EA014167B1 - Method for imaging of acoustic properties by nonlinear low frequency manipulation of high frequency scattering and propagation properties (variants) and acoustic instrument for implementing said method - Google Patents

Abstract

New methods of acoustic imaging are presented that provide images with reduced reverberation noise and images of nonlinear scattering and propagation parameters of the object, and estimation methods of corrections for wave front aberrations produced by spatial variations in the acoustic propagation velocity. The methods find use in a variety of applications such as acoustic imaging of geological structures, SONAR imaging of sub-sea objects, and medical ultrasound imaging. The methods are based on processing of the received signal from transmitted dual frequency band acoustic pulse complexes with overlapping high and low frequency pulses. The high frequency pulse is used for the image reconstruction and the low frequency pulse is used to manipulate the nonlinear scattering and/or propagation properties of the high frequency pulse. A 1-st method uses the scattered signal from a single dual band pulse complex for filtering in the fast time (depth time) domain to provide a signal with suppression of reverberation noise and with 1-st harmonic sensitivity and increased spatial resolution. In other methods two or more dual band pulse complexes are transmitted where the frequency and/or the pulse and/or amplitude of the low frequency pulse vary for each transmitted pulse complex. Through filtering in the pulse number coordinate and corrections for nonlinear propagation delays and optionally also amplitudes, a linear back scattering signal with suppressed pulse reverberation noise, a nonlinear back scattering signal, and quantitative nonlinear forward propagation and scattering parameters are extracted. The reverberation suppressed signals are further useful for estimation of corrections for wave front aberrations, where approximate estimates of aberration of corrections are given. The signals are further useful with broad transmit beams for multiple parallel receive beams to increase image frame rate with 2D and 3D IMAGING. The nonlinear signal is useful for imaging of differences in object properties, micro-calcifications, in growth of fibrous tissue or foam cells, or micro gas bubbles as found with decompression or injected as ultrasound contact agent. The methods are also useful with transmission imaging for generating the measured data for tomography and diffraction tomography image reconstructions.

Description

Field of Invention

The invention relates to methods and systems for constructing images from spatial changes in the acoustic properties of an object, in particular gas bubbles and high density scattering centers in this object. The methods find application in various fields with a wide variety of objects, for example, for constructing ultrasound images of biological tissues and fluids, acoustic images of geological structures, and for sonar detection of objects in water.

Description of the prior art

Acoustic imaging is used in a variety of applications, such as obtaining medical ultrasound images of internal organs, sonar images of fish, marine animals and other objects in the sea; obtaining images of geological structures for various purposes, such as the study of archaeological excavations and observation of oil wells. For various applications, a wide range of frequencies of the emitted acoustic pulse is used: from infrasound to construct images of some geological structures to ~ 100 MHz to build ultrasound images of some biological and microscopic structures. Despite the large selection of applications and the frequency spread, the methods for constructing an image are very similar for all applications. In the patent application, the authors generally called these imaging methods as acoustic imaging, although in most applications, especially in medical applications, inaudible ultrasonic frequencies in the range of ~ 20 kHz to ~ 100 MHz are used. Since these frequencies for imaging, such as medical imaging, lie in the ultrasound domain, we will also use the term ultrasound imaging, although not limiting the methods to ultrasound frequencies and medical applications.

Despite the widespread use of acoustic images, the images currently obtained have a noise background, require significant skills for their interpretation, and also provide limited quantitative information about objects. This creates problems for differentiating the structures of objects and quantifying the properties of these objects. When acquiring sonar images in water due to reflection from the seabed, it is often, for example, difficult to distinguish objects located close to the seabed or directly on the seabed, such as fish or other marine animals or mines. Also, when obtaining geological images, it may be difficult to determine the composition of the materials of geological structures. In medical applications, it can be difficult to distinguish structures such as tumors or atherosclerotic tissues from healthy tissues. The main reasons for this are outlined below.

Spatial changes in the linear acoustic properties of an object (density and compression ratio) are the physical basis for constructing acoustic images. However, due to large changes in acoustic properties in complex structures, the following effects exist that degrade image quality:

ί) the boundaries between materials with significantly different acoustic properties can lead to such a strong reflection of the acoustic pulse that multiple reflections will have a large amplitude. Such multiple reflections are called pulse reverberation and contribute to the propagating acoustic pulse in the form of a tail, which in the acoustic image appears as noise;

ίί) changes in the acoustic velocity in complex structures of objects lead to aberration of the direct propagation of the acoustic wavefront, leading to a disruption in the focusing of the main beam maximum and an increase in the lateral maximums of the beam.

Reduced focusing of the main peak of the beam by aberration of the wavefront reduces the spatial resolution of the acoustic imaging system. Both the reverberation of the pulse and the increase in the lateral maximums of the beam by aberration of the wavefront introduce cumulative noise into the image, which, in turn, reduces the ratio of the strongest scattering center to the weakest, which can be recognized next to each other, which determines the so-called contrast image resolution. This noise is also called acoustic noise because it is created by the emitted acoustic pulse itself. An increase in the power of the emitted pulse will not, therefore, lead to an increase in the signal-to-noise ratio of this type, in contrast to the noise of electronic receivers.

For example, in echocardiography, noise from reverberation of the pulse can obscure images of the apical region of the heart, making it difficult to find the apical thrombus and reduced compression of the apical myocardium. Still, for example, when constructing an image of the carotid artery, noise from reverberation can obscure the location and establishment of the boundaries of the thrombocide. As in these examples, the noise from the reverberation of the pulse limits the ability to find weak objects and the differentiation of small differences in the contrast of images in all aspects of the construction of acoustic images.

Obtaining images at the 2nd harmonic is a method that is aimed at reducing image quality deterioration due to the reverberation of the pulse in structures located near the source of acoustic pulses, such as the wall of the human body, since the content of the second harmonic in the pulse increases with depth and, therefore very low when the momentum passes through the structure

- 1 014167 ry located close to the source, like a wall of the body. However, the sensitivity of images at the 2nd harmonic is less (about 20 dB) than the images at the 1st harmonic, which limits the maximum depth of images, especially in dense structures, such as geological structures, and in biological objects, such as the liver, kidneys, mammary glands, as well as imaging blood flow rates. To construct 30 real-time images, a wide radiated beam is desired, which is coated with numerous parallel received beams to increase the volumetric image rate. It is difficult to obtain such wide emitted beams at the 2nd harmonic due to the reduced amplitude of the first harmonic in a wide beam, which creates difficulties for constructing images at the 2nd harmonic with numerous parallel received beams used to construct real-time 3D images. This is especially true for sparse acoustic arrays, in which the number of elements creating a radiated beam is limited.

In medical applications, tissue lesions, such as tumors and arteriosclerosis of the arterial wall, affect the acoustic parameters of the tissue, such as shear modulus, volume compression coefficient, and acoustic absorption. Changes in these properties are mainly caused by the growth of foam cells, fat or fibrous molecules of connective tissues, as well as due to the segregation of calcium in the tissue. The growth of connective tissue increases acoustic absorption and shear modulus, the latter causing increased palpation stiffness, which can be observed by touching the tissue. Much work has been done to evaluate the shear modulus by using bulk ultrasonic waves to determine the displacement of shear waves in the tissue, using methods often called elastography, as well as by remote ultrasound palpation. However, to date, these methods have found limited clinical application, and there is still a need for improved differentiation of such tissue changes under the influence of ultrasound.

To date, segregated microcalcifications in breast tumors are detected by x-ray mammography, which are indicators of malignant tumors. These microcalcifications are so small that the ultrasound signal reflected from them disappears in the signal from the tissue surrounding them and they cannot be detected by existing methods for constructing ultrasound images. Thus, there is a need to improve methods for constructing ultrasound images so that such microcalcifications can also be detected. Microcalcifications in an atherosclerotic plaque also provide information on its stability, and improved methods for constructing ultrasound images of such microcalcifications are also needed here.

Some diseases also affect the blood supply to the tissue, for example, through angiogenesis or necrosis of the microvascular system in malignant tumors, or reduced blood flow due to narrowing of blood vessels or thrombosis in both the coronary arteries of the heart and peripheral vessels. The blood flow rate in the microvascular system and small vessels is so small that it cannot be determined using conventional, non-invasive ultrasound methods based on the Doppler effect. To improve the quality of ultrasound images of the microvascular system, as well as to evaluate the blood supply to the tissue, ultrasound contrast agents in the form of solutions of small microbubbles (about 3 microns in diameter) were developed. These microbubbles are introduced into the blood stream, which leads to greatly increased and non-linear scattering of ultrasound from the blood. Therefore, they greatly increase the nonlinear scattering from the tissue containing such microbubbles, and in special cases in dense tissues, individual microbubbles can be observed that create the prospect of constructing a molecular ultrasound image with special contrasting bubbles in the tissue. Such microbubbles can provide beneficial image enhancement when introduced into other body fluids, such as intercellular fluid, to monitor lymph secretion to signal lymph nodes, or in the urinary system to target the attachment of vesicles to tumor tissues, or elsewhere. During decompression in space or during immersion, microbubbles of gas often form spontaneously in the tissue, causing decompression sickness, so there is a need for early detection of such gas bubbles to improve the decompression profile, to avoid decompression sickness of personnel during such work, as well as to monitor the formation of bubbles such as early warning during such work.

During oil production from oil wells, it is often desirable to observe changes in the geological structures around the wells, for example, the amount of oil or gas in the sandstone, the boundaries between oil, gas and water, as well as structural changes near the well. The acoustic properties of structures and, in particular, the nonlinear components of acoustic properties depend on the amount of gas, oil or water in a porous rock. Acoustic images of structures surrounding an oil well can be obtained using acoustic transducers in an oil well. Using methods for constructing acoustic images that provide numerical acoustic data from the object, therefore, it is possible to determine the amount of gas, oil or water in the structures surrounding the oil well.

- 2 014167

When detecting fish, or marine animals, or other objects near the seabed, it is often difficult to distinguish between the reflected signals from the seabed and the object, especially with side beams. Commercial fish or marine animals have a gas-filled swimming bladder or lungs, the acoustic properties of which are very different from the acoustic properties of the seabed, and are non-linear. These differences in acoustic properties, together with the methods in accordance with the present invention, can be used to differentiate overlapping reflected signals from such objects and from the seabed. These methods can also be used to strengthen small solid structures, such as mines, on softer seabed or in soil, as they are used to detect microcalcifications in the tumor.

Thus, there is a need for an improved method for constructing an acoustic image in which image noise is reduced and image contrast is enhanced associated with a change in object properties.

The methods performed in accordance with the present invention reduce image noise and enhance the image contrast associated with a change in the properties of an object by emitting groups of acoustic pulses into the object in the two-band frequency range with overlapping high-frequency and low-frequency pulses. Acoustic pulses in the two-band frequency range were previously used in the construction of medical ultrasound images for various purposes, where for cardiac intervention in M-mode and Doppler (British Journal of Heart Diseases, January 1984, No. 51 (1), pp. 61-69 ) at the same time, pulses with a frequency of 3 and 1.5 MHz with fixed phase relationships between the pulses were used to construct the optimal image of the heart in the M-mode (pulse 3 MG c) and for Doppler measurement of blood speed (pulse 1.5 MHz). A transducer with a concentric ring device was used, in which a 3-MHz ultrasonic pulse in the M-mode was emitted and received by the central disk of the transducer, while a 1.5-MHz Doppler ultrasonic pulse was emitted and received by the surrounding ring element.

The use of emitted pulses in the two-band frequency range for improved detection of ultrasonic contrast medium, micro bubbles, is also described in US patent No. 5410516. This patent describes the simultaneous emission of two ultrasonic pulses with different center frequencies, while the pulses scattered from the micro bubbles contain both total and the difference frequency from the emitted frequencies obtained in nonlinear scattering from micro bubbles, i.e. these sum and difference frequencies are used to detect microbubbles.

A similar use of pulses in the two-band frequency range for detecting ultrasonic contrast media is described in US Pat. No. 6,312,383, where the phase between two frequency bands varied between emissions. This case can be considered as a special case described in US patent No. 5410516, in which the phase change of the low-frequency pulse can be considered as a beating between the low frequency and the pulse repetition rate.

However, although both of the above patents use non-linear pulse scattering in the two-band frequency range to detect contrast material in the tissue, both of them do not consider the non-linear effect of the low-frequency pulse on the direct propagation speed of the high-frequency pulse, which in practical situations limits the signal suppression from the tissue relative to the signal from the contrast medium. The patents also do not indicate how information from nonlinear scattering from ordinary tissue or other objects can be used. The accumulated non-linear effects of direct propagation create similar characteristics of signals for strong linear scattering from tissue, as well as local non-linear scattering from micro bubbles and tissues. This effect obscures local nonlinear scattering from microbubbles and other parts of the object and limits the power ratio of contrast material to tissue (CTF).

