Search

Shear wave elasticity imaging (SWEI) usually assumes an isotropic material; however, skeletal muscle is typically modeled as a transversely isotropic material with independent shear wave speeds in the directions along and across the muscle fibers. To capture these direction-dependent properties, we implemented a rotational 3-D SWEI system that measures the shear wave speed both along and across the fibers in a single 3-D acquisition, with automatic detection of the muscle fiber orientation. We tested and examined the repeatability of this system's measurements in the vastus lateralis of 10 healthy volunteers. The average coefficient of variation of the measurements from this 3-D SWEI system was 5.3% along the fibers and 8.1% across the fibers. When compared with estimated respective 2-D SWEI values of 16.0% and 83.4%, these results suggest using 3-D SWEI has the potential to improve the precision of SWEI measurements in muscle. Additionally, we observed no significant difference in shear wave speed between the dominant and non-dominant legs along (p = 0.26) or across (p = 0.65) the muscle fibers.

Ultrasonic rotational 3-D shear wave elasticity imaging (SWEI) has been used to induce and evaluate multiple shear wave modes, including both the shear horizontal (SH) and shear vertical (SV) modes in in vivo muscle. Observations of both the SH and SV modes allow the muscle to be characterized as an elastic, incompressible, transversely isotropic (ITI) material with three parameters: the longitudinal shear modulus μsubL/sub, the transverse shear modulus μsubT/sub, and the tensile anisotropy χsubE/sub. Measurement of the SV wave is necessary to characterize χsubE/sub, but the factors that influence SV mode generation and characterization with ultrasonic SWEI are complicated. This work uses Green's function (GF) simulations to perform a parametric analysis to determine the optimal interrogation parameters to facilitate visualization and quantification of SV mode shear waves in muscle. We evaluate the impact of five factors: μsubL/sub, μsubT/sub, χsubE/sub, fiber tilt angle [Formula: see text], and F-number of the push geometry on SV mode speed, amplitude, and rotational distribution. These analyses demonstrate that the following hold: 1) as μsubL/subincreases, SV waves decrease in amplitude so are more difficult to measure in SWEI imaging; 2) as μsubT/subincreases, the SV wave speeds increase; 3) as χsubE/subincreases, the SV waves increase in speed and separate from the SH waves; 4) as fiber tilt angle [Formula: see text] increases, the measurable SV waves remain approximately the same speed, but change in strength and in rotational distribution; and 5) as the push beam geometry changes with F-number, the measurable SV waves remain approximately the same speed, but change in strength and rotational distribution. While specific SV mode speeds depend on the combinations of all parameters considered, measurable SV waves can be generated and characterized across the range of parameters considered. To maximize measurable SV waves separate from the SH waves, it is recommended to use an F/1 push geometry and [Formula: see text].

In the context of ultrasonic hepatic shear wave elasticity imaging (SWEI), measurement success has been determined to increase when using elevated acoustic output pressures. As SWEI sequences consist of two distinct operations (pushing and tracking), acquisition failures could be attributed to (i) insufficient acoustic radiation force generation resulting in inadequate shear wave amplitude and/or (ii) distorted ultrasonic tissue motion tracking. In the study described here, an opposing window experimental setup that isolated body wall effects separately between the push and track SWEI operations was implemented. A commonly employed commercial track configuration was used, harmonic multiple-track-location SWEI. The effects of imaging through body walls on the pushing and tracking operations of SWEI as a function of mechanical index (MI), spanning 5 different push beam MIs and 10 track beam MIs, were independently assessed using porcine body walls. Shear wave speed yield was found to increase with both increasing push and track MI. Although not consistent across all samples, measurements in a subset of body walls were found to be signal limited during tracking and to increase yield by up to 35% when increasing electronic signal-to-noise ratio by increasing harmonic track transmit pressure.

