Open Access
Add to BibSonomyAbstract
The high imaging resolution and motion sensitivity of optical-based shear wave detection has made it an attractive technique in biomechanics studies with potential for improving the capabilities of shear wave elasticity imaging. In this study we implemented laser speckle contrast imaging for two-dimensional (X–Z) tracking of transient shear wave propagation in agarose phantoms. The mechanical disturbances induced by the propagation of the shear wave caused temporal and spatial fluctuations in the local speckle pattern, which manifested as local blurring. By mechanically moving the sample in the third dimension (Y), and performing two-dimensional shear wave imaging at every scan position, the three-dimensional shear wave velocity distribution of the phantom could be reconstructed. Based on comparisons with the reference shear wave velocity measurements obtained using a commercial ultrasound shear wave imaging system, the developed system can estimate the shear wave velocity with an error of less than 6% for homogeneous phantoms with shear moduli ranging from 1.52 kPa to 7.99 kPa. The imaging sensitivity of our system makes it capable of measuring small variations in shear modulus; the estimated standard deviation of the shear modulus was found to be less than 0.07 kPa. A submillimeter spatial resolution for three-dimensional shear wave imaging has been achieved, as demonstrated by the ability to detect a 1-mm-thick stiff plate embedded inside heterogeneous agarose phantoms.
© 2016 Optical Society of America
1. Introduction
Palpation is widely used in clinical physical examinations for sensing the degree and distribution of the stiffness of biological tissues, since in many conditions the stiffness in pathological regions differs from that of the surrounding healthy tissue of similar type. However, the palpation technique is a qualitative and subjective method for detecting abnormality in tissue whose sensitivity and specificity may be insufficient for identifying small lesions present at an early stage of a disease.
There has been intensive development of noninvasive elasticity imaging techniques based on ultrasound for characterizing the mechanical properties of biological tissue during the past 2 decades [1,2]. These techniques have been exploited in numerous clinical applications, including the detection of prostate cancer [3], breast cancer [4], and the staging of liver fibrosis [5]. Shear wave elasticity imaging is an attractive technique since it is able to provide a quantitative assessment of the shear modulus of tissues based on the tracking of shear wave propagating therein [6,7]. The local shear modulus can be approximated directly from the local shear wave velocity, which typically ranges from 1 m/s to 10 m/s. Systems for three-dimensional shear wave imaging using a matrix array probe have been developed recently [8,9]. Moreover, applications such as the three-dimensional reconstruction of the shear modulus distribution of a sample as well as assessing the fiber orientation and mechanical anisotropy have been investigated [9].
The capabilities of shear wave elasticity imaging, in terms of its resolution, accuracy, and variance in shear modulus estimation, depends on several factors, such as the properties of the shear wave (e.g., its amplitude, bandwidth, and wavelength), the noise level, imaging frame rate, and motion sensitivity of the system [10]. A shear wave with a specific wavelength and bandwidth can be generated by suitably designing the profile and sequence of the ultrasound push beam [6,11]. Assuming that the signal-to-noise ratio (SNR) can be maintained, the capabilities of shear wave imaging improve when using higher frequency shear waves. However, the frequency-dependent attenuation tends to result in higher frequency shear waves having a lower SNR and reduced field of view. Furthermore, the properties of the generated shear waves are partially dependent on the mechanical properties of the tissue sample.
With high imaging resolution and motion sensitivity, optical-based shear wave detection has potential for improving the capabilities of shear wave elasticity imaging. Optical coherence tomography (OCT), which is analogous to ultrasound imaging, has been used to evaluate the mechanical properties of the bovine carotid artery ex vivo [12] and the mouse cornea in vivo [13]. The sensitivity of OCT in detecting the shear wave axial displacement can be as good as 10 nm [14], but the imaging penetration depth is limited to approximately 200 µm due to scattering (when using a full-field OCT setup). Although the imaging depth can be increased to approximately 3 mm by employing swept-source OCT, this reduces the imaging resolution to the micron scale and requires the use of one-dimensional scanning to produce a two-dimensional (lateral-depth view) elasticity image [15]. A feasibility study that applied optical speckle imaging for the acoustic radiation force and shear wave motion detection has been reported [16]. The displacement induced by the motions mentioned above caused transient decorrelation in the optical speckles that can be resolved by cross-correlating the time-consecutive speckle patterns. However, quantitative elasticity measurements have not been presented previously.