The presence of gas bubbles and microbubbles in a certain region also significantly increases the direct cumulative effect of nonlinear propagation and leads to the fact that linear scattering from objects outside this region greatly obscures the scattering from gas bubbles and microbubbles in the object. This phenomenon, for example, greatly affects the construction of images from the contrast medium in the myocardium using the pulses that pass through the ventricle of the heart with the contrast medium before entering the myocardium, and therefore can erroneously determine the blood flow in the ischemic myocardium. It also affects the differentiation between gas and oil beyond the area with gas in geological structures.

The present invention differs from the prior art in that it uses non-linear effects of the propagation speed of a high-frequency pulse caused by a low-frequency pulse, and in that an understanding of these effects is used to generate signals for constructing images based on high-frequency transmitted and scattered signals. This makes it possible to separate the accumulated nonlinear effect affecting the signals and local nonlinear parameters of the object, thereby allowing the determination of local nonlinear parameters of the object, which was not possible in the prior art. This invention also provides a method for separating the cumulative effect of acoustic absorption on signals, allowing parameters to be determined.

- 3 014167 local acoustic absorption of the object.

SUMMARY OF THE INVENTION

The methods are used in the construction of acoustic images using both backscattered signals and computer reconstruction based on measuring angular scattering and / or direct transmission.

Groups of two-way acoustic pulses with components of pulses in both low and high frequency ranges overlapping in time space are emitted in the direction of the region where the object whose image is to be built is located. In the process of generating the signal used to construct the image, nonlinear control is used by means of a low-frequency pulse of the scattering properties of the object and the propagation properties of the high-frequency pulse. The high-frequency components of the received signals are processed to construct the parameters of the image signals, and the low-frequency components of the received signals can be removed by filtering, for example, directly in the array of the receiving transducer.

The signal processing according to the invention is part of the complete processing necessary for imaging, and the additional processing necessary for forming the final image and not described in the present invention is widely known. To construct an image using backscattering, methods are used to form radial image lines, while 2Ό and 3E images are obtained by transverse scanning by a beam according to known methods. Radial lines of the images can be the envelope of the signal for structural images, Doppler measurements of radial displacement and displacement speeds of the scattering center, radial shear deformations or shear strain rates of the corresponding displacements of the scattering center or fast time (depth time) of the spectral parameters to construct the characteristics of the object. Parallel radiated and / or received beams can be used to build multiple radial image lines in parallel to accelerate frame rates. Using image reconstruction by computed tomography (CT), the methods provide improved measurements for said reconstruction with reduced pulse reverb noise and non-linear image parameters that provide complementary information.

The invention provides several methods for improved imaging with an increased number of pulses required to generate a pulse, with an additional reduction in image frame rate. The invention also provides a device for operating two or more methods and procedures for the optimal selection of methods that ensure the best operation of the device under given restrictions, such as frame rate, image quality, their combination, etc.

In the first method in accordance with the invention, the high-frequency pulse propagates against a negative spatial gradient of the low-frequency pulse oscillation, so that the decay of the high-frequency pulse has a higher propagation velocity than the pulse front. This leads to cumulative spatial compression of the high-frequency pulse as it propagates into the object, increasing the frequency and passband (i.e., shortening the length) of the high-frequency pulse, in addition to the nonlinear self-distortion of the high-frequency pulse, which leads to the appearance of harmonic components in the pulse. This increase in frequency, determined by the decrease in the pulse length, counteracts a decrease in the center frequency of the pulse due to frequency-dependent absorption in the object, thus providing a higher received center frequency than when this method is not used.

Since the amplitude of the low-frequency pulse is greatly reduced after the first reflection, numerous scattered pulses will not have the same compression (shortening) of the length caused by the nonlinear effect of the low-frequency pulse on the propagation speed of the high-frequency pulse, and due to absorption, their frequency will drop to lower frequencies than the frequencies first-order scattered pulses with the same propagation delay, and therefore can be filtered out, leading to a distinct suppression of noise caused by reverb pulse (multiple scattering), as when constructing images using the second harmonic, but with sensitivity as the 1st harmonic, providing images from a greater depth and using higher acoustic frequencies than when constructing images using the second harmonic, improving most spatial resolution. It is also easier to obtain wider emitted beams with the possibility of using more parallel received beams compared to obtaining images using the second harmonic, providing a higher frame rate for 2Ό- and especially for Ze-images. This is especially true when sparse gratings are used for the emitted beam, since it is difficult to provide sufficiently large amplitudes for adequate harmonic self-distortion of the pulse due to the limited number of grating elements. When the 2nd harmonic band of the received signal is used to construct the images, it is also interesting to take into account the nonlinear compression of the pulse, since the downward frequency shift due to absorption is neutralized, providing higher frequencies and shorter pulses at deep ranges with improved resolution. The invention also uses this type of pulse compression for observed pulses.

- 4 014167 of the shift of the object from the radiative force of the pressure pulses, for the separation of frequencies between the observed and pressure pulses. The invention also uses the location of the high-frequency pulse close to the maximum of the low-frequency pulse of the emitted pulse group to provide a higher emitted amplitude of the high-frequency pulse with a restriction in the mechanical index (MI) of the object.

In the second method in accordance with the invention, two or more groups of acoustic pulses are emitted in a two-band frequency range sequentially for each radial image line, the high-frequency pulse being placed close to the maximum or minimum of the low-frequency pulse, the frequency and / or phase and / or the amplitude of the low-frequency pulse is changed with each emitted group in order to nonlinearly control the acoustic scattering and direct propagation properties of the object for high-frequency components. Nonlinear control of the speed of direct propagation along with this method is used in the process of generating signals for imaging.

Using this method, it is possible, for example, to generate the 1st signal for imaging, equation (14), with significant suppression of the noise of the reverberation of the pulse and with the sensitivity of the 1st harmonic, to have the same advantages as a single pulse described above . The invention also offers the determination of non-linear propagation delays, which give the 1st numerical non-linear parameter of the image, equation (27), which is a non-linear parameter of direct propagation, in the form of a combination of differential and estimated delays of non-linear propagation, and also determination of the amplitude of the low-frequency pulse. The frequency of this pulse can be chosen so low (usually ~ 1 / 5-1 / 20 of the frequency of the high-frequency pulse) that differences in the absorption of acoustic power between different objects and individual objects can be neglected, and the amplitude of the low-frequency pulse can be determined by modeling or measurements in water or oil mixtures. The invention also provides a method for determining the local absorption coefficient through a combination of certain non-linear propagation delays in accordance with the second method (as well as in accordance with the 3rd and 4th methods described below), as well as the radial gradient of the non-linear propagation delays, the central frequency in the high frequency received the signal according to the first method and the radial gradient of the specified center frequency.

The reduced reverberation noise in the received signals contributes, in accordance with the invention, to the evaluation of corrections for wavefront aberration, as, for example, described in US Pat. . The invention also provides an approximate estimate of delay adjustments for wavefront aberration obtained from non-linear propagation delays, as estimates for signals from each element or subaperture signal defined in the description below.

In accordance with the invention, further in the process of generating signals for imaging, the received high-frequency signals are adjusted using estimates of non-linear propagation delays. Then you can strongly suppress the signal linearly scattered in this process from the object and get the 2nd image signal, equations (19) and (28), which is a nonlinearly scattered signal that displays the local nonlinear properties of the object at a scale smaller than the high-frequency wavelength, while the nonlinear propagation parameters show the nonlinear properties of the object on a scale greater than a pair of high-frequency wavelengths. Then the non-linear signal provides image contrast to rapid changes in the structure of the object with improved differentiation of these structures. Nonlinear scattering is especially strong at the boundaries between materials with a large difference in compliance, such as the boundaries between biological tissues and stiffer tissues, such as connective and muscular tissues, or solid materials, such as calcifications, or between parts of an object with low and high compliance, such as fat and microbubbles of gas in biological tissues, gas in porous rocks, swimming bubbles of fish or lungs of marine animals and surrounding water and tissue, thereby improving the characterization of structures p object.

Gas bubbles are often naturally present in an object, for example, in a swimming bladder of fish or lungs of a marine animal, gas bubbles in a porous rock, micro bubbles spontaneously formed in tissue during decompression, or micro bubbles introduced into the object as a contrast medium. The compression dynamics of bubbles in acoustic pressure waves is described by a differential equation showing resonant acoustic scattering with a frequency-dependent delay of the phase between the incident and scattered waves, as opposed to scattering from solids or fluids, where the change in the frequency of this phase can be practically neglected. The resonant frequency and, therefore, the phase delay of the scattered signal from the gas bubble for the high-frequency pulse is also controlled by the low-frequency pulse, in addition to the signal amplitude, which allows you to extract most of the scattered power from the gas bubbles by the second method (not only harmonic components) and significantly increases the LMS (ratio of contrast to noise) compared to existing methods. With a good definition of corrections in nonlinear propagation delays, as well as amplitude corrections, the methods in

- 5 014167 in accordance with the invention strongly suppress the signal linearly scattered from the object and significantly increase the RNS (contrast to object ratio) compared to existing methods. In contrast to prior art methods for detecting contrast media, such as acquiring harmonic images, pulse inversion, or Doppler power, higher acoustic frequencies can be used in the methods of the present invention compared to an improved bubble resonance frequency. spatial resolution. You can also use lower pulse amplitudes (lower mechanical index (MI)), as a result of which the bubbles, which are a medical contrast medium, are not destroyed. This is important when imaging certain microbubbles defined in the tissue, where only a limited group of microbubbles adheres to selected tissues, for example, tumor tissues, atherosclerotic plaques, blood clots, etc., where it is important to obtain an image of the vesicles without destroying them.

A cloud of microbubbles in biological tissue, fluids, or porous rocks has a strong nonlinear effect on the propagation velocity of a pulse passing through it, and in these cases it is extremely important to provide corrections for nonlinear propagation delay in order to well suppress the linearly scattered signal from an object outside the cloud of microbubbles. With these delay adjustments, the invention provides a separation between the accumulated non-linear forward propagation delay and local non-linear scattering, in contrast to other methods, such as harmonic image acquisition or image acquisition using pulse inversion, and provides a huge advantage for signal suppression for constructing an image from an object when an image is obtained from gas bubbles outside the cloud, for example, in a peripheral myocardium when triplets medical imaging, detection gas is a gas field in constructing the geologic imaging or detection of a fish or marine animals in the flock is fish or marine animals. If in this case no corrections are made for nonlinear propagation effects or these corrections are limited, then the linearly scattered signal from objects in areas where there are no gas bubbles will have the same properties as the signal scattered from gas bubbles, masking the detection of bubbles gas in these areas. This may erroneously indicate blood flow in the ischemic region of the myocardium, gas in geological structures, and fish or other marine animals in the water.

In addition to the numerous medical applications of microbubble images as an ultrasonic contrast medium, obtaining microbubble images in accordance with this method in cases of decompression encountered in space and diving can be used to observe the formation of such bubbles in order to study and develop decompression profiles, and also to have an early warning about the occurrence of decompression sickness.

In further processing of the data according to the second method, delay-corrected high-frequency signals are combined along the pulse number coordinate to obtain the 3rd image signal, equations (17) and (29), which is a linearly scattered signal. The linearly scattered signal has the same attenuation due to power absorption as the 2nd image signal, which is a nonlinearly scattered signal. By combining nonlinearly and linearly scattered signals and determining the amplitude of the low-frequency pulse, as shown above, the invention provides a second numerical nonlinear image parameter, equation (30), which is a numerical nonlinear scattering parameter. This 2nd non-linear numerical parameter represents spatial fluctuations of the non-linear parameters of the object at a scale approximately smaller than the wavelength in the high-frequency band, while the 1st numerical non-linear parameter, equation (27), is the space-averaged non-linear parameters of the object in a scale approximately larger than the wavelength in the high frequency band. Thus, backscattering and forward propagation are, therefore, two different numerical image parameters that can be observed for more information about the characteristics of the object. Numerical nonlinear parameters thereby improve the differentiation of the structure of the object, and also contribute to the characterization of the object using this method, which is very useful in the diagnosis of tumors and atherosclerotic plaques in medical applications, in obtaining information about gas, oil or water in geological structures, as well as in obtaining information on the number and size of fish and marine animals in the water. Calibration of changes in these numerical parameters with temperature also helps to evaluate local temperature using ultrasound, for example, to select hyperthermic or hypothermic treatment of tumors in medicine. It also provides new methods for quantifying the volume of contrast medium in tissue, blood flow through tissue, and the relative volumes of gas and oil in geological structures.

