Parotid glands are one of the most common sites for salivary gland tumors. Conventional imaging techniques have limited usefulness in the quantitative assessment of the parotid glands, making it difficult to differentiate between healthy tissue and tumors, as well as between benign and malignant tumors. Magnetic resonance elastography (MRE) is a non-invasive technique that may potentially overcome these limitations. Nevertheless, due to the size of the parotid gland, increased elastographic resolution is required. This may be achieved by applying shear waves at higher frequencies. However, it also results in stronger attenuation, making the illumination of the parotid challenging. Here, we describe a novel passive driver tailored to the anatomy of the human face, which minimizes the distance shear waves need to travel from the source to the area of interest and thus decreases shear wave attenuation, making high-frequency shear wave MRE feasible.

Background: MR elastography (MRE) has been shown to have excellent performance for noninvasive liver fibrosis staging. However, there is limited knowledge regarding the precision and test -retest repeatability of stiffness measurement with MRE in the multicenter setting. Purpose: To determine the precision and test -retest repeatability of stiffness measurement with MRE across multiple centers using the same phantoms. Materials and Methods: In this study, three cylindrical phantoms made of polyvinyl chloride gel mimicking different degrees of liver stiffness in humans (phantoms 1-3: soft, medium, and hard stiffness, respectively) were evaluated. Between January 2021 and January 2022, phantoms were circulated between five different centers and scanned with 10 MRE-equipped clinical 1.5-T and 3-T systems from three major vendors, using two-dimensional (2D) gradient-recalled echo (GRE) imaging and/or 2D spinecho (SE) echo-planar imaging (EPI). Similar MRE acquisition parameters, hardware, and reconstruction algorithms were used at each center. Mean stiffness was measured by a single observer for each phantom and acquisition on a single section. Stiffness measurement precision and same -session test -retest repeatability were assessed using the coefficient of variation (CV) and the repeatability coefficient (RC), respectively. Results: The mean precision represented by the CV was 5.8% (95% CI: 3.8, 7.7) for all phantoms and both sequences combined. For all phantoms, 2D GRE achieved a CV of 4.5% (95% CI: 3.3, 5.7) whereas 2D SE EPI achieved a CV of 7.8% (95% CI: 3.1, 12.6). The mean RC of stiffness measurement was 5.8% (95% CI: 3.7, 7.8) for all phantoms and both sequences combined, 4.9% (95% CI: 2.7, 7.0) for 2D GRE, and 7.0% (95% CI: 2.9, 11.2) for 2D SE EPI (all phantoms). Conclusion: MRE had excellent in vitro precision and same -session test -retest repeatability in the multicenter setting when similar imaging protocols, hardware, and reconstruction algorithms were used.

Structured AbstractIntroductionAdenomyosis is an under-recognised condition in which definitive diagnosis is only possible via histology after hysterectomy, an unacceptable option for those wishing to preserve fertility. Recent cellular/molecular studies indicate adenomyotic lesions may be fibrotic leading to increased uterine tissue stiffness. 3D Magnetic Resonance Elastography (MRE) is a novel imaging technique that allows in vivo measurement of tissue stiffness (via elastograms). 3D MRE has not been reported to study adenomyosis. The feasibility study aimed to utilise a novel 3D MRE protocol to measure global uterine stiffness and to investigate its potential application for non-invasive in vivo diagnosis of adenomyosis.Materials and Methods3D MRE protocol was conducted on one healthy volunteer (control) and four patients with suspected adenomyosis and heavy menstrual bleeding (HMB), diagnosed via transvaginal ultrasound and clinical history (REC:20/SS/0123 and 19/SS/0102). Two patients underwent hysterectomy, and representative uterine tissue samples were assessed for (i) histological presence of adenomyosis via H&E staining; (ii) cellular/molecular measures of tissue stiffness (collagen [picrosirius red], α-smooth muscle actin, e-cadherin); (iii) relationship between in vivo assessment of the uterus via MRI images and 3D MRE findings with in vitro uterine tissue histology from the same individuals.Results3D MRE was successfully used to acquire elastograms for four patients with adenomyosis (diffuse n=3, focal n=1) and one healthy volunteer. Calculated global uterine stiffness was higher in women with adenomyosis (2.93kPa; range 2.34 – 3.39kPa) compared to a healthy volunteer (2.04kPa). Areas of stiffness on 3D elastograms reflected adenomyotic changes visualised via conventional MRI, with the added benefit of also correlating with histology/immunohistochemical assessment for markers of tissue stiffness.DiscussionA novel 3D MRE protocol has been applied to obtain the global uterine stiffness in four women with HMB and suspected adenomyosis, and one healthy volunteer. 3D MRE has the potential to provide superior non-invasive tissue characterisation in vivo when compared to conventional MRI in the assessment of adenomyosis due to the correlation of imaging and tissue findings. Further studies are now needed to confirm the above exploratory findings, prior to performing a potential clinical trial.