Using a 3D rotational shear wave elasticity imaging (SWEI) setup, 3D shear wave data were acquired in the vastus lateralis of a healthy volunteer. The innate tilt between the transducer face and the muscle fibers results in the excitation of multiple shear wave modes, allowing for more complete characterization of muscle as an elastic, incompressible, transversely isotropic (ITI) material. The ability to measure both the shear vertical (SV) and shear horizontal (SH) wave speed allows for measurement of three independent parameters needed for full ITI material characterization: the longitudinal shear modulus μ(L), the transverse shear modulus μ(T), and the tensile anisotropy χ(E). Herein we develop and validate methodology to estimate these parameters and measure them in vivo, with μ(L) = 5.77 ± 1.00 kPa, μ(T) = 1.93 ± 0.41 kPa (giving shear anisotropy χ(μ) = 2.11 ± 0.92), and χ(E) = 4.67 ± 1.40 in a relaxed vastus lateralis muscle. We also demonstrate that 3D SWEI can be used to more accurately characterize muscle mechanical properties as compared to 2D SWEI.

Ultrasound shear wave elasticity (SWE) imaging measures tissue stiffness by tracking the shear wave speed (SWS), which is directly related to elastic moduli. One application of SWE is to quantify cervical softening during pregnancy in an effort to detect risk of spontaneous preterm birth and predict the success of induction of labor. Accurate SWS estimation in clinical data is challenging; existing robust methods are slow and do not provide a calibrated uncertainty metric. Here, we use a deep learning approach to estimate SWS from tracked displacement data and simultaneously produce a well-calibrated, quantitative metric of uncertainty. Our dataset consists of in vivo point-SWE cervix measurements acquired with a Siemens scanner at multiple timepoints during pregnancy in 28 subjects. We train a deep neural network to estimate the two parameters $m$ and σ of a lognormal distribution and derive the loss-function in a maximum likelihood framework. We use the median of 5 SWS estimators as ground truth and hold-out data from 3 subjects for testing the trained network. The average relative error in SWS was 13%, and the estimated uncertainty closely approximated the spread in true error, as measured by binned root-mean-square error. Our deep learning approach is the first SWS estimator in the toolbox that provides a well-calibrated and easily-interpretable uncertainty metric.

To characterize muscle tissue as a viscoelastic transversely isotropic material, we use rotational volumetric shear wave elasticity imaging (SWEI) and group speed viscoelastic analysis. We performed these measurements in vivo in the vastus lateralis of two healthy volunteers, both at rest and in varying degrees of contraction. We found a high degree of repeatability in our at rest elasticity measurements, determining that the stiffness along the fibers is higher than across the fibers. The viscosity at rest is approximately equal and low in both directions. In our contraction analysis, we found that stiffness increases substantially along the fibers and increases slightly across the fibers with increasing contraction. No clear trends were observed between viscosity and contraction level. Future work will improve the method of model-based viscoelastic characterization within the context of transversely isotropic materials and expand this analysis to more subjects.

Skeletal muscle exhibits transverse isotropy with a symmetry axis in the muscle fiber direction. Characterization of muscle mechanical properties with ultrasound shear wave elasticity imaging requires an accurate measurement of the 3D muscle fiber orientation (MFO): rotation and tilt. Existing approaches apply to 2D B-mode images, extract only fiber tilt, and detect individual fibers. Here, we present a Fourier-domain approach for calculating 3D MFO from 3D B-mode volumes acquired using two imaging setups: 1) a cylindrical volume acquired by rotating a linear transducer, and 2) a rectangular volume acquired by a rectilinear matrix array transducer. We imaged the vastus lateralis muscle of a healthy volunteer and also manually measured the orientation of individual fibers observed in these two B-mode volumes to assess heterogeneity. For rotation and tilt respectively, the standard deviations were 6.4° and 1.5° for the rotational B-mode (n=7) and 2.7° and 1.5° for the matrix B-mode (n=43). We validated our proposed algorithm on in silico and in vivo data: errors in rotation and tilt from the mean orientation were within 1° for both imaging setups and less than the in vivo MFO heterogeneity. Lastly, we performed a Bland- Altman analysis of rotation angles estimated from B-mode and from shear wave elastography data (n=35): bias was less than than 1°, and 95% limits of agreement was ±10°. Our Fourier-domain approach precisely computes the average 3D MFO from 3D ultrasound B-mode volumes of muscle, and these orientation estimates will be used in ongoing muscle characterization studies.

