Artifacts in magnetic resonance imaging from metals

Bennett, L. H.; Wang, P. S.; Donahue, Michael J.

doi:10.1063/1.361649

Cited by 53 publications

(33 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although we have restricted ourselves to magnetic susceptibility artifacts, there are other artifacts due to metallic objects, such as RF-induced eddy currents (12,13).…”

Section: Discussionmentioning

confidence: 99%

Integral method for numerical simulation of MRI artifacts induced by metallic implants

Balac

Caloz

Cathelineau

et al. 2001

Magnetic Resonance in Med

View full text Add to dashboard Cite

Numerical simulation is a valuable tool for the study of magnetic susceptibility artifacts from metallic implants. A major difficulty in the simulation lies in the computation of the magnetic field induced by the metallic implant. A new method has been designed and implemented to compute the magnetic field induced by metallic objects of arbitrary shape. The magnetic field is expressed pointwise in terms of a surface integral. In magnetic resonance imaging (MRI) even minor perturbations of the gradients can disturb the imaging process and may render the clinical image inaccurate or useless. Common causes of magnetic field perturbation are changes of magnetic properties in the sample due to metallic implanted objects, such as a dental prosthesis, orthopedic apparatus, etc. Metallic surgical instruments used in interventional MRI are also responsible for magnetic field perturbation. In the case of guided stereotactic surgery, image distortions are especially harmful since geometrical accuracy is required.In the past, several studies involving numerical simulation have been undertaken to help in the understanding of magnetic field perturbation and to establish experimental conditions that determine susceptibility artifacts (1-3). However, these simulation attempts have been restricted to simple test objects (cylinders, spheres, and ellipsoids) for which an analytical expression for the induced magnetic field is known. In general, a precise calculation of the magnetic field involves a boundary value problem with partial differential equations (PDE) derived from Maxwell's equations, and requires the use of PDE approximation schemes. Among the classical numerical methods are the finite element method (FEM), the finite difference method (FDM), and the boundary element method (BEM). For a comprehensive treatment of these numerical methods in electromagnetism see Ref. 4. Our approach, which is based on a surface integral representation formula for the magnetic flux density, is most similar to the BEM. In short, our method consists of expressing the problem in an integral form over the boundary of the implant, and dividing the boundary into elements in which the integral is numerically computed. It is able to deal with objects of any shape, provided that a mesh of the object boundary is available. This requirement is not at all restricting since mesh generation tools are widely used in computer assisted design (CAD). The advantages of our method compared to the FEM (5) or the FDM (6) are numerous. First, the computation depends only on the object boundary, so the discretization of space is reduced from 3D to 2D. Furthermore, in the FDM/FEM the exterior domain must be truncated and an approximation of the behavior of the field at infinity must be introduced on an artificial boundary. With the BEM, the behavior at infinity is always exactly satisfied. Finally, for 3D problems the FEM and the FDM lead to large linear systems to be solved, whereas in the proposed method the solution is obtained pointwise by evaluating a surfac...

show abstract

“…Although we have restricted ourselves to magnetic susceptibility artifacts, there are other artifacts due to metallic objects, such as RF-induced eddy currents (12,13).…”

Section: Discussionmentioning

confidence: 99%

Integral method for numerical simulation of MRI artifacts induced by metallic implants

Balac

Caloz

Cathelineau

et al. 2001

Magnetic Resonance in Med

View full text Add to dashboard Cite

show abstract

“…There are two sources of these artifacts: susceptibility differences between the metal and the surrounding tissue, and radiofrequency (RF) screening by the metal. The 6 7 susceptibility difference between the metal and the surrounding material [2,4,6] causes a local magnetic field inhomogeneity which is proportional to the measurement field B 0 , resulting in both a shorter relaxation time T 2 * and a local change in precession frequency.…”

Section: Introductionmentioning

confidence: 99%

“…The second effect of a metal object on nuclear magnetic resonance (NMR) and MRI is its interaction with the RF excitation pulses and the subsequent NMR signal [5,6,7].…”

mentioning

confidence: 99%

“…In the case of a piece of metal buried in the sample, the resultant RF inhomogeneity can result in a signal diminuition or enhancement, particularly near the metal [ ] 9 . These artifacts dominate over susceptibility artifacts for metals with high conductivity and low susceptibility, for example, copper or aluminum, whereas for materials with low conductivity and high susceptibility such as titanium they are small compared to susceptibility artifacts [5,6,7].…”

