Search

Diabetes mellitus can lead to diabetic kidney disease (DKD), which is a kind of chronic kidney diseases (CKDs). Compared with other renal diseases, DKD deteriorates more rapidly to end-stage renal disease, and the 5-year renal survival rate is about 61%. Therefore, monitoring of progression of DKD is vital for improving the survival rate and quality of life for diabetic patients. However, non-invasive and quantitative diagnostic tools are still lacking currently. It has been reported that the occurrence and development of DKD would lead to the changes of renal microvasculature and perfusion. Recently, ultrasound localization microscopy (ULM) has been demonstrated to break the diffraction limit of ultrasound and obtain the microvasculature morphology and blood flow speed map by localizing and tracking the flowing microbubbles. In this study, ULM was performed on kidneys of DKD rat models and control rats to investigate its feasibility in assessment of DKD in vivo. In addition, conventional clinical inspections (urine and serum tests) and conventional ultrasound imaging (Doppler and contrast-enhanced ultrasound imaging) were performed to verify the successful development of diabetes for DKD group. Statistical differences are observed between the clinical inspections-related parameters (urine and serum), CEUS-related parameters [time to peak (TTP) and area under curve (AUC)], and ULM-related parameter (mean blood flow speed of interlobular arterioles). In conclusion, this study shows the potential of ULM in the assessment of DKD in vivo.

This paper proposes a quantitative analysis method for lung ultrasound (LUS) images to evaluate the severity of COVID-19 pneumonia. Specifically, biomarkers related to the pleural line, including the thickness of pleural line (TPL) and the roughness of pleural line (RPL), and biomarkers related to the B-lines, including the accumulated width of B-lines (AWBL) and the acoustic coefficient of B-lines (ACBL), are extracted from LUS images to characterize the image patterns associated with the disease severity. 27 patients of COVID-19 pneumonia are enrolled in this study, including 13 moderate cases, 7 severe cases, and 7 critical cases. Patients of moderate cases are regarded as non-severe patients, and patients of severe and critical cases are regarded as non-severe patients. Biomarkers among different cases are compared, and the performances in the binary diagnosis of severe and non-severe patients are assessed using a support vector machine (SVM) classifier with all the biomarkers as the input. The classification performance is optimal using the SVM classifier (area under the receiver operating characteristics curve = 0.93, sensitivity = 0.93, specificity = 0.85). The proposed method may be a promising tool for the automatic grading and follow-up of patients with COVID-19 pneumonia.

Spinal cord tumor (SCT) is a critical disease with low incidence but high disability rate and morbidity. Surgical resection is a primary and effective treatment of SCT, but intraoperative detection of SCT highly depends on the clinical experience. Pathological studies show that microvascular proliferation (MP) occurs in some kinds of SCTs but it cannot be observed by conventional medical imaging modalities such as magnetic resonance imaging (MRI). Ultrasound microvascular imaging (UMI) has been developed to map tumor-related microvasculature. In this study, we introduced UMI into the visualization of microvessels in two types of human SCTs (i.e., ependymoma and schwannoma) and their corresponding normal tissues. For quantitative evaluation, microveseel density (MD) and power Doppler density (PDD) in the tumor region and normal tissue region of the two SCTs were calculated, respectively. Results show that the MDs and PDDs in the tumors are significantly higher than those in the normal tissues. Qualitative and quantitative results demonstrate that MP occurs in the tumor region compared with normal tissue, which is consistent with the results of immunohistochemical staining examination. In conclusion, this study demonstrate the feasibility of UMI in the intraoperative detection of human SCT.

The image reconstruction process in medical imaging can be treated as solving an inverse problem. The inverse problem is usually solved using time-consuming iterative algorithms with sparsity or other constraints. Recently, deep neural network (DNN)-based methods have been developed to accelerate the inverse-problem-solving process. However, these methods typically adopt supervised learning scheme, which requires ground truths, or labels of the solutions, for training. In many applications, it would be challenging or even impossible to obtain the ground truth, such as the tissue reflectivity function in ultrasound beamforming. In this study, a general framework based on self-supervised learning (SSL) scheme is proposed to train a DNN to solve the inverse problems. In this way, the measurements can be used as both the inputs and the labels during the training of DNN. The proposed SSL method is applied to four typical linear inverse problems for validation, i.e., plane wave ultrasound and photoacoustic image reconstructions, compressed sensing-based synthetic transmit aperture dataset recovery and deconvolution in ultrasound localization microscopy. Results show that, using the proposed framework, the trained DNN can achieve improved reconstruction accuracy with reduced computational time, compared with conventional methods., Comment: to be published in Proceedings of IEEE IUS 2021

