The benefits of compression methods in acoustic coherence tomography

Pulse compression methods improve the quality of ultrasonic medical images. In comparison with standard broadband pulse techniques, these methods enhance the contrast-to-noise ratio (CNR) and increase the probing depth without any perceptible loss of spatial resolution. The Golay compression technique is analyzed here in the context of ultrasonic computed tomography, first on a one-dimensional target and second on a very low-contrast phantom probed using a half-ring array tomograph. The imaging performances were assessed based on the image CNR. The improvement obtained (up to 40%) depends, however, on the number of coherently associated diffraction projections. Beyond a certain number, few advantages were observed. Advances in ultrasound computed tomography suggest that pulse compression methods should provide a useful means of optimizing the trade-off between the image quality and the probing sampling density.

A high-frequency ultrasound imaging system combining limited-angle spatial compounding and model-based synthetic aperture focusing

High-frequency ultrasound (HFUS) imaging systems are routinely used for medical diagnostics (skin, eyes) and for medical research (small animal imaging). Although systems with array transducers are already commercially available, imaging systems with single-element transducers are still of interest and available as well, because this type of transducer is less complex, less expensive, and technically mature. Nevertheless, drawbacks exist, for example, the need for mechanical scanning units and the limited depth of field. In this paper, we present a high-frequency (20 MHz) ultrasound imaging system equipped with a spherically focused transducer. Limited-angle spatial compounding is utilized to improve the image contrast, to suppress speckle and noise, and to reduce imaging artifacts. To overcome the limitation in depth of field, the system uses a novel synthetic aperture focusing technique based on the correlation of the recorded echo signals with the simulated point spread function of the imaging system. This method results in lower side lobe levels and greater noise reduction compared with delay-and-sum focusing, which is demonstrated by wire phantom measurements. When used in combination with limited-angle spatial compounding, as presented in this paper, the resulting image quality is superior to conventional single-element HFUS imaging systems and to array systems. Examples of measurements on tissue phantoms and small animals (ex vivo) are presented and discussed in detail.

A method to expedite data acquisition for multiple spatial-temporal analyses of tissue perfusion by contrast-enhanced ultrasound

For semiquantitative analyses of tissue perfusion using contrast-enhanced ultrasound the acquisition and processing of time intensity curves (TIC) is required. These TICs can be computed for each pixel of an image plane, yielding parametric images of classification numbers like "blood volume" and "flow rate." The expenditure of time for data acquisition and analysis typically limits semiquantitative perfusion imaging to a single image plane in 2-D. 3-D techniques, however, provide a higher diagnostic value since more information (e.g., of an entire lesion) is obtained. Moreover, spatial compounding, being a 2-D-technique where an object is imaged from different viewing angles, is known to improve image quality by reducing artifacts and speckle noise. Both techniques, 3-D and compounding, call for optimized acquisition and processing of TICs in several image planes (3-D) or in several (overlapping) sections of the same image plane (compounding) to decrease the time needed for data acquisition. Here, an approach of interleaved imaging is presented which is applicable, among others, to contrast perfusion imaging using the replenishment method. The total acquisition time is decreased by sequentially scanning image planes twice for short time spans - first, immediately after microbubble destruction to record the initial rise of the TICs, and second, a sufficient time thereafter to assess final values of the TIC. Data from both periods are combined to fit a model function from which parameters are extracted such as perfusion rate and blood volume. This approach was evaluated by in vitro measurements on a perfusion-mimicking phantom for both, individual images such as would be used for volume reconstruction in 3-D and compound images obtained from full angle spatial compounding (FASC, 360 degrees ). An error analysis is conducted to derive the deviation of the extracted parameters of the proposed method compared with the conventional one. These deviations are entailed by a reduction in acquisition time of the proposed method, which can be adjusted by several parameters, depending on the prevailing flow. Optimization strategies are proposed to find optimal values for those settings.

