Ultrasound RF Signals Research Articles

Assessment of Carotid Artery Vibrations by Using Optical Flow Methods on Ultrasound Images and Optical Flow

Purpose: This study aims to extract carotid wall vibrations non-invasively and evaluate changes in the carotid artery caused by age, BMI, and sex. Such evaluation can increase the possibility of detecting wall stiffness and atherosclerosis in the early stages and prevent heart attack and death. Materials and Methods: To extract small vibrations of the carotid wall, the image-tracking method, and optical flow with four different methods were used. The study involved twenty participants, comprising six females and fourteen males, with a mean age of 36.25 years, mean weight of 75.2 Kg, and mean BMI of 25. The posterior wall motion and vibration were extracted using ultrasound RF signals. Results: 4 optical flow methods, Gunnar-Farneback, Horn- Schunk, Lucas-Kanade, and Lucas-Kanade derivative were used for all samples, and covariance, correlation, P-value, and R-squared were estimated. Results showed the differences in parameters such as age and BMI with carotid wall vibration. These values for age are (Cov = -0.1891, Corr = -0.6770, P-value < 0.001, r2 = -0.4590) and for BMI are (Cov = -0.0564, Corr = -0.6510, P-value < 0.001, r2 = 0.4238), respectively. For gender as a new parameter, a comparison between men’s and women’s vibrations was estimated. The range of measured vibrations by optical flow methods is about 0.0002 µm to 0.1µm, and the mean standard deviation is 0.04967 µm. Conclusion: The results presented that gender affects the vibration of the vessel wall, which in men is more than in women. In addition, increasing age and BMI may increase the stiffness of the carotid wall and reduce vibrations that were evaluated previously. Using the Gunner-Farneback method as image tracking for small vibrations is the best way with the highest accuracy.

Unsupervised deep learning-based displacement estimation for vascular elasticity imaging applications

Objective. Arterial wall stiffness can provide valuable information on the proper function of the cardiovascular system. Ultrasound elasticity imaging techniques have shown great promise as a low-cost and non-invasive tool to enable localized maps of arterial wall stiffness. Such techniques rely upon motion detection algorithms that provide arterial wall displacement estimation. Approach. In this study, we propose an unsupervised deep learning-based approach, originally proposed for image registration, in order to enable improved quality arterial wall displacement estimation at high temporal and spatial resolutions. The performance of the proposed network was assessed through phantom experiments, where various models were trained by using ultrasound RF signals, or B-mode images, as well as different loss functions. Main results. Using the mean square error (MSE) for the training process provided the highest signal-to-noise ratio when training on the B-modes images (30.36 ± 1.14 dB) and highest contrast-to-noise ratio when training on the RF signals (32.84 ± 1.89 dB). In addition, training the model on RF signals demonstrated the capability of providing accurate localized pulse wave velocity (PWV) maps, with a mean relative error (MREPWV) of 3.32 ± 1.80% and an R2 of 0.97 ± 0.03. Finally, the developed model was tested in human common carotid arteries in vivo, providing accurate tracking of the distension pulse wave propagation, with an MREPWV = 3.86 ± 2.69% and R2 = 0.95 ± 0.03. Significance. In conclusion, a novel displacement estimation approach was presented, showing promise in improving vascular elasticity imaging techniques.

Improving axial resolution of medical ultrasound images by using noise-robust broadband filter based on singular value decomposition

Improving spatial resolution is a crucial issue in medical ultrasound. One of the improving methods is the post-processing of the received ultrasound RF signal. In the present paper, we proposed a design method for a noise-robust broadband filter based on the singular value decomposition of the received RF signal. To design a noise-robust filter, we proposed a logical method to determine the optimal truncated order of singular values, which was validated by applying the filter to noise-contaminated signals. Furthermore, the proposed filter applied to the wire phantom resulted in a better axial resolution than that obtained without the filter and with our previously designed Wiener filter.

Simulation of Ultrasound RF Signals Backscattered from a 3D Model of Pulsating Artery Surrounded by Tissue.