Ultrasound microvascular imaging (UMI) has been developed to visualize microvasculature to reveal the conditions and pathologies of diseases. The key of UMI is extraction of blood flow signals by suppressing tissue clutters and background noise. However, the suppression of noise is unsatisfactory when we only use common clutter filtering methods without additional processing of denoising. Spatiotemporal non-local means (stNLM) filtering method has been used to effectively reject noise in ultrasound localization microscopy. In this study, we implemented stNLM in UMI to better suppress the background noise for improved microvascular visualization using the spatiotemporal discontinuity of noise. For comparison, the proposed stNLM-based UMI method was compared with three methods, i.e., method without NLM filtering (conventional UMI), NLM filtering on power Doppler image (PDI), and NLM filtering in the spatial domain (sNLM) of filtered blood flow RF data. The denoising superiority of the proposed method was validated in a human spinal cord tumor in vivo. Quantitative indexes including contrast-to-noise ratio (CNR) and peak-to-side level (PSL) were calculated. Results show that stNLM-based UMI can achieve nearly 15 dB improvement in CNR and nearly 30 dB improvement in PSL compared with conventional UMI. The proposed stNLM-based UMI has the potential of achieving high-quality UMI.

Lateral estimation has been a challenge in ultrasound elastography mainly due to the low resolution, low sampling frequency and lack of phase information in the lateral direction. Synthetic transmit aperture (STA) can achieve high resolution thanks to two-way focusing for improved lateral estimation, with the disadvantages of low signal-to-noise ratio (SNR) and limited penetration depth. In this study, we propose Hadamard-encoded STA (Hadamard-STA), which has been demonstrated to enhance the SNR of STA imaging, for improvement of strain estimation in elastography. Simulations and experiments were conducted on a heterogeneous model and an elasticity phantom with the same Young's modulus contrast. White Gaussian noise with different noise levels was added on the simulated channel dataset of Hadamard-STA to investigate its de-noising performance. The results show that Hadamard-STA outperforms STA with higher lateral and axial strain estimation qualities and may be a competitive method for quasi-static elastography.

Studies have shown that the perivascular space (PVS) plays an important role in regulation of cerebrospinal fluid (CSF) flow and elimination of endogenous metabolites, and the fluid dynamics of the PVS is related to certain physiological and pathological conditions. Benefiting from high-frame-rate plane wave imaging and advanced clutter filters, ultrafast power Doppler imaging (uPDI) has been developed to visualize microvessels which are invisible using conventional power Doppler imaging. In this study, we explore the feasibility of uPDI in the detection of PVS by injecting microbubbles (MBs) into the cisterna magna of rats and acquiring the ultrasound data in the sections of cerebral ventricle and common carotid artery. Results showed that aggregation of MBs was observed in the cerebral ventricle, which proved that there is a pathway for the CSF flow between the cisterna magna and the cerebral ventricle. After unilateral ligation of the common carotid artery, region of the PVS became larger than before MBs injection and ligation. It suggests that unilateral ligation of the common carotid artery causes unilateral cerebral ischemia, which leads to influx of CSF with MBs into the PVS and further causes expansion of PVS. In conclusion, this study demonstrates that uPDI can be used in the detection of PVS and has the potential to monitor the hemodynamic of PVS in the pathological conditions.

Compared with conventional delay-and-sum (DAS) beamforming method, sparse regularization (SR) techniques can reconstruct higher quality ultrasound (US) images from plane wave (PW) transmission. However, these methods suffer from long reconstruction time, owing to the high computational complexity inherently associated with its iterative process. In this work, a deep neural network (DNN) is trained to reconstruct plane wave ultrasound (PWUS) images from RF channel data with significantly reduced computational time. To overcome the lack of gold standard of ideal US images, a self-supervised learning scheme is employed to utilize the RF channel data as both the inputs and the labels during the training process. Using simulation data, the proposed method achieves comparable spatial resolution and 3.2-dB higher CNR, compared with the SR method.

Ultrasound localization microscopy (ULM) has been developed in recent years to significantly improve the spatial resolution of ultrasound imaging by localizing the microbubbles (MBs) flowing in microvasculature. Nevertheless, challenges remain in ULM. In our previous work (IEEE Trans. Med. Imaging, 10.1109/TMI.2020.2986781), by using a modified sub-pixel convolutional neural network (CNN), we have implemented ULM with fast data-processing speed, high imaging accuracy, short data-acquisition time, and high flexibility characteristics. However, the signal intensities of MBs are not quantitatively recovered, as similar to many other ULM techniques. To overcome this limitation, in this work, we propose a new deep-learning (DL) method, termed as qULM-DL, to recover the true distributions (positions and intensities) of the MBs in ULM. A series of numerical simulations and in vivo experiment are performed to evaluate the performance of qULM-DL. The results show that qULM-DL can not only super-resolve the complex structures but also quantitatively recover MBs intensities in these structures.