Introduction: Low back pain (LBP) is a very costly and prevalent health disorder in the U.S., resulting in total costs exceeding $100 billion per year. In fact, its reported that up to 85% of people will experience LBP in their lives. One of the most common causes of LBP is degenerative disc disease (DDD), which has been shown to precede LBP and several other disorders in the spine associated with LBP. DDD is a complex cascade of biological and structural changes in the intervertebral disc (IVD) caused by altered mechanics and load distribution of the disc. There are many techniques that have been developed to characterize disc degeneration, but there is no way to directly assess the material properties of the IVD in vivo. It has been shown that the nucleus pulposus within the disc undergoes significant changes in shear modulus, even with early stage disc degeneration. Magnetic resonance elastography (MRE) is a sensitive, phase contrast-based imaging technique for non-invasively mapping the mechanical properties of tissues. Previous studies have demonstrated the feasibility of using MRE in the IVD in vitro to estimate the nucleus pulposus stiffness. The purpose of this study is to determine if MRE is capable of detecting a stiffness change in the intervertebral disc in an in vitro animal model of enzymatically-induced disc degeneration. The target audience of this research is MRI scientists involved in developing spinal imaging methods, radiologists involved in spinal imaging, clinicians involved in managing patients with low-back pain and disc-related spinal disorders, and basic scientists investigating DDD. Methods and Materials: (1) Intervertebral Disc Specimens. Three spinal motion segments (T12/L1, L2/L3, and L3/L4) of a goat lumbar spine were removed with musculature and entire IVD intact. All posterior elements of the motion segment were removed to increase flexibility of the each specimen, such that each motion segment consisted of partial upper and lower vertebral bodies and an intact IVD. (2) In Vitro Disc Degeneration Model. According to Roberts et al, trypsin or papain can be used to create an explant model for disc degeneration in bovine caudal discs. The L2/L3 goat IVD was injected with ~100 μL of trypsin at 20 mg/mL into the nucleus to induce disc degeneration, as reported in the literature. The other two discs served as normal controls. (3) Mechanical vibration. A piezoelectric mechanical driver was used to apply mechanical vibrations at frequencies in the 1-10 kHz range. The driver was positioned such that shear vibration was applied to the upper vertebral body of the motion segment while the lower vertebral body was fixed using a custom-built testing fixture. The specimen was put into a single-channel, 3-in diameter receive coil with the transverse, or axial, crosssection of the IVD parallel to the B0 direction. The vibration direction was parallel to the B0 direction and perpendicular to the spinal motion segment (fig. 1). (3) Wave imaging sequence. The disc specimens were imaged using a spin echo-based MRE sequence with the following parameters: 30 total cycles of 1000Hz motion-encoding gradients (2.4 Gauss/cm), motion sensitivity = 12.9μm/π, offsets = 4, 290/50-ms TR/TE, 8-cm FOV, one 8-mm slice, 256x96 matrix, 1 NEX. A standard 1.5T full-body MRI scanner (Signa 16X Software, GE Healthcare, Waukesha, WI) was used in the experiment. Motion encoding was done in the S/I direction, or along the direction of motion, in all scans. (4) Inversion algorithms. The resulting wave images were then masked, phase unwrapped, bandpass filtered (1-20 waves per FOV), directionally filtered (8 directions), and processed using a local frequency estimation (LFE) inversion algorithm to provide maps of shear stiffness. 10 The average shear stiffness of the nucleus pulposus region within the IVD was measured with a circular ROI placed in the center of the disc as shown in the magnitude and stiffness map images (fig. 2). Results and Discussion: As shown in Figure 2, shear waves were seen in axial cross-sections of both normal discs and the trypsininjected disc. In the normal control discs, the shear wave propagation appears distinctly different in the annulus compared to the nucleus region of the disc, with the nucleus showing a much shorter wavelength as seen in the filtered MRE wave images (fig. 2B). The wave image for the no motion case shows very little, incoherent motion as expected (fig. 2B). The wave data, for all but the no motion case, was inverted using the LFE inversion algorithm to give a shear stiffness map for the transverse cross-section of the discs (fig.2C). Although the wavelength was too long in the annular region to reliably approximate