Limited-angle spatial compound imaging of skin with high-frequency ultrasound (20 MHz)

In dermatology, high-frequency ultrasound (HFUS) is used for high-resolution skin imaging. The conventional B-scan type approach is to perform lateral scans perpendicular to the direction of sound propagation. Ultrasound spatial compounding enables improvement of the image contrast, suppression of speckle and noise, and reduction of imaging artifacts in comparison with conventional B-mode imaging, but it has not yet found its way into HFUS skin imaging applications. In this paper, the potential of HFUS spatial compounding for skin imaging is systematically evaluated. A new HFUS system with a sophisticated scanner for limited-angle (up to +/-40 degrees) spatial compound imaging was developed and implemented. Echo signals are acquired using a 20 MHz spherically focused single-element transducer with an axial and lateral resolution of 69 mum and 165 mum, respectively, in the focus. A calibration scheme for the estimation of unknown system parameters and precise image reconstruction has been developed. The implemented system has been evaluated using measurements of geometrically well-defined structures, speckle phantoms, and in vivo measurements. The results show the advantage of the proposed spatial compound skin imaging concept compared with conventional B-mode imaging in terms of image contrast, isotropy, and independence from the orientation of surfaces.

Three-dimensional reconstruction of fine vascularity in ultrasound breast imaging using contrast-enhanced spatial compounding: in vitro analyses

RATIONALE AND OBJECTIVES

Ultrasound image quality can be improved by imaging an object (here: the female breast) from different viewing angles in one image plane. With this technique, which is commonly referred to as spatial compounding, a more isotropic resolution is achieved while speckle noise and further artifacts are reduced. We present results obtained from a combination of spatial compounding with contrast-enhanced ultrasound imaging in three dimensions to reduce contrast specific artifacts (depth dependency, shadowing, speckle) and reconstruct vascular structures.

MATERIALS AND METHODS

We used a conventional ultrasound scanner and a custom made mechanical system to rotate an ultrasound curved array probe around an object (360 degrees, 36 transducer positions). For 10 parallel image planes, ultrasound compound images were generated of a flow-mimicking phantom consecutively supplied with water and contrast agent. These compound images were combined to form a volume dataset and postprocessed to obtain a sonographic subtraction angiography.

RESULTS

Image quality was significantly improved by spatial compounding for the native (ie, without contrast agent), and, in particular, for the contrast-enhanced case. After subtracting the native images from the contrast-enhanced ones, only structures supplied with contrast agent remain. This technique yields much better results for compound images than for conventional ultrasound images because speckle noise and an anisotropic resolution affect the latter.

CONCLUSIONS

With the presented approach contrast specific artifacts can be eliminated efficiently, and a subtraction angiography can be computed. A speckle reduced three-dimensional reconstruction of submillimeter vessel structures was achieved for the first time. In the future, this technique can be applied in vivo to image the vascularity of cancer in the female breast.

Full angle spatial compounding for improved replenishment analyses in contrast perfusion imaging: in vitro studies

For contrast enhanced perfusion imaging semi-quantitative methods (such as the bolus-, replenishment- or depletion-method) are commonly used to analyze the dynamic changes in concentration of contrast agent induced by insonification. In order to apply these methods and to decrease artifacts from tissue nonlinearity, perfusion imaging is conducted using decreased transmit power. However, echo signals from deeper structures are often too weak to be successfully analyzed. Furthermore, shadowing artifacts may occur as a result of high concentration of contrast agent in the beam path. Thus, those semi-quantitative methods often fail or yield ambiguous diagnoses. Imaging an object (e.g., the female breast) from multiple viewing angles (spatial compounding) may overcome these issues. In addition, spatial compounding achieves an isotropic resolution and reduces speckle and further common artifacts. In this paper we present results obtained from a combination of spatial compounding with contrast enhanced perfusion imaging. Applying the replenishment method, we extracted perfusion-related parameters and compared the conventional parametric images with the compound parametric images. We found that the compounded parametric images outperform the conventional images due to reduced noise and suppression of artifacts.