Laser speckle contrast imaging (LSCI) is another technique based on optical speckle imaging, where the spatial variation of the speckle pattern caused by the motion of the scattering particles, such as blood flow, is analyzed quantitatively to provide information about the motion. This technique has been used for monitoring blood flow dynamics in skin perfusion and the retinal vasculature [17]. Shear wave imaging based on laser speckle contrast analysis has been demonstrated in 20-mm-thick tissue-mimicking phantoms [18], with a spatial resolution of 2 mm and a shear wave velocity estimation error of 3.3% being achieved. Laser speckle contrast imaging has good sensitivity for detecting shear wave axial displacements [18], and the shear wave propagation can be detected with a high SNR even when the shear wave is induced as far as 7 cm from the optical detection region. However, that reported system requires the use of two-dimensional mechanical scanning to produce a two-dimensional elasticity image [18].
In this study we employed LSCI for two-dimensional tracking (in the X–Z plane) of the propagation of a shear wave generated by an acoustic radiation force in both homogeneous and heterogeneous agarose phantoms. Implementing one-dimensional mechanical scanning (in the Y direction) allowed the shear waves propagating in different X–Z planes of the phantom to be imaged, and thus the three-dimensional shear wave velocity distribution in the phantom could be reconstructed. The accuracy, imaging resolution, and imaging depth of the developed system are explored and discussed.
2. Materials and methods
2.1 Principles of laser speckle contrast imaging
A speckle pattern is formed when a coherent light source is used to illuminate a sample containing scattering particles. The pattern generated by the optical waves received at each pixel of the imaging sensor is determined by coherent interference between the scattered waves. The speckle size of the observed speckle pattern SS is expressed as [17]
where λ is the wavelength of the coherent light, M is the magnification and F is the f-number. Local disturbances within the sample induced by the application and propagation of an acoustic wave cause the scatterers to displace and the index of refraction of the sample to change, which consequently alters the phase of the optic wave and thus the interference pattern [19]. The local temporal and spatial fluctuations in the speckle pattern are integrated over a preset camera exposure time, and the effect manifests as local blurring in the speckle image. The spatial blurring is measured and defined by the speckle contrast K:where σs and <I> are the standard deviation and mean intensity of pixel values within an image region of interest, respectively. To increase the image contrast, speckle contrast difference ΔK was used in this study:where Kus off and Kus on are the speckle contrasts computed for the background frame (i.e., when the ultrasound push transducer is off) and the acoustic interference frame (i.e., during or after the ultrasound push transducer is turned on), respectively. The disturbance caused by shear wave propagation will result in a decrease in the speckle contrast difference.2.2 Experimental setup
The system and experimental setup are shown in Fig. 1. The light emitted by a laser operating at 785 nm (OBIS, 50 mW Coherent Inc., CA, USA) was expanded to 8 mm and transmitted through a 12-mm-thick agarose phantom placed inside a plastic container that had a wall thickness of 1 mm. The plastic container was secured on a translational stage, allowed the sample to move in Y direction. The speckle pattern formed by light scattering in the phantom was imaged with a CCD camera (Manta G-145B NIR, Allied Vision Technology, Germany). To satisfy the Nyquist sampling criterion and to increase the contrast of the speckle pattern imaged [20], the size of the speckle pattern was adjusted to be more than twice the size of the image pixels (pixel size of 6.45 µm) by setting the magnification of the system to M = 2, and reduce the size of the aperture.
