Metal Artifact Suppression
We recently purchased a metal artifact reduction software for one of our scanners. How does that work?
|
|
Metals have high intrinsic magnetic susceptibilities (χ) that produce significant local field disturbances with both in-plane and through-plane changes in resonance frequency. By altering resonance frequencies metals shift image pixels away from their true positions leading to significant geometric distortions including signal voids (black areas) and signal pile-ups (bright areas). Several methods for minimizing metal-related artifacts exist.
As the important first step in reducing metal-related artifacts, use the lowest field strength magnet possible. Susceptibility artifacts scale with field strength, so you are nearly always better off scanning that total hip replacement patient in that older 1.5T scanner rather than in your brand new 3.0T unit!
|
Even without purchasing specialized software, judicious use of standard techniques can be used to reduce the size of these artifacts. First and foremost, a fast (turbo) spin-echo pulse sequence with short echo spacing should be employed, as the multiple 180°-refocusing pulses applied at close intervals will partially correct for dephasing due to magnetic field inhomogeneities. Receiver bandwidth should also be increased to as large a value as possible consistent with an acceptable level of signal-to-noise. Reducing slice thickness and using parallel imaging acceleration will also help. Avoid spectral fat suppression and use STIR instead.
|
|
Before resigning to failure, it is also worth trying to image the area in at least three different planes. I have been pleasantly surprised on numerous occasions that while a metal artifact may make images uninterpretable in two planes, the third plane may be substantially free of artifact and allow a diagnosis to be made. This is purely a trial-and-error method, but may help you when you are in a pinch.
Major MR vendors also offer specially designed software techniques to minimize metal artifacts, sometimes called generically called MARS (Metal Artifact Reduction Sequences). Several methods are used to accomplish this goal.
Reducing In-Plane Distortions: View Angle Tilting (VAT)
Since its development in the 1980's, VAT has become a commonly used and well-established technique for reducing in-plane image distortions due to susceptibility effects. Philips original O-MAR (Orthopedic Metal Artifact Reduction) and Siemens original WARP techniques were 2D Turbo SE sequences based on VAT.
The principle of VAT is that magnetic susceptibility artifacts cause displacements in both the readout and slice-select directions. By applying an additional (VAT) gradient during signal readout along the slice-select direction equal magnitude to the one applied during RF-excitation, the off-resonance displacements are made to cancel each other exactly. The VAT readout gradient results in a "shearing" of the imaged pixels, equivalent to viewing the slice at a slight angle so that the displaced in-plane pixels do not overlap. The VAT method in theory thus fully compensates for in-plane pixel shifts in the readout direction.
Since its development in the 1980's, VAT has become a commonly used and well-established technique for reducing in-plane image distortions due to susceptibility effects. Philips original O-MAR (Orthopedic Metal Artifact Reduction) and Siemens original WARP techniques were 2D Turbo SE sequences based on VAT.
The principle of VAT is that magnetic susceptibility artifacts cause displacements in both the readout and slice-select directions. By applying an additional (VAT) gradient during signal readout along the slice-select direction equal magnitude to the one applied during RF-excitation, the off-resonance displacements are made to cancel each other exactly. The VAT readout gradient results in a "shearing" of the imaged pixels, equivalent to viewing the slice at a slight angle so that the displaced in-plane pixels do not overlap. The VAT method in theory thus fully compensates for in-plane pixel shifts in the readout direction.
Although adequately correcting for in-plane distortions, VAT produces image blurring due to geometric slice shear as well as the low-pass filtering effect of the additional VAT gradient. These effects can be minimized by using thinner sections, higher matrix resolution, and higher bandwidths.
Reducing Through-Plane Distortions: SEMAC and MAVRIC
Through-plane (i.e., slice-to-slice) magnetic field distortions due to metal are more challenging to correct. Two closely related techniques have been developed to accomplish this: SEMAC (Slice Encoding Magnetic Artifact Compensation) and MAVRIC (Multi-Acquisition Variable Resonance Image Combination).
The SEMAC technique utilizes both a VAT-compensation gradient (to reduce in-plane displacements) plus additional phase-encoding steps in the slice-select direction (to reduce through-plane displacements). This additional phase-encoding from overlapping volumes obtained at discrete frequency offsets provides information on how metallic susceptibility effects have distorted the original slice profile. Reconstruction software can then correct for signals shifted perpendicular to the imaging plane. Software choices include the number of overlap volumes to be acquired (more volumes = less artifact but longer imaging times). The current products Siemens Advanced WARP and Philips O-MAR XD are based on SEMAC with VAT. Imaging times can be significantly reduced by combination of SEMAC with Compressed Sensing (CS).
MAVRIC is based on a 3D fast spin-echo sequence and employs narrow band frequency-selective excitation coupled with a multispectral VAT-type readout. "Spectral bins" are created that collect narrow excitation spectrum data both on-resonance and for a set of approximately 30 partially overlapping offset frequencies to measure in-plane displacements. MAVRIC also employs several additional features including image interleaving (to allow more freedom in TE & TR selection), adaptive phase-encoding (reducing number of phase encode steps and hence imaging time for off-resonance excitations, and a deblurring post-processing algorithm to remove artifacts from the reconstruction of imaging planes. Hitachi currently offers a MAVRIC-based technique called HiMAR (Hitachi Metal Artifact Reduction).
