Anne E. C. M. Saris

Ultrafast vascular strain compounding using plane wave transmission

Deformations of the atherosclerotic vascular wall induced by the pulsating blood can be estimated using ultrasound strain imaging. Because these deformations indirectly provide information on mechanical plaque composition, strain imaging is a promising technique for differentiating between stable and vulnerable atherosclerotic plaques. This paper first explains 1-D radial strain estimation as applied intravascularly in coronary arteries. Next, recent methods for noninvasive vascular strain estimation in a transverse imaging plane are discussed. Finally, a compounding technique that our group recently developed is explained. This technique combines motion estimates of subsequently acquired focused ultrasound images obtained at various insonification angles. However, because the artery moves and deforms during the multi-angle acquisition, errors are introduced when compounding. Recent advances in computational power have enabled plane wave ultrasound acquisition, which allows 100 times faster image acquisition and thus might resolve the motion artifacts. In this paper the performance of strain imaging using plane wave compounding is investigated using simulations of an artery with a vulnerable plaque and experimental data of a two-layered vessel phantom. The results show that plane wave compounding outperforms 0° focused strain imaging. For the simulations, the root mean squared error reduced by 66% and 50% for radial and circumferential strain, respectively. For the experiments, the elastographic signal-to-noise and contrast-to-noise ratio (SNR(e) and CNR(e)) increased with 2.1 dB and 3.7 dB radially, and 5.6 dB and 16.2dB circumferentially. Because of the high frame rate, the plane wave compounding technique can even be further optimized and extended to 3D in future.

2-D Versus 3-D Cross-Correlation-Based Radial and Circumferential Strain Estimation Using Multiplane 2-D Ultrafast Ultrasound in a 3-D Atherosclerotic Carotid Artery Model

Three-dimensional (3-D) strain estimation might improve the detection and localization of high strain regions in the carotid artery (CA) for identification of vulnerable plaques. This paper compares 2-D versus 3-D displacement estimation in terms of radial and circumferential strain using simulated ultrasound (US) images of a patient-specific 3-D atherosclerotic CA model at the bifurcation embedded in surrounding tissue generated with ABAQUS software. Global longitudinal motion was superimposed to the model based on the literature data. A Philips L11-3 linear array transducer was simulated, which transmitted plane waves at three alternating angles at a pulse repetition rate of 10 kHz. Interframe (IF) radio-frequency US data were simulated in Field II for 191 equally spaced longitudinal positions of the internal CA. Accumulated radial and circumferential displacements were estimated using tracking of the IF displacements estimated by a two-step normalized cross-correlation method and displacement compounding. Least-squares strain estimation was performed to determine accumulated radial and circumferential strain. The performance of the 2-D and 3-D methods was compared by calculating the root-mean-squared error of the estimated strains with respect to the reference strains obtained from the model. More accurate strain images were obtained using the 3-D displacement estimation for the entire cardiac cycle. The 3-D technique clearly outperformed the 2-D technique in phases with high IF longitudinal motion. In fact, the large IF longitudinal motion rendered it impossible to accurately track the tissue and cumulate strains over the entire cardiac cycle with the 2-D technique.

A Comparison Between Compounding Techniques Using Large Beam-Steered Plane Wave Imaging for Blood Vector Velocity Imaging in a Carotid Artery Model