Ultrasonic reflection tomography vs. canonical body approximation: experimental assessment of an infinite elastic cylindrical tube

Comparisons were made between the results obtained using two quantitative ultrasound imaging methods on the solid cross section of a cylindrical tube that is infinite in the axial direction. The first method tested was the classical reflection tomography method based on the first-order Born approximation, which can only be used under conditions to obtain limited reconstruction of the external boundaries of the high contrast scatterer. The results were compared with those obtained using another inversion scheme based on the Intercepting Canonical Body Approximation (ICBA) in a large frequency range, which gives accurate complete geometrical information about the tube (thickness measurements). The numerical and experimental results obtained show the feasibility of the latter approach.

Ultrasonographic contrast-agent imaging of sub-millimeter vessel structures with spatial compounding:in vitroanalyses / Kontrastmittelgestützte Ultraschallabbildung von Sub-Millimeter-Gefäßstrukturen mittels Spatial Compounding:In-vitroAnalysen

In clinical diagnostics, ultrasonographic contrast-agent imaging gives access to medical parameters such as perfusion and vascularization. In addition to the artifacts that are typical for ultrasonic imaging, e.g., speckle noise and depth-dependent sensitivity and resolution, contrast-agent imaging shows more pronounced depth dependence and may suffer from shadowing artifacts that arise from high attenuation of the ultrasound waves by the contrast agent at high concentrations. By imaging an object from different viewing angles in one 2D image plane and summing the images obtained (spatial compounding), image quality can be increased and artifacts can be suppressed. In the present study, we combined both techniques to overcome the limitations of contrast-agent imaging. We used a commercially available ultrasound scanner and a custom-made high-precision mechanical system to rotate the ultrasound transducer fully around the object under investigation. Using this set-up, ultrasound data were acquired in reflection mode to generate a 360 degrees compound scan of a flow-mimicking phantom supplied with contrast agent.

An ultrasound research interface for a clinical system

Under a contract with the National Cancer Institute, we have developed a research interface to an ultrasound system. This ultrasound research interface (URI) is an optional feature providing several basic capabilities not normally available on a clinical scanner. The URI can store high-quality beamformed radio-frequency data to file for off-line processing. Also, through an integrated user interface, the user is provided additional control over the B-mode receive aperture and color flow ensemble size. A third major capability is the ability to record and playback macro files. In this paper, we describe the URI and illustrate its use on three research examples: elastography, computed tomography, and spatial compounding.

Distorted Born diffraction tomography applied to inverting ultrasonic field scattered by noncircular infinite elastic tube

This study focuses on the application of ultrasonic diffraction tomography to noncircular 2D-cylindrical objects immersed in an infinite fluid. The distorted Born iterative method used to solve the inverse scattering problem belongs to the class of algebraic reconstruction algorithms. This method was developed to increase the order of application of the Born approximation (in the case of weakly-contrasted media) to higher orders. This yields quantitative information about the scatterer, such as the speed of sound and the attenuation. Quantitative ultrasonic imaging techniques of this kind are of great potential value in fields such as medicine, underwater acoustics and nondestructive testing.

An ultrasound research interface for a clinical system

Under a contract with the National Cancer Institute, we have developed a research interface to an ultrasound system. This ultrasound research interface (URI) is an optional feature providing several basic capabilities not normally available on a clinical scanner. The URI can store high-quality beamformed radio-frequency data to file for off-line processing. Also, through an integrated user interface, the user is provided additional control over the B-mode receive aperture and color flow ensemble size. A third major capability is the ability to record and playback macro files. In this paper, we describe the URI and illustrate its use on three research examples: elastography, computed tomography, and spatial compounding.