Ultrasound localization microscopy (ULM) enables unprecedented subwavelength spatial resolution by localizing the ultrasound contrast microbubbles (MBs). In conventional ULM methods, low-concentration MB solutions are used to ensure that the MBs are sufficiently sparse to enable localization of individual MB. This leads to a long data-acquisition time (i.e., microbubble-accumulation time) to accumulate enough MBs events using plane wave (PW) imaging. Considering the continuity of the microvessels and redundancy of MB tracks, in this study, we present a method that leverages deep neural network (DNN) to predict microvasculature using significantly reduced data-acquisition time and achieve quality comparable to that obtained using full data-acquisition time. Phantom experiments show that the proposed method significantly improves the density of localized MBs compared with the conventional method. Besides, in vivo experiments demonstrate that the proposed method is capable of reconstructing ULM images using reduced data-acquisition time with small difference, compared with the conventional method.

Spinal cord (SC) is the relay station and reflex center of central nervous system, its function is therefore very important for nerve conduction pathway. Spinal tumor (ST) is the typical disease of SC. In clinic, the standard treatment of ST is resection and a series of sequelae occur with imprecise surgical removal. Therefore, the delineation of the lesion boundaries of ST is very critical in surgery. Magnetic resonance (MR) imaging and intraoperative frozen section biopsy (IFSB) are usually used to assist in determining ST boundaries or boundary type, However, they suffer from severe limitations, such as unclear boundaries of certain types of ST in MR imaging and the inaccuracy of IFSB. Ultrasound localization microscopy (ULM) with contrast microbubbles (MBs) have been developed by several groups and may have potential in determining ST boundaries during surgery. Therefore, we apply ULM to human SC to distinguish pathological tissue from normal tissue in SC. In this study, intramedullary cavernous malformation (ICM), a typical ST, is imaged to demonstrate the feasibility of ULM of SC. Maximum intensity projection (MIP) images of filtered MBs of normal SC and ICM are obtained for comparison. Results indicate that the boundaries of normal SC and ICM in ULM images are more definite than those in MIP images. Quantitatively, the microvessel intensity (MVI) in normal SC is higher than that in ICM, while other indexes, such as direction entropy (DE), mean velocity (MEV), minimum velocity (MinV), and maximum velocity (MaxV), are similar in normal SC and ICM, which suggests the potential of ULM in surgery of SC in terms of distinguishing pathological tissue.

Rupture of carotid atherosclerotic plaques is one of the major causes of cerebrovascular events. Polar strains in the cross-sectional carotid elastography have been demonstrated to identify vulnerable plaques. However, the vessel center used to calculate the polar strains is difficult to estimate precisely in carotid arteries with plaques. To solve this problem, principal strains which can be measured without identification of the vessel center are proved consistent with polar strains in healthy volunteers. In this study, we compared the performance of principal and polar strains in assessing the elasticity of carotid atherosclerotic plaques. Ultrasound radiofrequency data were acquired from 13 carotid plaques of 9 patients, in the transversal imaging views. The inter-frame axial and lateral strains were obtained using an affine model-based spatial angular compounding method, which were further transformed into the polar (radial and circumferential) and principal (minor and major) strains. The maximum, mean, median, standard deviation (SD) and 99th percentile were calculated from the absolute value of each strain for comparison. Statistically significant correlations were found between the maximum, mean, median, SD and 99th percentile calculated from minor principal strain and radial polar strain, as well as between the indices from major principal strain and circumferential polar strain. These findings demonstrate significant correlations between polar and principal strains, and principal strains may be useful indices for assessing the vulnerability of carotid atherosclerotic plaques.

Chronic liver diseases (CLDs) are a major killer of humans, and more than 1 billion people suffer from CLDs worldwide, which are likely to develop significant liver fibrosis (LF) and even cirrhosis. However, current noninvasive diagnostic methods for liver diseases are insensitive to early lesions, so accurate assessment of LF at early stage is very important for the treatment arrangement and fibrosis reversal before cirrhosis. Ultrasound localization microscopy (ULM) with contrast microbubbles (MBs) have been proposed by several groups and may have potential in evaluating LF at early stage. Unfortunately, breathing and heart beating can introduce motion artifacts in liver ULM, which brings the challenge for LF evaluation with ULM. Recently, rigid motion correction (MoCo) has been proposed to improve the performance of ULM of the brain and kidney. Considering typical non-rigid motion in the liver, the performance of rigid MoCo may be limited. Therefore, we propose a non-rigid MoCo method based on speckle tracking to improve the performance of liver ULM more effectively. Results of in vivo experiments indicate that ULM with non-rigid MoCo obtains better resolution and more continuous microvessels (MVs) than rigid MoCo.