In the presence of stationary, temporarily stationary objects, for example, two pulses can be emitted with different frequencies and / or phases and / or amplitudes of low-frequency components, then the scattered or emitted signals from these pulses can be combined to suppress the reverberation of pulses and the nonlinear parameters of the object can be determined. When an object and an acoustic probe move relative to each other, it is preferable to emit more than two pulses for each radial

- 6 014167 image to adequately suppress a linearly scattered signal or to suppress the pulse reverberation noise associated with multiple pulses. For example, you can emit a series of K-pulses having the same phase of high-frequency components, but with different frequencies, and / or phases, and / or amplitudes of low-frequency components for each pulse. The backscattered signals from these pulses are combined in a high-pass filter with pulse-by-pulse accumulation, in which pulse reverberations are suppressed and first-order components of the scattered signal pass. Using estimates and corrections for nonlinear propagation delays, as well as desirable adjustments for the amplitude performed in front of the high-pass filter, one can distinguish a local nonlinearly scattered signal from an object or a scattered signal from gas bubbles, as well as numerical nonlinear parameters of propagation and scattering of an object.

In this second method, the pulse reverberation noise (and to a small degree the non-linear signal components themselves) introduce errors in the correction of non-linear propagation delays. These errors limit the suppression of a linearly scattered signal when evaluating a non-linearly scattered signal. To effectively eliminate the effect of pulse reverberation noise on the estimation of non-linear delay corrections, the 2nd harmonic component of the scattered signals with the second method in accordance with the invention can be used, or at least three pulses with different frequencies can be emitted in accordance with the 3rd method of the invention, and / or phases and / or amplitudes of a low-frequency pulse, as described with reference to equations (42) to (44).

This third method also exhibits the effect of nonlinear scattering on the determination of nonlinear delay adjustments.

In the fourth method in accordance with the invention described with reference to equations (45) to (48), at least four pulses with different frequencies and / or phases and / or amplitudes of a low-frequency pulse can be emitted, allowing for the determination of non-linear delay corrections, linearly scattered signal and nonlinear scattered signal with minimal interference from each other and from reverberation noise.

You can also use the 2nd harmonic band (or any harmonic band) of the received signal to form an image according to any of the second, third and fourth methods instead of using the first harmonic band of the received signal, as described above. An advantage of this is the additional suppression of the noise of the reverberation of the pulse both in the signals for imaging and in the estimation of the nonlinear propagation delay, as described above. The disadvantage is lower sensitivity at deeper ranges, thereby reducing the image depth for the same image frequency.

Using electronic control of the beam direction, the same beam direction is usually used and the focus is emitted for all radiated pulses for each radial image line and depth range, and the received signals are combined to suppress signals linearly scattered from the object for this image line. Typical filter schemes used include FIR filters or filters with a variable pulse response time, such as orthogonal decomposition using, for example, Legendre polynomials with filtering performed along the pulse coordinate number for each depth.

Using mechanical control of the beam direction as with an annular array or obtaining 3E images, pulses are usually emitted with a change in the frequency and / or phase and / or amplitude of the low-frequency pulse with a continuous sweep in the direction of the beam, supplying a signal for each depth to the high-pass filter along the pulse coordinate number. The outputs of the high-pass filters are then stored for each depth and radial line of the image to evaluate the signals and image parameters that will be used to reconstruct the image along the radial line of the image in this depth range.

The invention also provides basic designs for imaging devices that operate in accordance with the methods of the invention. Since the number of pulses for each radial image line, together with the image quality and information increases with increasing order of the method, the frame rate decreases with increasing order of the method. In the most advanced model, the device can operate using more than one method with procedures for the optimal choice of methods for constructing the image with the best efficiency, taking into account the given restrictions. Typical limitations are the minimum frame rate, minimum image quality requirements, etc.

Finally, the invention provides a procedure for constructing converter arrays, which minimizes the nonlinear effect of the low frequency pulse on the propagation delay of the high frequency pulse. At small amplitudes (~ 50 kPa) of the components of the low-frequency pulse, such transducer gratings provide images of medical ultrasound contrast agents or microbubbles during decompression with limited but still interesting suppression of linearly scattered signals from the object, without adjusting the non-linear propagation delays of the high-frequency pulse introduced by the low-frequency impulse.

- 7 014167

Brief Description of the Drawings

FIG. 1 shows a first type of emitted pulses in accordance with the invention, this first type containing both a low-frequency and a high-frequency pulse, wherein the high-frequency pulse is located on the spatial gradient of the low-frequency pulse.

FIG. 2 shows how bandpass filtering is applied in depth to the received signal from the pulse shown in FIG. 1, can be used to significantly suppress pulse reverb noise.

FIG. 3 depicts a second type of emitted pulses in accordance with the invention, containing both a low-frequency pulse and a high-frequency pulse, in which the high-frequency pulse, for example, is located at the maximum of either the positive or negative period of the low-frequency pulse.

FIG. 4 depicts the forward propagation delays of a high-frequency pulse that are created by the low-frequency pulse depicted in FIG. 3.

FIG. 5 depicts a series of received high-frequency signals from successive radiated pulses as a function of fast time (depth) and slow time (pulse coordinate number).

FIG. 6 shows received linear and non-linear frequency lines along the frequency coordinate of the slow time.

FIG. 7a and 7b show how pulse reverberations experience less control of non-linear propagation through a low-frequency pulse than scattered first-order signals.

FIG. 7c shows a gain in signal processing depending on depth obtained by the method of suppressing reverberation of a pulse in a signal.

FIG. 8 shows a transducer array assembly for emitting low-frequency and high-frequency components.

FIG. 9 depicts the main array of transducers and the principle of creating a device for simultaneously measuring transmittance and angular scattering in an object.

FIG. 10 shows a block diagram of an estimator for signals and image parameters that can be obtained by this method.

FIG. 11 shows yet another block diagram of an estimator for signals and image parameters that can be obtained by this method.

FIG. 12 shows a block diagram of a block of an apparatus for imaging scattered signals in accordance with the invention.

FIG. 13 shows a block diagram of a block of a device for reconstructing a tomographic image from transmission and angular dispersion measurements in accordance with the invention.

FIG. 14 shows the location of acoustic transducers in an oil well for monitoring geological structures around an oil well.

Description of embodiments of the invention

Acoustic body waves in homogeneous materials in a linear mode obey the linear wave equation, where the velocity c _{0 of the} propagation of a body wave is determined by the density p ₀ and the volume compression coefficient to _{0 of a} homogeneous propagation medium. The volumetric compression coefficient in the linear approximation of volumetric elasticity is determined through the relative volume of compression of the material, as - = - & ψ = k _о p (1) where δν is the relative compression volume of a small volume of DE subjected to pressure p;

Ψ is the displacement of the particle in the material, so -5Ψ is the relative compression of the volume.

The scattering of acoustic body waves from objects is carried out by spatial fluctuations of the compression coefficient and density of the object. We denote the spatially varying density and compression coefficient for the low pressure amplitudes, respectively, as p ₀ (g) and k ₀ (g), where r is the spatial coordinate. Then, the linear coefficient of backscatter from a local point z is equal to ^ (H, _{L, L)} -. Ρ (4ΰ _{+ m} where p _0a (g) and _0a (g) - spatially averaged density and coefficient of volumetric compression in scale approximately a pair of wavelengths λ of acoustic pulse;

C is a unit vector in the direction of the incident wave;

c, is the unit vector in the direction of observation of the scattered wave, as shown in FIG. nine.

The scalar product is e ^ ₈ = С05у ₁₈ , where γ ₁₈ is the angle between the direction of the incident wave and the direction of the scattered wave. For backscattering, γ ,, = π and e _| s, = - 1. The wave number of the incident wave is k = w / s = 2pD, where ω is the angular frequency; c is the acoustic propagation velocity. The linear backscattered signal at a point r from a pressure wave with an amplitude p ₁ (g, ω) at an angular frequency ω is proportional

- 8 014167 flax to ² o0 (g) p1 (g, ω). Images are usually obtained using an emitted beam with a central frequency Οι and a Bb band, where the image signal is a filter that has passed through a filter band with a variant of ² ° ⁰ (g) in the g coordinates of the range of about 2k ₁ = 2o ₁ / s and a band of 2B ₁ / s. Since a softer material (a larger volume compression coefficient) usually has a lower density, the terms with the volume compression coefficient and density in equation (2) usually have opposite signs, so that for backscattering these quantities are added arithmetically, and the term with the volume compression coefficient exceeds backscattering member with a density of about 2.5 times.

Spatial changes in the average density and volume compression coefficient, p _0a (g) and to _Oa (g), lead to changes in the acoustic propagation velocity, i.e.

C0a (g) = (3)

This spatial change in the propagation velocity is responsible for wavefront aberrations, in particular those found in biological images in the body wall, but also in all objects, like the mammary gland and lymph glands, in which there is an area of fat, or connective tissue. Large changes in the speed of propagation in geological images are often found throughout the image.

Typical values of the volume compression coefficient for soft tissues and liquids are κ ₀ ~ 400 · 10 ¹² Pa ^-1 with a typical amplitude of the acoustic pulse p ~ 10 ⁶ Pa, which gives δν / Δν ~ 0.4 · 10 ^-3 . The compression of the volume leads to an increase in density as δρ / Λο ₀ = κ _0ρ ~ 0.4 · 10 ^-3 . Rocks have a lower compression ratio, while water and petroleum products have a compression coefficient similar to fabrics. Similarly, when an object is compressed, the volume compression coefficient decreases, which, taking into account the acoustic absorption in the object, modifies equation (1) [1]:

δν —— = -Θψ = (1 - β _η κ _ο ρ) κ _ο ρ + Λ ® κ _ο ρ (4) where β _η = (1 + Β / 2Α) ~ 5 is a nonlinear parameter corresponding to the traditionally used parameters B and A for a nonlinear elastic modulus [1].

The time convolution between the wavefront pressure and k is the frequency-dependent absorption of acoustic power in the material. The first term describes a non-linear coefficient of volume compression, depending on pressure, while the derivative of this term with respect to pressure at standard pressure p ₀ gives

ZLU «= du = (1 - 2β _η κ ₀ ρο) κ ₀ (5) which gives a relative change in the compression coefficient with pressure, since δκ _η / κ ₀ = -2β _η κ ₀ ρ ₀ .

Nonlinear changes in the volume compression coefficient are therefore 2β _η ~ 10 times larger than nonlinear changes in density created by pressure, where for p ~ 1 we get δκη / κ0 = -2βηκρ0 ~ 4 · 10 ^-3 .

A non-linear change in density and compression ratio leads to a non-linear change in both the scattering speed and the forward wave propagation velocity, and the invention uses these effects to reduce noise associated with pulse reverb, increase image contrast for various object structures, microcalcifications and gas bubbles and to create numerical acoustic parameters of the image of the object, microcalcifications, gas bubbles and regions filled with gas. Illustrative embodiments of the invention are described below with reference to the drawings.

The first method in accordance with the invention uses the compression and expansion in time of a high-frequency pulse by a low-frequency pulse to control the center frequency and bandwidth of the pulse propagating directly in the object. This compression over time is created due to the pressure dependence of the velocity of direct propagation, which can be approximately expressed as

where β _ηα and κ _0α are local spatially averaged values over a pair of wavelengths at zero pressure, as was determined above.

To further explain this principle, refer to FIG. 1a, which shows a transmitted pulse consisting of a low-frequency component 101 with amplitude p0 and added high frequency component 102 with amplitude p _s and the high-frequency component rides on a negative spatial gradient of the low frequency pulse to the high frequency pulse center located, e.g., near baseband pulse scratch . A high-frequency pulse is used to build an image, and a low-frequency pulse is removed by filtering in the receiver, for example in the receiver converter itself, or as described below.

- 9 014167

The pressure-dependent propagation velocity causes an accumulation-increasing distortion of the direct propagation of the pulse, determined by the actual pulse pressure, which is the sum of the pressures of the low-frequency pulse and the high-frequency pulse, which at a distance r of the propagation creates a distorted low-frequency pulse 103 with a distorted high-frequency pulse 104. The distortion of the high-frequency pulse can be divided by the compression of the pulse length of the zero points of the high-frequency pulse created by the local yes the appearance of a low-frequency pulse, and self-distortion of the pulse shape, caused by the instantaneous value of the high-frequency pressure. Item number 105 in FIG. 1b shows a time-squeezed distorted pulse (dashed line), while for comparison, an undistorted high-frequency pulse 106 (dotted line) is shown, while the addition of self-distortion gives a completely distorted pulse 107 (solid line). The compression of the pulse occurs due to the fact that a higher pressure of the low-frequency pulse on the slice of the high-frequency pulse leads to a higher value of the propagation velocity of the slice of the pulse compared to the speed of propagation with a lower value of the low-frequency pressure in front of the high-frequency pulse. This compression of the pulse leads to an increase in the center frequency and bandwidth of the high-frequency pulse, while distortion in the shape of the pulse introduces harmonic components of the fundamental frequency band of the high-frequency pulse. Both of them are used in the first method in accordance with the invention. Distortion of the propagation of the low-frequency pulse introduces the harmonic components of the emitted low-frequency band.