stiffness, the nucleus showed an average stiffness of 89 ± 14 kPa and 133 ± 25 kPa in the T12/L1 and L3/L4 normal discs respectively. These nucleus stiffness values are both very near the range reported in literature for the healthy human IVD nucleus stiffness. The trypsin-injected disc, L2/L3, showed a much longer shear wavelength in the nucleus (fig. 2B), and had a much higher average stiffness of 715 ± 145 kPa. Conclusions: These initial results suggest MRE is capable of detecting changes in the nucleus stiffness in this explant model for enzymaticallyinduced IVD degeneration. Additionally, the data suggests MRE can differentiate the nucleus and annulus regions of normal intervertebral discs, but this difference may be lost with degeneration. According to Roberts et al, this model for disc degeneration could be used to test novel injectable materials for reversing degenerative changes. Based on this work, it appears disc MRE could be used to compare and monitor the effects of novel intervertebral disc treatments using this explant degenerative model.

Target audience includes clinicians and scientists who are interested in MRI technologies for the assessment of hepatic fibrosis. Introduction: It is well known that liver biopsy is the reference standard for detecting hepatic fibrosis in patients with liver diseases, but liver biopsy is an invasive method and could cause sampling errors, inter-observer variation, patient refusal, pain, bleeding and death [1-3]. As a noninvasive technology for detecting liver fibrosis, MR Elastography (MRE) has shown the highest diagnostic accuracy when compared with other noninvasive methods: Fibroscan® and FIBROSpect II® [4]. MRE uses a mechanical driver to excite mechanical waves in a patient liver, use MRI to measure the wave speed, and calculate liver stiffness maps (elastograms) based on the wave speed. There are different types of drivers, including electromechanical drivers [5, 6], piezoelectric driver (with a rubber mat on patient) [7] and pneumatic drivers [8, 9]. Mechanical drivers are usually small and rigid. Piezoelectric driver uses a rigid rod to connect to the rubber mat, which has to maintain a certain rigidity to convert a point vibration from the tip of rod to its surface vibration. A conventional pneumatic liver driver is rigid too. Since human bodies are soft, large and contoured, the mechanical coupling between rigid drivers and human body is often not optimal. To improve the human-drive mechanical coupling, we have developed a pneumatic flexible driver[10]. The purpose of this study was to improve the efficiency of flexible driver, and to compare it with conventional pneumatic rigid driver in patients with a large range of fibrosis stage. Our hypothesis was that the optimized flexible driver is equivalent to the rigid driver in measuring liver stiffness. Methods: (1) Subjects: Our Institutional Review Board approved the study. A total of 23 patients with biopsy-proven fibrosis stage from F0 to F4. (2) Flexible driver : Detailed technique information could be found in [10]. In brief, a flexible driver can conform to and cover most of the chest wall (posterior, right and anterior sides) in the vicinity of liver (Fig. 1 and Fig. 2), with a 20 cm width elastic band wrapped around the human body, so that the human-driver mechanical coupling is optimized. For a comparison, a conventional rigid driver position is shown in Fig. 2. The optimized flexible driver has three components: a flexible rectangular bag, a 3-dimensional structure filling material with a energy reflection flexible back plate. (3) Haptic MRE: Technique details could be found in [8]. In brief, all patients underwent 2D-60Hz liver MRE performed in a 1.5-T MRI scanner (Signa HDx, GE, Wisconsin, USA). Region of interests (ROIs) were drawn in the liver where wave signal-to-noise ratio was higher and larger vessels were absent. Liver stiffness values (mean, standard deviation) were report within the ROIs. MRE was performed twice for each patient, using the optimized flexible driver and conventional rigid driver with comparable acoustic power level. (4) Statistic analysis: Linear regression was done to investigate the correlation between liver stiffness values measured by the flexible and rigid drivers. Bland-Altman analysis [11] was utilized to evaluate the difference between the two drivers. JMP Pro (SAS, USA) software was used in the analysis. Results: Linear regression analysis found that the liver stiffness values measured by the flexible driver and rigid driver were highly correlated (R = 0.97, Fig. 4). In the Bland-Altman analysis, the mean difference of the liver stiffness measurements between the rigid and flexible drivers is -0.11kPa, and