Marfan syndrome (MS) is a hereditary connective tissue disease characterized by the defection of collagen complex and decrease of the connection between collagen and elastic fibers. Thoracic aortic dissection (TAD) is a common cause of death for patients with MS. In this study, we performed pulse wave imaging (PWI) in a β-aminopropionitrile (BAPN) induced, MS-relevant mouse model of TAD and measured the pulse wave velocities (PWVs) to assess the aortic stiffness related to the occurrence of TAD in MS. Three-week-old male mice were fed with regular diet (control group, n = 6) and diet with BAPN (BAPN group, n = 10) for 20 days, respectively. BAPN inhibits the cross-linking of collagen and elastin in systemic arteries, similar to TAD in MS patients, and could induce the occurrence of TAD in mice. PWI of the proximal abdominal aorta was performed every two days with a SonixMDP system and an L40-8 probe. Focused wave imaging with a reduced beam density was used to obtain a high frame rate of 980 Hz. The distension velocities of the aortic wall were estimated using speckle tracking and PWVs at the systolic foot (PWV_sf) and dicrotic notch (PWV_dn) were measured in vivo. Histologic examinations were performed on the thoracic aorta to confirm the presence of TAD. Statistical significance was assessed using two-tailed t tests. After 20 days of feeding, 8 mice from the BAPN group were found to have TAD in the histologic examinations. No significant differences in PWV_sf and PWV_dn were found between the control group (n= 6) and the TAD group (n = 8) before BAPN diet (p = 0.60 for PWV_sf, p = 0.98 for PWV_dn). In the last measurements after BAPN diet, PWV_dn from the TAD group was lower than that from the control group (p < 0.05), while no significant difference was observed in PWV_sf (p=0.48). PWV_dn increased with the age of the mice for both the control and TAD groups, while the increment of PWV_dn between the last and first measurements was significantly lower in the TAD group (p < 0.05). Such difference was not found in PWV_sf (p = 0.85). These findings indicate that PWV_dn of the abdominal aorta can reflect the arterial stiffness change related to TAD and may be a promising index to monitor the aortic mechanical properties of MS.

Spatial angular compounding (SAC) has been developed to improve the performance of lateral estimation. It is a promising method for cross-sectional carotid elastography to image the strain distribution within carotid atherosclerotic plaque and identify the vulnerable plaque. However, in strain estimation with SAC, arterial motion and deformation during multi-angle acquisitions may cause motion artifacts. In this study, the influence of the key factors on motion artifacts, including the pulse repetition frequency (PRF) and number of steering angles (NSA), is investigated. Simulations and in vivo experiments are conducted using a linear array with plane wave transmission. The elastographic signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) are used to evaluate the quality of the strain image. The simulation results show that the SNR and CNR of lateral strain increase with the PRF when it is lower than 4 kHz, the CNR and SNR of lateral strain with an NSA of 7 are lower than those with an NSA of 3 when PRF 4 kHz. The in vivo experimental results show that the radial and circumferential strain images at a PRF of 3 kHz are smoother and more homogeneous than those at a PRF of 429 Hz. In conclusion, both the PRF and NSA have influence on motion artifacts in strain estimation with SAC. For carotid elastography, a PRF of more than 4 kHz and an NSA of 3 are suggested to obtain strain images with smaller motion artifacts.

Diseases such as diabetes mellitus can lead to chronic kidney disease (CKD). The early diagnosis of CKD is of significant importance to prevent it from rapid deterioration, while the current methods used in the clinic is susceptible. As one of the major complications of diabetes, diabetic kidney disease (DKD) leads to the significant changes of renal microvasculature and perfusion in the early stage. However, they cannot be observed by conventional imaging methods. Ultrasound localization microscopy (ULM) has been developed in recent years to significantly improve the spatial resolution of ultrasound imaging by localizing the microbubbles (MBs) flowing in microvasculature. This technique can not only obtain the microvasculature in deep tissue, but also provide an accurate blood velocity map. In this study, ULM was performed on an acute ischemic-reperfused rat kidney model and a normal rat kidney to investigate its in vivo feasibility in evaluation of DKD. The microvascular morphology and velocity maps of the kidneys were obtained. Results show that under current experimental conditions, a resolution of 36 μ m is reached. Much less MB events (70 k vs 160 k from 6, 000 frames) and much lower mean blood flow speed (typically < 10 mm/s vs ~30 mm/s) were detected in the ischemic-reperfused kidney than in the normal kidney. This study shows the potential of ULM in the assessment of DKD/CKD.

Coherent plane wave compounding (PWC) using tens of steered plane waves (PWs) can obtain high-quality ultrasound images but reduces the gain in frame rate. Recently a new strategy using convolutional neural network (CNN) was proposed to recover high-quality images from only 3 PWs. Considering the excellent performance of generative adversarial network (GAN) in image reconstruction, we propose to use GAN to reconstruct high-quality ultrasound images from 3 PWs. Phantom and in vivo experiments are performed. The results of GAN using 3 PWs, in terms of contrast ratio and lateral resolution are competitive with those of CNN using 3 PWs and coherent compounding using 31PWs, which demonstrates the feasibility of this method.