We previously have demonstrated the use of ultrasonic rotational 3D SWEI to measure all three mechanical properties necessary to fully characterize the vastus lateralis muscle in vivo as an incompressible transversely isotropic material: transverse shear modulus $\mu_{T}$ , longitudinal shear modulus $\mu_{L}$ , and tensile anisotropy $\chi_{E}$ . In this work we investigate how varying knee flexion angle affects the mechanical properties of the vastus lateralis. As knee flexion angle increased, $\mu_{L},\ \chi_{E}$ , and $\chi_{\mu}((\mu_{L}-\mu_{T})/\mu_{T})$ all increased, while $\mu_{T}$ remained consistent across knee angles. At all nine knee flexion angles investigated in a healthy volunteer $\chi_{E} > \chi_{\mu}$ , indicating that these are independent parameters.

Correctly calculating skin stiffness with ultrasound shear wave elastography techniques requires an accurate measurement of skin thickness. We developed and compared two algorithms, a thresholding method and a deep learning method, to measure skin thickness on ultrasound images. Here, we also present a framework for weakly annotating an unlabeled dataset in a time-effective manner to train the deep neural network. Segmentation labels for training were proposed using the thresholding method and validated with visual inspection by a human expert reader. We reduced decision ambiguity by only inspecting segmentations at the center A-line. This weak annotation approach facilitated validation of over 1000 segmentation labels in 2 hours. A lightweight deep neural network that segments entire 2D images was designed and trained on this weakly-labeled dataset. Averaged over six folds of cross-validation, segmentation accuracy was 57% for the thresholding method and 78% for the neural network. In particular, the network was better at finding the distal skin margin, which is the primary challenge for skin segmentation. Both algorithms have been made publicly available to aid future applications in skin characterization and elastography.

Received ultrasonic data are carrier-modulated broadband signals and are converted to different formats depending on the application. Common formats extracted from the raw radio-frequency (RF) data include a complex-valued analytic signal, the envelope/magnitude, demodulated in-phase and quadrature (IQ) components, and the phase angle. Deep neural networks (DNNs) have been applied to a variety of ultrasound signal processing tasks, yet how the format of input data affects DNN results has not been well-characterized. Here, we investigate how the data format affects DNN performance and robustness for two tasks: speckle reduction and displacement estimation. Simulated data were used for training, and multiple networks were trained for each task and each input format. Network loss was compared on test data with either added white noise or a different imaging frequency. For speckle noise reduction, networks using magnitude or IQ data were more robust to changes in imaging frequency than those using the carrier-modulated RF or analytic signals. Networks using magnitude were the least robust against added white noise. For displacement estimation, networks required an input data format with phase information to perform well. Performance for all input formats were equally affected by added noise, but the RF and analytic signals were the most robust to changes in center frequency.

Serum response factor (SRF), a key transcription factor, plays an important role in regulating cell functions such as proliferation and differentiation. Most proteins are unstable, and protein stability is regulated through the ubiquitin-proteasome system (UPS) or the autophagy lysosome pathway (ALP). Whether SRF is degraded and what mechanisms control SRF protein stability remain unexplored. Western blot analyses of cells treated with cycloheximide (CHX), a protein synthesis inhibitor, showed that SRF was degraded in a time-dependent manner. Moreover, we observed that SRF undergoes autophagy-dependent destruction, which is accelerated by serum deprivation. Through bioinformatics screening, we found that SRF contains the GSK3β phosphorylation motif (T/SPPXS): SPDSPPRSDPT, which is conserved from zebrafish to humans. Serum deprivation stimulated GSK3β activation that then potentiates SRF degradation through the autophagy lysosome pathway. Since SRF is important for numerous cellular activities, our results suggest that the autophagy-dependent SRF degradation pathway may provide a new avenue to modulate SRF-mediated cell functions.