the [lower, upper] 95% CI of the difference is [-0.31, 0.09] kPa (Fig. 5). Discussions and Conclusions: Because the flexible driver is soft, all the patients felt more comfortable with the flexible driver than the rigid driver. Fig. 3 shows examples of MRE scans of one patient with both drivers. Both drivers excited comparable wave amplitude in the liver (Fig. 3b, 3e), meaning the optimized flexible driver could be used at the same power level as the rigid driver. Also, the flexible driver excited a more uniform wave filed in the liver than the rigid driver did (Fig. 3b, 3e). In the wave field generated by the rigid driver, a severe wave interference was observed, resulting in an over-estimated stiffness region (hotspot, arrow) in the liver (Fig. 3e, 3f). These findings were popular in this study patient group, and consistent with previous observations [10]. The mean difference (0.11 kPa) between the flexible and rigid driver was larger than the value (0.0061kPa) found in the previous study [10]; a possible reason is that more patients had high stages of fibrosis in this study than the previous study. In practice, a difference of 0.11 kPa is considered within the standard deviation of liver stiffness measured by MRE. Therefore, liver stiffness values measured by the optimized flexible driver are equivalent to those measured by the conventional pneumatic rigid driver. References: [1] World J Gastroenterol, 2008. 14(21): p. 3396-402. [2] Gut, 2006. 55(4): p. 569-78. [3] Am J Gastroenterol, 2002. 97(10): p. 2614-8. [4] Chen, J., ISMRM 2014 submitted. [5] Gastroenterology, 2008. 135(1): p. 32-40. [6] Magn Reson Med, 2008. 60(2): p. 373-9. [7] Hirsch, S., Magn Reson Med, 2013. [8] Clin Gastroenterol Hepatol, 2007. 5(10): p. 1207-1213 e2. [9] Radiology, 2011. 259(3): p. 749-56. [10] ISMRM 1052, 2010. [11] Lancet, 1995. 346(8982): p. 1085-7.

Introduction: It has been reported that up to 85% of people will experience low-back pain (LBP) in their lives. Degenerative disc disease (DDD) and spinal instability are two of the most common causes of LBP. There are many imaging techniques that have been developed to characterize the disc and grade the level of degeneration, but there is no way to directly assess the material properties of the IVD in vivo. The nucleus pulposus has been shown to undergo substantial changes in shear stiffness with degeneration. Magnetic resonance elastography (MRE) is a sensitive, phase contrastbased imaging technique for non-invasively mapping the mechanical properties of tissues. Cortes et al demonstrated the feasibility of using MRE at very high frequencies in the IVD in vitro to estimate the nucleus pulposus stiffness. However, there has been no attempt at using MRE to measure the shear vibration response in a spine segment. The purpose of this study is to determine if MRE is capable of detecting spinal instability in an in vitro goat spine segment with induced disc degeneration and mechanical destabilization. The target audience of this research is MRI scientists involved in developing spinal imaging methods, radiologists involved in spinal imaging, clinicians involved in managing patients with low-back pain and disc-related spinal disorders, and basic scientists investigating DDD. Also, scientists working in MRE technique development may have interest in this research. Methods and Materials: (1) Intervertebral Disc Specimens. An entire goat thoraco-lumbar spine (T8S1) was removed with musculature and ligamentous structures intact. All posterior elements of the spine segment were removed to increase flexibility of the specimen, such that the entire spine segment consists of only vertebral bodies and discs. (2) Test Conditions. First, the control was taken as the spine segment before any alterations. Then, a series of four changes were made to the spine sequentially to cause varying amounts of mechanical instability: 1 – Trypsin solution was injected into the L2/L3 IVD to induce disc degeneration, as demonstrated in an in vitro animal model for disc degeneration. 2 – Partial cut through the posterior annulus of L1/L2 IVD (cut 1). 3 – Partial cut through the anterior annulus of T11/T12 IVD (cut 2). 