Transverse oscillations (TOs) have been demonstrated to improve the performance of lateral motion estimation, thus obtaining more reliable motion estimation for tissue characterization. TOs were introduced to diverging wave (DW) based ultrafast imaging for 2D motion estimation of rapidly moving targets such as the heart. Furthermore, coherent compounding can improve the quality of DW imaging. The objective of this study is to develop TO method for DW compounding (DWC) to improve the performance of 2D motion estimation for cardiac deformation imaging. Validation is carried out in simulations on a homogenous phantom with lateral rigid motion and a cardiac model with realistic motion. Results demonstrate that TOs can improve the performance of 2D motion estimation in DWC imaging.

Diverging wave (DW) transmit based on phased array offers a potential solution for cardiac deformation imaging at high frame rate. Diverging wave compounding (DWC) imaging has been developed to improve the imaging signal-to-noise ratio and contrast as well as the performance of cardiac strain estimation. We previously proposed to use coded excitation by chirp and Golay codes in DWC imaging to improve the B-mode image quality without sacrificing the frame rate. This study proposes to apply coded excitation in DWC imaging for cardiac deformation imaging and investigates its performance both in silico and in vivo. The results show that DWC with coded excitation achieves better motion estimation for cardiac deformation imaging.

Transient elastography (TE) is a non-invasive, rapid and reproducible technology, which performs liver stiffness measurement (LSM) for staging the liver fibrosis in the clinic. In the procedure of LSM, an M-mode strain image (MSI) is estimated from the M-mode radiofrequency (RF) data. On the hypothesis that the MSI contains richer information than LSM, we propose a deep learning (DL) method based on the MSI to improve the performance of fibrosis staging. A multicenter study was conducted where both TE and liver biopsy were performed on 421 patients and finally 245 patients with 2,713 MSIs were qualified. An optimal deep learning model with 8 layers (5 convolutional layers and 3 fully-connected layers) was built and evaluated on the dataset. LSMs were also used to assess liver fibrosis for comparison, while liver biopsy acted as the gold standard. The receiver operating characteristic (ROC) curve was plotted to evaluate the performance for assessing significant fibrosis (≥ F2), advanced fibrosis (≥ F3) and cirrhosis (F4). The AUC of the DL method was 0.850 for ≥ F2. The AUCs of the DL method were 0.948 for ≥ F3 and 0.934 for F4, respectively, slightly better than those of the LSM method (0.935 for ≥ F3 and 0.908 for F4, respectively). To conclude, the DL method performs better than the LSM method for advanced fibrosis and cirrhosis and could be used in transient elastography for assessment of liver fibrosis.

We propose a novel method for high efficiency wavefront control by utilizing insulator-metal and metal-insulator-metal metamaterial. Based on such method, high efficiency anomalous reflection and planar focus lens are achieved in microwave region. Moreover, we demonstrate that such wavefront control can also be used in near-field optics with realizing an almost perfect the conversion of propagating waves into surface waves.

Guided wave elastography (GWE) is promising to measure the local arterial stiffness. However, most studies focus on the longitudinal section, with a guided axial wave (GAW). In addition, the dispersion curve of GAW can be well approximated by immersed Lame wave model only at high frequencies (Li et al, UMB 2017). In the transverse section, guided circumferential wave (GCW) can be utilized to obtain the arterial Young's modulus (Li et al, JBM 2017). In this study, the effect of the lumen pressure on the dispersion curves of GAW and GCW was investigated, and a pressure quantification method from GCW at different pressures was developed.

In most studies on vascular shear wave imaging (SWI), longitudinal section is mainly concerned, and the radial motion and reflected waves are not considered in cross section. It is important to investigate the SWI in cross section when evaluating the anisotropy of the vessel wall or the complete plaque composition. In our previous study, a novel method based on coordinate transformation and directional filtering was proposed to achieve vessel cross-sectional shear wave imaging (VCS-SWI), where the radial motion was directly estimated in the polar coordinates of the vessel (referred to as the proposed method). The radial motion can also be calculated from the axial and lateral motions estimated in the Cartesian coordinate system of the transducer (referred as to Method 1), or from multiple axial motions with spatial angular compounding (referred as to Method 2). In this study, a comparison of the three methods for VCS-SWI was performed in simulations. The root-mean-square errors (RMSEs) of the estimated radial motions showed that the performance of radial motion estimation by using Method 1 is the worst due to the poor lateral motion estimation. With angular compounding, the radial motion can be well obtained, but at the cost of lower frame rate. The performance of the proposed method is similar to that of spatial angular compounding. In conclusion, the proposed method could obtain high-quality VCS-SWI without sacrificing the frame rate.

