Magnetic Resonance Elastography for Staging Liver Fibrosis in the Oncopig.

This pilot study investigated the feasibility of using magnetic resonance elastography (MRE) for the non-invasive detection and quantification of liver fibrosis in the Oncopig cancer model. Seven 8-week-old Oncopigs underwent alcoholic liver fibrosis induction and serial MRE imaging and liver biopsy at 1, 2, and 3 months post procedure. MRE was utilized to quantify liver stiffness, and liver fibrosis was histologically graded using the METAVIR system. The primary outcome measure was the capability to detect and quantify liver fibrosis using MRE with radiologic-pathologic correlation. Liver fibrosis induction, MRE imaging, and liver biopsy were successfully performed. MRE liver fibrosis was evident in 57% (4/7), 50% (3/6), and 40% (2/5) animal subjects 1, 2, and 3 months after fibrosis induction, with mean liver stiffness of 2.94, 3.25, and 2.91 kPa, respectively. Histological liver fibrosis was noted in 71% (5/7), 100% (5/5), and 100% (5/5) of animal subjects with available tissue samples. There was no significant statistical correlation between the MRE-measured liver stiffness and the METAVIR fibrosis scores. In conclusion, quantifiable liver fibrosis may be induced in the Oncopig. MRE has potential utility in non-invasively detecting liver stiffness in this large-animal preclinical model, though tissue biopsy was more sensitive in demonstrating disease.

1-Norm waveform analysis for MR elastography-based quantification of inhomogeneity: Effects of the freeze-thaw cycle and Alzheimer's disease.

BACKGROUND: Despite its success in the mechanical characterization of biological tissues, magnetic resonance elastography (MRE) uses ill-posed wave inversions to estimate tissue stiffness. 1-Norm has been recently introduced as a mathematical measure for the scattering of mechanical waves due to inhomogeneities based on an analysis of the delineated contours of wave displacement. PURPOSE: To investigate 1-Norm as an MRE-based quantitative biomarker of mechanical inhomogeneities arising from microscopic structural tissue alterations caused by the freeze-thaw cycle (FTC) or Alzheimer's disease (AD). METHODS: In this proof-of-concept study, we prospectively investigated excised porcine kidney (n = 6), liver (n = 6), and muscle (n = 6) before vs. after the FTC at 500-2000 Hz and excised murine brain of healthy controls (n = 3) vs. 5xFAD species with AD (n = 3) at 1200-1800 Hz using 0.5 T tabletop MRE. 1-Norm analysis was compared with conventional wave inversion. RESULTS: While the FTC reduced both stiffness and inhomogeneity in kidney, liver, and muscle tissue, AD led to lower brain stiffness but more pronounced mechanical inhomogeneity. CONCLUSION: Our preliminary results show that 1-Norm is sensitive to tissue mechanical inhomogeneity due to FTC and AD without relying on ill-posed wave inversion techniques. 1-Norm has the potential to be used as an MRE-based diagnostic biomarker independent of stiffness to characterize abnormal conditions that involve changes in tissue mechanical inhomogeneity.

Assessment of Mechanically Induced Changes in Helical Fiber Microstructure Using Diffusion Tensor Imaging.

Noninvasive methods to detect microstructural changes in collagen-based fibrous tissues are necessary to differentiate healthy from damaged tissues in vivo but are sparse. Diffusion Tensor Imaging (DTI) is a noninvasive imaging technique used to quantitatively infer tissue microstructure with previous work primarily focused in neuroimaging applications. Yet, it is still unclear how DTI metrics relate to fiber microstructure and function in musculoskeletal tissues such as ligament and tendon, in part because of the high heterogeneity inherent to such tissues. To address this limitation, we assessed the ability of DTI to detect microstructural changes caused by mechanical loading in tissue-mimicking helical fiber constructs of known structure. Using high-resolution optical and micro-computed tomography imaging, we found that static and fatigue loading resulted in decreased sample diameter and a re-alignment of the macro-scale fiber twist angle similar with the direction of loading. However, DTI and micro-computed tomography measurements suggest microstructural differences in the effect of static versus fatigue loading that were not apparent at the bulk level. Specifically, static load resulted in an increase in diffusion anisotropy and a decrease in radial diffusivity suggesting radially uniform fiber compaction. In contrast, fatigue loads resulted in increased diffusivity in all directions and a change in the alignment of the principal diffusion direction away from the constructs' main axis suggesting fiber compaction and microstructural disruptions in fiber architecture. These results provide quantitative evidence of the ability of DTI to detect mechanically induced changes in tissue microstructure that are not apparent at the bulk level, thus confirming its potential as a noninvasive measure of microstructure in helically architected collagen-based tissues, such as ligaments and tendons.

Tabletop MR elastography for investigating effects of the freeze-thaw cycle on the mechanical properties of biological tissues.

