Zhang, Xiaoming

Lung density is directly associated with lung pathology. Computed Tomography (CT) evaluates lung pathology using the Hounsfield unit (HU) but not directly the lung density. We recently developed a lung ultrasound surface wave elastography (LUSWE) technique for measuring the surface wave speed of superficial lung tissue. The objective of this study is to analyze lung density of superficial lung tissue using deep neural network (DNN) and wave speed measurements obtained from LUSWE. The synthetic training dataset (7,88,000 in total) in terms of wave speeds of superficial lung tissue at different excitation frequencies from LUSWE, viscoelasticity and density of lung tissue was used to train the DNN. DNN is composed of 3 hidden layers of 1024 neurons and trained for 10 epochs with a batch size of 4096 and a learning rate of 0.001 with Adam optimizer. Test dataset in terms of wave speeds at different excitation frequencies (100, 150, and 200 Hz) as well as elasticity was used to predict the lung density and evaluate its accuracy. The obtained results showed that predictions matched well with test dataset (validation accuracy is 0.997). This method may be useful to analyze lung density based on the LUSWE measurements.

Purpose: Abnormal ocular biomechanical properties may be important for understanding the risk of glaucoma. However, there are no clinical methods for measuring standard material properties in patients. In this feasibility study we demonstrated proof-of-principle for a novel method, ultrasound surface wave elastography (USWE), to determine the in vivo Young’s modulus of elasticity of corneas in normal human eyes. Methods: In total, 20 eyes of 10 healthy subjects (mean age, 51.4±7.2; ±SD; range, 43 to 64 y) were studied. A spherical-tipped probe (3-mm diameter) was placed on closed eyelids and generated a gentle harmonic vibration at 100 Hz for 0.1 second. Wave speed propagation in the cornea was measured by USWE, and Young’s modulus was calculated from the wave speed. Associations between Young’s modulus and intraocular pressure (IOP), age, central corneal thickness, and axial length were explored by the Pearson correlation. Statistical significance was determined by using generalized estimating equation models to account for possible correlation between fellow eyes. Results: Mean IOP was 12.8±2.7 mm Hg. Mean wave speed in the cornea was 1.82±0.10 m/s. Young’s modulus of elasticity was 696±113 kPa and was correlated with IOP (r=0.57; P=0.004), but none of the other variables (P>0.1). Conclusions: USWE is a novel noninvasive technique for measuring ocular biomechanical properties. Corneal Young’s modulus in normal eyes is associated with IOP, consistent with measurements in cadaver eyes. Further work is needed to determine elasticity in other ocular tissues, particularly the sclera, and if elasticity is altered in glaucoma patients.

ABSTRACT Carpal tunnel pressure is a key factor in the etiology of carpal tunnel syndrome. Numerous approaches have been conducted to measure carpal tunnel pressure. However, most techniques are invasive and take time and effort. We have developed an innovative approach to noninvasively assess the tunnel pressure by using the ultrasound surface wave elastography (USWE) technique. In a previous study it was shown that the shear wave speed in a tendon increased linearly with increasing tunnel pressure enclosed the tendon in a simple tendon model. This study aimed to examine the relationship between the carpal tunnel pressure and the shear wave speeds inside and outside the carpal tunnel in a human cadaveric model. The result showed that the shear wave speed inside the carpal tunnel increased linearly with created carpal tunnel pressure, while the shear wave speed outside the carpal tunnel remained constant. These findings suggest that noninvasive measurement of carpal tunnel pressure is possible by measuring the shear wave speed in the tendon. After fully establishing this technology and being applicable in clinic, it would be useful in the diagnosis of carpal tunnel syndrome. For that reason, further validation with this technique in both healthy controls and patients with carpal tunnel syndrome is required. © 2017 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 36:477–483, 2018.

Ultrasonography is not widely used in clinic for lung assessment because ultrasound cannot image deep lung tissue. In this abstract, we present a novel technique, lung ultrasound surface wave elastography (LUSWE), for measuring the elastic properties of superficial lung tissue. In LUSWE, a small, local, and 0.1 second harmonic vibration is generated on the chest of a subject. The speed of surface wave on the lung is measured by using an ultrasound probe. We are evaluating LUSWE for assessing patients with interstitial lung disease (ILD). LUSWE may be useful for assessing ILD because most ILD patients have typical fibrotic scars in the peripheral and subpleural regions of the lung. In a large clinical study of ILD patients, we measure both lungs through six intercostal spaces for patients and controls. The surface wave speed is measured at 100 Hz, 150 Hz, and 200 Hz. In an example, the surface wave speed is 1.88 ± 0.11 m/s and 3.3 ± 0.37 m/s, respectively, for a healthy subject and an ILD patient at 100 Hz and in the same intercostal space. LUSWE may compliment the clinical standard high-resolution computed tomography for assessing ILD. LUSWE may be useful for assessing other lung disorders.

The purpose of this Letter to the Editor is to demonstrate an effective method for estimating viscoelasticity based on measurements of the Rayleigh surface wave speed. It is important to identify the surface wave mode for measuring surface wave speed. A concept of start frequency of surface waves is proposed. The surface wave speeds above the start frequency should be used to estimate the viscoelasticity of tissue. The motivation was to develop a noninvasive surface wave elastography (SWE) technique for assessing skin disease by measuring skin viscoelastic properties. Using an optical based SWE system, the author generated a local harmonic vibration on the surface of phantom using an electromechanical shaker and measured the resulting surface waves on the phantom using an optical vibrometer system. The surface wave speed was measured using a phase gradient method. It was shown that different standing wave modes were generated below the start frequency because of wave reflection. However, the pure symmetric surface waves were generated from the excitation above the start frequency. Using the wave speed dispersion above the start frequency, the viscoelasticity of the phantom can be correctly estimated.