This non-linear distortion of the direct propagation of the pulse is the same effect that creates harmonic components in the directly propagating pulse, which is linearly backscattered from the object and used in harmonic imaging of objects, described below with respect to equations (10) - (14) and FIG. . 9. The amplitude of the harmonic components in the pulse first increases with the propagation range, then decays with the propagation range due to the absorption of the acoustic power of the high-frequency pulse and the beam divergence. The low-frequency band can be chosen so narrow (~ 1 / 5-1 / 20 of the high-frequency bandwidth) that the absorption of the low-frequency pulse can almost be neglected for the actual image ranges. The decrease in the amplitude of the low-frequency pulse is thereby mainly determined by the beam divergence, which can be limited by choosing the lattice design; the effect of the nonlinear propagation of the low-frequency pulse on the high-frequency pulse by means of the grating design can thereby be made sufficiently large over the entire image range, as well as in other cases in accordance with the invention, as, for example, shown in FIG. 3. All this provides increased sensitivity at deep ranges using the methods performed in accordance with the invention, in comparison with obtaining images using the second harmonic, a phenomenon to which we will return in connection with equation (14) and FIG. 7th. In some cases, the use of a second harmonic band of the received high-frequency signals together with the methods according to the invention may have a definite advantage, since it is used for improved suppression of pulse reverb noise in images and for determining image parameters, as described below. To clarify the essential points, we first analyze the case when there is no beam divergence (i.e. plane waves) and there is no ultrasound absorption. The pressure difference of the low-frequency pulse at a wavelength of λ! a high-frequency pulse is given by the expression Dr ₀ = X ₁ 5p ₀ / dg. The gradient creates a difference in the propagation speed at a high frequency wavelength equal to Ac, _| = -in _{1 |} , _| to _01. , with ₀ , _| Ar ₀ . which is the reason for the compression of the wave with a wavelength of g propagation. It should be noted that the same Dc _a leads to an increase in the low-frequency pressure gradient with r in proportion to the decrease in the high-frequency wavelength, so that the pressure drop along the compressed high-frequency wavelength is maintained equal to Dr ₀ , as shown in FIG. 1a. After the time 1 = g / s _{0a of the} propagation, we thus obtain a compression of the high-frequency wavelength Δλ ^ Δο ^ ΓΔ ^ / ο ^ -β ^^ Δρ ^. In the absence of absorption and beam divergence, we obtain Dr ₀ ^ 1dr ₀ / dg = p ₀₀ k ₀ D ₁₀ = 2pr ₀ D ₁₀ A ₀₀ , where p ₀₀ is the amplitude, κ ₀₀ = 2π / λ ₀₀ is the wave number and λ ₀₀ is the length waves of the emitted low-frequency pulse on the surface of the lattice to distortion of propagation. After the propagation time 1 = g / s _0a, we thus obtain the compression of the high-frequency wavelength Δλ ₁ = -β _η κ _0ί1 Δρ ₀ ι ^, = -β _Ι1 κ _0ί1 ρ ₀₀ Ε ₀ ολ ₁ οΓ, which gives an increase in compression by high frequency equal to

Compression / expansion is accompanied by a proportional change in the pulse bandwidth (approximately the inverse of the pulse length). In accordance with these formulas, the frequency becomes infinite for x _{4 |} = 1 / in _1a to _0a p ₀₀ to _{0 ()} - a phenomenon called acoustic shaking, in which the negative gradient of low-frequency oscillations becomes infinite. In reality, the absorption of acoustic power prevents this situation.

- 10 014167

Acoustic absorption is also responsible for sliding down the frequency of the center frequency of the high-frequency pulse while maintaining the pulse bandwidth. An increase in the frequency due to compression is observed only for the outgoing pulse, in which the amplitude of the low-frequency pulse is large enough, while a slide down in frequency due to absorption is observed for both the outgoing and the scattered pulse, in which the propagation length is obtained for backscattering, equal to 2g. For a Gaussian pulse envelope, sliding downward in frequency is given by

Δ / _{1α -} 0, 3ba <r (8) where B1 is the 6 dB bandwidth of the high-frequency acoustic pulse; α = / 8.686 () ^-1 .

It should be noted that by constructing an image using transmission computed tomography, which is described further in connection with FIG. 9, we obtain only one-sided propagation, in which glide down in frequency due to absorption is equal to half the value given by equation (8).

A typical imaging range of K. for imaging using backscattering is limited by acoustic absorption, which increases linearly with frequency. The distance K, therefore, correlates with the high-frequency acoustic wavelength λ ₁ = ο _0α / _{ί 1} as Κ ~ 200λ ₁ -300λ ₁ . For p = ₀ to 1 MPa, β _η = 5, k = _0a 400-10 ^-12 Pa ^-1, Κ = 250λ ₁ and λ ₁₀ / λ ₀₀ = 10 we obtain Δί _1ρ = 0,458ί ₁₀ that ί ₁₀ = 10 gives Δί _1ρ = 4.58. Acoustic shake found for ζ _4ι / λ | ₀ = 796. which is 2.65 times larger than the maximum range of image construction. For B ₁₀ = 5, we obtain a slide down in frequency due to the absorption of the high-frequency center frequency equal to Δί _1α = 4, in accordance with equation (8), and the resulting slip compensates for the increase in frequency due to compression, so that the accepted frequency for backscattering of the first order in the absorbing medium approximates ί ₁₀ = 10 for all depths, i.e. like radiated.

Since the compression of the pulse is created by a low-frequency pulse, where, as described above, for real ranges of imaging, power absorption can be neglected, the compression of the pulse length and the corresponding band broadening are practically independent of absorption for all real ranges of imaging. Gliding down the center frequency of the high-frequency pulse due to absorption occurs, however, due to the absorption of the high-frequency pulse. Sliding down in frequency is proportional to the square of the absolute width of the pulse band B1, and sliding down in frequency due to absorption keeps the bandwidth unchanged. Therefore, in the example above, the combined effect of non-linear compression of the pulse and down-conversion due to absorption is a pulse with an approximately constant center frequency that supports lateral resolution (beam width) but with a bandwidth (approximately the opposite pulse length), which increases with depth, which improves the resolution of the range with depth. Note that from equation (8) it follows that an increase in B ₁ with depth leads to an increasing sliding downward in frequency with depth due to absorption. Let us return to a more detailed analysis of this case when considering equations (35) and (36).

The up-conversion of the high-frequency pulse associated with direct propagation can be used to improve image resolution at deeper ranges. It can also be used to improve penetration with better resolution at deep ranges where, for example, a sufficiently low frequency can be emitted, which is increased to a higher frequency at deeper ranges by means of a low frequency pulse, thereby reducing the total absorption along the pulse path for the obtained frequency pulse in the high frequency band at deep ranges. In this aspect, phase slip of the low-frequency and high-frequency pulses with the propagation distance can also be used, with special solutions of the low-frequency beam profile relative to the high-frequency beam profile, as described below in connection with FIG. 8. This slip provides, for example, the possibility that a high-frequency pulse located on a negative spatial gradient of low-frequency oscillations in the near / mid ranges will slide to zero or even to a positive spatial gradient of low-frequency oscillations at a far range. This reduces the compression of the pulse to almost zero or even stretches the high-frequency pulse by exposing it to a low-frequency pulse at deep ranges, which (together with slip due to absorption) reduces the center frequency and bandwidth of the pulse in the high-frequency band for deep ranges with increased penetration.

The amplitude of the low-frequency pulse decreases strongly at the first reflection, and the influence of the low-frequency pulse on the compression / expansion of the high-frequency pulse can be practically neglected after the first reflection, as described in relation to FIG. 7 further, while sliding downward in the frequency of the high-frequency center frequency due to absorption prevails for the entire propagation distance of a multiply scattered pulse. Therefore, the frequency distance between the first-order scattered signal (single scattering) and the pulse reverberation noise (multiple scattering) in the image can increase with depth, as shown in FIG. 2. In this drawing, the number

- 11 014167 of position 201 shows the expected change in the center frequency of the first harmonic component of the backscattered signal of the 1st order, while the bandwidth B ₁ (g) of the signal increases with depth, as shown by the boundary lines 202, due to compression of the high-frequency pulse by a low-frequency pulse. The central frequency of the pulse reverberation noise due to absorption decreases with depth, which is depicted by the line G _ha , (g). marked with the position number 203 in this drawing. The most powerful sources that produce pulse reverberation noise are usually close to the acoustic source, creating a bandwidth In _gay (g) of the pulse reverb noise, which is almost the same value as the bandwidth of the emitted high-frequency pulse, as illustrated by the limit lines 204. The second harmonic band of the first-order backscattered signal due to self-distortion of the directly propagating pulse (104) is shown with a central frequency of 2Г1 (g) position number 205 and a bandwidth of В ₂ (g)> В1 (g) limit lines 206. The amplitude of the second harmonic component first increases with depth, and then decreases due to the absorption and divergence of the beam of the first harmonic high-frequency component.

The first image signal with a strong suppression of pulse reverb noise in accordance with the invention can be obtained using a receiver filter that suppresses lower frequencies with a cutoff frequency that slides with a range depth. In FIG. 2 illustrates by way of example a receiver bandpass filter with a moving center frequency ге _ge (g), depicted by line 207, with a bandwidth of _{ге ge} (g), which can vary with depth, as shown by boundary lines 208 in this figure. In the near and middle ranges, the frequency difference between the noise of the pulse reverb and the first harmonic band of the scattered signal of the first order is so small that it is impossible to separate the two components in the frequency domain. However, the 2nd harmonic band of the scattered signal of the 1st order increases rapidly in amplitude and has low reverberation noise, so that setting the frequency of the receiver filter near the second harmonic band for small and medium depths, as shown in the drawing, gives a received signal with strong noise reduction pulse reverb at these depths. The first image depth is usually the distance to the object where the second harmonic components of the emitted high-frequency pulse are detected. Otherwise, the receiver filter may skip the first harmonic component at very shallow depths and go to the second harmonic component in the near and middle image ranges. With increasing ζ, the frequency separation between the first harmonic component of the scattered signal of the first order and the pulse reverberation noise increases, while the receiver filter can be slid in frequency and its bandwidth can be increased, as shown in the drawing, to include in the frequency of the received signal components of the first harmonic band of the scattered signal of the 1st order. This provides a greater amplitude of the received signal, since the first harmonic components are more powerful and less attenuated with depth than the second harmonic components, which also supports the sensitivity of the imaging system for greater depths. The value of _Ge can increase with depth, since the width of the first harmonic pulse band increases due to the described compression of the pulse length, and in order to include both the first and second harmonic components to increase the signal power, a reduction of the bandwidth to reduce the receiver noise can be provided for deep ranges. Also in converter arrays in which a high-frequency pulse slides from a negative to a positive spatial gradient of a low-frequency pulse, as described above, i.e. from compression of the pulse to its expansion, the width of the pulse band decreases, which can be comparable with a reduced value In _ge . It should be noted that the pulse reverberation noise lies in the low-frequency range, so instead of a band-pass filter, you can use a receiver high-pass filter in which the cutoff frequency slides with ζ to include most of the first harmonic band in the signal, supporting strong suppression of the pulse reverb noise.

Using this method, one can thus preserve the sensitivity of the first harmonic for deep ranges while simultaneously suppressing the noise of the reverberation of the pulse, as when constructing images with the second harmonic, providing deeper images of dense objects such as the liver, kidneys, breasts, etc. p., with higher frequencies and better resolution. Reducing the length of the high-frequency pulse due to compression also improves the resolution of the range in the image, which to some extent has not been previously observed. When solving the beam profile described above, in which the phase between the low-frequency and high-frequency pulses slides with depth so that the high-frequency pulse expands at great depths, it is still possible to obtain sufficiently separated pulse reverberation noise and first-order scattering using a suitable solution in the frequency space so that the noise of the reverberation of the pulse can be largely suppressed using the receiver filter. This modification of this method provides a deeper penetration, while maintaining high frequencies for better resolution in the middle and near areas. It should be noted, however, that for some applications, the advantage is the use of a second harmonic band of the received signal over the entire image range in order to obtain the best pulse noise reverberation noise reduction, as well as the best spatial resolution of the image.

The advantage of pulse compression is thus improved range resolution at deep ranges due to the compression of the pulse length, as well as increasing second harmonic frequencies at deep ranges, resulting in narrower beams and better lateral resolution. The improved separation between the second harmonic band and the noise reverb noise band at deep ranges also improves the suppression of pulse reverb noise compared to standard second harmonic imaging.

