Patient-to-patient variation of susceptibility-induced B₀ field in bilateral breast MRI

Multishot Diffusion-Weighted MRI of the Breasts in the Supine vs. Prone Position

BACKGROUND

Diffusion-weighted imaging (DWI) may allow for breast cancer screening MRI without a contrast injection. Multishot methods improve prone DWI of the breasts but face different challenges in the supine position.

PURPOSE

To establish a multishot DWI (msDWI) protocol for supine breast MRI and to evaluate the performance of supine vs. prone msDWI.

STUDY TYPE

Prospective.

POPULATION

Protocol optimization: 10 healthy women (ages 22-56), supine vs. prone: 24 healthy women (ages 22-62) and five women (ages 29-61) with breast tumors.

FIELD STRENGTH/SEQUENCE

3-T, protocol optimization msDWI: free-breathing (FB) 2-shots, FB 4-shots, respiratory-triggered (RT) 2-shots, RT 4-shots, supine vs. prone: RT 4-shot msDWI, T2-weighted fast-spin echo.

ASSESSMENT

Protocol optimization and supine vs. prone: three observers performed an image quality assessment of sharpness, aliasing, distortion (vs. T2), perceived SNR, and overall image quality (scale of 1-5). Apparent diffusion coefficients (ADCs) in fibroglandular tissue (FGT) and breast tumors were measured.

STATISTICAL TESTS

Effect of study variables on dichotomized ratings (4/5 vs. 1/2/3) and FGT ADCs were assessed with mixed-effects logistic regression. Interobserver agreement utilized Gwet's agreement coefficient (AC). Lesion ADCs were assessed by Bland-Altman analysis and concordance correlation (ρc ). P value <0.05 was considered statistically significant.

RESULTS

Protocol optimization: 4-shots significantly improved sharpness and distortion; RT significantly improved sharpness, aliasing, perceived SNR, and overall image quality. FGT ADCs were not significantly different between shots (P = 0.812), FB vs. RT (P = 0.591), or side (P = 0.574). Supine vs. prone: supine images were rated significantly higher for sharpness, aliasing, and overall image quality. FGT ADCs were significantly higher supine; lesion ADCs were highly correlated (ρc  = 0.92).

DATA CONCLUSION

Based on image quality, supine msDWI outperformed prone msDWI. Lesion ADCs were highly correlated between the two positions, while FGT ADCs were higher in the supine position.

EVIDENCE LEVEL

2.

TECHNICAL EFFICACY

Stage 1.

Automated Placement of Scan and Pre-Scan Volumes for Breast MRI Using a Convolutional Neural Network

Graphically prescribed patient-specific imaging volumes and local pre-scan volumes are routinely placed by MRI technologists to optimize image quality. However, manual placement of these volumes by MR technologists is time-consuming, tedious, and subject to intra- and inter-operator variability. Resolving these bottlenecks is critical with the rise in abbreviated breast MRI exams for screening purposes. This work proposes an automated approach for the placement of scan and pre-scan volumes for breast MRI. Anatomic 3-plane scout image series and associated scan volumes were retrospectively collected from 333 clinical breast exams acquired on 10 individual MRI scanners. Bilateral pre-scan volumes were also generated and reviewed in consensus by three MR physicists. A deep convolutional neural network was trained to predict both the scan and pre-scan volumes from the 3-plane scout images. The agreement between the network-predicted volumes and the clinical scan volumes or physicist-placed pre-scan volumes was evaluated using the intersection over union, the absolute distance between volume centers, and the difference in volume sizes. The scan volume model achieved a median 3D intersection over union of 0.69. The median error in scan volume location was 2.7 cm and the median size error was 2%. The median 3D intersection over union for the pre-scan placement was 0.68 with no significant difference in mean value between the left and right pre-scan volumes. The median error in the pre-scan volume location was 1.3 cm and the median size error was -2%. The average estimated uncertainty in positioning or volume size for both models ranged from 0.2 to 3.4 cm. Overall, this work demonstrates the feasibility of an automated approach for the placement of scan and pre-scan volumes based on a neural network model.

Generalized magnetostatic target field method for shielded magnetic field coils in a separable coordinate system

A theoretical method is described to analytically calculate a pair of surface current densities, which produce a desired static magnetic field in one region of the space and zero magnetic field in another. The analysis is based on the known relationship between a surface current density and a stream function, the equivalence of stream functions and surface magnetic dipole density, and the scalar potential representation of the associated magnetic field in free space. From these relations, we formulate the magnetostatic problem, which is often treated as a vector field problem, as a scalar field problem in which a two-dimensional scalar field (stream function) is related to a three-dimensional one (magnetic scalar potential) via the differentiation of the electrostatic Green's function 1/|r-rs|. It is shown that, in a coordinate system in which a separated form of the Green's function exists (separable coordinate system), there exists a simple relationship between a harmonic component of a stream function and a harmonic component of the magnetic scalar potential. The method is applied to calculate idealized surface current patterns for actively shielded, linear gradient field coils in the Cartesian, cylindrical, and spherical coordinates.