mentioning

confidence: 99%

See 1 more Smart Citation

SQUID-detected microtesla MRI in the presence of metal

Mößle

Han

Myers

et al. 2006

Journal of Magnetic Resonance

119

View full text Add to dashboard Cite

In magnetic resonance imaging (MRI) performed at fields of 1 T and above, the presence of a metal insert can distort the image because of susceptibility differences within the sample and modification of the radiofrequency fields by screening currents.Furthermore, it is not feasible to perform nuclear magnetic resonance (NMR) spectroscopy or acquire a magnetic resonance image if the sample is enclosed in a metal container. Both problems can be overcome by substantially lowering the NMR frequency. Using a microtesla imaging system operating at 2.8 kHz, with a superconducting quantum interference device (SQUID) as the signal detector, we have obtained distortion-free images of a phantom containing a titanium bar and threedimensional images of an object enclosed in an aluminum can; in both cases high-field images are inaccessible.

show abstract

“…Conventional MRI techniques, however, are not applicable to these materials for two fundamental reasons. First, the penetration depth of radiofrequency radiation, which is inversely proportional to the square root of the frequency, is only several micrometers in metals for the strong magnetic fields (Ͼ1 T) used in conventional MRI scanners (13). The nuclear spins within an electrically conductive sample that reside deeper than the penetration depth will not be sufficiently excited, and thus, will not contribute to the MR images.…”

mentioning

confidence: 99%

Flow in porous metallic materials: A magnetic resonance imaging study

Harel

Michalak

et al. 2008

Magnetic Resonance Imaging

View full text Add to dashboard Cite

Purpose:To visualize flow dynamics of analytes inside porous metallic materials with laser-detected magnetic resonance imaging (MRI). Materials and Methods:We examine the flow of nuclearpolarized water in a porous stainless steel cylinder. Laserdetected MRI utilizes a sensitive optical atomic magnetometer as the detector. Imaging was performed in a remotedetection mode: the encoding was conducted in the Earth's magnetic field, and detection is conducted downstream of the encoding location. Conventional MRI (7T) was also performed for comparison.Results: Laser-detected MRI clearly showed MR images of water flowing through the sample, whereas conventional MRI provided no image. Conclusion:We demonstrated the viability of laser-detected MRI at low-field for studying porous metallic materials, extending MRI techniques to a new group of systems that is normally not accessible to conventional MRI. MAGNETIC RESONANCE IMAGING (MRI), conventionally performed in a strong homogenous magnetic field, is a versatile imaging modality for materials research (1). Its advantage of noninvasiveness allows imaging of opaque porous materials not possible by optical methods. Sederman et al (2) used MRI to study the structureflow correlation in packed beds. Seymour et al (3) studied biofouling of a homogeneous model porous media. Callaghan and Khrapitchev (4), and Blü mich et al (5) developed various pulse sequences to investigate timedependent flow velocities in porous media. Swider et al (6) employed high-resolution MRI to study both localized and global flow in porous biomaterials. Recently, Granwehr et al (7) developed a time-of-flight MRI scheme that utilized remote detection to study the flow behavior of xenon gas in porous rocks. In addition, different fluids can be selectively encoded based on their respective chemical shifts (8). The samples involved in these studies with conventional MRI are limited to electrically nonconductive materials.Flow dynamics of liquids in porous metallic materials is of extensive practical interest, as these materials are widely used in filtration, catalysis, and biomedicine (9 -12). Conventional MRI techniques, however, are not applicable to these materials for two fundamental reasons. First, the penetration depth of radiofrequency radiation, which is inversely proportional to the square root of the frequency, is only several micrometers in metals for the strong magnetic fields (Ͼ1 T) used in conventional MRI scanners (13). The nuclear spins within an electrically conductive sample that reside deeper than the penetration depth will not be sufficiently excited, and thus, will not contribute to the MR images. Even if some nuclear spins could be excited within a region sufficiently below the penetration depth of the metallic sample, these spins would not be detected at the pickup coil, because their oscillating field would be similarly screened out. Second, the large magnetic-susceptibility gradients intrinsically associated with porous metallic materials significantly distort the field homogeneity a...

show abstract

Artifacts in magnetic resonance imaging from metals

Cited by 53 publications

References 11 publications

Integral method for numerical simulation of MRI artifacts induced by metallic implants

Integral method for numerical simulation of MRI artifacts induced by metallic implants

SQUID-detected microtesla MRI in the presence of metal

Flow in porous metallic materials: A magnetic resonance imaging study

Contact Info

Product

Resources

About