- Metallic implants distort the magnetic field in MRI, resulting in image artifacts

- The right choice of scan settings and parameters can reduce these artifacts

- Artifacts can be corrected further by the use of more advanced methods, such as View-Angle Tilting for in-plane distortions and MSI methods for through-plane distortions

Students, researchers, and healthcare professionals interested in MRI close to metal implants.Audience members will have learned about the challenges of MRI close to metal implants, the main associated artifacts and their causes, and methods for correcting these artifacts.Arthroplasty procedures, such as hip and knee replacements, are becoming more common in the United States, a trend which is projected to continue. Often, further operations or monitoring is necessary due to complications related to the implant. MRI close to the implant can be useful for determining whether such revision surgery is needed, since it gives good soft tissue contrast. However, MRI close to such devices is difficult due to the metal’s interaction with the magnetic fields, which can cause severe imaging artifacts. In recent years, powerful methods have been developed to overcome these artifacts. This talk will discuss the main methods for reducing artifacts when imaging close to metal, both in terms of choosing scan parameters and using methods designed for artifact reduction.

Close to metal implants, the distortion in the main field can be so great that it makes spins precess at varying rates within a single voxel. This leads to signal loss from T2* dephasing. This signal drop-off can be greatly reduced by applying spin-echo imaging instead of gradient echo imaging.

Fat suppression techniques that rely on the frequency difference between fat and water will also often fail close to metal, since the frequency variation caused by the field inhomogeneity can be much larger than the expected fat-water frequency difference. This can be solved by using STIR imaging, which relies on differences in T1, rather than frequency, between fat and water.

During signal readout, the field distortions caused by the metal result in errors in the readout gradient. This will lead to signals being mapped to the wrong location, resulting in image distortion, signal loss, and pile-up. These in-plane distortions can be resolved by using a large readout bandwidth and a View-Angle Tilting (VAT) gradient. View-Angle Tilting involves replaying the slice-select gradient during the readout.

Similarly, the metal will cause errors in the slice-select gradient, resulting in an unintended shape of the region that gets excited by the pulse. These through-plane, or slice, distortions can lead to the excited slices being curved, thinned, thickened, or split. The large frequency variations can also cause regions to simply not be excited, since they are at a frequency outside the excitation bandwidth. Such errors can be reduced by using a large excitation bandwidth and a strong slice-select gradient. Even better correction can be achieved using Multi-Spectral Imaging (MSI) methods, such as MAVRIC and SEMAC. These resolve the slice distortions with 3D phase encoding, mapping each excited voxel to its correct location.

Using Spin Echo imaging instead of Gradient Echo imaging greatly reduces signal drop-off close to the metal [1]. Through-slice and in-plane artifacts remain, resulting in distortion, signal loss, and pile-up artifacts.

STIR imaging [2] gives better fat suppression than conventional fat-saturated imaging. However, it yields lower signal-to-noise (SNR) than other methods, and is usually not a good option for contrast-enhanced imaging, since it will suppress the tissue that takes up contrast, due to the shortened T1.

The use of a VAT gradient largely removes in-plane distortions, such as signal pile-up, due to metal [3]. Artifacts from through-plane distortions, such as slice thinning and thickening, remain. The use of a VAT gradient can result in blurring, which can be minimized by using a large readout bandwidth and reducing the readout duration to less than that of the main lobe of the RF pulse [4].

Using the MAVRIC or SEMAC MSI techniques [5,6] resolves the through-plane artifacts, making the image useful for diagnosis close to the metal. This comes at the cost of increased scan time from the additional phase encoding. Some artifacts remain very close to the metal. These are due to the very strong gradient errors, induced very close to the metal, disrupting the Fourier encoding process [7].

The methods presented can reduce signal drop-off, improve fat suppression, and correct in-plane and through-plane artifacts close to metal implants. Other methods for reducing artifacts near metal have been proposed, such as field mapping techniques for estimating the field errors [8] and ultrashort TE techniques to not allow for much dephasing before the signal is acquired [9]. MSI methods have proven very useful for correcting artifacts close to metal but do come at the cost of increased scan time. Several methods have been proposed to reduce MSI scan time. These include both conventional acceleration methods such as partial Fourier and parallel imaging [10,11] as well as using compressed sensing [12] or making use of the shape of the distorted slices to use a more efficient Field of View (FOV), saving scan time [13]. Since the phase encoding direction does not experience distortion, recent work has explored phase encoding along all directions within feasible scan times, with promising results [14]. Recently, a 2D MSI method has been developed [15], which can save scan time and has been used for thermal monitoring close to metallic implants [16]. Methods have also been developed to resolve signal voids, due to strong effects on local gradients close to the metal, in order to visualize the implant geometry [17].Metallic implants can produce severe artifacts in MRI, complicating the monitoring of tissue close to the implant. Careful selection of conventional imaging techniques and scan parameters can reduce these artifacts to some extent. Furthermore, using more advanced techniques, such as View-Angle Tilting or MSI methods can reduce these artifacts even further.