A homemade single-element 20 MHz focused ultrasound transducer (focal length = 17.5 mm, aperture diameter = 7 mm) was used to generate the acoustic radiation force in the phantom. The acoustic attenuation in the agarose phantom (α) was measured as 1.47 dB/cm at 20 MHz. The acoustic axis was perpendicular to the optical axis, with the focal point of the ultrasound transducer lying on the focal plane of the imaging lens, and was approximately 1 mm (in the X direction) from the illuminated area. The ultrasound transducer (referred to as the push transducer herein) was driven by a sine-wave pulse of 100-µs duration. This arrangement generated an outward-propagating cylindrical shear wave. Within the limited field of view of the camera, the imaged shear wave propagated mainly in the X direction.
A trigger sequence, illustrated in Fig. 2(a), was employed for each acquisition session, with a total of 60 frames being recorded. The first three frames captured in a session were used as the background images: the first background frame was used to calculate Kus off and the other two were used to monitor the system noise. In frames 4 and 5, the effect of the acoustic radiation force on the speckle pattern was imaged by initiating the image acquisition process at 100 µs after the push transducer has been triggered. In the subsequent image frames (frames 6 to 60), the disturbance in the speckle pattern induced by the propagation of the shear wave was imaged by adjusting the time delay between two trigger signals, where the time delay had an increment step of 100 µs. This protocol achieved an effective frame rate of 10,000 frames/s for shear wave imaging. The control system, consisting of a microcontroller and software program, provided the trigger signals for the camera and the push transducer at predetermined time instances.
Fig. 2 Illustrations of (a) the acquisition process (striped block: trigger for the camera, solid block: trigger for the push transducer) and (b) the imaging process.
2.3 Image processing
The local disturbance in an image of the speckle pattern was calculated by computing the speckle contrast difference for each spatial kernel, depicted as ΔK(x,z,t) in Fig. 2(b), where x, z, and t indicate the indices of the kernel in the X and Z directions and the frame number, respectively. The kernel size used in this study was 145 µm (in both the X and Z directions), and the percentage overlap between the adjacent kernels was 80%. To increase the SNR, the speckle contrast difference kernels were summed and averaged in the Z direction (ΔZ = 610 µm). For each ΔZ, a spatiotemporal map ΔK(x,t) showing the shear wave propagation distance in the X direction versus time was deduced. Analyzing the wavefront slope in the spatiotemporal map based on the time-of-flight (TOF) algorithm—that is an estimation of the time difference (td) between the shear wave wavefront at two known locations along the X direction using normalized cross-correlation—yielded the shear wave velocity (vs), which is equal to , where ΔX denotes the distance between the wavefronts of the shear wave at the two locations used in the TOF algorithm. Before the TOF algorithm was applied, a direction filter was applied to the spatiotemporal map for all cases in this study to remove any non-forward-propagating wavefront. A two-dimensional map (in the X–Z plane) of the shear wave velocity was subsequently constructed by estimating the shear wave velocity for every ΔZ along the Z direction. This process was repeated at every scanned Y position to yield the three-dimensional shear wave velocity distribution.
2.4 Result validation
A commercial ultrasound elasticity imaging system (Aixplorer, Supersonic Imaging, Aix-en-Provence, France) was used to estimate the shear modulus of the homogeneous phantoms used in this study, with the results used as the reference for assessing the accuracy of the shear moduli measured by our system.
2.5 Phantoms
Eight homogeneous agarose phantoms were constructed, comprising four pairs with different stiffness values produced by using agarose at concentrations of 0.4%, 0.5%, 0.6%, and 0.7%. One phantom of each pair was made with silicon dioxide (at a mass concentration of 0.15%) and the other with agar powder (at a mass concentration of 0.5%) added as the optical or ultrasound scattering agents. Then the phantoms were imaged by the LSCI system or the Aixplorer ultrasound imaging system, respectively, and results were compared against each other. The reduced scattering coefficient (µs') of the phantom (with 0.15% silicon dioxide added as optical scatterers), as measured and calculated using the integrating sphere technique [21] and the inverse adding-doubling method (available at http://omlc.org/software/iad), was 0.97 cm–1 for wavelength (λ) of 785 nm.