PURPOSE: We aimed at characterizing the effects of the freeze-thaw cycle (FTC) on ex vivo specimens of porcine muscle, liver, kidney, and brain using tabletop magnetic resonance elastography (MRE) combined with rheological modeling. While frozen tissue banks potentially facilitate access to large amounts of well-preserved biospecimens, the impact of the FTC on their viscoelastic properties remains elusive. METHODS: In this proof-of-concept study, fresh specimens from porcine lumbar muscle (n = 6), liver (n = 6), kidney (n = 6), and brain (n = 6) were examined before and after the FTC using 0.5T tabletop MRE at 500 Hz, 1000 Hz, 1500 Hz, and 2000 Hz. Seven standard rheological models (Maxwell, Springpot, Voigt, Zener, Jeffrey, fractional Voigt, fractional Zener) were employed to calculate frequency independent viscoelastic parameters. RESULTS: The Zener rheological model showed the best fit quality for tissues before and after FTC in the investigated frequency range. Global rheological behavior after the FTC was softer for all tissues. Differences in mechanical parameters between tissues were preserved after the FTC and showed similar trends as before the FTC. Moreover, rheological fit quality improved after the FTC - a result that will be beneficial in investigating frozen tissue bank samples. CONCLUSION: Multifrequency tabletop MRE allows rheological characterization of tissue samples before and after the FTC. Our results encourage further biomechanical characterization of frozen tissue bank samples, which may provide valuable information on the diagnostic potential of elastographic methods.

Investigating the heterogeneity of viscoelastic properties in prostate cancer using MR elastography at 9.4T in fresh prostatectomy specimens.

PURPOSE: To quantify the heterogeneity of viscoelastic tissue properties in prostatectomy specimens from men with prostate cancer (PC) using MR elastography (MRE) with histopathology as reference. METHODS: Twelve fresh prostatectomy specimens were examined in a preclinical 9.4T MRI scanner. Maps of the complex shear modulus (|G*| in kPa) with its real and imaginary part (G' and G" in kPa) were calculated at 500 Hz. Prostates were divided into 12 segments for segment-wise measurement of viscoelastic properties and histopathology. Coefficients of variation (CVs in %) were calculated for quantification of heterogeneity. RESULTS: Group-averaged values of cancerous vs. benign segments were significantly increased: |G*| of 12.13 kPa vs. 6.14 kPa, G' of 10.84 kPa vs. 5.44 kPa and G" of 5.45 kPa vs. 2.92 kPa, all p < 0.001. In contrast, CVs were significantly increased for benign segments: 23.59% vs. 26.32% (p = 0.014) for |G*|, 27.05% vs. 37.84% (p < 0.003) for G', and 36.51% vs. 50.37% (p = 0.008) for G". DISCUSSION: PC is characterized by a stiff yet homogeneous biomechanical signature, which may be due to the unique nondestructive growth pattern of PC with intervening stroma, providing a rigid scaffold in the affected area. In turn, increased heterogeneity in benign prostate segments may be attributable to the presence of different prostate zones with involvement by specific nonmalignant pathology.

Longitudinal study of sub-regional cerebral viscoelastic properties of 5XFAD Alzheimer's disease mice using multifrequency MR elastography.

PURPOSE: To study sub-regional, longitudinal changes occurring inside brains of 5XFAD mice, an Alzheimer's disease (AD) model, based on viscoelastic parameters derived using MR elastography and their spatial variation. METHODS: Female 5XFAD and non-transgenic B6SJLF1/J mice as controls (n = 9 for both groups) were used for the study. Scans were performed inside a 9.4T preclinical MRI scanner using SampLe Interval Modulation-magnetic resonance elastography (SLIM-MRE). Experiments were performed at ages 2, 4, and 6 mo, and by using three actuation frequencies: 900, 1000, and 1100 Hz. Multifrequency dual elasto-visco (MDEV) reconstruction was used to combine 3D multifrequency MRE data and calculate magnitude G∗ , and phase angle φ, of the complex shear modulus G∗ . Mean values were measured for the overall brain and sub-regions associated with the early onset of AD, to check for the effect of aging and mouse model. Spatial coefficient of variation (CV) of both parameters across different age-groups were analyzed. RESULTS: G∗ and φ values reduced with age for overall brain in 5XFAD mice with significant difference in mean G∗ between 5XFAD and control mice at 6 mo (P = .029). Analyzing values from the hippocampal region highlighted drop in mean G∗ and φ values. The CV of G∗ inside hippocampus enabled differentiation at 4 mo with it being significantly lower in 5XFAD mice (P = .0007). CONCLUSION: Multifrequency 3D MRE revealed longitudinal viscoelastic changes in 5XFAD mice and the CV of G∗ in brain sub-regions may qualify as biomarker for early diagnosis of AD.

