Iben Kraglund Holfort

Broadband minimum variance beamforming for ultrasound imaging

A minimum variance (MV) approach for nearfield beamforming of broadband data is proposed. The approach is implemented in the frequency domain, and it provides a set of adapted, complex apodization weights for each frequency subband. The performance of the proposed MV beamformer is tested on simulated data obtained using Field II. The method is validated using synthetic aperture data and data obtained from a plane wave emission. Data for 13 point targets and a circular cyst with a radius of 5 mm are simulated. The performance of the MV beamformer is compared with delay-and-sum (DS) using boxcar weights and Hanning weights and is quantified by the full width at half maximum (FWHM) and the peak-side-lobe level (PSL). Single emission {DS boxcar, DS Hanning, MV} provide a PSL of {-16, -36, -49} dB and a FWHM of {0.79, 1.33, 0.08} mm. Using all 128 emissions, {DS boxcar, DS Hanning, MV} provides a PSL of {-32, -49, -65} dB, and a FWHM of {0.63, 0.97, 0.08} mm. The contrast of the beamformed single emission responses of the circular cyst was calculated as {-18, -37, -40} dB. The simulations have shown that the frequency subband MV beamformer provides a significant increase in lateral resolution compared with DS, even when using considerably fewer emissions. An increase in resolution is seen when using only one single emission. Furthermore, the effect of steering vector errors is investigated. The steering vector errors are investigated by applying an error of the sound speed estimate to the ultrasound data. As the error increases, it is seen that the MV beamformer is not as robust compared with the DS beamformer with boxcar and Hanning weights. Nevertheless, it is noted that the DS does not outperform the MV beamformer. For errors of 2% and 4% of the correct value, the FWHM are {0.81, 1.25, 0.34} mm and {0.89, 1.44, 0.46} mm, respectively.

P2B-12 Minimum Variance Beamforming for High Frame-Rate Ultrasound Imaging

This paper investigates the application of adaptive beamforming in medical ultrasound imaging. A minimum variance (MV) approach for near-field beamforming of broadband data is proposed. The approach is implemented in the frequency domain, and it provides a set of adapted, complex apodization weights for each frequency sub-band. As opposed to the conventional, Delay and Sum (DS) beamformer, this approach is dependent on the specific data. The performance of the proposed MV beamformer is tested on simulated synthetic aperture (SA) ultrasound data, obtained using Field II. For the simulations, a 7 MHz, 128-element, phased array transducer with lambda/2-spacing was used. Data is obtained using a single element as the transmitting aperture and all 128 elements as the receiving aperture. A full SA sequence consisting of 128 emissions was simulated by sliding the active transmitting element across the array. Data for 13 point targets and a circular cyst with a radius of 5 mm were simulated. The performance of the MV beamformer is compared to DS using boxcar weights and Hanning weights, and is quantified by the Full Width at Half Maximum (FWHM) and the peak-side-lobe level (PSL). Single emission {DS Boxcar, DS Hanning, MV} provide a PSL of {-16,-36, -49} dB and a FWHM of {0.79,1.33, 0.08} mm = {3.59lambda, 6.05lambda, 0.36lambda}. Using all 128 emissions, {DS Boxcar, DS Hanning, MV} provide a PSL of {-32, -49,-65} dB, and a FWHM of {0.63, 0.97, 0.08} mm = {2.86lambda, 4.41lambda, 0.36lambda}. The contrast of the beamformed single emission responses of the circular cyst were calculated to { -18, -37, -40} dB. The simulations have shown that the frequency sub-band MV beamformer provides a significant increase in lateral resolution compared to DS, even when using considerably fewer emissions. An increase in resolution is seen when using only one single emission. Furthermore, it is seen that an increase of the number of emissions does not alter the FWHM. Thus, the MV beam- former introduces the possibility for high frame-rate imaging with increased resolution.

Pulse wave velocity in the carotid artery

The pulse wave velocity (PWV) in the carotid artery (CA) has been estimated based on ultrasound data collected by the experimental scanner RASMUS at DTU. Data is collected from one test subject using a frame rate (FR) of 4000 Hz. The influence of FRs is also investigated. The PWV is calculated from distension wave forms (DWF) estimated using cross-correlation. The obtained velocities give results in the area between 3-4 m/s, and the deviations between estimated PWV from two beats of a pulse are around 10%. The results indicate that the method presented is applicable for detecting the local PWV. Additional studies with data collections from several test subjects are required to determine the accuracy of the approach. Based on a spectrum analysis it appears that there is no gain from using FRs above 1000 Hz, but it is shown that FRs below 1000 Hz do not give accurate PWVs.