A New Least Squares Based Calibration for an Ultrasound Tomography Scanner Using Radial Image Processing

In the context of the development of an ultrasound scanner where a probe turns around the part of the body to be studied, a new tomographic technique has been developed: the Ultrasound Reflection-mode Tomography Using Radial Image Processing (URTURIP Technique). This technique is used when a unique B-scan image is insufficient to correctly describe a cross-section. It utilises B-scan images obtained under different angles of view in the same plane to reconstruct a better cross-sectional image. Before using the URTURIP Technique, a calibration step is required to accurately determine the centre of rotation of the probe. To solve the calibration problem, a reference method has been developed, but it is very time-consuming. This paper presents a faster method based on a least squares fitting and is compared to the reference method in terms of accuracy and time-consuming.

Cancellous and cortical bone imaging by reflected tomography

This paper deals with the inverse scattering problem observed when ultrasonic waves are used to analyze biological media. The objective is to image cancellous and cortical bone by ultrasonic reflected tomography (URT). Because strong contrast and high absorbance bodies such as bones cannot be imaged at usual ultrasonic high frequencies (> 1 MHz), we adapted for low-frequency URT (< 1 MHz) our tomographic set-up and reconstruction and acquisition tools, previously developed for weakly scattered media. Indeed, when the frequency of the transducer decreases, the penetration length of the wave increases, which unfortunately makes resolution poor, inappropriate for bone imagery. To improve resolution, we extend the generalized inversion in the complementary bandwidth of the electro-acoustic set-up (Papoulis deconvolution). This resolution enhancement for human porous vertebrae and human and animal femur showed that high-resolution images can be obtained with low-frequency URT.

An automatic method for determining the centre of rotation of a mechanically scanned UCT system

A method will be described for determining the centre of rotation of a mechanically scanned reflection ultrasound computed tomography system. It is based on the principle of obtaining opposing images of a test object containing many point targets. The method is automatic in the sense that the centre of rotation is calculated by a computer without the need for an operator to make direct measurements on the mechanical system. For the particular reflection UCT system described here, the centre of rotation is obtained in 3-5 min with a repeatability (+/-2 SD) of +/-0.3 mm. Ways in which even higher accuracy can be obtained are discussed. The basic principle of the method is applicable to any concentric imaging system for which a good approximation to an ideal point target can be produced.

Experimental study of high frame rate imaging with limited diffraction beams

Limited diffraction beams have a large depth of field and have many potential applications. Recently, a new method (Fourier method) was developed with limited diffraction beams for image construction. With the method and a single plane wave transmission, both 2D (two-dimensional) and 3D (three-dimensional) images of a very high frame rate (up to 3750 frames/s for a depth of 200 mm in biological soft tissues) and a high signal-to-noise ratio (SNR) can be constructed with relatively simple and inexpensive hardware. If limited diffraction beams of different parameters are used in both transmission and reception and transducer aperture is shaded with a cosine function, high-resolution and low-sidelobe images can be constructed with the new method without montage of multiple frames of images [the image quality is comparable to that obtained with a transmit-receive (two-way) dynamically focused imaging system]. In this paper, the Fourier method was studied with both experiment and computer simulation for 2D B-mode imaging. In the experiment, two commercial broadband 1D array transducers (48 and 64 elements) of different aperture sizes (18.288 and 38.4 mm) and center frequencies (2.25 and 2.5 MHz) were used to construct images of different viewing sizes. An ATS539 tissue-equivalent phantom of an average frequency-dependent attenuation of 0.5 dB/MHz/cm was used as a test object. To obtain high frame rate images, a single plane wave pulse (broadband) was transmitted with the arrays. Echoes received with the arrays were processed with both the Fourier and conventional dynamic focusing (delay-and-sum) methods to construct 2D B-mode images. Results show that the quality (resolution and contrast) of constructed images is virtually identical for both methods, except that the Fourier method is simpler to implement. Both methods have also a similar sensitivity to phase aberration distortions. Excellent agreement among theory, simulation, and experiment was obtained.