Figure 3 shows schematics of the three heterogeneous phantoms. The phantoms shown in Fig. 3(a) and 3(b) include a stiff plate oriented in the X and Y directions, respectively. The stiff-plate inclusion shown in Fig. 3(b) and the stiff-cylinder inclusion shown in Fig. 3(c) are located approximately 3 mm (in the Y direction) from the phantom boundary. The concentration of agarose in the background was 0.4% for all three heterogeneous phantoms, whereas the plate and cylinder inclusions shown in Fig. 3(a)–(c) were made with agarose at concentrations of 0.5%, 0.6%, and 0.5%, respectively.
Fig. 3 Schematics of three heterogeneous phantoms containing a stiff-plate inclusion in the (a) X direction and (b) Y direction, and a stiff-cylinder inclusion (c).
3. Results
3.1 Experiments with the homogeneous phantoms
Shear wave imaging is firstly performed by adjusting the aperture to a fully opened position (i.e., aperture size of 15 mm), which allows high spatial frequency components to be imaged. An animation of shear wave propagation was produced by calculating ΔK(x,z,t) for every spatial kernel and image frame in an acquisition session. Figure 4 shows speckle contrast difference maps for the 0.6%-agarose phantom before and after the push transducer had been triggered. The decreased local speckle contrast evident in Fig. 4(b)–(d) indicates the location of the shear wave wavefront at 0.3 ms, 1.3 ms, and 2.3 ms after the acoustic radiation force had been applied, respectively, while the variation in local speckle contrast in Fig. 4(a) corresponds to the system noise. The spatiotemporal map (obtained at Z = 1.6 mm) for the four investigated homogeneous phantoms are shown in Fig. 5. It is apparent that the wavefront slope in the spatiotemporal map was steepest for the 0.7%-agarose phantom, which indicates that the shear wave velocity was highest among the four phantoms. To evaluate the lateral resolution (in the X direction) and the accuracy in estimating the shear wave velocity of the imaging system, shear wave imaging was performed on homogeneous phantoms using an aperture size of 6 mm. The imaged X–Z plane was located at Y = 2 mm, where Y = 0 mm is the boundary of the phantom. For each acquisition session, 27 shear wave velocity maps were calculated using the TOF algorithm for ΔX values ranging from 320 µm to 1075 µm. The spatial mean and spatial standard deviation of the shear wave velocity maps are shown in Fig. 6, with the error bars indicating the standard deviation for five acquisition sessions. The estimated shear wave velocity became approximately constant (with changes in the estimated velocity of less than 0.5% compared to the case where a larger ΔX was used) when ΔX was 552 µm, 581 µm, 639 µm, and 697 µm for the phantoms made with 0.4%, 0.5%, 0.6%, and 0.7% agarose, respectively. The Aixplorer system was used to obtain an animation of the shear wave propagation for the homogeneous phantoms, and the same TOF algorithm (using ΔX = 800 µm for all four cases) was used to estimate the shear wave velocity for each phantom. The results are summarized in Table 1.
Fig. 4 Temporal series of speckle contrast difference maps showing (a) the background noise and (b)–(d) the propagation of the shear wave at 0.3 ms, 1.3 ms, and 2.3 ms, respectively, after applying the acoustic radiation force. Aperture size = 15 mm.
Fig. 5 Spatiotemporal maps for phantoms made with (a) 0.4%, (b) 0.5%, (c) 0.6%, and (d) 0.7% agarose. The black dashed lines indicate the wavefront slope. Aperture size = 15 mm.
Fig. 6 (a) Spatial mean and (b) spatial standard deviation values of the shear wave velocity map for the homogeneous phantom experiments. The error bars are the standard deviations for five acquisition sessions. Aperture size = 5mm.
Table 1. Shear Wave Velocity Estimated Using LSCI and the Aixplorer System
Assuming that the density of the phantom is 1000 kg/m3, the shear moduli estimated using LSCI based on five acquisition sessions were 1.52 ± 0.02 kPa (mean ± standard deviation), 2.85 ± 0.04 kPa, 4.42 ± 0.07 kPa, and 7.99 ± 0.07 kPa for the homogeneous phantoms made with 0.4%, 0.5%, 0.6%, and 0.7% agarose, respectively.