Arterial stiffness is an independent predictor of cardiovascular events. The motion of arterial tissues during the cardiac cycle is important as a mechanical deformation representing vessel elasticity and is related to arterial stiffness. In addition, arterial pulsation is the main source of endogenous tissue micro-motions currently being studied for tissue elastography. Methods based on artery motion detection are not applied in clinical practice these days, because they must be carefully investigated in silico and in vitro before wide usage in vivo. The purpose of this paper is to propose a dynamic 3D artery model capable of reproducing the biomechanical behavior of human blood vessels surrounded by elastic tissue for endogenous deformation elastography developments and feasibility studies. The framework is based on a 3D model of a pulsating artery surrounded by tissue and simulation of linear scanning by Field II software to generate realistic dynamic RF signals and B-mode ultrasound image sequential data. The model is defined by a spatial distribution of motions, having patient-specific slopes of radial and longitudinal motion components of the artery wall and surrounding tissues. It allows for simulating the quantified mechanical micro-motions in the volume of the model. Acceptable simulation errors calculated between modeled motion patterns and those estimated from simulated RF signals and B-scan images show that this approach is suitable for the development and validation of elastography algorithms based on motion detection.

Advantages of cooled fiber for monitoring laser tissue ablation through temporal and spectral analysis of RF ultrasound signal: A case study

The promising minimally invasive laser thermal therapy technique may be improved if thermal lesions induced into the tissue can be carefully monitored in extension and morphology during the treatment. According to results obtained in several recent experimentations, solutions that avoid tissue carbonization during the treatment have been proposed, in order to allow deeper and longer lasting light penetration in treated tissue and to reduce failures of the applicator tip and fiber optic, dangerous for patients. In the work the advantages in using a cooled fiber are shown, in order not only to induce efficient lesions but also in performing an accurate monitoring by ultrasound. Indeed, one important limit of the ultrasound control is caused by the gas bubbles generation, which represent an acoustic barrier that invalidate the ultrasonic image representation of the treated tissue. Ultrasonic radiofrequency signals were acquired from the same bovine liver ex vivo sample by using both bare and cooled fiber and processed to produce B-mode and spectral parametric images by implementing TUV (Thermotherapy Ultrasonic View) algorithm. Radiofrequency signals, B-mode and TUV images were analysed and compared in order to evaluate the different tissue heating processes during ablation and the different lesion extensions induced into the tissue after the treatment. Cooled fiber avoided carbonization and strongly reduced gas bubbles generation inducing a larger lesion and allowing a more effective ultrasound monitoring. Moreover by correlating optical images of the lesions and the corresponding Integral TUV images, by using Dice and Jaccard coefficients, it was proven that TUV algorithm is able to characterize the tissue portions differently modified by ablation exhibiting better performances in the case of cooled fiber and revealing to be a potential tool capable to improve the laser delivery settings control.

Performance comparison of rigid and affine models for motion estimation using ultrasound radio-frequency signals.

Tissue motion estimation is widely used in many ultrasound techniques. Rigid-model-based and nonrigid-modelbased methods are two main groups of space-domain methods of tissue motion estimation. The affine model is one of the commonly used nonrigid models. The performances of the rigid model and affine model have not been compared on ultrasound RF signals, which have been demonstrated to obtain higher accuracy, precision, and resolution in motion estimation compared with B-mode images. In this study, three methods, i.e., the normalized cross-correlation method with rigid model (NCC), the optical flow method with rigid model (OFRM), and the optical flow method with affine model (OFAM), are compared using ultrasound RF signals, rather than the B-mode images used in previous studies. Simulations, phantom, and in vivo experiments are conducted to make the comparison. In the simulations, the root-mean-square errors (RMSEs) of axial and lateral displacements and strains are used to assess the accuracy of motion estimation, and the elastographic signal-tonoise ratio (SNRe) and contrast-to-noise ratio (CNRe) are used to evaluate the quality of axial strain images. In the phantom experiments, the registration error between the pre- and postdeformation RF signals, as well as the SNRe and CNRe of axial strain images, are utilized as the evaluation criteria. In the in vivo experiments, the registration error is used to evaluate the estimation performance. The results show that the affinemodel- based method (i.e., OFAM) obtains the lowest RMSE or registration error and the highest SNRe and CNRe among all the methods. The affine model is demonstrated to be superior to the rigid model in motion estimation based on RF signals.