Skin stiffness correlates with the progression of sclerotic skin diseases. Ultrasound shear wave elasticity imaging techniques can measure skin stiffness, but an accurate skin thickness measurement is required to compute the elastic modulus. We explored different automated methods to segment the skin for use in real-time skin elastography. Local gradient-based methods could not robustly segment the skin on our B-mode images, so we developed a new thresholding method to detect the edges of the skin. We also used our thresholding method to generate labels to train a deep neural network. We compared the performance of thresholding and the trained network for central thickness estimation on a held-out test set. Our thresholding method correctly segmented 58% of images, and the neural network correctly segmented 82%. More than half of thresholding failures on the test set were from overestimation of the bottom skin boundary. The neural network had significantly less overestimation failures and similar rates of failure due to bubbles and underestimation.

In 10 patients with cutaneous sclerotic skin conditions, shear wave elasticity imaging (SWEI) was used to measure shear wave speed in the skin in vivo. The dermis was segmented using an thresholding-based algorithm and group speeds were found in both the dermis and the subcutaneous tissue. Phase velocities were calculated in the layer, and a linear regression was fit to obtain a dispersion slope. A statistically significant non-zero dispersion was found in the skin in vivo. An ordinary least squares multiple linear regression was performed to assess what factors affected the dispersion slope. Thickness did not have a statistical effect on the dispersion slope in each model, and subcutaneous speed was only found to have a statistical effect at one of two investigated frequency-thickness products. This work shows SWEI based dispersion in the skin in vivo and demonstrates the importance of subcutaneous speed in future models.

Endothelial dysfunction, characterized by impaired activation of endothelial nitric oxide (NO) synthase (eNOS) and ensued decrease of NO production, is a common mechanism of various cardiovascular pathologies, including hypertension and atherosclerosis. Laminar blood flow-mediated specific signaling cascades modulate vascular endothelial cells (ECs) structure and functions. We have previously shown that flow-stimulated Gab1 (Grb2-associated binder-1) tyrosine phosphorylation mediates eNOS activation in ECs, which in part confers laminar flow atheroprotective action. However, the molecular mechanisms whereby flow regulates Gab1 tyrosine phosphorylation and its downstream signaling events remain unclear. Here we show that platelet endothelial cell adhesion molecule-1 (PECAM1), a key molecule in an endothelial mechanosensing complex, specifically mediates Gab1 tyrosine phosphorylation and its downstream Akt and eNOS activation in ECs upon flow rather than hepatocyte growth factor (HGF) stimulation. Small interfering RNA (siRNA) targeting PECAM1 abolished flow- but not HGF-induced Gab1 tyrosine phosphorylation and Akt, eNOS activation as well as Gab1 membrane translocation. Protein-tyrosine phosphatase SHP2, which has been shown to interact with Gab1, was involved in flow signaling and HGF signaling, as SHP2 siRNA diminished the flow- and HGF-induced Gab1 tyrosine phosphorylation, membrane localization and downstream signaling. Pharmacological inhibition of PI3K decreased flow-, but not HGF-mediated Gab1 phosphorylation and membrane localization as well as eNOS activation. Finally, we observed that flow-mediated Gab1 and eNOS phosphorylation in vivo induced by voluntary wheel running was reduced in PECAM1 knockout mice. These results demonstrate a specific role of PECAM1 in flow-mediated Gab1 tyrosine phosphorylation and eNOS signaling in ECs.

Endothelial nitric oxide (NO) synthase (eNOS/NOS3)-derived NO represents an important vasculoprotective molecule in the cardiovascular system. The transient receptor potential ion channel vanilloid subtype 4 (TRPV4) mediates vascular mechanosensitivity, in part through an increase in blood flow-mediated NO production and endothelium-dependent relaxation. Here, we hypothesized that non-mechanical TRPV4 activation can mimic laminar flow-induced signaling and thereby limits monocyte adhesion and atherosclerosis. We observed that GSK1016790A, a novel, potent and specific agonist of TRPV4, mimics laminar flow, thereby inducing robust phosphorylation (Ser1177) of eNOS in a TRPV4 and calcium-dependent manner. Mechanistically, GSK1016790A induced eNOS phosphorylation, NO production and vasorelaxation via calcium/calmodulin-dependent kinase kinase (CaMKK)/AMP-activated protein kinase (AMPK) pathway. In functional experiments, GSK1016790A inhibited TNFα-induced monocytes adhesion to human endothelial cells by inhibiting the expression of intercellular adhesion molecule 1 (ICAM1) and vascular cell adhesion molecule 1 (VCAM1), and this inhibition was abolished by an eNOS inhibitor L-NG-Nitroarginine Methyl Ester (L-NAME) or AMPK inhibitor Compound C. Mice given GSK1016790A showed increased eNOS (2.5- fold increase versus vehicle, n=12, P