In vivo measurement of the mechanical properties of thin-walled soft tissues (e.g., mitral valve, artery and bladder) and in situ mechanical characterization of thin-walled artificial soft biomaterials in service are of great challenge. Those thin-walled structures are usually pre-stressed to achieve and/or improve their functional performance, which further complicate the inverse analysis to identify the mechanical properties. In this study, we investigate the properties of guided waves generated by focused acoustic radiation force in immersed pre-stressed plates and tubes, and show that they can address this challenge.

Spatial angular compounding (SAC) with plane wave has been proposed to improve the performance of motion and strain estimation at high frame rate. A previous study shows that the accuracy of 2D estimation depends on the number of steering angles, maximum steering angle, and accuracy of axial estimation at different angles (He et al, Ultrasonics 2017). We have demonstrated (Pan et al, TUFFC 2015) that optical flow (OF) with affine model obtains better axial estimation than normalized cross-correlation based block matching (BM) with rigid model. The objective of this study was to investigate whether OF combined with SAC performs better in plane wave imaging (PWI).

Cardiac mock models are helpful to develop new imaging methodologies. Although several mock model setups have been proposed, a good model for the development of intra-cardiac flow imaging is missing. The challenge lies in the fact that mimicking realistic flow patterns requires a realistic left ventricular shape as well as valve systems and a fluid with appropriate rheological properties. Moreover, in order to enable an independent reference method, the model either has to be optically transparent (i.e. for OPIV) or has to be MRI-compatible (i.e. phase-contrast MRI). The goal of this study was therefore to develop such a cardiac mock model.

Thermal strain imaging (TSI) is promising for lesion targeting and dose control in thermal-based treatment. We previously developed HIFU and temperature sensitive cerasomes (HTSCs) for local chemotherapy of cancer (Liang et al, Acs Nano 2015). However, the temperature within tissues cannot be noninvasively measured in vivo. The aim of this study was to investigate the feasibility of TSI in noninvasive monitoring of HIFU-medicated doxorubicin (DOX) loaded HTSCs delivery.

Recently, gradient metasurfaces, have paved a new direction to the field of metamaterial with many remarkable achievements, including generalized Snell's law [1], new bridge to couple surface wave [2], meta-holograms [3], planar optical devices [4, 5], etc‥ In spite of these great developments, there are some aspects of the metasurfaces needed to be improved. For example, most of the previous systems only operate with the single functionality, which obviously does not satisfy the high integration requirement of the optical devices in modern science and technology. In this paper, we experimentally realized the bifunctional operation based on the polarization-dependent metasurfaces in both reflection and transmission geometry. In the former case, the functionality of the metasurface can be freely switched between focusing and surface wave excitation by rotating the polarization direction of the input linear polarized electromagnetic wave. In the latter case, the metasurface can serve as either a focusing lens or a beam deflector under different polarizations. Compared to the previous works, our device can two distinct functionalities with very high efficiency. The finding may suggest a new avenue to the designs of bifunctional devices.

Metamaterialsare electromagnetic composites constructed by subwavelength sized microstructures, and thus can exhibit tailored effective permittivity and permeability. Effective-medium theory (EMT) plays a crucial role since it serves as a bridge to link theories, which are frequently conducted on model hypothetical systems, and experiments which are always performed on realistic metamaterial systems with complex microstructures. Therefore, a good EMT, which can precisely predict the effective parameters of complex metamaterials, is always highly desired. Based on a mode-expansion theory under single-mode approximation, we derived the scatterings parameters for a general one-dimensional photonic grating composed by two different materials, and then established an effective-medium theory (EMT) for such a composite using the derived scatterings parameters. Our EMT well describes the gating structures with general material and geometrical parameters, and recovers previous two formulas which are valid only at certain limiting conditions. The theory is justified by full wave simulations and microwave experiments. Then we generalize our theory into two-dimensional case. Mapping such an inhomogeneous system to an effective homogeneous medium exhibiting the same scattering coefficients, we established an effective medium theory for an array of subwavelength cylinders. Our theory is justified by comparing with full wave simulations on realistic structures, and is found to work well in a much more expanded parameter space than the standard Maxwell-Garnett theory.

Metasurfaces in a metal-insulator-metal configuration have been widely used in photonics, with applications ranging from perfect absorption to phase modulation, but why and when such structures can realize what functionalities are not yet fully understood. Here, we establish a complete phase diagram in which the optical properties of such systems are fully controlled by two simple parameters (i.e., the intrinsic and radiation losses), which are, in turn, dictated by the geometrical or material properties of the underlying structures. Such a phase diagram can greatly facilitate the design of appropriate metasurfaces with tailored functionalities demonstrated by our experiments and simulations in the terahertz regime. In particular, our experiments show that, through appropriate structural or material tuning, the device can be switched across the phase boundaries yielding dramatic changes in optical responses. Our discoveries lay a solid basis for realizing functional and tunable photonic devices with such structures.

