Previously, it was demonstrated by simulations that modes of the shear waves can be observed in the brain during MR elastography (MRE) with high shear wave displacement values at mode frequencies and change in stiffness causes shift in mode frequency¹. Purpose of this study is to validate experimentally that greater shear wave displacement than applied displacement at mode frequency can be observed and mode frequency shifts when stiffness alters. Thus, observing shift in mode frequency may mean change in stiffness. In addition, safety issues in MRE are evoked due to high shear wave displacement at resonance. Three homogeneous phantoms were prepared in spherical flasks having 1L volume using the same agar-agar powder and following the preparation steps described in a previous study², with agar-agar powder concentrations of 0.65%, 0.75% and 0.85%. From measurement results², the shear modulus values of these phantoms are estimated as 6, 7.7 and 10 kPa, respectively. Experiments were conducted in a 3T Siemens Tim Trio MRI scanner. Using an actuator coil driven by sinusoidal current under the B₀ field, phantoms were rotationally vibrated about the y-axis, as shown in Figure 1. The motion was induced continuously and the frequency of excitation was swept from 25 to 45.5 Hz, with 0.5 Hz steps. Peak-to-peak current applied to the actuator coil was kept constant during all experiments in the range of 1.75-1.85A. Two acquisitions were made in each scan using a GRE pulse with motion encoding gradient (MEG), by switching polarity of the zeroth and first moment nulled MEG, having same frequency with excitation frequency, in an interleaved fashion. The shear wave displacement images were formed by taking root of sum of squares (RSS) of two phase difference images at steady state of shear waves having π/2 phase difference obtained from two scans for each frequency. Peak of the RSS image was used as a measure of shear wave displacement magnitude. Detector coils were mounted on the actuator system to measure the amount of displacement applied by the actuator to the phantoms³. As shown in Figure 1, setup was placed approximately 30° tilted with respect to B₀ direction to measure the induced voltage on the detector coil properly³. Shear wave displacement was normalized by the actuator displacement, which was obtained from the measured induced voltage on the detector coil. Central transversal slice of the phantoms were selected for imaging and MEG was in the slice selection direction. All experiments were repeated three times in 1-6 weeks to test the repeatability. To demonstrate an example, results for measurements of shear wave displacement in the phantom with 0.85% agar-agar concentration, induced voltage on detector coil, computed displacement of actuator and normalized shear wave displacement are shown in Figure 2. Normalized displacement images for the same phantom experiment can be seen in Figure 3. In Figure 4, normalized displacement plots for all phantoms are depicted. It is expected to observe peak normalized displacement at mode frequencies, which were found to be 27.5, 29.5 and 34 Hz for 0.65%, 0.75% and 0.85%, respectively, and repeated experiments yielded the same mode frequencies except third experiment for 0.85% which resulted with a mode frequency of 34.5 Hz. It can be stated that results are coherent with previous findings in simulations¹. To the best of our knowledge, this study analyzes, for the first time, the frequency response of the actuator coil by sweeping the excitation frequency in MRE. From Figure 2, it is observed that frequency response of actuator system is not constant although current applied to actuator coil is kept almost constant. We argue that this is due to interaction of shear wave and actuator system resonances. Consequently, output displacement is a combination of two peak displacements, which correspond to mode of phantom and mode of actuator system loaded with phantom. At the mode frequency of shear wave, displacement of actuator is minimum, hence creating a high shear wave displacement for small input displacement, whereas second peak output displacement is due to peak displacement applied by the actuator. Using normalization, we were able to isolate shear wave resonance by removing the effect of actuation system. An important outcome of this study is that the safety issues in MRE should be reconsidered⁴, as we can observe 10-20 times shear wave displacement of applied displacement in Figure 4. Repeatability of determining mode frequency has been validated, thus tracking any shift in mode frequency due to possible stiffness changes is feasible. Our future work will include applications of our findings to human brain studies.