The received high-frequency signal is isolated from the total scattered / radiated signal (also including low-frequency components) by filtering, for example, directly in the receiver converter. However, the distortion of the nonlinear propagation of the low-frequency pulse component, as shown in FIG. 1a creates harmonic components of the low-frequency band and with a small separation between the low-frequency and high-frequency bands, and these harmonic components can generate energy from the low-frequency pulse in the high-frequency band. Such harmonic components can be removed by emitting a low frequency pulse without a high frequency pulse and placing the received signal in the high frequency band. The received signal is then subtracted from the received high frequency pulse signals present in the emitted pulse group. This procedure can also be used with other methods in accordance with this invention to reduce the energy received from the emitted low-frequency pulse in the high-frequency band.

By placing a high-frequency pulse on the positive peak of the low-frequency pulse of the emitted group, it is possible to reduce the mechanical index (MI) of the high-frequency pulse (due to the lower negative amplitude of the general group of pulses), which allows radiating high-frequency signals with large amplitudes. This leads to greater harmonic distortion of the high-frequency pulse, increases the sensitivity of the harmonic image with a high-frequency pulse, since the receiver filter described above is made to select the harmonic bands of the high-frequency pulse to suppress the noise of the reverberation of the pulse.

It should be obvious that one direction from each direction of the beam can emit more than one group of pulses shown in FIG. 1, followed by processing the received signal from each pulse, as in FIG. 2, where these processed received signals are used with well-known further processing to create signals for image construction, such as, for example, structural anatomical images, Doppler images of the speed of moving scattering centers and all signals that can be obtained from them, as well as frequency analysis using depth / time to characterize scattering centers, etc., known to any person skilled in the art. 2Ό and 30 images are formed by transverse scanning by a beam with possible parallel-emitted and / or pleasant beams. In this case, there are no the same restrictions on the width of the emitted beam as with obtaining images using the second harmonic, which allows the use of wider emitted beams and more parallel received beams for a higher frame rate of 2Ό- and 3О-images, in comparison with images obtained using harmonics.

A pulse compression method for a high-frequency pulse is also useful in cases in which the emitting force of ultrasonic pulses is used to locally apply pressure to an object, for example, to measure the shear deformations associated with the shear modulus of an object, or to improve the attachment of target vesicles - contrast agents to selected tissues . Using the present method, it is possible to increase the frequency separation between the pressing pulses and the pulses for monitoring by placing the latter on the negative spatial gradient of the low-frequency pulse to increase the receiving frequency of the pulses for observation. The pressure pulses can be emitted as longer pulses with a zero low-frequency pulse or as a sequence of short high-frequency pressure pulses placed near the positive spatial gradient of the low-frequency pulse for stretching and down-converting high frequency pressure pulses. In both cases, the pulses for observation acquire higher frequencies than the pressure pulses, and can be emitted shortly after or during the sequence of the pressure pulses, while the separation of the echo signals between the pressure pulses and the pulses for observation is carried out by filtering deep (fast) time space, as noted above.

In the second method in accordance with the invention, two or more groups of pulses are emitted with frequency components in the low-frequency and high-frequency bands, which overlap in time and in which the amplitude and / or phase and / or frequency of the low-frequency pulses varies from pulse to pulse. This method gives another type of received signal with strongly suppressed pulse reverberation noise, similarly to the previous method in accordance with the invention, and also provides imaging of nonlinear scattering parameters in the object, in particular obtaining microcalcifications, gas bubbles and regions filled with gas, and also gives numerical

- 13 014167 nonlinear parameters of scattering and propagation of the object.

We begin the description of this method with reference to FIG. For, which shows the emitted pulse, consisting of a low-frequency component 301 and a high-frequency component 302, and the high-frequency component travels on the positive crest of the low-frequency pulse with an amplitude of p ₀ . A high-frequency pulse is used to build an image, and a low-frequency pulse is removed by filtering at the receiver. Since these two pulses propagate through the object together, the amplitude of the low-frequency pulse affects the scattering coefficient of the object for high-frequency components by non-linear changes in the compression coefficient and density as

The term with a nonlinear compression coefficient is now ~ 2β _η · 2.5 ~ 25 times larger than the term with a nonlinear density, and these two terms usually have opposite signs. The construction of images using high-frequency pulses with a central frequency of ω ₁ thus gives a band-pass filter in the coordinate range of this parameter in the frequency range of about 2Ε ₁ = 2ω ₁ / ο, as described above.

Further, for example, a 2nd pulse is emitted, as shown in FIG. 3b, where the polarity of the low-frequency pulse 303 is reversed compared to the pulse 301, while the high-frequency pulse 304 occupies the same temporary position in the group of pulses as the pulse 302 shown in FIG. 3a, so that the low-frequency amplitude at the location of the high-frequency pulse is now -p ₀ . Since the nonlinear scattering parameter in equation (9) is linear in the added pressure, ± p ₀ , the nonlinear scattered signal from the high-frequency pulse 304 in FIG. 3b has the opposite sign with respect to the nonlinear scattered signal from the high-frequency pulse 302 shown in FIG. 3 a. At the same time, the linear component of the high-frequency scattering from the object, as shown by equation (2), is not affected by the low-frequency pulse.

Due to non-linear changes in the propagation velocity with pressure according to equation (6), the propagation velocity of the high-frequency pulse will vary between the values corresponding to the pulses shown in FIG. 3a and 3b, as Dc / c _0a = 2c _pa to _0a p ₀ . The backscattered signal from the high-frequency pulses is thus time-shifted for the positive and negative polarities of the low-frequency pulses. The velocity c _{0a of} propagation for soft tissues, water and oil products has average values of ~ 1.5 mm / μs. Rocks have about 2 times higher speeds. The delay time of the backscattered signal from the scattering center at a distance r is defined as ί (τ) = e ₀ (r) + t (r) ^{ίο (Γ) = 2} & ^{= 10} where 8 is the distance along the beam axis;

1 ₀ (g) is the delay time for p ₀ = 0;

t (g) is the added nonlinear propagation delay time due to nonlinear control of the propagation velocity of the high-frequency pulse by the low-frequency pulse;

p ₀ is the amplitude of the low-frequency pulse at the location of the high-frequency pulse as a function of depth.

In what follows we shall call m (r) the delay time of the nonlinear propagation or the delay of the nonlinear propagation.

A high-frequency pulse will also exhibit cumulative self-distortion, as described with respect to FIG. 1b, which increases the harmonic bands of the high-frequency pulse for a certain distance, followed by their decrease due to the absorption of the high-frequency pulse at deeper ranges. Due to the low pulse frequency of the low-frequency band (usually 1 / 5-1 / 20 of the high frequency), the delay of nonlinear propagation introduced by the low-frequency pulse will prevail for much greater depths. 2 in multiplier! ₀ (g) appears due to the summation of the propagation delay time of the outgoing, emitted, pulse and the delay time of the backscattered pulse. The low-frequency component will have a sufficiently high amplitude to affect the propagation velocity of the outgoing pulse, and therefore this factor 2 is absent in t (r). Since the non-linear delay time control is carried out on the outgoing pulse, this delay time control is the same for the scattered signal in all directions, as well as in the forward direction, which is the forward propagation velocity control, which we will return to in connection with the description of FIG. nine.

- 14 014167

When the phase relationship of high and low frequency components are practically constant along the beam, this time shift will vary monotonically with the local spatial mean value in _step to _a, as shown in FIG. 4, where reference numeral 401 denotes t ₊ (g), where the high-frequency pulse travels on the positive crest of the low-frequency pulse, and position number 402 indicates t _- (g), where the high-frequency pulse travels on the negative valley of the low-frequency pulse. The value of τ _Σ (τ) denotes the difference delay between these two pulses and is shown by the position number 403.

For the amplitude of the low frequency pulse p ₀ ~ 1 MPa obtain Ae / e = _{0a 0a pas} to p ₀ ~ 2 × 10 ^-3, that distance K = 300X 300e _0a = ₁ T ₁ where T ₁ = 1/1 ₁ - high-frequency pulse period, i.e. we obtain the added delay time τ (Κ) from equation (10) equal to ~ 0.6T ₁ , i.e. close to the period at the center frequency of the high frequency band. This range of images is typical for constructing medical ultrasound images, while in water it is possible to have a larger range, and in geological applications the maximum range for imaging may be less than this value.

It should be noted that the difference in the non-linear propagation delay for the positive and negative low-frequency pulses in FIG. 3 is twice that size. Therefore, even with lower amplitudes p ₀ , for example, up to p ₀ ~ 50 kPa, we obtain a significant non-linear propagation delay in accordance with equation (10), which must be compensated for adequate extraction of the non-linearly scattered signal, as described below. To construct images using computed tomography in transmitted beams, we have one-way propagation without reducing the signal during scattering, which provides at least a double range of computed tomography images in transmitted beams and, therefore, a double finite non-linear propagation delay, which we will return in connection with the description of FIG. nine.

We now describe how the invention establishes image signals by non-linear control of a low-frequency pulse by the scattering and propagation parameters of an object for a high-frequency pulse in accordance with the second method of the invention. First, we study the case of obtaining images using backscattering, where x _k (1) is the backscattered signal received from the emitted pulse with number k, see examples in FIG. 5. Time 1 reflects the depth of the scattering centers and is designated as fast time, while the number k of the coordinate of the pulse measures slower changes in the object and is indicated as the coordinate of the slow time. The sampling speed along the slow time coordinate is the pulse repetition rate of 1 _pg (= 1 / T _pg £, where T _pg £ is the time interval between the emitted pulses and is usually chosen a little longer than the time T _max necessary to obtain a scattered signal from the deepest range image lag. to build a medical imaging T _max ~ 300T _1, whereas for imaging in the water get higher values, and for the construction of geological images are usually obtained lower values of T _max. in Jun hedgehog schematically illustrates the received signals to the slow time five samples 501-505 as a function of the fast time signals vary as a function of slow time coordinate due to the following effects.:

changes in p _{0k of the} low-frequency pulse as a function of k. In some cases, according to the invention, the amplitude of the low-frequency pulse is p _0k ~ (-1) ^k . This leads to a change in the received signal in the coordinate of the slow time for a fixed 1, caused by nonlinear scattering and propagation with a frequency of slow time ~ Г _{рг £} / 2, as described below;

movement between scattering centers and the array of transducers in the direction of the high-frequency beam range. This leads to a Doppler shift of the received signal in the coordinate of the slow time for a fixed 1;

movement between scattering centers and the array of transducers in the direction transverse to the beam direction.

This phenomenon can, for example, be observed during mechanical scanning in the transverse direction to the direction of the acoustic beam or when moving scattering centers, such as a heart wall or leaking oil, which leads to an expansion of the signal frequency along the coordinate of the slow time.

A mathematical model of the backscattered first harmonic band of a high-frequency signal x _k (1), where a low-frequency pulse switches the polarity between the emitted pulses, i.e. p _0k ~ (-1) ^k , and the scattering centers in this case move, can be written as

where w _a = -2sh1O _g / e ₀ is the average Doppler shift for scattering centers moving with an average radial velocity υ _Γ from the transducer in each cell of the range along the beam;

magnitude and _1k (1) is the complex envelope of the linear backscattered signal;

and _pc (1) is the complex envelope of the nonlinear backscattered signal from the high-frequency pulse with number k with a positive amplitude p _{0 of the} low-frequency pulse.

- 15 014167

Envelopes change with the coordinate number k of the pulse, since the scattering centers and the beam move relative to each other, and the scattering centers can move at different speeds within the cell of the range, both effects leading to an increase in the signal frequency in the coordinate of slow time. The quantity τ (ΐ) is the delay of nonlinear propagation as a function of the coordinate of the fast time-range for the positive amplitude of the low-frequency pulse. For simplicity, we used the analytical form of the received signal, where the physical radio frequency acoustic signal is ¹ .

An analytical signal can be obtained from a physical signal as

where H {} denotes the Hilbert transform of the signal;

x _k ° ^{/ o} (1) is the complex envelope of the signal.

The second harmonic band can be represented by a formula similar to equation (11), where the angular frequency is 2ω ₁ , the Doppler frequency is 2ω, |. and the nonlinear scattered signal is very small and can be neglected, except for the case of scattering from microbubbles. The second harmonic band suppressed the noise of the reverberation of the pulse, which can help in determining the delay of nonlinear propagation, to which we will return with respect to equation (21).

Switching the nonlinear delay, (-1) ^to t (1) between pulses is most pronounced in the phase, since the phase switching is (-1) ^to w ₁ t (1), compared with switching the envelopes 1 - (- 1) ^to t (1) since the signal bandwidth is limited. To observe the effect of delay switching on envelopes, we divide the envelopes into even and odd functions around ΐ, which allows us to express the complex envelope of the received signal as

440 = {(H (OS)) + (-1) '(4 (<, m (a)) + ^m ) + ^and <! K (1 - m)} g = 1, n ^and ek (A ^g ) = H ^and <* (* + ^m ) - - m)} (12) where the superscript e denotes even components and the superscript o denotes odd components with respect to τ about ΐ.

