field inhomogeneities tend to be proportional to the static field strength; therefore imaging at higher field strengths (for example, 3.0T instead of 1.5T) will result in accentuation of artifacts generated by implant [1-3]. The type of pulse sequence also influences the degree of artifact generated by a particular implant, with gradient echo sequences, due to their lack of a refocusing pulse, generating significantly more artifact than fast spin echo sequences, which employ multiple refocusing pulses. Fat suppression techniques also vary in their utility about metallic implants, with spectrally selective techniques, such as conventional chemical fat saturation tending to result in more artifacts than alternate less frequency dependent techniques such as inversion recovery [1-3]. Component design also plays a role in the degree of artifact generated, with certain metals and particular geometric configurations generating a greater amount of artifact.

Parameter Modifications

When imaging metal utilizing conventional pulse sequences, various modifications of typical pulse acquisition parameters may be employed in order to reduce the amount of artifact generated. Increasing the receiver bandwidth increases the strength of the readout gradient, which results in decreased artifact due to the fact that spatial distortion is inversely proportional to gradient strength. Using a wider bandwidth will also reduce interecho spacing, allowing for longer echo train lengths resulting in reduced scan time. A wider receiver bandwidth, however, results in decreased SNR. Decreasing the voxel size improves spatial resolution, but decreases SNR. Increasing the number of excitations increases SNR, offsetting the reduction in SNR resulting from modifying the bandwidth and voxel size; however, this results in increased scan time [1-3].

Advanced Techniques for Metal Suppression

Single point MRI, which eliminates the slice select and readout encoding processes by fully phase encoding the images, can remove susceptibility artifact related to metal, though acquisition times tend to be excessively long. One example of this type of sequence is single point ramped imaging with T1 enhancement (SPRITE), in which a single point of the FID is sampled per TR, while gradients are kept on and progressively ramped down with each successive TR. While SPRITE has been demonstrated to effectively reduce susceptibility artifact with experimental phantoms, there are multiple barriers to utilizing the sequence clinically, foremost among these being a prohibitively long image acquisition time [3].

View angle tilting (VAT) reapplies the slice selection gradient during the readout, which shears the image in the plane of the slice and readout directions. Because slice displacement and in plane displacement have a constant ratio, the reapplication of the slice gradient results in the slice displacements exactly cancelling in plane displacements [1-3]. WARP VAT (Siemens) is a metal reduction sequence combining view angle tilting with a high bandwidth fast spin echo acquisition in order to obtain the metal reduction benefits of both techniques [4].

Prepolarized MRI utilizes application of a relatively low magnetic field of minimal spatial homogeneity to polarize spins prior to signal acquisition, which is performed utilizing an even lower but more homogeneous field. Because the readout field is on the order of mT, the effects of susceptibility are greatly reduced as compared to conventional images obtained at typical clinical field strengths. These types of scanners, however, are not in routine use clinically, and tend to have small bores unsuitable for imaging of larger anatomic regions [3].

Slice encoding for metal artifact correction (SEMAC) utilizes view angle tilting and additional phase encoding steps in the slice select direction in order to reduce artifact related to metal susceptibility. Frequency perturbations about metallic implants may result in excitation of protons within slices adjacent to the intended slice of interest upon application of the slice selective RF pulse, which results in signal arising from outside of the slice of interest mistakenly being registered as having arisen from the intended imaging slice position. SEMAC applies additional phase encoding in the Z axis in order to accurately resolve the spatial origin of signal arising from the slice of interest and immediately adjacent slices. A postprocessing algorithm is then utilized to assign signal arising from the excited slab to the correct slice location. Because this technique alone only corrects misregistration in the Z axis, view angle tilting is added to address disturbances in the frequency direction. SEMAC typically results in a significant decrease in the amount of susceptibility artifact associated with metallic implants; however, the additional encoding steps result in increased scan time. Techniques such as parallel and partial Fourier imaging may be applied to reduce scan time, particularly given the fact that the robust spatial encoding also results in an increase in SNR [2, 3, 5-9].