Diagnostic performance of tomoelastography of the liver and spleen for staging hepatic fibrosis.

OBJECTIVES: To determine the diagnostic performance, cut-off values, and optimal drive frequency range for staging hepatic fibrosis using tomoelastography by multifrequency MR elastography of the liver and spleen. METHODS: This prospective study consecutively enrolled a total of 61 subjects between June 2014 and April 2017: 45 patients with chronic liver disease and proven stage of fibrosis and 16 healthy volunteers. Tomoelastography was performed at 1.5 T using six drive frequencies from 35 to 60 Hz. Cut-off values and AUC were calculated. Shear wave speed (in m/s) of the liver and spleen was assessed separately and in combination as a surrogate of stiffness. RESULTS: For compound multifrequency processing of the liver, cut-off and AUC values by fibrosis stage were as follows: F1, 1.52 m/s and 0.89; F2, 1.55 m/s and 0.94; F3, 1.67 m/s and 0.98; and F4, 1.72 m/s and 0.98. Diagnostic performance of the best single drive frequencies (45 Hz, 55 Hz, 60 Hz) was similar (mean AUC = 0.95, respectively). Combined analysis of the liver and spleen slightly improved performance at 60 Hz in F4 patients (mean AUC = 0.97 vs. 0.95, p = 0.03). Full-field-of-view elastograms displayed not only the liver and spleen but also small anatomical structures including the pancreas and major vessels. CONCLUSION: Tomoelastography provides full-field-of-view elastograms with unprecedented detail resolution and excellent diagnostic accuracy for staging hepatic fibrosis. Our analysis of single-frequency tomoelastography suggests that scan time can be further reduced in future studies, making tomoelastography easier to implement in clinical routine. KEY POINTS: • Tomoelastography provides full-field-of-view elastograms of the abdomen with unprecedented detail resolution and excellent diagnostic accuracy for staging hepatic fibrosis. • Diagnostic performance of single-frequency tomoelastography at higher frequencies (45 Hz, 55 Hz, 60 Hz) and compound multifrequency processing are equivalent for staging hepatic fibrosis. • Combined assessment of hepatic and splenic stiffness slightly improves diagnostic performance for staging hepatic fibrosis.

Prostate cancer assessment using MR elastography of fresh prostatectomy specimens at 9.4 T.

PURPOSE: Despite its success in the assessment of prostate cancer (PCa), in vivo multiparametric MRI has limitations such as interobserver variability and low specificity. Several MRI methods, among them MR elastography, are currently being discussed as candidates for supplementing conventional multiparametric MRI. This study aims to investigate the detection of PCa in fresh ex vivo human prostatectomy specimens using MR elastography. METHODS: Fourteen fresh prostate specimens from men with clinically significant PCa without formalin fixation or prior radiation therapy were examined by MR elastography at 500 Hz immediately after radical prostatectomy in a 9.4T preclinical scanner. Specimens were divided into 12 segments for both calculation of storage modulus (G' in kilopascals) and pathology (Gleason score) as reference standard. Sensitivity, specificity, and area under the receiver operating characteristic curve were calculated to assess PCa detection. RESULTS: The mean G' and SD were as follows: all segments, 8.74 ± 5.26 kPa; healthy segments, 5.44 ± 4.40 kPa; and cancerous segments, 10.84 ± 4.65 kPa. The difference between healthy and cancerous segments was significant with P ≤ .001. Diagnostic performance assessed with the Youden index was as follows: sensitivity, 69%; specificity, 79%; area under the curve, 0.81; and cutoff, 10.67 kPa. CONCLUSION: Our results suggest that prostate MR elastography has the potential to improve diagnostic performance of multiparametric MRI, especially regarding its 2 major limitations: interobserver variability and low specificity. Particularly the high value for specificity in PCa detection is a stimulating result and encourages further investigation of this method.

An investigation into the relationship between inhomogeneity and wave shapes in phantoms and ex vivo skeletal muscle using Magnetic Resonance Elastography and finite element analysis.

Soft biological tissues such as skeletal muscle and brain white matter can be inhomogeneous and anisotropic due to the presence of fibers. Unlike biological tissue, phantoms with known microstructure and defined mechanical properties enable a quantitative assessment and systematic investigation of the influence of inhomogeneities on the nature of shear wave propagation. This study introduces a mathematical measure for the wave shape, which the authors call as the 1-Norm, to determine the conditions under which homogenization may be a valid approach. This is achieved through experimentation using the Magnetic Resonance Elastography technique on 3D printed inhomogeneous fiber phantoms as well as on ex-vivo porcine lumbus muscle. In addition, Finite Element Analysis is used as a tool to decouple the effects of directional anisotropy from those of inhomogeneity. A correlation is then established between the values of 1-Norm derived from the wave front geometry, and the spacing (d) between neighboring inhomogeneities (spherical inclusions or fibers and fiber intersections in phantoms and muscle). Smaller values of 1-Norm indicate less wave scattering at the locations of fiber intersections, which implies that the wave propagation may be approximated to that of a homogeneous medium; homogenization may not be a valid approximation when significant scattering occurs at the locations of inhomogeneities. In conclusion, the current study proposes 1-Norm as a quantitative measure of the magnitude of wave scattering in a medium, which can potentially be used as a homogeneity index of a biological tissue.