3.2 Experiments with the heterogeneous phantoms containing stiff-plate inclusions
Figure 7(a) and 7(c) show camera raw images of the heterogeneous phantom with a stiff-plate inclusion in the X direction captured using aperture sizes of 15 mm and 6 mm, respectively. The inclusion had a size of 1 mm × 9 mm (in the X and Y directions, respectively) and it was approximately located between X = 1.5 mm and 2.5 mm. Figure 7(b) and 7(d) show the 3.75 mm × 3.14 mm shear wave velocity maps for the heterogeneous phantom, imaged at Y = 1.5 mm and processed using ΔX = 610 µm, corresponding to the case shown in Fig. 7(a) and 7(c). The mean shear wave velocities estimated for the regions indicated in Fig. 7(b) and 7(d) are summarized in Table 2.
Fig. 7 (a)(c) Camera images and (b)(d) shear wave velocity maps for the heterogeneous phantom with a stiff-plate inclusion in the X direction captured with (a)(b) aperture size = 15 mm and (c)(d) aperture size = 6mm. The yellow dashed box indicates the region where the stiff-plate inclusion is located.
Table 2. Shear Wave Velocities Estimated for Different Regions of the Heterogeneous Phantom with a Stiff-plate Inclusion in the X Direction
The heterogeneous phantom with a stiff-plate inclusion in the Y direction (where Y = 0 mm is the boundary of the phantom) was imaged by mechanically moving the phantom toward the imaging lens in steps of 0.25 mm. The experiment was conducted twice, with aperture sizes of 15 mm and 6 mm. For scans at each Y position, three acquisition sessions of shear wave imaging for the X–Z view of the phantom were performed. The spatial mean of the shear wave velocity maps for each scan Y position is shown in Fig. 8, with the error bars indicating the standard deviation for these three acquisition sessions. The ∆X value used in computing the shear wave velocity map was 720 µm. Figure 8 indicates that the imaging depth of the system was approximately 5.5 mm.
Fig. 8 Spatial mean values of the shear wave velocity map for the heterogeneous phantom with a stiff-plate inclusion the in Y direction. The error bars are the standard deviations for three acquisition sessions.
3.3 Three-dimensional shear wave imaging of a heterogeneous phantom with a stiff-cylinder inclusion
Figure 9(a)–(t) show the shear wave velocity maps (for the X–Z view) of the heterogeneous phantom with a stiff-cylinder inclusion for scans at Y positions from 1.5 mm to 6.25 mm, in steps of 0.25 mm, respectively. The aperture size used in this case study was 8 mm, which was selected based on the trade-off between the X-direction spatial resolution and the imaging sensitivity. The cylinder inclusion had a diameter of approximately 3 mm and it was located between X = 1.5 mm and X = 4.5 mm. The presence of the cylinder inclusion is evident in Fig. 9(g)–(s), which is consistent with the known structure of the heterogeneous phantom. The spatial variation in the measured shear wave velocity map of the cylinder inclusion is resulted from the energy loss of the forward-propagating shear wave at the boundary between the soft and the stiff agarose. In addition, the energy of the shear wave decays as the propagation distance increases, which also results in increased in spatial standard deviation particularly at X > 3mm. Due to the limited imaging depth of the system, shear wave imaging was not performed beyond Y = 6.25 mm, where the background agarose image reappeared. Nevertheless, a partial three-dimensional outline of the cylinder inclusion is still observable in Fig. 9.
Fig. 9 (a)–(t) Shear wave velocity maps (for the X–Z view) of the heterogeneous phantom with a stiff-cylinder inclusion for scans at Y positions of 1.5 mm to 6.25 mm, in steps of 0.25 mm, respectively.