The even components do not change when the sign of τ changes, while the odd components change the sign. It should be noted that the (-1) ^k = exp {PK1} = exp {1set _{pi (pi} w _£ / 2}, where _{ω ρι £ = 2π / Τ ρΓ} £ represents the angular frequency of pulse repetition, and hence the angular sampling frequency in coordinates of slow time. By introducing this expression, equation (12) can be further expressed as

4 ° (0 = | (“No. - ^and pc) SO8SH1G (1) southwest)

With '*'^'2>' (13) + <( «ή * - A) sovogI -r (and _'1k - ztodDTs with in

For a fixed fast time ΐ the signal in the coordinate by the slow time consists of 4 components (frequency lines), as shown in FIG. 6a, at which position number 601 depicts a frequency line A equal to (uY-uA) • ω ^ ДЦЦЦ ¹ "" ¹ ' ¹ ' ¹ "''¹; centered around the average Doppler shift ω4, the position number 602 represents the frequency line B equal to -ί (πΧ-Η ^ο Πτ8ίηωιτ (ΐ ^ ^{1 (ωά + ωρΓ £ / 2) Τρι £} \ centered around ω4 + ωρΓ £ / 2, position number 603 shows the frequency line C equal to (and ^e pk- ^o 1k) · oοδω1τ (ΐ) e ¹ ^^ ² ^^, centered around ωά + ω _{ρι £} / 2, and the position number 604 represents the frequency line Ό equal to - 1 (and ^e-pc and ^{about 1k) · δίηωιτ (ΐ ^ ιω <} Πρι £ \ centered around ω _ά.

The lines are obtained by mixing linear and nonlinear scattering with switching the delay of nonlinear propagation, (-1 ^ τ (ΐ) between pulses. If there is no switching of nonlinear delay, i.e., at τ = 0, the odd components vanish and the even components are equal The initial envelope: Lines 602 and 604 disappear, and all the linearly scattered power is contained in line 601, centered around ω _ά , while the nonlinearly scattered power is contained in line 603, centered around ω ₄ + ω _{ρΓ £} / 2.

- 16 014167

The effect of switching the delay of nonlinear propagation affects the phase of the signal most strongly due to the limited signal bandwidth, since this leads to a frequency shift with a shift of ω _ρΓ (/ 2. It also leads to a frequency shift with a shift of ω _ρΓ (/ 2 by participating in the envelopes ! - (- 1) ^to t (!) Through the odd envelope components for linear scattering, and ° 1k (!), And nonlinear scattering, and ° pc (!), While switching the envelope delay does not reveal the effect of a frequency shift to even components Envelopes: A phase delay shift results in Moving parts are linearly scattered power from a line centered around ω4, to the line 602 centered around ω3 + ωρΓ (/ 2 represented by the even component i1k ^e. joint switching in the phase and the envelope shifts part of the power from centered around ωά to centered around ω _ά + ω _ριΐ / 2 and back to the center near ω _ά as part of the line 604 represented by the odd component and ₁ °. Similarly, switching the nonlinear delay shifts part of the nonlinear power dissipation from line 603 to line 604 through the phase switching represented by even component and ^e _pc , while the combined phase and envelope switching shifts a part of the power from a center near ω _ά + ω _ρΓί / 2 to a center near ω _ά and back to a center near ω ₃ + ω _ρΓί / 2 as part of the line 602 represented by an odd component and ° _pc .

We note that with an increase in τ, the amplitude of the 601 line decreases ~ C05O | T (1). and the amplitude of line 602 increases ~ δίηω ₁ τ (1) and reaches a maximum when ω ₁ τ (ί) = π / 2, which also gives zero for line 601. This means that a significant part of the linearly dissipated power by switching the phase delay deviates from centering near ω _ά to centering near ω _ά + ω _ριΐ / 2, but some part of the linearly scattered power through the odd components and _1k (!) mixes back with centering near ω _ά as part of line 604. The same effect is observed for nonlinear scattering in which, with increasing τ, the amplitude of line 603 drops ~ C08 <Л1Т (!), and the amplitude of line 604 increases ~ δίηω ₁ τ (1) and reaches a maximum when a part of the nonlinearly dissipated power by switching the delay in the phase moves away from centering near ω _ά + ω _ρΓί / 2 to centering near ω _ά , but some part of the nonlinearly dissipated power through the odd components and _pc (!) mixes back with centering around ω ₃ + ω _ρΓ (/ 2 as part of line 602.

If the range cell also covers moving blood or other fluids, such as oil, then the result of linear scattering from the moving fluid will be additional and usually wider spectra in which power is distributed between line 605 with initial Doppler shifts ω ₃ = -2ωιυ _Γ / ^ where υ _Γ is the distribution of fluid velocity, and by line 606 with frequencies ω _ά + ω _ρΓί / 2. Mixing a signal from a fluid with switching the delay of nonlinear propagation obeys the same laws as for linear scattering from an object in lines 601 and 602. However, the nonlinear scattering from the fluid is so weak that it disappears in noise.

The signal model in equations (11) - (13) includes only a first-order scattered signal in which the outgoing high-frequency pulse follows the low-frequency pulse. Taking into account multiple scattering, also called reverberations, of the outgoing high-frequency pulse, in this model we obtain some modifications of the multiply scattered signal. The amplitude of the low-frequency scattered pulse is low and, therefore, the influence of its nonlinear delay on the propagation velocity of the scattered high-frequency pulse, equations (6), (10), can be neglected. This is especially important for the reverberation of the outgoing pulse in the body wall, where FIG. 7a shows an example of the structure of a lattice 701 of transducers and reflectors of the body wall. Item number 702 indicates a strong reflector in front of the grille. The emitted pulse follows the path indicated by reference numeral 703, where when it first hits the reflector 702, the pulse partially passes further as 704 and partially reflects as 705. The reflected pulse is then reflected again from the surface of the converter or other strong reflectors to create a reflected pulse 706, which again partially passes further and partially reflected, etc.

The reflected initial pulse from the deeper reflector 707 is shown at 708 in FIG. 7b for a positive low-frequency pulse and position number 709 for a negative low-frequency pulse, and these pulses have different delays, determined by changes in the delay of nonlinear propagation according to equation (10). The pulse twice reflected from the reflector 702 is indicated by the reference number 710 for a positive low-frequency pulse. After the first reflection from 702, the pulse amplitudes decrease, while the decrease in the amplitude of the low-frequency pulse reduces the delay time control for the reverberation pulses in accordance with equation (10), compared with a directly propagating pulse. The pulse twice reflected from the reflector 702 for the negative low-frequency pulse will therefore only slightly differ in delay from 710, and therefore it is shown in the drawing as 711, overlapping with 710. The reverberations of the pulse of the outgoing pulse in the body wall will therefore be observed as a frequency line centered around ω = 0, shown as a dotted line 607 in FIG. 6a, which also shows the mixed delay line centered around ω _ρΓ (/ 2, which has a very small amplitude.

The inventors are now able to describe and discuss how several new image parameters can be extracted from transmitted and scattered acoustic signals in accordance with the second method of the invention in order to form new acoustic images of object structures and velocities of scattering centers of improved quality. The isolation of these parameters can be demonstrated based on the received sequence x _to (1) of the signals.

The first image signal extracted in accordance with the second method of the invention is based on a scattered signal with reverberation correction obtained by bandpass filtering the received sequence in the slow time region near ω _ρΓί / 2, as, for example, shown by the bandpass filter 610 in FIG. 6a. This filter greatly attenuates the reverb line 607 and highlights lines 602 and 603 in which the linear scattering components are dominant compared to the nonlinear scattering components. By combining the received high-frequency signals, for example, in a filter that attenuates the low-frequency components of the slow time, however, skipping the components of the slow time in the band, as in the traditional Doppler method of imaging, you can get a group of linearly scattered signals with reverberation correction

The amplitudes of these signals monotonically increase with τ.

By selecting the appropriate baseband frequency field (e.g., ω ~ ω _{0 1/10)} can be very low attenuation of low-frequency field absorption within the range of the image, which varies slightly between different objects and individual articles in predetermined ranges of depth. Thus, it is possible to set the low-frequency field and the pulse amplitude so as to obtain a monotonic increase in τ (τ), as in FIG. 4, in turn, to obtain a monotonically increasing, depth-varying gain function in processing signals for linear scattering with reverberation correction, shown by reference number 712 in FIG. 7s

By setting the low-frequency field so that ω ^ (1 ^) = ^ 2, where T _max is the maximum time range, we obtain in T _max a gain curve close to the maximum. This gain-varying gain function in signal processing, along with user-controlled gain-varying gain, takes part in acoustic imaging devices, reducing the need for user intervention in controlling gain depth. Movements of an object, such as a myocardium, can lead to Doppler shifts that can pass through the filter along with the components in equation (14). This can be an advantage, since the noise of the reverberation of the impulse of the fixed body wall is greatly attenuated, the movement contributes to the first-order signal from the structures of objects to pass through the filter, thereby improving image quality, for example, the apical region of the heart. The outputs from the filter will also contain non-linearly scattered components of lines 602 and 603, namely

which have a maximum for τ = 0 and decrease with depth with increasing τ.

However, the nonlinearly scattered signal components in equation (15) will be negligible compared to the linearly scattered signal components in equation (14).

The signals after the bandpass filter can be used for further Doppler processing to create Doppler spectra and radial Doppler lines of the image of the velocities of the scattering centers in accordance with known methods in which a full 2Ό or 3Ό image is created by transverse beam sway. This method is especially useful for Doppler estimation of myocardial displacements and displacement deformations (radial displacement gradient), since the reverberation noise greatly interferes with such estimates. To measure the fluid velocity, it should be noted that the object interference signal in equation (14) is located near ω _ρΓ (/ 2, and in order to suppress the object interference for estimating the fluid signal, either a slow-pass band-pass filter should be used near ω _ρΓ (/ 2, or shift the signal in frequency from ω _ρΓ (/ 2 to ω = 0 and use the standard high-frequency interference suppression filter before determining the Doppler frequencies of the fluid signal.

You can use several different and overlapping directions of the beams for each k, for example, as obtained by continuous swinging of the beam using an annular lattice. You can also then use the IIR filter to filter in slow time (bandpass, lowpass, highpass), where the lowpass filter is illustrated in equation (77) for similar processing.

- 18 014167

The first signal in accordance with the second method of the invention for use in the radial image line to enhance linear backscattering with suppression of the reverberation of the pulse can thus be obtained as the envelope of one of the signals ί) = (16)

BUT:

where Εην {} is the envelope operator;

r = e1 / 2 is the depth range along the beam, and the formula for ζ ₁ (ΐ) is an example of a strip combination, as in equation (19).

The full 2Ό or 3O image is then obtained by transverse scanning of the beams.

Impulse reverberations are reduced due to the cumulative effect of controlling the non-linear propagation velocity on the pressure, i.e., the same effect that produces harmonic distortion in a directly propagating impulse, which is used in obtaining harmonic images. However, in the present method, non-linear propagation is created by a low-frequency pulse with such a low absorption that the sensitivity of this method is similar to the sensitivity for the 1st harmonic imaging. This fact allows the use of higher image frequencies than in the case of the 2nd harmonic imaging, with improved resolution, and in particular allows you to get better images at deep ranges in dense objects such as the liver, kidneys and mammary glands.

Since the radiated beam in this method is the 1st harmonic beam, it is easier to obtain a wider radiated beam in it than in the case of the 2nd harmonic radiated beam. This allows the use of parallel received beams to increase the frame rate with the construction of 3Ό acoustic images.

To separate the nonlinear scattering components and the linear scattering components, it is necessary to correctly organize the delay (compensation of the time shift) of the received signals, so that the mixing of the frequency shift of the nonlinear propagation delay switching disappears for the linear signal, and the line 602 disappears, leaving only the nonlinear scattering line 603 in the band near ω _ρΓ (/ 2. The delay correction depends on the amplitudes and / or phases of the low-frequency pulses relative to the high-frequency pulses and, in addition, also varies with depth according to equation (10) and Fig. 4.

In accordance with one aspect of the invention, delay corrections can be evaluated by maximizing signal strength

K-1 = ( ¹⁷ ) k = 0

Usually, there is no reference signal, so to determine the delay corrections, you need to use one of the signals as the reference signal, while the delay correction for this signal becomes zero. Therefore, only k-1 independent delay corrections can be determined in accordance with the methods described below. The sum is a low-pass filter in the area of slow time with a frequency transfer function

NDi) = (1 ₈ )

З1п πω / ω _ρΓ /

Other low-pass filter options may also be used, for example, indicated by 611 in FIG. 6b. This can for example be achieved by administering various weights w _k in the sum of equation (17) as FIR filter, but also in accordance with known methods can be used IIR filter, especially for continuous beam oscillation, as with the annular lattice .

