Synopsis

Keywords: Elastography, Brain

Conventional gradient echo (GRE) based Magnetic Resonance Elastography (MRE) suffers from long TE and scan time at low mechanical excitation (<25Hz). Displacement Encoding with a Stimulated Echo (DENSE) can record displacement with short TR, useful for low-frequency excitation. Here, 3D excitation was used to compensate for potential signal loss. A multiphase acquisition scheme with stack-of-star radial sampling was also used along with interleaved multi-slab acquisition and Hadamard encoding for MRE acceleration. Phantom experiments were conducted for validation. 32 slices brain MRE scan at 20Hz can be achieved in 11min30s, 5 folds faster than conventional GRE-based MRE.

Summary of Main Findings

A 3D GRE-based fast Multiphase DENSE MRE sequence with stack-of-star radial sampling, interleaved multi-slab acquisition and Hadamard encoding was proposed for MRE acceleration. Compared with conventional GRE-based MRE, the proposed sequence can achieve 5-fold acceleration.

Introduction

Magnetic resonance elastography (MRE) is a useful method to detect the biomechanical properties of soft tissues noninvasively^{1, 2}. Conventional gradient-echo (GRE) based MRE sequence has to be synchronized with mechanical excitation³, elongating the scanning time when the excitation frequency is low (<25Hz). Displacement encoding with stimulated echo (DENSE) can record displacement with short TR. Therefore, DENSE is ideal for low-frequency encoding than conventional GRE-MRE at low mechanical excitation scenarios. It has been shown that multiphase DENSE MRE with radial sampling has the advantage of acceleration and artifact suppression⁴. In this study, we proposed a 3D GRE-based fast multiphase DENSE MRE sequence with stack-of-star radial sampling, interleaved multi-slab acquisition^{5, 6}, and Hadamard encoding⁷ for acceleration of MRE. Phantom experiments were conducted for validation. A human brain MRE scan was also performed to demonstrate its potential for MRE acceleration.

Methods

The DENSE method can be divided into two parts: Encode and Decode. Spatial modulation of magnetization block was used for encoding the initial place. In the imaging block, a decode gradient was used to generate a DENSE echo.
$$M_{x y}^{D E N S E}=\frac{1}{2} M_{0} \sin \theta \cdot e^{-\frac{T M}{T_{1}}} \cdot e^{i \mathrm{~K} \Delta \mathrm{r}}$$ where $$$M_0$$$ is the steady state longitudinal magnetization, $$$\theta$$$ is the flip angle, $$$TM$$$ is the mixing time, $$$K$$$ is the area under the motion encoding gradient and $$$\Delta r$$$ represents the displacement.

The proposed sequence timing diagram is shown in Figure 1. Motion encoding block (MEB) encoded the initial position and motion decoding blocks (MDB) decoded and generated the DENSE echo. In MEB, the encoding gradients were set up in three encoding directions simultaneously for Hadamard encoding. 1 MEB was followed by N times MDB (N = slice phase encoding (SPE) number * slice oversample rate / 2). One MEB plus N MDB is called an imaging section, which filled the SPE-Spokes k-space a half from the center to the edge. In one MDB, 8 times in plane golden-angle radial sampling acquisition was performed for imaging 4 motion phases for the 2 imaging slabs respectively. A broad-band minimum-phase RF pulse is used for 3D excitation. The decoding gradient was also set up ib three encoding directions simultaneously. The Hadamard DENSE signal was composited of three direction displacements. With 4 times Hadamard encoding, the origin displacement can be decoded using the least square method.
$$M_{x y}^{DENSE}=\frac{1}{2} M_{0} \sin \theta \cdot e^{-\frac{T M}{T_{1}}} \cdot e^{i\left(\mathrm{~K}_{\mathrm{x}} \Delta \mathrm{x}+\mathrm{K}_{\mathrm{y}} \Delta \mathrm{y}+\mathrm{K}_{\mathrm{z}} \Delta \mathrm{z}\right)}$$ $$\left[\begin{array}{l}\Phi_{1} \\\Phi_{2} \\\Phi_{3} \\\Phi_{4}\end{array}\right]=\left[\begin{array}{llll}+1 & +1 & +1 & +1 \\-1 & +1 & +1 & +1 \\+1 & -1 & +1 & +1 \\+1 & +1 & -1 & +1\end{array}\right]\left[\begin{array}{c}K_{x} \Delta x \\K_{y} \Delta y \\K_{z} \Delta z \\\Phi_{0}\end{array}\right]=E\left[\begin{array}{c}K_{x} \Delta x \\K_{y} \Delta y \\K_{z} \Delta z \\\Phi_{0}\end{array}\right]$$ $$\left[\begin{array}{c}\Delta x \\\Delta y \\\Delta z \\\sim\end{array}\right]=\operatorname{inv}(E)\left[\begin{array}{l}\Phi_{1} \\\Phi_{2} \\\Phi_{3} \\\Phi_{4}\end{array}\right] \cdot *\left[\begin{array}{c}1 / K_{x} \\1 / K_{y} \\1 / K_{z} \\\sim\end{array}\right]$$ where $$$K_x, K_y, K_z$$$ are the areas of the encoding gradients on three axes respectively and $$$\Delta x, \Delta y, \Delta z$$$ represent the displacements of three directions. $$$\Phi_i$$$ is the $$$i^{th}$$$ time composite phase with Hadamard encoding.