Plane wave medical ultrasound imaging using adaptive beamforming

In this paper, the adaptive, minimum variance (MV) beam-former is applied to medical ultrasound imaging. The significant resolution and contrast gain provided by the adaptive, minimum variance (MV) beamformer, introduces the possibility of plane wave (PW) ultrasound imaging. Data is obtained using Field II and a 7 MHz, 128-elements, linear array transducer with lambda/2-spacing. MV is compared to the conventional delay-and-sum (DS) beamformer with Boxcar and Hanning weights. Furthermore, the PW images are compared to the a conventional ultrasound image, obtained from a linear scan sequence. The four approaches, {Linear Scan, DS Boxcar, DS Hanning, MV}, have full width at half maximum of {0.82, 0.71, 1.28, 0.12} mm and peak side-lobe levels of {-40.1,-16.8,-34.4,-57.0} dB.

Adaptive receive and transmit apodization for synthetic aperture ultrasound imaging

This paper suggests a framework for utilizing adaptive, data-dependent apodization weights on both the receiving and transmitting aperture for Synthetic Aperture (SA) ultrasound imaging. The suggested approach is based on the Minimum Variance (MV) beamformer and consists of two steps. A set of uniquely designed receive apodization weights are applied to pre-summed element data forming a set of adaptively weighted images; these are in SA literature conventionally referred to as low-resolution images. The adaptive transmit apodization is obtained by applying MV across the full set of single emission images before summation. The method is investigated using simulated SA ultrasound data obtained using Field II. Data of 13 point targets distributed at depths from 40 mm to 70 mm, and a 5.5 MHz, 64-element linear array transducer have been used. The investigation has shown that the introduction of adaptive apodization weights on the transmitting aperture provides a main-lobe reduction (estimated at -30 dB) by a factor of 1.8 compared to the method using adaptive apodization weights on the receiving aperture only.

In-vivo validation of fast spectral velocity estimation techniques

Spectral Doppler is a common way to estimate blood velocities in medical ultrasound (US). The standard way of estimating spectrograms is by using Welchs method (WM). WM is dependent on a long observation window (OW) (about 100 transmissions) to produce spectrograms with sufficient spectral resolution and contrast. Two adaptive filterbank methods have been suggested to circumvent this problem: the Blood spectral Power Capon method (BPC) and the Blood Amplitude and Phase Estimation method (BAPES). Previously, simulations and flow rig experiments have indicated that the two adaptive methods can display sufficient spectral resolution for much shorter OWs than WM. The purpose of this paper is to investigate the methods on a larger population and letting a clinical expert evaluate the spectrograms. Ten volunteers were scanned over the right common carotid artery and four different approaches were used to estimate the spectrograms: WM with a Hanning window (WMhw), WM with a boxcar window (WMbw), BPC and BAPES. For each approach the window length was varied: 128, 64, 32, 16, 8, 4 and 2 emissions/estimate. Thus, from the same data set of each volunteer 28 spectrograms were produced. The artery was scanned using the experimental ultrasound scanner RASMUS and a B-K Medical 5 MHz linear array transducer with an angle of insonation not exceeding 60deg. All 280 spectrograms were then randomised and presented to a radiologist blinded for method and OW for visual evaluation: useful or not useful. WMbw and WMhw estimated less useful spectrograms compared to the adaptive methods at OW below 64. The BAPES method performed better than BPC at OW of 16 and 8. Furthermore, BAPES was the only method that estimated spectrograms equally well for OW of 16 compared to OW of 128. All four approaches failed at OW of 4 and 2. The preliminary results indicate that the OW can be reduced to 32 when using the BPC method and to 16 when using the BAPES method for spectral blood estimation. This will liberate processing time in spectral US examination and could be used to increase the frame rate of the interleaved B-mode image.