Effects of parameters on the accuracy and precision of ultrasound-based local pulse wave velocity measurement: a simulation study.

Quantification of arterial stiffness, such as pulse wave velocity (PWV), is increasingly used in the risk assessment of cardiovascular disease. Pulse wave imaging (PWI) is an emerging ultrasound-based technique to noninvasively measure the local PWV instead of the global PWV, as in conventional methods. In PWI, several key parameters, including the frame rate of ultrasound imaging, motion estimation rate (MER), number of scan lines, image width, PWV value, and sonographic signal-to-noise ratio (SNRs), play an important but still unclear role in the accuracy and precision of PWV measurement. In this study, computer simulations were performed to investigate the fundamental effects of these parameters on the PWV measurement. The pulse waveform was estimated by speckle tracking on ultrasound RF signals acquired at a frame rate of 2083 Hz from a location on the common carotid artery of a healthy subject. By applying different time delays on the estimated waveform based on specific PWI parameters, the pulse waveforms at others locations were simulated. Ultrasound RF signals of the artery during the pulse wave propagation were generated from a 2-D convolutional image formation model. The PWI technique was applied to estimate the PWV at different values of frame rate, MER, number of scan lines, image width, PWV, and SNRs. The performance of the PWV estimation was evaluated by measuring the relative error, coefficient of variation (CV) and coefficient of determination (R(2)). The results showed that PWVs could be correctly measured when the frame rate was higher than a certain value (i.e., minimum frame rate), below which the estimated error increased rapidly. The minimum frame rate required for PWV estimation was found to increase with the value of PWV. An optimal MER was found (i.e., about 200 Hz) and allowed better performance of PWV measurement. The CV of PWV estimation decreased and R(2) increased with number of scan lines and image width, indicating that the performance of the PWV estimation could be improved with a larger number of scan lines and image width. For a given sufficiently high frame rate, a higher PWV value was found to deteriorate the PWV estimation, as indicated by an increasing CV and decreasing R(2). The simulation results were in good agreement with the theoretical analysis. Finally, high-quality PWV estimation could be obtained as long as the SNRs was higher than about 30 dB. The quantitative effects of the key parameters obtained from this study might provide important guidelines for parameter optimization in ultrasound-based local PWV measurement in vivo.

Effects of key parameters on the accuracy and precision of local pulse wave velocity measurement by ultrasound imaging.

Quantification of arterial stiffness, such as pulse wave velocity (PWV), is increasingly used in the risk assessment of cardiovascular disease. Pulse wave imaging (PWI) is an emerging ultrasound-based technique to noninvasively measure the local PWV, instead of the global PWV as in conventional methods. In PWI, several key parameters, including the frame rate, number of scan lines, image width and PWV, play an important but still unclear role in the accuracy and precision of PWV measurement. In this study, computer simulations were performed to investigate the fundamental effects of these parameters on the PWV estimation. By applying different time delays on the pre-obtained pulse waveform based on specific PWI parameters, the pulse wave propagation along the artery was simulated and the ultrasound RF signals were generated from a convolutional image formation model. The PWI technique was applied to calculate the PWV at different values of key parameters. The performance is evaluated by measuring the bias, standard deviation (SD) and coefficient of determination (R(2)) of the estimated PWVs. 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 SD decreases while R(2) increases with number of scan lines and image width, indicating a better performance of the PWV estimation with a larger number of scan lines and image width. A higher value of PWV is found to deteriorate the PWV estimation. The quantitative effects of the key parameters obtained from this study may provide important guidelines for optimization of PWI parameters in vivo.

Monitoring of high-intensity focused ultrasound lesion formation using decorrelation between high-speed ultrasonic images by parallel beamforming

High-intensity focused ultrasound (HIFU) is a type of therapeutic treatment in which ultrasound is focused to a target tissue such as cancer to be thermally coagulated. To enhance the safety and accuracy of HIFU treatment, a noninvasive method of monitoring the thermal lesion formation is important. A potential method for this purpose is ultrasound imaging. In this study, high-speed imaging by parallel beamforming was performed using ultrasound RF signals acquired during HIFU exposure, and the distribution of the cross-correlation coefficient between RF frames was calculated to estimate the tissue coagulation. Using high-speed imaging, HIFU can be irradiated quasi-continuously. The result shows that the decorrelation was observed at and around the focal spot of HIFU exposure. The decorrelation was induced by the change in RF signals owing to tissue coagulation, which was confirmed by the fact that the emergence and increase in the number of decorrelated pixels at and around the focal spot corresponded to the start and progress of tissue coagulation.