The original MAVRIC technique did not utilize slab-selection gradients (relying on surface coils to restrict active signal in the z-direction). In certain locations like the hip and shoulder, significant through-plane aliasing artifacts often occurred, because coil sensitivies were not sufficient to provide volume selectivity. This limitation could be overcome by adding a z-gradient during both multispectral excitation and readout. The resultant sequence, called MAVRIC-SL, a GE product, is thus a hybrid between MAVRIC and SEMAC.
MAVRIC is based on a 3D fast spin-echo sequence and employs narrow band frequency-selective excitation coupled with a multispectral VAT-type readout. "Spectral bins" are created that collect narrow excitation spectrum data both on-resonance and for a set of approximately 30 partially overlapping offset frequencies to measure in-plane displacements. MAVRIC also employs several additional features including image interleaving (to allow more freedom in TE & TR selection), adaptive phase-encoding (reducing number of phase encode steps and hence imaging time for off-resonance excitations, and a deblurring post-processing algorithm to remove artifacts from the reconstruction of imaging planes. Hitachi currently offers a MAVRIC-based technique called HiMAR (Hitachi Metal Artifact Reduction).
The original MAVRIC technique did not utilize slab-selection gradients (relying on surface coils to restrict active signal in the z-direction). In certain locations like the hip and shoulder, significant through-plane aliasing artifacts often occurred, because coil sensitivies were not sufficient to provide volume selectivity. This limitation could be overcome by adding a z-gradient during both multispectral excitation and readout. The resultant sequence, called MAVRIC-SL, a GE product, is thus a hybrid between MAVRIC and SEMAC.
References
Cho Z, Kim D, Kim Y. Total inhomogeneity correction including chemical shifts and susceptibility by view angle tilting. Med Phys 1988; 15:7–11.
Hargreaves BA, Worters PW, Pauly KB et al. Metal-induced artifacts in MRI. AJR Am J Roentenol 2011; 197:547-555. (A recent excellent review of metal artifacts and suppression techniques).
Hey S, Hoogenraad D, Elanchezhian V, et al. Orthopedic Metal Artifact Reduction (O-MAR): Distortion correction in the presence of a orthopedic implant. Philips White Paper, 2016. (Contains nice diagrams showing how VAT and SEMAC correct for slice distortions)
Jungman PM, Agten CA, Pfirrmann CW, Sutter R. Advances in MRI around metal. J Magn Reson Imaging 2017; 46:972-991. [DOI LINK]
Koch KM, Brau AC, Chen W, et al. Imaging near metal with a MAVRIC-SEMAC hybrid. Magn Reson Med 2011; 65:71-82. (The MAVRIC-SL technique)
Koch KM, Lorbiecki JE, Hinks RS, King KF. A multispectral three-dimensional acquisition technique for imaging near metal implants. Magn Reson Med 2009; 61:381–390. (The MAVRIC technique).
Lee EM, Ibrahim E-SH, Dudek N, et al. Improving MR image quality in patients with metallic implants. RadioGraphics 2021; 41: (in press) [DOI LINK] Paper has many practical techniques such as patient positioning and pulse sequence/parameter choice in addition to metal-suppression software.
Lu W, Pauly KB, Gold GE, Pauly JM, Hargreaves BA. SEMAC: slice encoding for metal artifact
correction in MRI. Magn Reson Med 2009; 62:66–76.
Olsen RV, Munk PL, Lee MJ et al. Metal artifact reduction sequence: early clinical applications. Radiographics 2000; 20:699-712.
Talbot BS, Weinberg EP. MR imaging with metal-suppression sequences for evaluation of total joint arthroplasty. RadioGraphics 2016; 36:209-225.
Cho Z, Kim D, Kim Y. Total inhomogeneity correction including chemical shifts and susceptibility by view angle tilting. Med Phys 1988; 15:7–11.
Hargreaves BA, Worters PW, Pauly KB et al. Metal-induced artifacts in MRI. AJR Am J Roentenol 2011; 197:547-555. (A recent excellent review of metal artifacts and suppression techniques).
Hey S, Hoogenraad D, Elanchezhian V, et al. Orthopedic Metal Artifact Reduction (O-MAR): Distortion correction in the presence of a orthopedic implant. Philips White Paper, 2016. (Contains nice diagrams showing how VAT and SEMAC correct for slice distortions)
Jungman PM, Agten CA, Pfirrmann CW, Sutter R. Advances in MRI around metal. J Magn Reson Imaging 2017; 46:972-991. [DOI LINK]
Koch KM, Brau AC, Chen W, et al. Imaging near metal with a MAVRIC-SEMAC hybrid. Magn Reson Med 2011; 65:71-82. (The MAVRIC-SL technique)
Koch KM, Lorbiecki JE, Hinks RS, King KF. A multispectral three-dimensional acquisition technique for imaging near metal implants. Magn Reson Med 2009; 61:381–390. (The MAVRIC technique).
Lee EM, Ibrahim E-SH, Dudek N, et al. Improving MR image quality in patients with metallic implants. RadioGraphics 2021; 41: (in press) [DOI LINK] Paper has many practical techniques such as patient positioning and pulse sequence/parameter choice in addition to metal-suppression software.
Lu W, Pauly KB, Gold GE, Pauly JM, Hargreaves BA. SEMAC: slice encoding for metal artifact
correction in MRI. Magn Reson Med 2009; 62:66–76.
Olsen RV, Munk PL, Lee MJ et al. Metal artifact reduction sequence: early clinical applications. Radiographics 2000; 20:699-712.
Talbot BS, Weinberg EP. MR imaging with metal-suppression sequences for evaluation of total joint arthroplasty. RadioGraphics 2016; 36:209-225.