Multiacquisition variable resonance image combination (MAVRIC) is an advanced metal artifact reduction sequence which utilizes multiple image acquisitions over a range of frequency bins centered about the Larmor frequency, with subsequent post processing yielding a composite image in which the effects of susceptibility are markedly reduced. Rather than exciting a slice or slab, MAVRIC uses a frequency selective excitation, limiting the range of frequency offsets imaged at one time, which results in decreased in plane displacement. Phase encoding is performed in the slice direction to avoid distortion in this direction. Image acquisition is performed over a range of frequency bins and the individual images are subsequently combined utilizing a sum of squares operation to yield a composite image [2, 3, 5, 10]. The main difference between the MAVRIC and SEMAC techniques is that MAVRIC excites limited frequency bands, while SEMAC excites limited spatial bands. A hybrid of the 2 techniques also exists, combining the MAVRIC spectral bins with SEMACs slab selectivity [11, 12].

CONCLUSIONS

Metallic implants are commonly encountered during clinical MR imaging evaluation, but result in a variety of artifacts, which tend to reduce diagnostic quality on conventional clinical pulse sequences, particularly in the region immediately adjacent to the implant. Appropriate modification of scan acquisition parameters can result in significant reduction of artifact on conventional imaging sequences, and the utilization of advances sequences specifically designed for use in the setting of metal can result in further artifact reduction.

Acknowledgements

No acknowledgement found.

References

1. Gupta, A., et al., Metal artifact reduction: standard and advanced magnetic resonance and computed tomography techniques. Radiol Clin North Am, 2015. 53(3): p. 531-47.

2. Hargreaves, B.A., et al., Metal-induced artifacts in MRI. AJR Am J Roentgenol, 2011. 197(3): p. 547-55.

3. Koch, K.M., et al., Magnetic resonance imaging near metal implants. J Magn Reson Imaging, 2010. 32(4): p. 773-87.

4. Lazik, A., et al., Usefulness of metal artifact reduction with WARP technique at 1.5 and 3T MRI in imaging metal-on-metal hip resurfacings. Skeletal Radiol, 2015. 44(7): p. 941-51.

5. Chen, C.A., et al., New MR imaging methods for metallic implants in the knee: artifact correction and clinical impact. J Magn Reson Imaging, 2011. 33(5): p. 1121-7.

6. Hargreaves, B.A., et al., Accelerated slice encoding for metal artifact correction. J Magn Reson Imaging, 2010. 31(4): p. 987-96. 7. Lee, Y.H., et al., Usefulness of slice encoding for metal artifact correction (SEMAC) for reducing metallic artifacts in 3-T MRI. Magn Reson Imaging, 2013. 31(5): p. 703-6.

8. Lee, Y.H., et al., Feasibility of fat-saturated T2-weighted magnetic resonance imaging with slice encoding for metal artifact correction (SEMAC) at 3T. Magn Reson Imaging, 2014. 32(8): p. 1001-5. 9. Lu, W., et al., SEMAC: Slice Encoding for Metal Artifact Correction in MRI. Magn Reson Med, 2009. 62(1): p. 66-76.

10. Liebl, H., et al., In vitro assessment of knee MRI in the presence of metal implants comparing MAVRIC-SL and conventional fast spin echo sequences at 1.5 and 3 T field strength. J Magn Reson Imaging, 2015. 41(5): p. 1291-9.

11. Choi, S.J., et al., Metal artifact reduction with MAVRIC SL at 3-T MRI in patients with hip arthroplasty. AJR Am J Roentgenol, 2015. 204(1): p. 140-7.

12. Gutierrez, L.B., et al., MR imaging near metallic implants using MAVRIC SL: initial clinical experience at 3T. Acad Radiol, 2015. 22(3): p. 370-9.

Figures

Sagittal T2 FSE image of the cervical spine acquired at bandwidth of 122 Hz/px (A) versus a bandwidth of 390 Hz/px (B) demonstrates resultant decrease in the extent of susceptibility artifact related to the patients anterior fixation hardware.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)