Vascular endothelial dysfunction and inflammation are hallmarks of atherosclerosis. It is important to understand the detailed mechanisms for regulation of endothelial function and inflammation. Emerging evidence suggests that the transcription factor Kruppel-like Factor 2 (KLF2) is a key mediator of the anti-inflammatory and anti-thrombotic properties of the endothelium. KLF2 also mediates laminar blood flow-dependent endothelial atheroprotective functions. Despite intense investigation of KLF2 function, little is known of the molecular mechanisms for regulation of KLF2 transcriptional activation. Histone deacetylase 5 (HDAC5) HDAC5 is a transcriptional repressor, and we have previously shown that HDAC5 inhibits flow-mediated endothelial nitric oxide synthase (eNOS) gene expression (Wang & Jin, Blood 2010). Here, we demonstrate that HDAC5 physically associates with KLF2 and negatively regulates KLF2 transcriptional activation. Using co-transfected cell line and GST-fusion protein pull-down assays, we found that HDAC5 directly interacts with KLF2. We also detected the association of endogenous HDAC5 and KFL2 in endothelial cells. To understand the nature of this interaction, we further mapped the HDAC5 and KLF2 interaction domains. Using luciferase reporter assay, we observed that HDAC5, through association with KLF2, repressed KLF2 transcriptional activation. Moreover, HDAC5 inhibited KLF2-dependent eNOS promoter activity and eNOS expression in endothelial cells, which reveals epigenetic mechanism whereby flow mediates eNOS expression through stimulating nuclear export of HDAC5. Collectively, these results define a novel transcriptional pathway that regulates KLF2-dependent gene expression and suggest a new molecular target to prevent vascular inflammation and endothelial dysfunction associated with cardiovascular disease.

Background Vascular hyperpermeability is critical components of many human diseases, including inflammation, cancer and diabetic retinopathy. Vascular endothelial growth factor (VEGF) is a potent vascular permeability factor. Better understanding of the molecular mechanisms by which VEGF induces vascular permeability may provide novel therapeutic targets for these diseases. We and others have recently demonstrated that protein kinase D1 (PKD1), a newly discovered serine/threonine protein kinase, is implicated in VEGF signaling. We hypothesized that PKD1 plays a crucial role in VEGF-induced vascular permeability in vivo and vitro by promoting endothelial cell stress fiber formation. Methods and results Using Tie2-Cre transgenic mice and PKD1 flox mice, we generated endothelium-specific PKD1 knockout (PKD1-ecKO) mice. PKD1-ecKO mice were viable and born at the expected Mendelian ratio. Comparing with wild type mice, PKD1-ecKO mice exhibited the dramatic reduction of VEGF-induced vascular permeability in lung and ear with Miles assay, expressed as Evans blue leakage. Selective pharmacological PKD inhibitor also significantly attenuated in vivo vascular permeability induced by VEGF intravenously injection in mice. In cultured endothelial monolayer, down-regulation of PKD1 by siRNA, GFP-PKD1 dominant negative mutant plasmid or PKD inhibitor significantly reduced VEGF-induced endothelial permeability, measured by bovine serum albumin leakage through endothelial monolayer cultured on transwell system. Moreover, we found that the inhibition of PKD1 diminished F-actin stress fiber formation and the dismantlement of endothelial adherens junctions that lead to intercellular gap formation and leakage in response to VEGF. We further observed that PKD1 acted as a downstream mediator of Rho activation in VEGF-induced stress fiber formation. Specifically, Rho inhibitor Y27632 blocked VEGF-induced PKD1 phosphorylation, while down-regulation of PKD1 expression and activation did not affect VEGF-stimulated Rho activation in endothelial cells. Conclusion PKD1 is required for VEGF-induced vascular permeability in vivo and in vitro through promoting F-actin stress fiber formation and disrupting endothelial adherens junctions.