An efficient block matching and spectral shift estimation algorithm with applications to ultrasound elastography

An efficient block matching and spectral shift estimation algorithm for freehand quasi-static ultrasound elastography is described in this paper. The proposed method provides a balance between computational speed and robustness against displacement estimation error and bias; a fundamental aspect of elastography. The new algorithm was tested on an extensive set of simulated 1-D RF ultrasound signals, replicating various strain profiles. Additionally, real 2-D scans were conducted on an ultrasound phantom with prescribed elastic properties; the algorithm output was further validated with a comparison to a finite element model (FEM) of the phantom. Clinical data from a breast cancer study and histology slides were used to demonstrate the in vivo use of the new elastography technique. The algorithm showed a significant computational savings (at least 60 times faster) over existing spectral shift analysis methods. Accurate strain images were produced in as little as 2 s with the scope for further speed enhancements through parallel processing; making real-time implementation a future possibility. Moreover, it demonstrated a robustness toward displacement estimation error when compared with conventional gradient-based techniques, and was able to perform at high strain values (>5%) while showing relative insensitivity to various parameters settings, such as sample rate and block window size; a desirable performance for a practical clinical tool.

Gamma mixture classifier for plaque detection in intravascular ultrasonic images

Carotid and coronary vascular incidents are mostly caused by vulnerable plaques. Detection and characterization of vulnerable plaques are important for early disease diagnosis and treatment. For this purpose, the echomorphology and composition have been studied. Several distributions have been used to describe ultrasonic data depending on tissues, acquisition conditions, and equipment. Among them, the Rayleigh distribution is a one-parameter model used to describe the raw envelope RF ultrasound signal for its simplicity, whereas the Nakagami distribution (a generalization of the Rayleigh distribution) is the two-parameter model which is commonly accepted. However, it fails to describe B-mode images or Cartesian interpolated or subsampled RF images because linear filtering changes the statistics of the signal. In this work, a gamma mixture model (GMM) is proposed to describe the subsampled/interpolated RF images and it is shown that the parameters and coefficients of the mixture are useful descriptors of speckle pattern for different types of plaque tissues. This new model outperforms recently proposed probabilistic and textural methods with respect to plaque description and characterization of echogenic contents. Classification results provide an overall accuracy of 86.56% for four classes and 95.16% for three classes. These results evidence the classifier usefulness for plaque characterization. Additionally, the classifier provides probability maps according to each tissue type, which can be displayed for inspecting local tissue composition, or used for automatic filtering and segmentation.

Ultrasound-Guided Characterization of Interstitial Ablated Tissue Using RF Time Series: Feasibility Study

This paper presents the results of a feasibility study to demonstrate the application of ultrasound RF time series imaging to accurately differentiate ablated and nonablated tissue. For 12 ex vivo and two in situ tissue samples, RF ultrasound signals are acquired prior to, and following, high-intensity ultrasound ablation. Spatial and temporal features of these signals are used to characterize ablated and nonablated tissue in a supervised-learning framework. In cross-validation evaluation, a subset of four features extracted from RF time series produce a classification accuracy of 84.5%, an area under ROC curve of 0.91 for ex vivo data, and an accuracy of 85% for in situ data. Ultrasound RF time series is a promising approach for characterizing ablated tissue.

A PRELIMINARY STUDY ON TEST EFFICIENCY OF ARTERIAL FUNCTION PARAMETERS IN NORMAL YOUTHS BY HIGH FREQUENCY ULTRASOUND COMBINED WITH EXERCISE STRESS TESTING

ObjectivesTo investigate the difference of carotid elasticity function by two-dimensional speckle tracking imaging (X-strain) combined with ultrasound RF signal vascular stiffness (Quantitative Arterial Stiffness, QAS) technique, before and after exercise...