4. Discussion
The setup of the LSCI system is simpler than that of the optical coherence elastography system, with only one-dimensional mechanical scanning required to achieve three-dimensional shear wave imaging. In the current system setup it takes approximately 8 s to obtain a 60-frame animation of the shear wave propagation. This time is mainly determined by the frame rate of the camera used. Thus, real-time shear wave imaging can be implemented by using a high-speed camera [16].
While transmission-mode LSCI was implemented in this study, reflection-mode LSCI—which has been widely demonstrated for imaging blood flow dynamics—can also be implemented for shear wave imaging. However, the imaging sensitivity may be reduced by the anisotropy characteristic of scattering in soft tissue. Determining the camera exposure time is important in LSCI. Increasing the exposure time will reduce the background noise (due to it being temporally integrated) at the expense of degrading the spatial resolution of the final shear modulus map, as the spatiotemporal curve is widened. This will also limit the ability of the system to detect a rapidly propagating shear wave. The absolute intensity of the speckle pattern does not influence the tracking of a shear wave in LSCI, which can therefore provide both the mechanical and optical absorption properties of a sample, as demonstrated previously [22].
The resolution of shear wave imaging in the Y direction is mainly determined by the effective depth of field of the imaging lens. The push transducer is placed at the focal plane of imaging lens, hence, the disturbance caused by the propagation of shear wave in the focal plane will significantly blur the speckle pattern, whereas a disturbance induced outside of the depth of field of the imaging lens will cause less blurring. Therefore, by mechanically moving the sample in the Y direction, different X–Z planes of the phantom can be evaluated by tracking the shear wave propagation in each plane. In this study, a 1-mm-thick stiff-plate inclusion in the Y direction is resolved, as shown in Fig. 8. However, the shear wave velocity of the inclusion is underestimated. This is due to the push transducer lying directly above the stiff-plate inclusion when shear wave imaging was performed between Y = 3 mm and Y = 4 mm. The aperture diameter of the push transducer was larger than the thickness of the stiff-plate inclusion, which resulted in poor induction of the acoustic radiation force and the shear wave. In this study, the spatial resolution of the imaging system is empirically tested and verified using small inclusion phantoms. A complete investigation on the spatial resolution using simulation may be required, however, was beyond the scope of the current study. We hope to address this issue in our future work.
As shown in Fig. 1, an aperture was placed at the Fourier plane of the imaging lens for implementing speckle size adjustment and spatial filtering. Reducing the size of the aperture increases the speckle size and blocks light exiting the phantom at a wider angle (due to multiple scattering), and consequently a clear speckle pattern can be observed; that is, with higher speckle contrast. This implementation is used to improve the sensitivity of shear wave detection and the accuracy of the shear wave velocity estimation. As shown in Fig. 8, the shear wave velocity of the background agarose was overestimated when the aperture size was 15 mm. However, a trade-off in the spatial resolution is observed in Fig. 7(b) and 7(d), as the outline of the inclusion is more pronounced when the aperture size was 15 mm.
In ultrasound shear wave elasticity imaging, the shear wave axial displacement is measured as positive and negative phase differences, which allows properties of a shear wave such as its wavelength and bandwidth to be measured. These parameters are essential for estimating the tissue viscosity. In the present system, setting the camera exposure time to 0.5 ms provided a good trade-off between SNR and spatial resolution. However, the positive and negative shear wave axial displacements, which both produced a decrease in speckle contrast, were merged and integrated. Thus, viscosity measurement based on the detection of a transient shear wave using LSCI is an issue that needs to be addressed in future investigations.
The spatial resolution and imaging depth shown in this study are achieved for low optical scattering sample. With increased optical scattering within the sample, spatial resolution and the imaging depth will degrade. Therefore, the proposed method is aimed for in vitro studies, and one of the potential applications is to perform three-dimensional shear wave imaging for analyzing the mechanics of a cell matrix. Studies have shown that the local microenvironment of a cell, particularly the extracellular matrix (ECM) component, plays vital roles in cell development and tissue homeostasis. In addition, ECM stiffening has been found to promote cancer cell invasion and migration [23]. Thus, information about the elasticity of the ECM may provide useful information about cancer progression. Cell culture samples (made of cells and hydrogels) have relatively low optical scattering. In addition, it is technically feasible to apply the transmission setup for cell culture samples. Therefore, it is believed that the proposed method is more suitable for in vitro stiffness studies.