Delay corrections t _to (1) introduce frequency mixing, which moves all the linearly dissipated power to line 621 in FIG. 6b, centered around ω = 0, since the delay corrections in equation (17) also remove the Doppler deviation ω ₄ , as described below. Mixing the frequencies of the delay corrections also moves the nonlinearly scattered power to line 622, centered around ω = ω _ριί / 2.

The nonlinearly scattered signal can therefore be obtained after delay corrections in the form

4s (0 = Σ O +? * (*)) (19)

L = 0 where the summation is a band pass filter near ω _ρΓί / 2, which, for example, at 11 (1) = 1 takes the form

Η _η (ω) = ^{5] η} cos Ίΐω / ω ^ ι while other values for 11 (1) = 1 lead to a modification of this filter, for example, a filter denoted by position number 610 in FIG. 6b.

The linear combination in equations (17) and (19) gives one output signal in the slow time domain, while it is obvious to a person skilled in the art how to change the equations as FIR or IIR filters, which create a sequence of output signals in the slow time region in the same way as described in connection with equations (14) and (16). In this case, such sequences of signals in the slow time domain are used for further Doppler processing of signals in the slow time domain for imaging Doppler velocity or deformation of displacement or strain rate of scattering centers of objects and fluid in accordance with known methods.

To evaluate delay corrections, in accordance with one embodiment of the invention, the received time / depth interval T can be divided into shorter time intervals {T ₁ ,, = 1, ..., I}, which are likely to overlap, so that T <E; T ₁ , and determine the optimal delay correction for each T ₁ interval separately. The signal power ζι (ΐ) in the interval T _{1 is} then given by the functional

Chi / _t / h ' _s ^) ² = £ _m / g1' (7-1 ₍ -t _r ^ & k1 ~ bk) = / _Τι (ί + T _1k ) x> (ί + 7 ^) and delay correction determined by maximizing 1 ₁₁ with respect to τ _ιρ .

Examples of maximization procedures are given below.

The non-linear self-distortion of the high-frequency pulse described in FIG. 1 with respect to equation (6), leads to a drop in the harmonic conversion in the first harmonic high-frequency band with the propagation distance. This conversion drop is slightly different for negative p0 and positive ρ0, leading to a small difference in the amplitude of the 1st harmonic high-frequency signal linearly scattered from the object for positive and negative ρ ₀ . In addition, inaccuracies in radiating amplifiers create difficulties in the emission of precisely specified amplitudes, creating inaccuracies in the amplitude of the received signals. For improved suppression of a linearly scattered high-frequency signal in equation (19), it is possible to organize the possibility of changing the IF) from k, so that the signal power ζ _ηο (ΐ) is minimized in each T ₁ interval. Such values of Is can, for example, be found by minimizing the power in ζ _ηο (ΐ) in each interval with respect to And _to , after adjusting the delay of the signals. This comes down to minimizing the following functionality:

Chgg - / t, ^- Λ 221; ku. ~ 221 ,, / Κι / ΐιΐΝΐί - Τ _ιΛ ) - A, 22

- bk) = (-1) ^{k + 1} i £ _(gg t - Τ _ιλ;)

Examples of minimization procedures are given in equations (55) - (58). Changes in the values of And _{to are} also responsible for changes in the amplitudes of radiation of high-frequency components, when the amplitude and / or polarity, and / or phase, and / or frequency of the low-frequency pulse changes from one emitted pulse to another. The delays τ _ιρ can then be very simply obtained by maximizing 1 _k , while And _{k is} obtained by minimizing Ι _ηι .

Maximization gives an estimate of the average delay correction for each T ₁ interval. For the best correction in accordance with equation (17), one should give the values of these delay estimates to the point inside the corresponding intervals T ₁ and obtain interpolated estimates of the delay correction τ _to at each point of the fast time sample ΐ between the selected points. The selected points can be, for example, the midpoints of the intervals or the center of gravity of the power in the received signals in the corresponding intervals or similar points. Several interpolation methods can be used, such as linear interpolation, spline interpolation of any degree, and Fourier interpolation.

The signal linearly scattered from the object, shown by lines 605 and 606 in FIG. 6a, during the delay correction process, moves to line 623 shown in FIG. 6b, in the same way as a linearly scattered signal from the fluid, after which the specified signal can be processed in accordance with well-known methods of Doppler signal processing. For the reverberation signal in line 607, frequency mixing in the delay corrections leads to the propagation of energy to line 624, centered on - <bd, and to line 625, centered on ω _ρΓί / 2-ω ₄ . The reverberation noise will thus introduce errors in the estimation of nonlinear propagation delay corrections with a maximum of 1 _k in equation (21), and mixing the frequency shifts of the delay corrections introduces reverberation noise in the same slow-frequency band as non-linear scattering after delay corrections (line 625), and therefore introduces noise into the estimate of the nonlinear scattering signal, for example, in accordance with equation (19). The image signals described in equations (24) to (30) below will thereby be affected more by the noise of the reverberation of the pulse than the image signal based on equations (14) and (16).

- 20 014167

With reference to the discussion in connection with FIG. 1 and 2, it should be obvious that not only the first harmonic band of the received signal can be used, but also other bands, in particular the second harmonic band, as can be seen from the analysis of equations (14) - (22). The second harmonic band is mainly formed by direct self-distortion of the pulse, indicated by the reference number 104 in FIG. 1, which locally scatters both linearly and nonlinearly, as shown in equation (9). Since self-distortion is small in multiply scattered pulses, the second harmonic band of the received high-frequency signals will have significant suppression of reverberation noise, as described in relation to FIG. 2. In case of strong reverberation noise, to determine the delays of nonlinear propagation, it is preferable to use the second harmonic components of backscattered signals in equation (21). The use of the second harmonic band for the signal in equations (14) and (16) leads to additional suppression of the noise of the reverberation of the pulse, especially when one of the many scattering centers moves. Using the second harmonic band in equation (19) gives the second harmonic band of the local nonlinearly scattered signal. You can also use the second harmonic band to determine the non-linear propagation delays, and then apply delay corrections to the first harmonic band of the received signal to determine the first harmonic band of the non-linearly scattered signal, as above. The methods described in US patent No. 6485423 and 6905465 may also be useful when used in conjunction with the present invention to reduce the effect of pulse reverberations on the determination of non-linear delays.

Next will be shown the third, equations (42) - (44), and the fourth, equations (45) - (48), methods in accordance with the invention, in which the effect of the noise of the reverberation of the pulse is suppressed directly when determining propagation delays. Before describing these methods, it should be shown how it is possible to isolate numerous signals of image parameters using the second or fourth methods of the invention.

Maximizing C with respect to ζ _1ο , as described by equation (17), introduces delay corrections both for the delay time of nonlinear propagation and for the average Doppler delay time caused by the displacement of the object / transducer in the time interval T ₁ . The reason for this is that the summing filter described by equations (17) and (18) has a slow time response with a maximum at ω = 0, and a maximum of _{11 is} achieved when all the linear energy is shifted and centered around ω = 0. When using a slow-speed low-pass filter with a flat response near ω = 0 and strong attenuation near ω-ρΓΓ / 2, a maximum of ₁₁ can be obtained at which there are only non-linear propagation delays, and the linearly scattered energy of lines 602 and 603 is shifted into the transmission band of the filter , without adjustments for Doppler delay. However, such a filter is difficult to perform with a limited number of slow time pulses used for each radial line of the image, and, therefore, the roughest estimate can be obtained using a low-pass filter with a pronounced maximum at ω = 0. The delay correction estimate then will be the total propagation delay as the sum of the nonlinear propagation delay and the average Doppler shift delays (i.e., Doppler delay) of the relative displacement of the scattering centers of the object and the transducer.

As noted in connection with equation (17), it is possible to determine the K-1 delays of nonlinear propagation, i.e., one less than the number of signals. With three signals or with more than three signals, the exact definition of delay corrections for the signal modeled by Eqs. (11) - (13) and the filter with the maximum frequency response at ω = 0 gives delay corrections free of errors, like total propagation delays, which are the sum of the nonlinear propagation delays and the Doppler delays to ^t (1) = (-1) 4 (() + = (-1) 4 (() + kT _a (23) where m _{and z} = 2y (1) T _HGR / c is the Doppler delay of the bias due to the radial bias at _r (1) T _{prh of} the scattering center between the emitted groups and pulses.

Doppler phase shift and Doppler frequency can be found from

a) 0l (() = - ^ 1 (^ (() + ^ - 6 ()) / 2 = - (2к - 1Ц1Т- _й / 2

b) = {0l (() - Фа, к-1 (В} / Тр / = = - 2Ш11> Д () / с

This Doppler estimate is interesting in determining the radial displacement (from the phase in equation (24a)) and the speed (from the angular frequency in equation (24b)) of objects, for example, the myocardium, as well as the radial displacement of displacement and the strain rate as the radial gradient of radial displacement and velocities scattering centers. To determine the Doppler shifts of scattering centers in interference noise, like blood or other fluids, you can first use a high-pass filter to suppress the high frequencies before Doppler corrections, as described with respect to FIG. 12 and is known in the art. The determination of the displacement deformation of the scattering centers along the radial direction of the beam can be obtained from the derivative of fck (0 with respect to fast time. Similarly, we can estimate the radial strain rate of the deformation of scattering centers from the derivative of fck (0 with respect to fast time).

- 21 014167

The nonlinear propagation delay is found as τ (ί) = {^ (ί) _ (ί)} / 2 {—1) * - (25)

Since the definition of t _1k contains errors, an error determination can be reduced by averaging the equations (24) and (25) for several adjacent values of k. When a transverse displacement of the scattering centers or with rapid mechanical swing acoustic beam may be a characteristic change ;, r _k from to due to the exchange of the structure of the object in the beam from pulse to pulse, and for averaging it is necessary to limit the number (k) of pulses.

Since p ₀ (r) can be determined from preliminary measurements due to the low absorption of the low-frequency pulse, the first numerical non-linear image parameter representing the properties of non-linear direct propagation in the material can be determined from the non-linear propagation delays. The increment in the delay corrections between adjacent intervals T ₁ represents a nonlinear forward propagation parameter which may be recorded as Τι ⁼ ~ 1., a n-1, k ~ βηια ^ ιαΡο ^ (26) where β _ηια and _1a are spatial averages over the range interval corresponding to T _; ; and p _01k is the average amplitude of the low-frequency component in the same range interval, corresponding to the emitted pulse with number k.

The first numerical non-linear parameter / image signal (non-linear propagation image parameter) is then obtained from equation (26) as

The second image signal, which is used to construct images in accordance with the invention, is the envelope a _ps (g) of the nonlinearly scattered signal x _{|| s} (2g / s) in equation (19). This envelope is correlated with the nonlinear scattering parameters of the material as „ _e (r) = Epu {r _ps (2r / e)} ~ ² C (r) ro (r) C (r) exp {- 2 /, ά ₈ μ { ₈ )} (28) where ν _η (τ) is the passband filtered about 2 k ₁ , as discussed in connection with equations (2) and (9), and averaged in the transverse direction along with the amplitude p _{0 of the} low-frequency pulse across high-frequency beam profile for the g-band

The exponential term describes the attenuation of the absorption of a high-frequency acoustic pulse in an object and is compensated by a user-adjustable compensation C (t) for time / depth amplification in the acoustic device.

The absorption factor can be found by comparing a _ps (g) with the envelope of the 3rd image signal, a linearly scattered signal, / | _c (1) in equation (17) after corrections for the delay, which correlates with the parameters of linear scattering and acoustic absorption in the object as

where \ '| (r) is the passband filtered about 2 _k1 and averaged in the transverse direction across the high-frequency beam profile for the r-band, as discussed above.

When p ₀ (g) is known, for example, by calculation or measurement, as described above, you can combine the signals in equations (28) and (29) to obtain the 2nd numerical non-linear parameter / image signal, numerical parameter / signal non-linear scattering object as _h 'PS (z) Y _f {z) ^P3 and _1c (r) ro (t) y (t)

The corrected time shift then contains propagation information about the value in _n1a to _1a averaged over the local interval T ₁ , while the signal x _{|| c} (1) of nonlinear scattering contains information about local, spatial fluctuations of β _η «in the interval T ₁ . You can also use the interpolated values of t _1k along the fast time to t _k (1), taking into account the fact that equations (26) and (27) are the derivatives along the fast time samples to obtain a smoother version pr (1) of the nonlinear parameter pr, Images. Similarly, the values of pr, can be assigned to points inside the intervals T; and interpolating the values between these points to build the image in the same way as described above for interpolating the propagation delay.