Monitoring of Lesion Induced by High-Intensity Focused Ultrasound Using Correlation Method Based on Block Matching

High-intensity focused ultrasound (HIFU) is a noninvasive therapeutic application that focuses ultrasound to the target tissue, such as a malignant tumor, and thermally coagulates it. Monitoring methods for evaluating the formation of thermal lesions induced by HIFU are required to perform safe and accurate HIFU treatment. However, the conventional ultrasonic B-mode image incurs difficulties in assessing the formation of HIFU-induced lesions. In this study, ultrasound RF signals were acquired during HIFU exposure. A correlation coefficient was used to evaluate the changes occurring in the RF signals backscattered from the thermal lesion. Also, a block matching algorithm has been implemented to compensate the tissue motion during HIFU exposure. The experimental results show that correlation coefficients in the focal spot decreased significantly with HIFU exposure, which indicates that the backscattered RF signals changed owing to tissue coagulation.

Monitoring of Lesion Induced by High-Intensity Focused Ultrasound Using Correlation Method Based on Block Matching

High-intensity focused ultrasound (HIFU) is a noninvasive therapeutic application that focuses ultrasound to the target tissue, such as a malignant tumor, and thermally coagulates it. Monitoring methods for evaluating the formation of thermal lesions induced by HIFU are required to perform safe and accurate HIFU treatment. However, the conventional ultrasonic B-mode image incurs difficulties in assessing the formation of HIFU-induced lesions. In this study, ultrasound RF signals were acquired during HIFU exposure. A correlation coefficient was used to evaluate the changes occurring in the RF signals backscattered from the thermal lesion. Also, a block matching algorithm has been implemented to compensate the tissue motion during HIFU exposure. The experimental results show that correlation coefficients in the focal spot decreased significantly with HIFU exposure, which indicates that the backscattered RF signals changed owing to tissue coagulation.

Time-domain calculation of spectral centroid from backscattered ultrasound signals

Spectral centroid estimation from backscattered ultrasound RF signals is the preliminary step for quantitative ultrasound analysis in many medical applications. The traditional approach of estimating the spectral centroid in the frequency domain takes a long time because discrete Fourier transform (DFT) processing for each RF segment is required. To avoid this, we propose time-domain methods to estimate the spectral centroid in this paper. First, we derive the continuous-time-domain equations for the spectral centroid estimation using Parseval's theorem and Hilbert transform theory. Then, we extend the method to the discrete-time domain to ease the implementation while maintaining the same accuracy as the calculation in the frequency domain. From the result, we observe that it is not practical to apply the discrete-time equations directly, because a high sampling rate is needed to approximate the time derivative in the discrete-time domain. Therefore, we also derive the feasible version of the discrete-time equations using a circular autocorrelation function, which has no constraints on the sampling rate for real RF signals acquired from pulse-echo ultrasound systems. Simulation results using numerical phantoms show that the time-domain calculation is approximately 4.4 times faster on average than the frequency-domain method when the software's built-in functions were used. The average estimation error compared with that of the frequency-domain method using DFT is less than 0.2% for the entire propagation depths. The proposed time-domain approach to estimate the spectral centroid can be easily implemented in real-time ultrasound systems.

Pulse wave imaging of the human carotid artery: an in vivo feasibility study

Noninvasive quantification of regional arterial stiffness, such as measurement of the pulse wave velocity (PWV), has been shown to be of high clinical importance. Pulse wave imaging (PWI) has been previously developed by our group to visualize the propagation of the pulse wave along the aorta and to estimate the regional PWV. The objective of this paper is to determine the feasibility of PWI in the human carotid artery in vivo. The left common carotid arteries of eight (n = 8) healthy volunteers (male, age 27 + 4 years old) were scanned in a long-axis view, with a 10-MHz linear-array transducer. The beam density of the scan was reduced to 16 beams within an imaging width of 38 mm. The frame rate of ultrasound imaging was therefore increased to 1127 Hz at an image depth of 25 mm. The RF ultrasound signals were then acquired at a sampling rate of 40 MHz and used to estimate the velocity of the arterial wall using a 1-D cross-correlationbased speckle tracking method. The sequence of the wall velocity images at different times depicts the propagation of the pulse wave in the carotid artery from the proximal to distal sides. The regional PWV was estimated from the spatiotemporal variation of the wall velocities and ranged from 4.0 to 5.2 m/s in eight (n = 8) normal subjects, in agreement with findings reported in the literature. PWI was thus proven feasible in the human carotid artery, and may be proven useful for detecting vascular disease through mapping the pulse wave and estimating the regional PWV in the carotid artery.