4 – Completely cut through T11/T12 disc (cut 3). There was also a no motion case collected for comparison. (3) Mechanical vibration. A passive, pneumatic driver was used to apply mechanical vibrations at 120 Hz to the spine segment. Using a custom spine testing fixture, the spine segment was fixed on both ends and then the driver was positioned such that shear vibration was applied to the T12 vertebral body (fig. 1A). The spine was rigidly fixed to the surface of the driver (fig. 1B). A two-channel, rectangular receive coil was placed around the spine with the length of the spine parallel to the B0 direction (fig. 1C). The vibration direction was perpendicular to the B0 direction and the spinal column. (4) Wave imaging sequence. The disc specimens were imaged using a spin echo-based MRE sequence with the following parameters: 2 total cycles of 120 Hz motion-encoding gradients (2.4 Gauss/cm), motion sensitivity = 13.9μm/π, offsets = 4, 166.7/41-ms TR/TE, 30-cm FOV, one 10-mm slice, 256x128 matrix, 2 NEX. A standard 1.5T full-body MRI scanner (Signa 16X Software, GE Healthcare, Waukesha, WI) was used in the experiment. For all experiments, a sagittal slice was taken through the center of the spine segment. Motion encoding was done in the R/L (through-plane) direction (5) Data analysis. The resulting phase images were then masked and phase unwrapped. The unwrapped phase images were then bandpass filtered (0.1-20 waves per FOV) and interpolated to 8 offsets, which gives the interpolated wave images (fig. 2C). The unwrapped phase images were also Fourier transformed over time. The magnitude of the 1 harmonic (120 Hz) of this FFT is shown in Figure 2C. Line profiles were drawn from T10 to L3 on these FFT magnitude images for each testing condition to see the effect of each on the 120 Hz shear vibration. Results and Discussion: As shown in Figure 2, shear waves were seen in the entire spine segment for all conditions tested. The line profiles of the magnitude of the FFT 1 harmonic for all spine conditions are plotted in Figure 3. The induced disc degeneration and first 2 disc cuts changed the shear wave propagation and caused minor blips and jumps in the line profiles. But, the transected disc (cut 3) showed a very clear boundary in the magnitude of the FFT, as well as the wave images. Conclusions: These initial results suggest MRE may be capable of detecting spinal instability. Additionally, the data shows that low frequency MRE (~120 Hz) can be used to vibrate the entire spine from a single vibration source. Further work is needed to determine and improve the detectability of small changes in local spine mechanical properties.

Introduction: A woman born in the United States today has a 1 in 8 chance of having invasive breast cancer during her lifetime. An estimated 39,970 breast cancer deaths (39,520 women, 450 men) were expected in 2011. Death rates for breast cancer have steadily decreased in women since 1990, which represents progress in both earlier detection and improved treatment. Early diagnosis of breast cancer is critical, because treatment is most effective at the early stages of the disease. However, mammography, the principle tool for diagnosing breast cancer, has radiation exposure risks and is significantly less accurate in young women with mammographically dense breast tissue; women who also have a higher risk of developing breast cancer than women with less dense breast tissue. Other techniques, like ultrasound and contrast-enhanced MRI, have low specificity for breast cancer. In recent years there has been interest in exploring the potential of MR elastography as a method to augment the diagnostic specificity of breast MRI. In past implementations of breast MRE the required mechanical drivers have been placed in direct contact and to some extent compressing the breast. However, a recent human study has shown that compression can affect the mechanical properties of normal and diseased breast tissues. This may explains the inconsistencies in the reported stiffness values of breast tissues from different elastography methods in the literature. We have developed and tested a new breast MRE technique that does not require any tissue compression to avoid its possible effect on breast tissue stiffness. In this study, we further improved the noncompressive breast MRE technique with a more efficient breast MRE driver and a shorter MRE volumetric imaging sequence. Our purpose was to design the technique to be comfortable for the patients and capable of producing volumetric elastograms of both breasts simultaneously. Target audience includes clinicians and scientists interested in new technologies for MRI-based breast cancer imaging. Methods: (1) Noncompressive Breast Driver: The noncompressive breast MRE driver was designed to be narrower than the previous design (2 x 0.6 x 22 cm vs. 3.5 x 0.8 x 20 cm, width x thickness x length), which increased the efficiency of the driver due to its decreased volume. Our study was approved by our Institutional Review Board (IRB). Subjects were scanned in the prone position, feet first, in a 1.5 T MRI scanner (GE, Signa, Wisconsin, USA). The MRE driver was positioned in between the sternum and the bridge of a commercial breast RF coil (Liberty 9000 