Investigation of sound speed errors in adaptive beamforming

Previous studies have shown that adaptive beam-formers provide a significant increase of resolution and contrast, when the propagation speed is known precisely. This paper demonstrates the influence of sound speed errors on two adaptive beamformers; the minimum variance (MV) beamformer and the amplitude and phase (APES) beamformer. Simulations of a single point target are carried out in Field II, and a percentage error is applied on the speed of sound. As the error increases, MV and APES provide amplitude drops of 17 dB and 3 dB on the signal strength. Two approaches to overcome this amplitude drop is proposed; diagonal loading (DL) and forward-backward (FB) averaging of the covariance matrix. The investigations show that DL provides a slightly decreased resolution and amplitude compared to FB. It is noted that APES provides more robust estimates than MV at the mere expense of a slight decrease of resolution. From the investigations, it is concluded the performance of the adaptive beamformers are not outperformed by the conventional delay-and-sum beamformer.

APES beamforming applied to medical ultrasound imaging

Recently, adaptive beamformers have been introduced to medical ultrasound imaging. The primary focus has been on the minimum variance (MV) (or Capon) beamformer. This work investigates an alternative but closely related beamformer, the Amplitude and Phase Estimation (APES) beamformer. APES offers added robustness at the expense of a slightly lower resolution. The purpose of this study was to evaluate the performance of the APES beamformer on medical imaging data, since correct amplitude estimation often is just as important as spatial resolution. In our simulations we have used a 3.5 MHz, 96 element linear transducer array. When imaging two closely spaced point targets, APES displays nearly the same resolution as the MV, and at the same time improved amplitude control. When imaging cysts in speckle, APES offers speckle statistics similar to that of the DAS, without the need for temporal averaging.

Transverse correlation: An efficient transverse flow estimator - initial results

Color flow mapping has become an important clinical tool, for diagnosing a wide range of vascular diseases. Only the velocity component along the ultrasonic beam is estimated, so to find the actual blood velocity, the beam to flow angle has to be known. Because of the unpredictable nature of vascular hemodynamics, the flow angle cannot easily be found as the angle is temporally and spatially variant. Additionally the precision of traditional methods is severely lowered for high flow angles, and they breakdown for a purely transverse flow. To overcome these problems we propose a new method for estimating the transverse velocity component. The method measures the transverse velocity component by estimating the transit time of the blood between two parallel lines beamformed in receive. The method has been investigated using simulations performed with Field II. Using 15 emissions per estimate, a standard deviation of 1.64% and a bias of 1.13% are obtained for a beam to flow angle of 90 degrees. Using the same setup a standard deviation of 2.21% and a bias of 1.07% are obtained for a beam to flow angle of 75 degrees. Using 20 emissions a standard deviation of 3.4% and a bias of 2.06% are obtained at 45 degrees. The method performs stable down to a signal-to-noise ratio of 0 dB, where a standard deviation of 5.5% and a bias of 1.2% is achieved.

P3C-3 In-Vivo Vector Velocity Imaging Using Directional Cross-Correlation

Previous investigations have shown promising results in using the directional cross-correlation method to estimate velocity vectors. The velocity vector estimate provides information on both velocity direction and magnitude. The direction is estimated by beamforming signals along directions in the range [0deg; 180deg] and identifying the direction that produces the largest correlation across emissions. An estimate of the velocity magnitude is obtained from the spatial shift between signals beamformed along the estimated direction. This paper expands these investigations to include estimations of the vector velocities of a larger region by combining the estimations along several scan lines. In combination with a B-mode image, the vector velocities are displayed as an image of the investigated region with a color indicating the magnitude, and arrows showing the direction of the flow. Using the RASMUS experimental ultrasound scanner, measurements have been carried out in a water tank using a 7 MHz transducer. A 6 mm tube contained the flow and a Danfoss, MAG 3000, magnetic flow meter measured the volume flow. The tube has a parabolic flow profile with a peak velocity of 0.29 m/s. During the experiments fixed beam-to-flow angles at {60deg, 75deg, 90deg} have been applied. The images are obtained using a pulse repetition frequency of 15 kHz, and the images contain 33 lines with 60 emissions for each line. Corresponding to the three fixed beam-to-flow angles, the angle estimates along the center scan line have a bias of {-3.9deg, -12.8deg, -18.1deg} and standard deviation of {10.0deg, 18.2deg, 32.2deg}. The estimates of the velocity magnitude have bias of {4.4%, 8.1%, -5.4%} and standard deviation of {9.7%, 14.3%, 13.4%} relative to the peak velocity. The amount of in-tube angle estimates in the range of plusmn10deg from the true angle are {74%, 77%, 66%}. In-vivo measurements are carried out on a human volunteer. These measurements include the common carotid artery and the femoral bifurcation