Conventional color Doppler imaging is limited, since it only provides velocity estimates along the ultrasound beam direction for a restricted field of view at a limited frame rate. High-frame-rate speckle tracking, using plane wave transmits, has shown potential for 2-D blood velocity estimation. However, due to the lack of focusing in transmit, image quality gets reduced, which hampers speckle tracking. Although ultrafast imaging facilitates improved clutter filtering, it still remains a major challenge in blood velocity estimation. Signal dropouts and poor velocity estimates are still present for high beam-to-flow angles and low blood flow velocities. In this paper, ultrafast plane wave imaging was combined with multiscale speckle tracking to assess the 2-D blood velocity vector in a common carotid artery (CCA) flow field. A multiangled plane wave imaging sequence was used to compare the performance of displacement compounding, coherent compounding, and compound speckle tracking. Zero-degree plane wave imaging was also evaluated. The performance of the methods was evaluated before and after clutter filtering for the large range of velocities (0-1.5 m/s) that are normally present in a healthy CCA during the cardiac cycle. An extensive simulation study was performed, based on a sophisticated model of the CCA, to investigate and evaluate the performance of the methods at different pulse repetition frequencies and signal-to-noise levels. In vivo data were acquired of a healthy carotid artery bifurcation to support the simulation results. In general, methods utilizing compounding after speckle tracking, i.e., displacement compounding and compound speckle tracking, were least affected by clutter filtering and provided the most robust and accurate estimates for the entire velocity range. Displacement compounding, which uses solely axial information to estimate the velocity vector, provided most accurate velocity estimates, although it required sufficiently high pulse repetition frequencies in high blood velocity phases and reliable estimates for all acquisition angles. When this latter requirement was not met, compound speckle tracking was most accurate, because it uses the possibility to discard angular velocity estimates corrupted by clutter filtering. Similar effects were observed for in vivo data obtained at the carotid artery bifurcation. Investigating a combination of these two compounding techniques is recommended for future research.Conventional color Doppler imaging is limited, since it only provides velocity estimates along the ultrasound beam direction for a restricted field of view at a limited frame rate. High frame rate speckle tracking, using plane wave transmits, has shown potential for 2D blood velocity estimation. However, due to the lack of focusing in transmit, image quality gets reduced, which hampers speckle tracking. Although ultrafast imaging facilitates improved clutter filtering, it still remains a major challenge in blood velocity estimation. Signal drop-outs and poor velocity estimates are still present for high beam-to-flow angles and low blood flow velocities. In this work, ultrafast plane wave imaging was combined with multi-scale speckle tracking to assess the 2D blood velocity vector in a common carotid artery (CCA) flow field. A multi-angled plane wave imaging sequence was used to compare the performance of displacement compounding, coherent compounding and compound speckle tracking. Zero-degree plane wave imaging was also evaluated. The performance of the methods was evaluated before and after clutter filtering for the large range of velocities (0 to 1.5 m/s) that are normally present in a healthy CCA during the cardiac cycle. An extensive simulation study was performed, based on a sophisticated model of the CCA, to investigate and evaluate the performance of the methods at different pulse repetition frequencies and signal-to-noise levels. In vivo data were acquired of a healthy carotid artery bifurcation to support the simulation results. In general, methods utilizing compounding after speckle tracking, i.e., displacement compounding and compound speckle tracking, were least affected by clutter filtering and provided the most robust and accurate estimates for the entire velocity range. Displacement compounding, which uses solely axial information to estimate the velocity vector, provided most accurate velocity estimates, although it required sufficiently high pulse repetition frequencies in high blood velocity phases and reliable estimates for all acquisition angles. When this latter requirement was not met, compound speckle tracking was most accurate, because it uses the possibility to discarded angular velocity estimates corrupted by clutter filtering. Similar effects were observed for in vivo data obtained at the carotid artery bifurcation. Investigating a combination of these two compounding techniques is recommended for future research.

2D versus 3D cross-correlation-based radial and circumferential strain imaging in a 3D atherosclerotic carotid artery model using ultrafast plane wave ultrasound