In conventional shear wave elastography materials are assumed to be linear, elastic, homogeneous, and isotropic. These assumptions are important to account for in certain tissues because they are not always appropriate. Many tissues such as skeletal muscle, the kidney, and the myocardium are anisotropic. Shear waves can be used to investigate the directionally dependent mechanical properties of anisotropic media. To study these tissues in a systematic way and to account for the effects of the anisotropic architecture, laboratory-based phantoms are desirable. We will report on several phantom-based approaches for studying shear wave anisotropy, assuming that these materials are transversely isotropic. Phantoms with embedded fibers were used to mimic anisotropic tissues. Homogeneous phantoms were compressed to induce transverse isotropy according to the acoustoelastic phenomenon, which is related to nonlinear behavior of the materials. The fractional anisotropy of these phantoms was quantified to compare with measurements made in soft tissues. In addition, soft tissues are also viscoelastic, and we have developed a method to model viscoelastic transversely isotropic materials with the finite element method (FEM). The viscoelastic property estimation from phantom experiments and FEM simulations will also be discussed.

Increased stiffness of arteries has recently gained acceptance as an independent risk factor for cardiovascular and many other diseases. Pulse wave velocity (PWV) is widely used for estimating the stiffness of an artery. However, PWV is an average indicator of artery stiffness between the two measuring points, and therefore does not identify local stiffness. In addition, the thickness of the artery is needed to calculate artery elastic modulus. We have discovered that there is a series of ring resonant frequency in arteries that can be used for estimating local arterial elastic modulus. These resonances can be excited by remote palpation with selectively placed ultrasonic radiation force. Experiments on ring resonance were carried out on an excised artery as well as a rubber tube. This family of resonances occurs at relatively low frequency, around 356 Hz for the artery. Estimation of the Young’s modulus is 135 kPa by the ring frequency 356 Hz without the requirement of the arterial thickness. The estimated modulus is very consistent with the other ring resonant frequency measurements. The theory of ring resonance is developed and verified with the rubber tube experiments.

We have proposed generating a bending wave in the arterial wall using ultrasound radiation force and measuring the wave velocity along the arterial wall [Zhang et al., IEEE Trans. Ultrason. Ferroelectr. Freq. Control 52, 642–652 (2005)]. Here, we report the results of in vivo studies on pigs. The pig was anesthetized, and a micromanometer tip catheter was inserted into the femoral artery to measure luminal pressure. A water bath was created on the animal’s groin to allow unimpeded access of the ultrasound beams to the femoral artery. The femoral artery was first located using a 13-MHz linear-array transducer. Then, a vibro-acoustography image was obtained to ensure precise positioning of the excitation force relative to the artery. The artery was excited by the force transducer and the resulting vibration of the arterial wall was measured by a sensing Doppler transceiver. Measured wave velocity was 3.1 m/s at 300 Hz. With this new method wave velocity over a distance of 5 mm, and therefore stiffness of arteries, can be measured locally and non-invasively. Measurement time is short in a few tens of milliseconds, which allows pressure dependence and pharmacological effect on the wall properties to be measured at different cardiac times.

Arterial wall stiffness can be associated with various diseases. The stiffness of an artery can be assessed by measurement of the pulse wave velocity (PWV). PWV is directly related to the Youngs modulus by the well-known Moens–Korteweg equation. Usually, PWV is estimated using the foot-to-foot method. However, the foot of the pressure wave is not very clear due to reflected waves. Also, the pressure wave is normally at a low frequency, hence, the time resolution is low. PWV is an average indicator of artery stiffness between the two measuring points, therefore it is not easy to identify local stiffness. We propose producing a very short pulse wave in the arterial wall using ultrasound radiation force and measuring its propagation speed along the artery by laser. The temporal resolution of this method is in the range of microseconds, which allows PWV to be measured accurately over a few millimeters. Experiments were carried out on a silicone tube in gelatin. PWV was measured by two scanning methods: (1) fixed source and scanning detector, (2) scanning source and fixed detector. Results: PWV was measured at 2 mm/40 μs by both methods. The Doppler technique was also tested which is potentially suitable for clinical applications.

A coupled vibration model, which takes into account the elastic arterial wall, the interior blood, and exterior tissue, is presented. The arterial wall is considered as a cylindrical elastic shell and is modeled with the thick shell theory due to the large thickness-to-radius ratios. Both the blood and the tissue are considered as ideal fluids and are modeled with acoustic wave theory. A newly developed method, the wave propagation approach, is extended for the coupled vibration of arterial walls. The theoretic model is evaluated against those available in the literature. The experiments were carried out on a silicone rubber tube in a water tank with an ultrasound stimulated system. This system uses the radiation force of ultrasound to vibrate the tube at low frequency and records the resulting response by a laser vibrometer. Both the excitation and measurement are remote and noncontact. The measured fundamental frequency is in good agreement with the theoretical prediction.