The proposed sequence was implemented and tested on a 3T scanner (uMR 790, United Imaging Healthcare, Shanghai, China). A custom-built electromagnetic actuator was used to generate the vibration. The imaging parameters were TR=12.5ms, TE=4.2ms, matrix size=80*80, slice=32, voxel size=3*3*3 mm, spokes /slab=125, BW=500Hz/Pixel, $$$G_{MEG}$$$=35mT/m, $$$G_{dura}$$$=0.85ms. A phantom with two stiff inclusions,(I) was stiffer than (II) (Figure 2), was constructed for sequence validation at 60Hz⁸. A healthy volunteer was recruited for testing the proposed sequence at 20Hz.

Results

Phantom experiments result showed that the displacement field recorded using the proposed was similar to that from GRE-MRE sequence. (Figure 2) Moreover, the two stiff inclusions can also be detected and distinguished from each other in the inversion shear modulus map (Figure 3). The human brain experiment showed the proposed sequence can clearly record the displacement and the potential in GRE-based MRE acceleration with low mechanical excitation. (Figure 4)

Conclusion

We proposed a 3D GRE-based fast multiphase DENSE MRE sequence with stack-of-star radial sampling, interleaved multi-slab acquisition and Hadamard encoding for MRE acceleration. The phantom experiment result showed wave images acquired were similar to that of GRE-MRE. The human brain scan showed its potential for acceleration with low-frequency mechanical excitation.

Acknowledgements

Funding support from the National Natural Science Foundation of China (grant 32271359) and the Natural Science Foundation of Shanghai (grant 22ZR1429600) is acknowledged.

References

[1] S. K. Venkatesh, "Clinical Applications of Liver Magnetic Resonance Elastography: Chronic Liver Disease," in Magnetic Resonance Elastography, 2014, ch. Chapter 4, pp. 39-60.
[2] L. V. Hiscox et al., "Magnetic resonance elastography (MRE) of the human brain: technique, findings and clinical applications," Phys Med Biol, vol. 61, no. 24, pp. R401-R437, Dec 21 2016, doi: 10.1088/0031-9155/61/24/R401.
[3] J. Strasser, M. T. Haindl, R. Stollberger, F. Fazekas, and S. Ropele, "Magnetic resonance elastography of the human brain using a multiphase DENSE acquisition," Magn Reson Med, vol. 81, no. 6, pp. 3578-3587, Jun 2019, doi: 10.1002/mrm.27672.
[4] R. K. Wang et al., "Fast magnetic resonance elastography with multiphase radial encoding and harmonic motion sparsity based reconstruction," (in English), Phys. Med. Biol., Article vol. 67, no. 2, p. 13, Jan 2022, Art no. 025007, doi: 10.1088/1361-6560/ac4a42.
[5] P. Garteiser et al., "Rapid acquisition of multifrequency, multislice and multidirectional MR elastography data with a fractionally encoded gradient echo sequence," NMR Biomed, vol. 26, no. 10, pp. 1326-35, Oct 2013, doi: 10.1002/nbm.2958.
[6] C. Guenthner, S. Sethi, M. Troelstra, A. S. Dokumaci, R. Sinkus, and S. Kozerke, "Ristretto MRE: A generalized multi-shot GRE-MRE sequence," NMR Biomed, vol. 32, no. 5, p. e4049, May 2019, doi: 10.1002/nbm.4049.
[7] C. L. Dumoulin, S. P. Souza, R. D. Darrow, N. J. Pelc, W. J. Adams, and S. A. Ash, "Simultaneous acquisition of phase-contrast angiograms and stationary-tissue images with Hadamard encoding of flow-induced phase shifts," (in English), Journal of magnetic resonance imaging : JMRI, vol. 1, no. 4, pp. 399-404, 1991 1991, doi: 10.1002/jmri.1880010403.
[8] Y. Feng et al., "Viscoelastic Characterization of Soft Tissue-Mimicking Gelatin Phantoms using Indentation and Magnetic Resonance Elastography," (in English), J. Vis. Exp., Article no. 183, p. 12, May 2022, Art no. e63770, doi: 10.3791/63770.