Bayesian regularization applied to ultrasound strain imaging.

Noise artifacts due to signal decorrelation and reverberation are a considerable problem in ultrasound strain imaging. For block-matching methods, information from neighboring matching blocks has been utilized to regularize the estimated displacements. We apply a recursive Bayesian regularization algorithm developed by Hayton et al. [Artif. Intell., vol. 114, pp. 125-156, 1999] to phase-sensitive ultrasound RF signals to improve displacement estimation. The parameter of regularization is reformulated, and its meaning examined in the context of strain imaging. Tissue-mimicking experimental phantoms and RF data incorporating finite-element models for the tissue deformation and frequency-domain ultrasound simulations are used to compute the optimal parameter with respect to nominal strain and algorithmic iterations. The optimal strain regularization parameter was found to be twice the nominal strain and did not vary significantly with algorithmic iterations. The technique demonstrates superior performance over median filtering in noise reduction at strains 5% and higher for all quantitative experiments performed. For example, the strain SNR was 11 dB higher than that obtained using a median filter at 7% strain. It has to be noted that for applied deformations lower than 1%, since signal decorrelation errors are minimal, using this approach may degrade the displacement image.

Sub-sample displacement estimation from digitized ultrasound RF signals using multi-dimensional polynomial fitting of the cross-correlation function

A widely used time-domain technique for motion or delay estimation between digitized ultrasound RF signals involves the maximization of a discrete pattern-matching function, usually the cross-correlation. To achieve sub-sample accuracy, the discrete pattern-matching function is interpolated using the values at the discrete maximizer and adjacent samples. In prior work, only 1-D fit, applied separately along the axial, lateral, and elevational axes, has been used to estimate the sub-sample motion in 1-D, 2-D, and 3-D. In this paper, we explore the use of 2-D and 3-D polynomial fitting for this purpose. We quantify the estimation error in noise-free simulations using Field II and experiments with a commercial ultrasound machine. In simulated 2-D translational motions, function fitting with quartic spline polynomials leads to maximum bias of 0.2% of the sample spacing in the axial direction and 0.4% of the sample spacing in the lateral direction, corresponding to 38 nm and 1.31 μm, respectively. The maximum standard deviations were approximately 1% of the sample spacing in both the axial and the lateral directions, corresponding to 193 nm axially and 4.43 μm laterally. In simulated 1% axial strain, the same function fitting leads to mean absolute displacement estimation errors of 255 nm in the axial direction and 4.77 μm in the lateral direction. In experiments with a linear array transducer, 2-D quartic spline fitting leads to maximum bias of 458 nm and 6.27 μm in the axial and the lateral directions, respectively. These results are more than one order of magnitude smaller than those obtained with separate 1-D fit when applied to the same data set. Simulations and experiments in 3-D yield similar results when comparing 3-D polynomial fitting with 1-D fitting along the axial, lateral, and elevational directions.

Detection of Brachytherapy Seeds Using 3-D Transrectal Ultrasound

Detection of brachytherapy seeds plays a key role in dosimetry for prostate brachytherapy. However, seed localization using B-mode transrectal ultrasound (TRUS) still remains a challenge for prostate brachytherapy, mainly due to the small size of brachytherapy seeds in the relatively low-quality B-mode TRUS images. In this paper, we propose a new solution for brachytherapy seed detection using 3-D ultrasound. A 3-D reflected power image is computed from ultrasound RF signals, instead of conventional B-mode images. Then, implanted seeds are segmented in 3-D local search spaces that are determined by a priori knowledge, e.g., needle entry points and seed placements. Needle insertion tracks are also detected locally by the Hough transform. Experimental results show that the proposed solution works well for seed localization in a prostate phantom implanted according to a realistic treatment plan with 136 seeds from 26 needles.