8-ch. breast coil, USA Instruments, Inc., Aurora, OH). (2) MRE Imaging Sequence: Imaging was performed with a 3D GRE MRE sequence similar to one previously reported. The parameters included vibration frequency = 40 Hz; FOVx/y/z = 30-34/30-34/14.4-18 cm; 4 phase offsets; motion-encoding gradient (MEG) amplitude = 2.8 G/cm; TR = 31.3 ms; TE = 27.2 ms (fat/water in-phase echo time); flip angle = 15o; BW = 31.25 kHz; axial imaging plane covering both breasts in the SI direction; acquisition matrix = 96X96X40; reconstruction matrix = 256X256X36; NEX = 1; SENSE acceleration factor = 2 (RL direction); total scan time = 9’54’’ (free breathing). (3) Calculation of Elastograms: The vector curl of the measured wave data was calculated using 3x3x3 derivative kernels on the wrapped phase data acquired in three orthogonal directions. A 3D local frequency estimation (LFE) inversion was performed on the curl data with 2D directional filters (cut-off frequencies of 2 and 128 cycles/FOV) to calculate the volumetric elastograms of the two breasts. Regions of interest were drawn in the adipose and glandular tissue of all subjects and in the tumor for the patient volunteer to measure the stiffness of the tissues. Results: Seven volunteers without known breast diseases and a 41-yearold female patient with a biopsy-proven invasive ductal carcinoma were enrolled in this preliminary study. The results are shown in Fig. 1. In the seven normal volunteers, the stiffness of adipose tissue ranged from 0.25 to 0.41 (mean = 0.33) kPa and glandular tissue ranged from 0.46 to 0.9 (mean = 0.64) kPa. For the patient, the stiffness of adipose tissue was 0.41 ± 0.1 kPa, glandular tissue was 0.90 ± 0.18 kPa and the invasive ductal carcinoma was 1.42 ± 0.17 kPa. Fig. 2 shows images of the patient with invasive ductal carcinoma who underwent contrast-enhanced breast MRI and the noncompressive breast MRE exams. CE-MRI shows that in the left breast, there is a heterogeneously enhancing mass in the right subareolar breast tissue corresponding to the biopsy-proven malignancy, with a size of 3.2 x 2.0 x 2.4 cm (Fig. 2(a), arrow). No abnormal indications were seen in the right breast. Breast MRE shows that the glandular tissue is heterogeneous in stiffness, and the carcinoma is much stiffer than the surrounding breast tissue (Fig. 2(b), arrow). Discussion: All of the subjects who underwent the noncompressive breast MRE exam felt it was comfortable. Patient comfort in these exams was maintained, in part, by the design of the MRE driver, which does not compress the breast tissue. We also developed a custom pad, not specific for MRE, that replaces the original commercial pad used on top of the breast RF coil to support the patient. The small, noncompressive breast MRE technique may reduce any anxieties the patient may have for traditional breast MRE due to the lack of breast compression. In addition to maintaining patient comfort, this noncompressive driver design also can avoid the possible problem of breast tissue compression changing the observed tissue stiffness while also providing a possible “one size fits all” solution to fitting the MRE driver to patients with different breast sizes since this driver does not require any patient-specific adjustments. This driver design should be compatible with any existing RF coils that have a bridge to support the patient’s sternum. The driver was able to deliver significant shear wave motion into both breasts in all subjects. The 3D GRE MRE acquisition successfully imaged the full volumetric breast wave field in all subjects and detected the invasive ductal carcinoma in the patient. Conclusion: The noncompressive breast MRE technique is comfortable and compatible with a commercial breast RF coil. It has proven to be reproducible in a small cohort of healthy volunteers and identified an invasive ductal carcinoma as being significantly stiffer than normal adipose and glandular tissue. Future work will include recruiting a larger number of healthy and patient volunteers to assess the potential for this technique to characterize and differentiate suspicious breast lesions. Acknowledgement: Diane Sauter, Jacqueline Duhn, and Pamela Trester for assistance in performing the noncompressive breast MRE exams. References: 1. Warner, E., N Engl J Med, 2011. 365(11). 2. Breast Cancer Facts & Figures 2011-2012, American Cancer Society. 3. Hooley, R.J., Radiology, 2012. 4. Siegmann, K.C.,Eur Radiol, 2010. 20(2). 5. McKnight, A.L.,AJR Am J Roentgenol, 2002. 178(6). 6. Sadigh, G., Breast Cancer Res Treat, 2012. 134(3). 7. Barr, R.G., J Ultrasound Med, 2012. 31(6). 8. Chen, J., ISMRM 2987, 2012. 9. Glaser, K.J., ISMRM 4669, 2009. 10. Manduca, A., Med Image Anal, 2001. 5(4).