In this paper, the previously proposed compressed sensing based synthetic transmit aperture (CS-STA) was compared with synthetic transmit aperture (STA), multi-element synthetic transmit aperture (ME-STA) and the conventional focused mode in two different phantoms. The quantitative results of full width at half-maximum (FWHM) and contrast-to-noise ratio (CNR) demonstrated the merits of CS-STA. Besides, the results from two different phantoms verified the reproducibility of CS-STA. Moreover, the clinical potentials of CS-STA were presented through preliminary in-vivo studies on human bicipital muscle of arm and thyroid. At last, CS-STA is demonstrated to exhibit certain advantages through the comparison with coherent plane wave (CPW) compounding method.

High frame rate (≤200 Hz) is required in pulse wave imaging (PWI) to estimate regional pulse wave velocity (PWV) of arteries, and can be achieved by sparse imaging or plane wave imaging at the expense of spatial resolution or signal-to-noise ratio (SNR). However, both high temporal and spatial resolution are required when mapping local PWV variations along inhomogeneous arteries. In this study, the whole imaging field of view (FOV) was divided into 4 sub-sectors, each with 34 beams including a common beam. Ultrasound radiofrequency (RF) data was then acquired sequentially in each sector. A high line density (128 beams, 38 mm width) and high frame rate (∼507 Hz, 35 mm depth) RF frame sequence was reconstructed from the sub-sector frames by aligning the estimated vessel motion along the common beam. The composite RF frames were then post-processed using PWI technique. Local PWV was estimated within a sliding kernel of ∼10 mm along the vessel wall. For the homogeneous phantom, local PWV measurements of the proposed method show good agreement with the sparse imaging, and undergo less fluctuation compared with plane wave imaging. In the low SNR condition, the proposed method obtains smaller standard deviation (SD) than sparse imaging. For inhomogeneous phantoms, significant difference of local PWV measurements between the stiffer and softer regions is found by using the proposed method, the sparse imaging or plane wave imaging. The proposed method is capable of obtaining both high line density and high frame rate data and mapping local PWV variations.

We briefly summarize our recent efforts in employing meta-surfaces to control electromagnetic waves, including realizing high-efficiency photonic spin-hall effect and surface-plasmon couplers, and controlling phases with graphene-based meta-surfaces.

Rupture of atherosclerotic plaques is the key factor leading to ischemic stroke. In conventional ultrasound, B-mode imaging and/or color flow/pulsed wave Doppler can be used to evaluate the stenosis of blood vessels but not the vulnerability of plaques. In this paper, we propose an ultrasound strain rate imaging based indicator called mean maximum strain rate (MMSR) to evaluate the vulnerability of carotid atherosclerotic plaques validated using multi-sequence magnetic resonance (MR) imaging characterization. This indicator was tested on 47 plaques from 34 human subjects. Statistical significance was found (p=0.01) between stable plaques and vulnerable plaques. The sensitivity and specificity at the cost effective point of receiver operating characteristic (ROC) curve is 76% and 81%, respectively. This indicator is shown to be effective in evaluating the vulnerability of carotid atherosclerotic plaques.

The predictive value of pulse wave velocity (PWV) for various cardiovascular diseases has been demonstrated in many studies. By means of ultrasound imaging, local PWV, instead of global PWV as in conventional methods, can be measured noninvasively. The accuracy and precision of local PWV measurement have not been fully investigated and several key parameters (i.e. frame rate, number of scan lines and image width) are shown to play an important but still unclear role on PWV measurement. In this study, the quantitative effects of these parameters were investigated with theoretical analysis and in-vivo experiments. The theoretical relationship between the standard deviation (SD) of the PWV estimation and some key parameters (number of scan lines, image width and PWV value) and the variance of characteristic time-point identification (or time-shift estimation) was derived first. PWVs were then estimated from in-vivo RF signals at different values of frame rate, number of scan lines, and image width. The results show that PWVs can be correctly measured when the frame rate is higher than a certain value, below which the estimated PWVs become inaccurate. The performance of PWV measurement is found to improve with increased number of scan lines and image width. Overall, a larger number of scan lines and image width with a sufficiently high frame rate is preferred. The in-vivo results are in in good agreement with the theoretical analysis as well as our previous simulation results. Our quantitative findings can provide important guidelines for parameters optimization in ultrasound-based local PWV measurement.

We design an anisotropic ultrathin metamaterial to allow perfect transmissions of electromagnetic (EM) waves for two incident polarizations within a common frequency interval. The transparencies are governed by different mechanisms, resulting in significant differences in transmission phase changes for two polarizations. The system can thus manipulate EM wave polarizations efficiently in transmission geometry, including polarization conversion and rotation. Microwave experiments performed on realistic samples are in excellent agreement with numerical simulations.