5. Conclusions
We have demonstrated the capability of LSCI in tracking the propagation of shear waves in two dimensions, which can be used to generate a shear wave velocity map using the conventional TOF algorithm. By implementing one-dimensional mechanical scanning, two-dimensional shear wave velocity maps obtained at different scan positions allow reconstruction of the three-dimensional shear wave velocity distribution. The accuracy, sensitivity, and spatial resolution of the presented system for shear modulus measurements have been demonstrated for homogeneous phantoms with different stiffness values, and for heterogeneous phantoms that include stiff objects with different shapes and orientations. The presented results demonstrate that shear wave imaging with submillimeter spatial resolution is achievable, with an estimated shear modulus variance of less than 0.07 kPa.
Funding
Ministry of Science and Technology of Taiwan (MOST) (104-2221-E-002-105-).
References and links
1. J. Ophir, S. K. Alam, B. S. Garra, F. Kallel, E. E. Konofagou, T. Krouskop, C. R. B. Merritt, R. Righetti, R. Souchon, S. Srinivasan, and T. Varghese, “Elastography: Imaging the elastic properties of soft tissues with ultrasound,” J. Med. Ultrason. (2001) 29(4), 155–171 (2002). [CrossRef] [PubMed]
2. K. Nightingale, R. Bentley, and G. Trahey, “Observations of tissue response to acoustic radiation force: opportunities for imaging,” Ultrason. Imaging 24(3), 129–138 (2002). [CrossRef] [PubMed]
3. K. Hoyt, B. Castaneda, M. Zhang, P. Nigwekar, P. A. di Sant’agnese, J. V. Joseph, J. Strang, D. J. Rubens, and K. J. Parker, “Tissue elasticity properties as biomarkers for prostate cancer,” Cancer Biomark. 4(4-5), 213–225 (2008). [PubMed]
4. A. Itoh, E. Ueno, E. Tohno, H. Kamma, H. Takahashi, T. Shiina, M. Yamakawa, and T. Matsumura, “Breast disease: clinical application of US elastography for diagnosis,” Radiology 239(2), 341–350 (2006). [CrossRef] [PubMed]
5. L. Sandrin, B. Fourquet, J. M. Hasquenoph, S. Yon, C. Fournier, F. Mal, C. Christidis, M. Ziol, B. Poulet, F. Kazemi, M. Beaugrand, and R. Palau, “Transient elastography: a new noninvasive method for assessment of hepatic fibrosis,” Ultrasound Med. Biol. 29(12), 1705–1713 (2003). [CrossRef] [PubMed]
6. A. P. Sarvazyan, O. V. Rudenko, S. D. Swanson, J. B. Fowlkes, and S. Y. Emelianov, “Shear wave elasticity imaging: a new ultrasonic technology of medical diagnostics,” Ultrasound Med. Biol. 24(9), 1419–1435 (1998). [CrossRef] [PubMed]
7. K. Nightingale, S. McAleavey, and G. Trahey, “Shear-wave generation using acoustic radiation force: in vivo and ex vivo results,” Ultrasound Med. Biol. 29(12), 1715–1723 (2003). [CrossRef] [PubMed]
8. J. L. Gennisson, J. Provost, T. Deffieux, C. Papadacci, M. Imbault, M. Pernot, and M. Tanter, “4-D ultrafast shear-wave imaging,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control 62(6), 1059–1065 (2015). [CrossRef] [PubMed]
9. M. Wang, B. Byram, M. Palmeri, N. Rouze, and K. Nightingale, “Imaging transverse isotropic properties of muscle by monitoring acoustic radiation force induced shear waves using a 2-D matrix ultrasound array,” IEEE Trans. Med. Imaging 32(9), 1671–1684 (2013). [CrossRef] [PubMed]