Introduction: Osteoporosis is a skeletal disorder characterized by low bone mass and microarchitectural deterioration of bone tissue with a consequent increase in the fragility of bone and hence susceptibility to fracture[1]. About 8% of the U.S. population suffer from osteoporosis, 80% of them are women. Pathologic fractures are the hallmark of osteoporosis and happen when the bone cannot sustain minimal trauma, such as falling from standing height, or the bone is so weak that even a normal load can exceed its strength. Fractures are more common than heart disease or cancer in women, affecting 1 in 2 women and 1 in 6 men older than 50 in North America [2], causing approximately 2.3 million fractures annually at a cost of more than 23 billion dollars per year in the USA and Europe [3]. Clinically, dual-energy X-ray absorptiometry (DEXA), quantitative ultrasound (QUS), quantitative CT (QCT) and MRI-based techniques are used to evaluate bone mineral density (BMD) for the diagnosis and management of the disease. However, BMD is not the only component contributing to bone strength, which depends on the combination of bone mineral mass and the bone microstructure/architecture[4]. It has been shown that a deterioration in collagen content or collagen cross-linking can increase fracture risk [5]. Therefore, assessing the overall biomechanical properties of bone could be more valuable than measuring BMD alone. We have shown that MR Elastography (MRE) is a promising noninvasive method of evaluating the biomechanical properties of ex vivo bone[6]. The purpose of this study was to create an ex vivo osteoporosis model of bone and test the feasibility of using MRE to detect changes in bone stiffness. We hypothesize that bone stiffness decreases in this osteoporosis model and this decrease can be detected by MRE. Methods: (1) Osteoporosis model. Acetic acid exposure is known to soften bone by decreasing its calcium content. We used this method to create an ex vivo osteoporosis model. Five porcine ribs bought from a local grocery store were prepared with musculature removed (Fig. 1a). The weight of the five ribs ranged from 22.0 to 28.2 (mean = 24.4) grams, length ranged from 11.5 to 12.8 (mean = 12.1) cm and the thickness ranged from 1.2 to 1.7 (mean = 1.4) cm. The bones were soaked in a container with 1000 mL of distilled white vinegar (5% acidity, Supervalu, Inc. USA) and the container with the bones inside was kept in a constant 90°C water bath for 48 hours. MRE was performed on the five bones at room temperature (22°C) before the 48-hour acetic acid treatment for baseline measurements, and was then repeated after the treatment was finished and the bones were cooled down to room temperature. All of the bones were kept moist during the MRE exams using a wet paper tower wrapped around them. (2) Mechanical vibration. A piezoelectric stack mechanical driver was fabricated to apply mechanical vibrations to the bone at 1500 Hz. As seen in Fig. 1b, the driver was secured to a horizontal supporting bar, and was connected to the bone by a rigid clamp. The driver polarity was perpendicular to the supporting bar and the bone. The bone was inserted into the clamp and secured about 3 cm from one end of

(a) (b) (c) (d) (e) Fig. 3 MRE exam of volunteer v3. (a) Magnitude image of GREMRE. (b) Elastogram of both breasts. (c) AP, (d) RL and (e) SI axis of curl filtered shear wave field. The mean (range) stiffness of fat was 0.28(0.24 0.30) kPa. The mean (range) stiffness of glandular tissue was 0.61(0.37-0.79) kPa. Fig. 4. Stiffness of glandular and fat tissue in the six volunteers. Fig.1.Non-contact soft breast MRE driver. Improved Noncontact 3-Dimensional Breast MR Elastography Jun Chen, Roger Grimm, Kevin Glaser, Kugel Jennifer, Kay Pelletier, and Richard Ehman Radiology Department, Mayo Clinic, Rochester, MN, United States