Simultaneous 3D MR elastography of the in vivo mouse brain.

The feasibility of sample interval modulation (SLIM) magnetic resonance elastography (MRE) for the in vivo mouse brain is assessed, and an alternative SLIM-MRE encoding method is introduced. In SLIM-MRE, the phase accumulation for each motion direction is encoded simultaneously by varying either the start time of the motion encoding gradient (MEG), SLIM-phase constant (SLIM-PC), or the initial phase of the MEG, SLIM-phase varying (SLIM-PV). SLIM-PC provides gradient moment nulling, but the mutual gradient shift necessitates increased echo time (TE). SLIM-PV requires no increased TE, but exhibits non-uniform flow compensation. Comparison was to conventional MRE using six C57BL/6 mice. For SLIM-PC, the Spearman's rank correlation to conventional MRE for the shear storage and loss modulus images were 80% and 76%, respectively, and likewise for SLIM-PV, 73% and 69%, respectively. The results of the Wilcoxon rank sum test showed that there were no statistically significant differences between the spatially averaged shear moduli derived from conventional-MRE, SLIM-PC, and SLIM-PV acquisitions. Both SLIM approaches were comparable to conventional MRE scans with Spearman's rank correlation of 69%-80% and with 3 times reduction in scan time. The SLIM-PC method had the best correlation, and SLIM-PV may be a useful tool in experimental conditions, where both measurement time and T2 relaxation is critical.

High Field Sodium MRI Assessment of Stem Cell Chondrogenesis in a Tissue-Engineered Matrix.

The development of non-invasive assessment techniques in vitro and in vivo is essential for monitoring and evaluating the growth of engineered cartilage tissues. Magnetic resonance imaging (MRI) is the leading non-invasive imaging modality used for assessing engineered cartilage. Typical MRI uses water proton relaxation times (T1 and T2) and apparent diffusion coefficient (ADC) to assess tissue growth. These techniques, while excellent in providing the first assurance of tissue growth, are unspecific to monitor the progress of engineered cartilage extracellular matrix components. In the current article, we present high field (11.7 T, (1)H freq. = 500 MHz) sodium MRI assessment of tissue-engineered cartilage at the early stage of tissue growth in vitro. We observed the chondrogenesis of human bone marrow derived stromal cells seeded in a gradient polymer-hydrogel matrix made out of poly(85 lactide-co-15 glycolide)--PuraMatrix™ for 4 weeks. We calculated the sodium concentration in the engineered constructs using a model of sodium MRI voxels that takes into account scaffold volume, cell density and amount of glycosaminoglycan (GAG). The sodium concentration was then converted to the fixed charge density (FCD) and compared with FCD derived from biochemical GAG analysis. Despite the small amount of GAG present in the engineered constructs, the sodium MRI derived FCD is found to be correlated (Pearson correlation coefficient R = 0.79) with the FCD derived from biochemical analysis. We conclude that sodium MRI could prove to be an invaluable tool in assessing engineered cartilage quantitatively during the repair or regeneration of cartilage defects.

A DTI study to probe tumor microstructure and its connection with hypoxia.

Solid tumors have chaotic organization of blood vessels, disruptive nerve paths and muscle fibers that result in a hostile and heterogeneous microenvironment. These tumor regions are often hypoxic and resistant to radiation therapy. The knowledge of partial pressure of oxygen concentration (pO2), in conjunction with the information about tissue organization, can predict tissue health and may eventually be used in combination with intensity-modulated radiation therapy (IMRT) for targeted destruction of radiation-resistant areas, while sparing healthy tissues. Diffusion tensor imaging (DTI) based parameter fractional anisotropy (FA) can be used to assess organization of tissue microstructure, whereas the pO2 can be measured using electron paramagnetic resonance oxygen imaging (EPROI). This study is our first step to connect these two important physiological parameters. We calculated FA in fixed fibrosarcoma (FSa) grown in hind leg of nude mice (n = 6) using preclinical 9.4 T MRI. The FA in tumor region (0.34 ± 0.014) was found to be lower when compared to normal surrounding region (0.36 ± 0.013). We hypothesized that the change in FA is directly correlated with the change in oxygen concentration in tumor. We present preliminary in vivo results showing a positive correlation (R = 0.85, p = 0.017) between the FA and pO2 values acquired for MCa4 tumor (n = 1) using DTI and EPROI.