Three-dimensional vascular strain estimation is crucial for assessment of the location of high strain regions in the carotid artery (CA) and the identification of vulnerable plaque features. This study compares 2D vs. 3D displacement estimation in terms of radial and circumferential strain using simulated ultrasound images of a 3D atherosclerotic CA model at the bifurcation embedded in surrounding tissue. The 3D finite element model (FEM) of a patient-specific, pulsating atherosclerotic CA (pulse pressure 60 mmHg) was generated with ABAQUS FEM software. Global longitudinal motion was superimposed to the model. Radiofrequency (RF) ultrasound data were simulated in Field II by moving point scatterers (vessel wall) according to the deformation patterns of the model. A linear array transducer (fc = 9 MHz, pitch = 198 μm, 192 elements) was used which transmitted plane waves at 3 alternating angles (+19.5°, 0°, -19.5°) at a pulse repetition rate of 10 kHz. Simulations with 20 ms (systole) and 100 ms (diastole) inter-frame (IF) time were performed for 191 equally spaced (0.1 mm) longitudinal positions of the internal CA containing fatty and calcified areas. After delay-and-sum beamforming, IF axial displacements were estimated using a coarse-to-fine normalized cross-correlation method. The axial displacement at 0° was used as the vertical displacement component. Projection of the -19.5° and +19.5° axial displacements yielded the horizontal displacement component. A polar grid and the lumen center were determined in the end-diastolic frame of each longitudinal position and used to convert the tracked vertical and horizontal displacements into radial and circumferential displacements. Least squares strain estimation was performed to determine accumulated radial and circumferential strain. The performance of the 2D and 3D method was compared by calculating the root-mean-squared error (RMSE) of the estimated strains with respect to the reference strains obtained from the model. More accurate strain images were obtained using the 3D displacement estimation for the entire cardiac cycle. The 3D technique clearly outperforms the 2D technique in phases with high IF longitudinal motion.

New developments in paediatric cardiac functional ultrasound imaging

Ultrasound imaging can be used to estimate the morphology as well as the motion and deformation of tissues. If the interrogated tissue is actively deforming, this deformation is directly related to its function and quantification of this deformation is normally referred as ‘strain imaging’. Tissue can also be deformed by applying an internal or external force and the resulting, induced deformation is a function of the mechanical tissue characteristics. In combination with the load applied, these strain maps can be used to estimate or reconstruct the mechanical properties of tissue. This technique was named ‘elastography’ by Ophir et al. in 1991. Elastography can be used for atherosclerotic plaque characterisation, while the contractility of the heart or skeletal muscles can be assessed with strain imaging. Rather than using the conventional video format (DICOM) image information, radio frequency (RF)-based ultrasound methods enable estimation of the deformation at higher resolution and with higher precision than commercial methods using Doppler (tissue Doppler imaging) or video image data (2D speckle tracking methods). However, the improvement in accuracy is mainly achieved when measuring strain along the ultrasound beam direction, so it has to be considered a 1D technique. Recently, this method has been extended to multiple directions and precision further improved by using spatial compounding of data acquired at multiple beam steered angles. Using similar techniques, the blood velocity and flow can be determined. RF-based techniques are also beneficial for automated segmentation of the ventricular cavities. In this paper, new developments in different techniques of quantifying cardiac function by strain imaging, automated segmentation, and methods of performing blood flow imaging are reviewed and their application in paediatric cardiology is discussed.

Semi-3D strain imaging of an atherosclerotic carotid artery by multi-cross-sectional radial strain estimations using simulated multi-angle plane wave ultrasound

Three-dimensional vascular strain estimation is crucial for assessment of the location of high strain regions in the carotid artery. This study introduces a semi-3D radial strain imaging method which is tested in a 3D model of an atherosclerotic carotid artery. A 3D finite element model (FEM) of a patient-specific, pulsating atherosclerotic carotid artery (pulse pressure 60 mmHg) was generated with ABAQUS FEM software. Radiofrequency (RF) ultrasound data were simulated in Field II by point scatterers (≈vessel wall) moving according to the deformation patterns of the model. RF element data were simulated for a linear array transducer (fc= 9 MHz, pitch = 198 μm, 192 elements) which transmitted plane waves at 3 alternating angles (+20°, 0°, -20°) at a pulse repetition rate of 12 kHz. Simulations with 25 ms inter-frame time were performed for 25 equally spaced (0.5 mm) elevational positions of the internal carotid artery containing fatty and calcified areas. After delay-and-sum beamforming, inter-frame axial displacements were estimated using a coarse-to-fine normalized cross-correlation method. The axial displacement at 0° was used as the vertical displacement component. Projection of the -20° and +20° axial displacements yielded the horizontal displacement component. Tracking was performed to accumulate displacements for each transversal position. A polar grid and the lumen center were determined in the end-diastolic frame of each elevational position and used to convert the tracked axial and lateral displacements into radial displacements. Least squares strain estimation was performed to determine accumulated radial strain. The root-mean-squared error (RMSE) of the estimated strains was calculated with respect to the ground truth strains obtained from the model. Fair agreement between the ground truth and the estimated radial strain was observed for all volumes over the entire pressure cycle. The RMSE between the ground truth and estimated strain was 1.4% at the maximum systolic pressure and revealed a ≈-7% strain region corresponding to a fatty region and a ≈-2% strain region corresponding to a calcified region. These preliminary results show the feasibility of 3D carotid strain imaging using plane wave imaging.