High resolution low frequency ultrasonic tomography

Ultrasonic reflection tomography results from a linearization of the inverse acoustic scattering problem, named the inverse Born approximation. The goal of ultrasonic reflection tomography is to obtain reflectivity images from backscattered measurements. This is a Fourier synthesis problem and the first step is to correctly cover the frequency space of the object. For this inverse problem, we use the classical algorithm of tomographic reconstruction by summation of filtered backprojections. In practice, only a limited number of views are available with our mechanical rig, typically 180, and the frequency bandwidth of the pulses is very limited, typically one octave. The resolving power of the system is them limited by the bandwidth of the pulse. Low and high frequencies can be restored by use of a deconvolution algorithm that enhances resolution. We used a deconvolution technique based on the Papoulis method. The advantage of this technique is conservation of the overall frequency information content of the signals. The enhancement procedure was tested by imaging a square aluminium rod with a cross-section less than the wavelength. In this application, the central frequency of the transducer was 250 kHz so that the central wavelength was 6 mm whereas the cross-section of the rod was 4 mm. Although the Born approximation was not theoretically valid in this case (high contrast), a good reconstruction was obtained.

Variable density linear acoustic inverse problem

The linear acoustic inverse problem is solved simultaneously for density (rho) and compressibility (kappa) using the basic ideas of diffraction tomography (DT). The key to solving this problem is to utilize frequency diversity to obtain the required independent measurements. The receivers are assumed to be in the far field of the object, and plane wave incidence is also assumed. The Born approximation is used to arrive at a relationship between the measured pressure field and two terms related to the spatial Fourier transform of the two unknowns, rho and kappa. The term involving compressibility corresponds to monopole scattering and that for density to dipole scattering. Measurements at several frequencies are used and a least squares problem is solved to reconstruct rho and kappa at the same time. It is observed that the low spatial frequencies in the spectra of rho and kappa produce inaccuracies in the results. Hence, a regularization method is devised to remove this problem. Several results are shown. Low contrast objects for which the above analysis holds are used to show that good reconstructions are obtained for both density and compressibility after regularization is applied.

Ultimate limits in ultrasonic imaging resolution

According to elementary theory, the resolution of an ultrasonic imaging system increases with the ultrasonic frequency. However, frequency is limited by frequency-dependent attenuation. For imaging at any required depth, resolution improvement beyond the limit imposed by ultrasonic frequency can be obtained by increasing the ultrasonic intensity. This is itself, however, dependent on safety considerations and the effects of nonlinearity. In homogeneous media, image resolution increases with decreasing f-number. Particularly at low f-numbers, however, tissue inhomogeneity leads to a deterioration in image quality. Inhomogeneity may also be considered in terms of phase aberration. It has been found that for a given aperture, image degradation due to phase aberration is worse at higher frequencies. Schemes have been proposed for correction of this problem, but so far model systems do not lend themselves to clinical application. Deconvolution is unsatisfactory, speed correction is impracticable and synthetic aperture scanning and holography are virtually useless in biological tissues. Ultrasound-computed tomography has had only limited success. Speckle reduction can improve target detectability, but at the expense of resolution. Time-frequency control provides a useful partial solution to the problem of resolution reduction resulting from attenuation. It is clear that improved resolution would result in significant clinical benefits. An optimisation system for aperture size and ultrasonic frequency is proposed with signal averaging for resolution enhancement of a defined object area. This would have a compact ultrasonic beam and would allow frame rate to be traded for resolution, by means of signal averaging.

Limitations of ultrasound imaging and image restoration

The definition of medical ultrasound images is strongly limited by the need for low examination frequencies which is imposed by the high attenuation of acoustic waves in tissues. The filtering effect of imaging systems is described and quantified for echography, transmission tomography and reflection tomography. Improvement of image definition is demonstrated to be the result of a numerical restoration of the received echoes implemented, in the present case, by a simplified Kalman filter. The improvement in definition obtained is emphasized on simulated data and tissue images. The comparison between the results obtained from the three techniques shows that: if only echography provides a real-time acquisition of signals, tomographic methods lead to faster processing associated with a better signal-to-noise ratio on the reconstructed images, and reflection tomography offers the best definition.