Rapid calculation of static magnetic field perturbation generated by magnetized objects in arbitrary orientations

PURPOSE

Most previous work on the calculation of susceptibility-induced static magnetic field (B₀ ) inhomogeneity has considered strictly unidirectional magnetic fields. Here, we present the theory and implementation of a computational method to rapidly calculate static magnetic field vectors produced by an arbitrary distribution of voxelated magnetization vectors.

THEORY AND METHODS

Two existing B₀ calculation methods were systematically extended to include arbitrary orientations of the magnetization and the magnetic field; they are (1) Fourier-domain convolution with k-space-discretized (KD) dipolar field, and (2) generalized susceptibility voxel convolution (gSVC). The methods were tested on an analytical ellipsoid model and a tilted human head model, as well as against experimentally measured B₀ fields induced by a stainless-steel implant located in an inhomogeneous region of a clinical 3T MRI magnet.

RESULTS

Both methods were capable of correctly calculating B₀ fields inside a magnetized ellipsoid in all tested orientations. The KD method generally required a larger grid and longer computation time to achieve accuracy comparable to gSVC. Measured B₀ fields due to the implant showed a good match with the gSVC-calculated fields that accounted for the spatial variation of the applied magnetic field including the radial components.

CONCLUSION

Our method can provide a reliable and efficient computational tool to calculate B₀ perturbation by magnetized objects under a variety of circumstances, including those with inhomogeneous magnetizing fields, anisotropic susceptibility, and a rotated coordinate system.

Changing Paradigms in Breast Cancer Screening: Abbreviated Breast MRI

Breast magnetic resonance imaging (MRI) is the most sensitive imaging method for breast cancer detection. In this review we discuss the vastly superior performance of MRI compared to traditional breast cancer screening modalities of mammography, tomosynthesis and ultrasound. We discuss an abbreviated breast MRI (AB-MRI) protocol utilizing Dixon sequences which is compliant with American College of Radiology (ACR) guidelines for accreditation of breast MRI but with significantly reduced scan times. Adaptation of such an AB-MRI protocol significantly increases patient throughput and may allow MRI to serve as a stand- alone breast cancer screening tool.

Rapid, theoretically artifact-free calculation of static magnetic field induced by voxelated susceptibility distribution in an arbitrary volume of interest

PURPOSE

To demonstrate a computationally efficient and theoretically artifact-free method to calculate static field (B₀ ) inhomogeneity in a volume of interest induced by an arbitrary voxelated susceptibility distribution.

METHODS

Our method computes B₀ by circular convolution between a zero-filled susceptibility matrix and a shifted, voxel-integrated dipolar field kernel on a grid of size N_S +N_T - 1 in each dimension, where N_S and N_T are the sizes of the susceptibility source and B₀ target grids, respectively. The computational resource requirement is independent of source-target separation. The method, called generalized susceptibility voxel convolution, is demonstrated on three susceptibility models: an ellipsoid, MR-compatible screws, and a dynamic human heartbeat model.

RESULTS

B₀ in an ellipsoid calculated by generalized susceptibility voxel convolution matched an analytical solution nearly exactly. The method also calculated screw-induced B₀ in agreement with experimental data. Dynamic simulation demonstrated its computational efficiency for repeated B₀ calculations on time-varying susceptibility. On the contrary, conventional and alias-subtracted k-space-discretized Fourier convolution methods showed nonnegligible aliasing and Gibbs ringing artifacts in the tested models.

CONCLUSION

Generalized susceptibility voxel convolution can be a fast and reliable way to compute susceptibility-induced B₀ when the susceptibility source is not colocated with the B₀ target volume of interest, as in modeling B₀ variations from motion and foreign objects.

Distortion correction in diffusion-weighted imaging of the breast: Performance assessment of prospective, retrospective, and combined (prospective + retrospective) approaches

PURPOSE

To compare the effectiveness of prospective, retrospective, and combined (prospective + retrospective) EPI distortion correction methods in bilateral breast diffusion-weighted imaging (DWI) scans.

METHODS

Five healthy female subjects underwent an axial bilateral breast DWI exam with and without prospective B₀ inhomogeneity correction using slice-by-slice linear shimming. In each case, an additional b=0 DWI scan was performed with the polarity of the phase-encoding gradient reversed, to generate an estimated B₀ map; this map or a separately acquired B₀ map was used for retrospective correction, either alone or in combination with the prospective correction. The alignment between an undistorted, anatomical reference scan with similar contrast and the corrected b=0 DWI images with different correction schemes was assessed.

