The measurement method termed Magnetic Resonance Elastography (MRE) permits a direct measurement of elastic displacements within media that are ...
Broad-Spectrum Beam Magnetic Resonance Elastography A. J. Romano1, P. J. Rossman2, R. C. Grimm2, J. A. Bucaro1, R. L. Ehman2 1
Naval Research Laboratory, Washington, DC, United States, 2Mayo Clinic and Foundation, Rochester, MN, United States
Abstract The measurement method termed Magnetic Resonance Elastography (MRE) permits a direct measurement of elastic displacements within media that are subject to forced vibration. Standard MRE methods, however, allow for sensitization to only a single frequency of vibration and must be phase locked to the mechanical stimulus. In this paper, we present a novel adaptation of standard MRE methods which permits an extremely rapid measurement of elastic displacements along a beam of interrogation due to broad-spectrum vibration. Temporal and spatial Fourier transforms are then performed on the displacement information to form dispersion images for elastic modulus, attenuation, and anisotropic characterization. Introduction Broad-Spectrum Beam Magnetic Resonance Elastography is an extremely rapid data acquisition technique which permits a measurement of displacements at each of 256 pixels along a single line of interrogation (of arbitrary orientation) with high temporal sampling. This permits the measurement of displacements due to broadspectrum excitation along the beam in much less time than is required to perform a single standard 3-D or 2-D monochromatic MRE measurement [1]. Temporal and spatial Fourier transforms are then performed on the displacements resulting in dispersion curves. The phase velocities of the various wave-types manifest as families of separate curves in k-ω space, from which can be determined elastic moduli, attenuation, and characterization of anisotropy. Methodology This method involves use of a 2-dimensional column excitation (beam) positioned through the region of interest. A broad-spectrum vibration is applied to the phantom. Two beams phase locked to the band limited stimulus are motion encoded with opposite polarity and Fourier transformed. The phase of each beam is then calculated and the difference is taken, yielding the displacements. Motion sensitization is accomplished using a band-limited time domain gradient pulse during the TE. Temporal and spatial Fourier transforms are then performed on the displacements for dispersion analysis. Results Experiments were performed on two test phantoms, one made of 3% Agar and the other of 26% Bovine gel. A broad-spectrum mechanical excitation pulse was constructed by superposing sine waves over a band from 40 Hz to 1 Khz (in 10 Hz increments), and a broad-spectrum sensitization waveform (sinc pulse was utilized. In Figure 1a and b, we see the measured out-of-plane displacement (due to a shear wave actuator) along the beam for Agar and the B-gel, respectively, while in Figure 2a and b, we see the corresponding dispersion curves. Total time required for the data acquisition was 3.4 minutes for the Agar and 6.8 minutes for the B-gel. From the dispersion curves, it is seen that the Agar has no apparent attenuation and a phase velocity of around 7.7 m/s. The B-gel, on the other hand, is highly attenuative and has a frequency dependent phase velocity ranging from around 3.8 m/s at 100 Hz to around 4.7 m/s at 600 Hz. Conclusions We have outlined the methodology and demonstrated the application of a novel displacement measurement method called Broad-Spectrum Beam MRE. This new method is extremely rapid relative to standard MRE and is robust in the presence of noise, for the determination of elastic moduli, attenuation, and characterization of anisotropy, since the dispersion relations are not dependent on inversion methods [2-3]. This work supported in part by the Office of Naval Research and the National Institutes of Health grant CA75552. References 1. Muthupillai, R., et al., Science, 1854-1857, 1995. 2. Romano, A. J., et al., IEEE Trans. Ultrason., Ferroelect., Freq. Contr., 47(6), 1575-1581, 2000. 3. Romano, A. J., et al., Proc. 10th meeting of the ISMRM, 3, 2603, 2002. 40
0.1
0
FxFt
Frequency (Hz)
Distance (meters)
0.0
1280
40
0
30
a)
a) -40
0.2 0.0
Time (seconds)
0.05
0.0
40
0.07
0
FxFt
b) 0.14 0.0
4021.24
1280
40
0
30
b) -1280 -5744.62
-40
Time (seconds)
0.05
Figure 1. Broadband out-of-plane displacement in a) Agar and b) B-gel.
Proc. Intl. Soc. Mag. Reson. Med. 11 (2003)
20
0
Wavenumber (m-1)
Frequency (Hz)
Distance (meters)
-1280 -4021.24
20
0
5744.62
Wavenumber (m-1)
Figure 2. Dispersion relations of Fig. 1 a) Agar, b) B-gel.
1083