Robust blood velocity estimation using point-spread-function-based beamforming and multi-step speckle tracking

Speckle tracking enables 2D blood velocity estimation; however, imaging at a high frame rate is crucial, since speckle quickly decorrelates due to high velocity gradients. Recently, ultrafast imaging has become possible, which allows imaging at very high frame rates. This improves the accuracy of high velocity blood flow estimation, but hinders the estimation of low velocities, since interframe displacements drop below the data sampling resolution. A lot of effort has already been put into the development of methods to estimate subsample displacements, but still it remains challenging. Ultrafast imaging provides the opportunity to generate ultrasound radio frequency (RF) data at any location covered by the transmit beam. In other words, the spatial sampling of the data can be reconsidered. In this work, a method is proposed which incorporates knowledge of the point spread function (PSF) of the imaging system into the data sampling grid, to improve subsample displacement estimation. By using the dimensions of the PSF to choose the data sampling grid, the peak in the cross-correlation function is captured in a standardized way, which facilitates an optimal match with a 2D cubic interpolation method. Performance of the new approach is evaluated based on simulated data from a carotid artery as well as on in vivo data from the common carotid artery of a healthy volunteer. For a large range of pulse repetition frequencies and beam-to-flow angles, simulation results showed improved performance with respect to results obtained using conventionally spaced RF and envelope data. These findings are corroborated by in vivo results, showing the most consistent flow using the proposed method.

Vector flow imaging using high frequency versus conventional frequency plane wave ultrasound

In the Western World, the high prevalence of stenotic plaques in the carotid arteries results in many strokes and transient ischemic events. In daily clinical practice, the severity of stenosis is estimated based on the maximum velocity measured before and after the stenosis using pulsed wave Doppler. However, because of the angle dependency of Doppler techniques, novel flow imaging techniques are being developed which can quantify the local true velocity vector. Ultrafast imaging combined with speckle tracking allows for 2D vector flow imaging in the carotid. This study compares the performance of 2D vector flow imaging of a high frequency (> 20 MHz) transducer and a clinically utilized (9 MHz) transducer.

4D strain imaging using single and dual probe acquisitions of a patient specific carotid bifurcation phantom

Atherosclerosis in the carotid artery (CA) elevates the risk for cerebral events. Strain imaging has demonstrated to be a technique capable of identifying plaque composition. For strain imaging of carotid cross-sections, compound techniques have been developed to solve the poor strain estimation quality perpendicular to the ultrasound beam direction. This study assesses the performance of radial and circumferential strain imaging in 4D derived from multi-plane acquisitions with a single transducer with and without compounding and using two orthogonally placed transducers.

High frequency ultrafast flow and strain imaging in the carotid bifurcation: An ex vivo phantom study

In the Western World, the prevalence of plaques in the carotid is high resulting in stroke and transient ischemic events. Quantification of the plaque size and compositions and the remaining flow and velocity profile is crucial for adequate therapy. We developed a high frequency based method for characterization of plaques and flow. A realistic bifurcation phantom with plaque based on CT scans of a patient was constructed. A realistic pulsatile flow was imposed on the phantom resembling a human flow and pressure profile. Using a Verasonics experimental echomachine with a Visualsonics M250 transducer (Fc = 20MHz), plane wave ultrasound data were acquired at 12.000 fps. A cross-correlation based coarse to fine method was applied on the beamformed data to quantify the axial and lateral strain in the arterial wall and to determine the velocity magnitude and direction of the blood. The strain and flow data were presented side by side for 80 frames covering the full pressure cycle. Strain in the non-diseased ca...