RESULTS

The average cross-correlation coefficient between the DWI images and the anatomical reference scan was increased from 0.82 to 0.92 over the five volunteers when combined prospective and retrospective distortion correction was applied. Furthermore, such correction substantially reduced patient-to-patient variation of the image alignment and the variability of the average apparent diffusion coefficient in normal glandular tissue.

CONCLUSION

Combined prospective and retrospective distortion correction can provide an efficient way to reduce susceptibility-induced image distortions and enhance the reliability of breast DWI exams. Magn Reson Med 78:247-253, 2017. © 2016 International Society for Magnetic Resonance in Medicine.

A new approach to shimming: the dynamically controlled adaptive current network

PURPOSE

Magnetic field homogeneity is important in all aspects of magnetic resonance imaging. A new approach to increase field homogeneity is presented that allows dynamic and adaptive control over the flow of current over a single surface using a network of actively controlled solid-state switches.

METHODS

Computer simulations were completed demonstrating the potential of this approach. Wire patterns were produced using the boundary element method to remove magnetic field inhomogeneities over multiple regions of interest. Field maps and regions of interest histograms were compared with and without the shim present. A prototype was constructed confirming the feasibility of this approach within the magnetic resonance environment. Metal-oxide-semiconductor field-effect transistors were used. Two field maps were acquired with the prototype producing gradient and offset field profiles, respectively. The experimental field profiles were compared with simulation.

RESULTS

The wire patterns significantly increased field homogeneity over all regions of interest investigated. The field profiles produced by the prototype matched simulation. No imaging artifacts were produced.

CONCLUSIONS

An approach to control the shape of a current distribution over a single surface has been described. This method has the potential to improve field homogeneity over any desired region of interest and is particularly well suited for dynamic applications. The method is feasible with current technology and construction techniques.

Requirements for static and dynamic higher order B0 shimming of the human breast at 7 T

The increased magnetic susceptibility effects at higher magnetic fields increase the demands for shimming of the B0 field for in vivo MRI and MRS. Both static and dynamic techniques have been developed to compensate for susceptibility-induced field inhomogeneities. In this study, we investigate the impact of and need for both static and dynamic higher order B0 shimming of magnetic field homogeneities in clinical breast MRI at 7 T. Both global and local field variations at lipid-tissue interfaces were observed in the magnetic field using TE-optimized B0 mapping at 7 T. With static B0 shimming, a field homogeneity of 39 ± 11 Hz (n = 48) was reached in a single breast using second-order shimming. Further compensation of the residual local field inhomogeneities caused by lipid-tissue interfaces does not seem to be feasible with shallow spherical harmonic fields. For bilateral shimming, the shimming quality was significantly less at 62 ± 15 Hz (n = 22) over both breasts, even after (simulated) fourth-order shimming. In addition, a substantial time-dependent field instability of 30 Hz peak to peak, with significant higher order field contributions, was observed during regular breathing. In conclusion, TE-optimized B0 field mapping reveals substantial field variations in the lipid-rich environment of the human breast, in both space and time. The static field variations could be partially minimized by third-order B0 shimming, providing sufficient lipid suppression. However, in order to fully benefit from the increased spectral dispersion at high fields, the significant magnetic field variations during breathing need to be considered.

Dynamic slice-dependent shim and center frequency update in 3 T breast diffusion weighted imaging

PURPOSE

To demonstrate dynamic slice-dependent shim update as a simple method to reduce susceptibility-induced B0 inhomogeneity and associated pixel shift artifacts in diffusion-weighted echo planar imaging (DW-EPI) in 3 T breast imaging.

METHODS

Dynamic slice-dependent update of linear shim and center frequency was implemented in a dual-echo B0 mapping sequence and a DW-EPI sequence. Multi-slice axial B0 maps and diffusion-weighted images were obtained from four volunteers with both conventional and dynamic shim methods. The two shim methods were compared in terms of B0 homogeneity and EPI pixel shift artifacts.

RESULTS

In all volunteers the B0 maps showed significantly improved homogeneity; the left-right asymmetry was reduced by 79% and within-slice B0 standard deviation was reduced by 20% on the average. The improvements were better than what was previously reported for conventional (static) third-order shim in bilateral breast. Anatomy-referenced apparent diffusion coefficient (ADC) maps showed reduced overall image registration error obtainable with dynamic shim.

CONCLUSIONS

Dynamic shim is an effective method to improve B0 shimming and DW-EPI image quality in 3 T bilateral breast imaging. Magn Reson Med 71:1813-1818, 2014. © 2013 Wiley Periodicals, Inc.