10. T. Deffieux, J.-L. Gennisson, B. Larrat, M. Fink, and M. Tanter, “The variance of quantitative estimates in shear wave imaging: theory and experiments,” IEEE Trans. Ultrason. Ferroelectr. Freq. Control 59(11), 2390–2410 (2012). [CrossRef] [PubMed]
11. H. Zhao, P. Song, M. W. Urban, J. F. Greenleaf, and S. Chen, “Robust shear wave speed measurement using comb-push ultrasound radiation force,” in Proceedings of IEEE International Ultrasonics Symposium (IEEE 2011), pp. 1270–1273. [CrossRef]
12. M. Razani, T. W. H. Luk, A. Mariampillai, P. Siegler, T. R. Kiehl, M. C. Kolios, and V. X. D. Yang, “Optical coherence tomography detection of shear wave propagation in inhomogeneous tissue equivalent phantoms and ex-vivo carotid artery samples,” Biomed. Opt. Express 5(3), 895–906 (2014). [CrossRef] [PubMed]
13. R. K. Manapuram, S. R. Aglyamov, F. M. Monediado, M. Mashiatulla, J. Li, S. Y. Emelianov, and K. V. Larin, “In vivo estimation of elastic wave parameters using phase-stabilized swept source optical coherence elastography,” J. Biomed. Opt. 17(10), 1005011 (2012). [CrossRef] [PubMed]
14. A. Nahas, M. Tanter, T. M. Nguyen, J. M. Chassot, M. Fink, and A. Claude Boccara, “From supersonic shear wave imaging to full-field optical coherence shear wave elastography,” J. Biomed. Opt. 18(12), 121514 (2013). [CrossRef] [PubMed]
15. S. Wang and K. V. Larin, “Shear wave imaging optical coherence tomography (SWI-OCT) for ocular tissue biomechanics,” Opt. Lett. 39(1), 41–44 (2014). [CrossRef] [PubMed]
16. E. Bossy, A. R. Funke, K. Daoudi, A.-C. Boccara, M. Tanter, and M. Fink, “Transient optoelastography in optically diffusive media,” Appl. Phys. Lett. 90(17), 174111 (2007). [CrossRef]
17. D. A. Boas and A. K. Dunn, “Laser speckle contrast imaging in biomedical optics,” J. Biomed. Opt. 15(1), 011109 (2010). [CrossRef] [PubMed]
18. Y. Cheng, R. Li, S. Li, C. Dunsby, R. J. Eckersley, D. S. Elson, and M. X. Tang, “Shear wave elasticity imaging based on acoustic radiation force and optical detection,” Ultrasound Med. Biol. 38(9), 1637–1645 (2012). [CrossRef] [PubMed]
19. L. V. Wang, “Ultrasound-mediated biophotonic imaging: a review of acousto-optical tomography and photo-acoustic tomography,” Dis. Markers 19(2-3), 123–138 (2004). [CrossRef] [PubMed]
20. S. J. Kirkpatrick, D. D. Duncan, and E. M. Wells-Gray, “Detrimental effects of speckle-pixel size matching in laser speckle contrast imaging,” Opt. Lett. 33(24), 2886–2888 (2008). [CrossRef] [PubMed]
21. T. Moffitt, Y. C. Chen, and S. A. Prahl, “Preparation and characterization of polyurethane optical phantoms,” J. Biomed. Opt. 11(4), 041103 (2006). [CrossRef] [PubMed]
22. K. Daoudi, A.-C. Boccara, and E. Bossy, “Detection and discrimination of optical absorption and shear stiffness at depth in tissue-mimicking phantoms by transient optoelastography,” Appl. Phys. Lett. 94(15), 154103 (2009). [CrossRef]
23. P. Lu, V. M. Weaver, and Z. Werb, “The extracellular matrix: a dynamic niche in cancer progression,” J. Cell Biol. 196(4), 395–406 (2012). [CrossRef] [PubMed]