We proposed a simple scheme to make a continuous metallic film on a semi-infinite substrate optically transparent, thus obtaining a completely transparent electrode in a desired frequency range. By placing a sub-wavelength composite layer consisting of dielectric and metallic stripes on top of the metallic one, we found that the back-scattering from the metallic film can be almost perfectly canceled by the composite layer under certain conditions, leading to transparency of the whole structure. Since our mechanism does not require any openings on the opaque metallic plate, the proposed structure retains the full electric and mechanical properties of a natural metal. The present mechanism is insensitive to structural disorders and broad variation of incidence angles. Meanwhile, we performed proof-of-concept experiments in the terahertz domain to verify our theoretical predictions, using carefully designed metamaterials to mimic plasmonic metals in optical regime. Experiments are in excellent agreements with full-wave simulations.

We show that a flat metasurface with a parabolic reflection-phase distribution can focus an impinging plane wave to a point image in reflection geometry. Our system is much thinner than conventional geometric-optics devices, and does not suffer the energy-loss issues encountered by many metamaterial devices working in transmission geometry. We designed realistic microwave samples and performed near-field scanning experiments to verify the focusing effect. Experimental results are in excellent agreement with full wave simulation and theoretical analyses.

We have designed and realized an optically non-existing transformation-medium, based on a specific plasmonic metamaterial. Such a medium can perfectly transport impinging waves at any incident angle under both TE and TM polarizations without phase accumulations. Effective medium theory, mode expansion theory and full-wave simulations are in consistency with each other and facilitated the design of realized samples. We demonstrated experimentally that a curved microwave sample fabricated based on our design can work as a hyperlens, with experimental results in excellent agreement full-wave simulations.

The disadvantages of two existing traction power supply modes are analyzed and advanced co-phase traction power supply system based on three-phase to single-phase converter is studied in the paper. Then, the chain model of network and the constant power model of locomotive are established. According to the above results, further analysis of the circulating-impedance and the feasibility study of droop control in this system are presented. The circulating-current analysis results of the advanced co-phase traction power supply system and the droop control of the system are given in the end of the paper.

An iterative strategy combining Kirchhoff Approximation (KA) with the hybrid finite element-boundary integral (FE-BI) method is employed in this paper to consider the interactions between the plasma coated target and the underlying rough sea surface. This hybrid FE-BI-KA scheme is an improved and generalized version of previous KA-MoM, whereas the newly presented technique can deal with inhomogeneous target. The multi-hybrid method is applied to investigate electromagnetic scattering from plasma coated target above rough sea surface. Discussion on the effects of plasma layer on scattering pattern are carried out by using this method, and several examples present the reduction of BSC near the backward direction when coated by plasma layer. Also the influences of both electron collision frequency and plasma angular frequency (which is determined by electron density) on the plasma stealth are considered respectively.

An approach to studying composite EM scattering from 2-D rough surface and half-buried 3-D target is proposed in this paper, from accurate geometry modeling to RCS analysis. First, two dimensional rough surface is simulated with spectral method, which then is converted into a continuous curved surface through inverse engineering. Target model and rough surface model are combined together to be a very accurate composite CAD model, after which surface mesh is generated with the information of incidence wave frequency. Then physical optics method (PO) and multi-level fast multipole algorithm (MLFMA) are respectively employed to study the scattering problem. The results of single rough surface and combinative model are given and the results by the two different methods are compared.

It is consume much time for hidden surface removal when compute the scattering of electrically large object by physical optics method. In this paper, a fast hidden surface removal algorithm is presented. First, the planar triangles covering the body surface are divided into several groups. The characteristic vectors of each groups is selected. Then, the dot product of the characteristic vectors and orientation vector of the incident wave are computed and the group is judged if illuminated by the incident wave. At last, the group which is not all the planar triangles are illuminated is judge by each planar triangle one by one. The computed results shown that our algorithm can improve the computation efficiency when the number of the planar triangles is very large.

In this paper, a hybrid method combining time domain integral equation (TDIE) with finite difference time domain(FDTD) is presented to study scattering properties of arbitrary thin wire structure horizontally positioned above a layered half space. The time domain reflections of electromagnetic waves from layered half space are obtained by one dimensional FDTD method. Then, transient response of thin-wire induced by two excitation sources (original incidence and reflected wave) is achieved by employing TDIE. Numerical examples for thin wires with different structures are given and analysed.

A new iterative strategy combining analytical Kirchhoff Approximation (KA) with the powerful simulation tool finite elementboundary integral (FE-BI) method is presented in this paper to study the interactions between an object and the underlying rough surface. This hybrid FE-BI-KA scheme is an improved and generalized version of MOM-KA. Instead of using MOM, FE-BI is used to model scattering from the above object with the original incidence and scattered wave by the rough surface as the exciting source. Whereas scattering from rough surface is studied by the classical KA in order to reduce the computation resource and time required