Methods for reducing pulse reverb, as described in connection with FIG. 2 and Equation (14) are useful in conjunction with methods for determining corrections for wavefront aberration, for example, as described in US Pat. Nos. 6,485,423, 6905465 and US Patent Application No. 10/894387, to reduce the negative impact of reverb noise on determining an aberration correction. To correct aberration, you can use an array of acoustic transducers with a two-dimensional distribution of elements, while corrections are applied to the signal of each element before final summing in the beam former, or in many cases, signals from neighboring elements can be combined into a signal

- 22 014167 subaperture channels, in which aberration corrections are applied to the subaperture signals before the final beam summation, and not directly to the signals of individual elements. The best results in determining aberration corrections are obtained when the noise reverberation noise suppression is applied to all element (or subaperture) signals before the aberration corrections are determined, and this suppression can also be applied to the summed beam signal, since the effects of pulse reverberations are reduced by correlation between element signals and the summed beam signal, as described in the above patents.

Representing the spatial change in the acoustic propagation velocity in the form с _0а (г) = с ₀ + Дс _0а (г), where с ₀ is the constant propagation velocity adopted by ~ 1.54 mm-μs when calculating the delays of the beam former according to the assumption of a homogeneous material, we can approximate the delay of the wavefront aberration as (31) where r is the position vector of the element on the surface of the lattice, the actual element or subaperture;

r _G - the position vector of the focus of the beam and

Г (г, г _Г ) - the path of the beam from the center of the r element to the focus of r _Г.

To correct wavefront aberrations for elements arranged in grams, should be organized delay emitted pulses to the emitted beam and the received signals to the received beam with the correction delay T _wr (r) = - t _ab (g) as described in U.S. Patent № 6,485,423.

There is a strong correlation between changes in density and compression coefficient in different materials, which according to equation (3) also indicates a correlation between changes in compression coefficient and propagation velocity as ~ - 0s (ur _g (t) - Τΐρ _αν Α (32)

Where in _c is the proportionality coefficient determined experimentally, and it should be noted that in _{pa a} = pr is the numerical parameter of nonlinear propagation in equation (27), which is now determined for each signal of an element or subaperture, depending on the position of r on the surface lattice.

The parameters β _η0κ0 and pr _aud are spatially averaged parameters for all elements and delay intervals. Substituting this expression into equation (31), we obtain an approximate estimate for aberration corrections in the form

Table (d) in Σ ^t * ( "I (c) - etc. _{A" s)} (33) where the latter is independent of the amount of

This assessment can also be used as an initial assessment for the procedures described in the above patents and patent applications.

The change in pr ₁ (r) with r mainly arises due to propagation through the wall of the body, and since it is the changes in the parameter m _b (r) with the position of the element or subaperture in r that create aberrations, good results can be obtained by summing for the intervals ί in equation (33) only slightly beyond the wall of the body. In this region of a near field, one can create such a low-frequency field that p ₀ (§) is approximately constant in the body wall for each element, so that this term can be taken outside the integral for m (r) in equation (10). Thus, it can be seen that it is possible to compare the delays of correction of aberration directly with the delays of nonlinear propagation behind the wall of the body and subtract the spatially averaged value of m (r), i.e. τ _3νκ , for all elements, namely ₍₃₄₎ m (r) = - / _ГДДб) ^^^ ηα (ί) κοα (3) ρ _ο (5) where p _aud (r) is the spatially averaged value of the low-frequency field p ₀ (§) along the propagation path T (r, _Kb ), starting from the position of the element through the body wall of thickness _Kb

As noted above, non-linear propagation delays using the second method can be determined from the second harmonic band of the received signals to reduce errors caused by pulse reverb noise. In the third and fourth methods presented below, in accordance with the invention, the noise of the reverberation of the pulse is avoided directly in the processing of signals, while in practical cases, the use of the second harmonic band improves accuracy using these methods.

From equation (8) it can be seen that from the derivative of the central frequency of the received high-frequency signal, one can estimate the local absorption coefficient a (g) as a function of the range of r. However, in the absence of pulse compression, sliding the frequency down can cause the receiver converter to limit the bandwidth of the received signal. Therefore, by using

- 23 014167 pulse compression in equation (7) so that it compensates for the frequency glide down due to absorption, you can put the received signal in the middle of the receiver transducer band to avoid clipping the signal bandwidth from below due to the limitations of the receiver transducer band, making it possible determining local power absorption from measurements of the received center frequency of the high-frequency signal and non-linear propagation delays m (g), as defined in equation (10). The parameter m (r) is evaluated, for example, according to the second method described above, or according to the third or fourth methods described below, in addition to emitting a group of pulses, as in FIG. 1, where the high-frequency pulse is placed near the maximum spatial gradient of the low-frequency pulse.

With reference to the discussion in connection with equations (7) and (8) in this latter case, we obtain a compression of the high-frequency pulse length equal to 6λι = -β _ηί1 κ ₀ 3ρ ₀₀ ^ ₀ λι ₀ ρο (Γ) / ρ ₀₀ 6 _Γ at a distance propagation bg, where p ₀ (r) is the local amplitude of the low-frequency pulse, including changes associated with the convergence / divergence of the beam and a small absorption; p ₀₀ and k ₀₀ , respectively, the amplitude and wave number of the low-frequency pulse on the surface of the grating; and λ ₁₀ is the high-frequency wavelength on the surface of the lattice. The received center frequency of the high-frequency signal is related to the wavelength as Γ ₁ (Γ) = ο _0α / λ ₁ (Γ), which gives

6 £ 1 = - £ 1 ² / ^ α6λ1-0.36Β1 ² (Γ) α (Γ) 6Γ = βηαΚ0α ^ 0λ10ρ0 (Γ) Γ1 ² 6Γ / ^ _ΕΙ -0.36Β1 ² (Γ) α (Γ) 6Γ .

Pulse width strip corresponds to the length of pulse A with B1 = _0a /K.1. According to the reasoning given above, the pulse length is compressed at a distance bg of propagation as

6Η-βι. _{: Κ,:} ρ KN-ρ (Γ) / ρ bg where Κι ₀ - pulse length on the array surface.

Down-conversion due to absorption preserves the signal bandwidth and gives

6Β -Β / _s: b1V to β ρ to .k ρ (ιίΒ 'Bg / s.

As a result of this analysis, we obtain the following differential equations for ί and Βι

a) ⁼ Λ α α α * (λ λι _Ο ρο (· τ) ^ - 0.36 B ^ (g) a (g) ⁼ P ^ oaRoocaoK1oPo (g) ^ (35)

Equation (35) can be integrated directly, which allows one to determine the local absorption coefficient from the measured delay t (g) of nonlinear propagation, its gradient and the gradient of the central high frequency ί (g) as

where m (r) is given in equation (10);

61/61 - = (1 -) ^ (1- ^ 0 (1 -) / ^ (1) and with K10V10 = _Ov (0) and c ₍₀₎ ω, the angular frequency of the central low-frequency pulse.

By adjusting the amplitude of the low-frequency pulse, one can avoid modulating the received high-frequency signal by the edges of the high-frequency receiving band so that equation (35b) is satisfied, which allows its integration into equation (36b), which, in turn, allows one to determine a (d) from equations (35a) ) and (36a).

The non-linear parameter β _η becomes very small for solid materials, as does the compression coefficient k. Thus, in particular, at the boundaries between soft and harder materials, for example, fabrics with a high density of connective fibrous molecules, calcifications or other materials with a high density, non-linear scattering becomes very strong. Similarly, strong nonlinear scattering at the boundaries is obtained into softer materials, such as fat, foam cells, and especially microbubbles of gas in the tissue, where nonlinear scattering is further enhanced, as described below. The same is true for the boundary between solid rock and fluid or gas in geological structures, swimming bladder in fish or lungs in marine animals in water, mines in the ground or on the soft bottom of the sea, etc. Non-linear image enhances, therefore, visualization of such structures. The invention is thus useful for visualizing microcalcifications in soft tissue, for example, for imaging tumors in the mammary gland and other tissues or atherosclerotic plaques that are difficult to observe using existing acoustic imaging methods. In addition, with less dramatic changes in the compression ratio of the material, namely, when compliance decreases with the growth of connective tissue or when compliance increases with the growth of fat or foam cells, non-linear parameters determined by these methods give an increased image contrast for changes in tissue compared to existing methods of building images. Image parameters presented in equations (27) and (30) allow us to obtain numerical estimates of changes in tissue.

- 24 014167

For gas bubbles that are either naturally present in the object in the form of hydrocarbons, or decompression bubbles when immersed, or are introduced into the object in the form of an ultrasonic contrast medium, the dynamics of their scattering is described by a differential equation that gives resonance scattering with a frequency-dependent phase delay between an incident and scattered wave is completely opposite to scattering from a fluid or solid objects, where the change in phase delay with frequency is very small. A low-frequency pulse controls the diameter of the microbubbles (small diameter with positive p0 and large diameter with negative p0), and therefore the resonant frequency of the microbubble. This, in turn, controls the phase delay of the scattered signal for the high frequency pulse in addition to the amplitude of the scattered signal. The control is especially strong for high-frequency pulses near the resonant frequency of the microbubbles, as described in US patent application No. 10/851820, filed May 21, 2004

The corrected non-linear signal ζ _ηο (ΐ), for example, in accordance with equation (19) will contain almost all of the high-frequency dissipated power from the bubbles - contrast medium (both linear and non-linear components). For a scattering object that contains gas microbubbles, the present invention thus significantly increases ΟΝΚ (contrast to noise ratio) compared to existing methods for imaging such bubbles by extracting almost all of the scattered power of the high-frequency signal from the microbubbles (especially strong linear components, not just nonlinear components). Corrections for switching non-linear propagation delays by a low-frequency pulse provide suppression of linearly dissipated power from the object, resulting in a large RNS (contrast to object ratio). The method thus separates the effects of nonlinear direct propagation and local nonlinear scattering and uses local control of the change in the phase frequency of the signal scattered from the gas bubbles to obtain a strong local signal from gas bubbles with strong suppression of the local signal from the object.

This differs from existing methods of imaging using contrast agents, in which nonlinear propagation leads to a cumulative effect on the directly propagating pulse, which, in addition, enhances linear scattering from the tissue during the detection process so that this linear signal from the fabric masks the signal from microbubbles (as well as a nonlinearly scattered signal from the tissue). Similar effects occur when gas bubbles are detected behind a cloud of gas bubbles in geological structures and objects behind a school of fish with swimming bladders or marine animals with lungs.

When a pulse passes through a cloud of gas bubbles (also through swimming bubbles), they create an increased, cumulative non-linear propagation delay for a directly propagating pulse, as well as a non-linear change in the pulse amplitude, a phenomenon that emphasizes the need to adjust non-linear propagation delays and changes in the pulse amplitude to get a good suppression of a linearly scattered signal from an object outside the cloud. The present invention, therefore, has a distinct advantage over the known microbubble imaging methods. For example, in the harmonic imaging method, the increased, cumulative harmonic distortion for the pulse that passes through the cloud of microbubbles manifests itself as strong harmonic components in linear scattering from tissue outside the cloud. This, for example, can lead to strong harmonic scattering from the myocardium for an impulse that passed through a cloud of contrast medium in the ventricle of the heart, masking the scattered signal from the contrast medium - micro bubbles in the myocardium. This effect may erroneously indicate blood circulation in the myocardial region with very low perfusion or with its absence, and also indicate gas in the geological structure without gas. Using the present method for scattering from the myocardium behind a cloud, the influence of a cloud of microbubbles in the ventricle of the heart on a directly propagating pulse is eliminated by correcting non-linear propagation delays. The invention distinguishes between local nonlinear scattering and the cumulative effect of nonlinear direct propagation and therefore prevents the measurement of local nonlinear scattering from leading to such erroneous indications of the absence of microbubbles in the myocardium.

With respect to non-linear harmonic imaging methods, the present invention can use a wider radiated pulse, and therefore, a higher resolution of the image range can be achieved. In addition, higher frequencies can be used for imaging, resulting in a significant increase in resolution compared to other methods of imaging gas bubbles both in the range direction and in the transverse direction. The mode of operation of the present invention is less sensitive to the amplitude of the pulses used for imaging, compared with harmonic imaging methods. Together with the indicated suppression of the received linearly scattered signal with the resulting increase in RNS, this facilitates the non-destructive detection and construction of high-resolution images of individual bubbles — a contrast medium with a low mechanical index (MI).

- 25 014167

In medical applications, the construction of images of increased sensitivity and high resolution of an ultrasound contrast medium has a great prospect in obtaining images of changes in the microvascular system, for example, with neoangiogenesis or necrosis in tumors or with decreased blood circulation in the myocardium, where some standard methods have been developed for using the filling time with a contrast agent. The numerical parameters in equations (27) and (30) provide quantitative information about the density of the contrast medium in the tissue and, therefore, provide improved data on the relative volume of the microvascular system. The methods also have applications in estimating the relative volume of gas in geological structures, the density and size of fish with a swimming bladder and marine animals with lungs. By destroying the contrast agent - bubbles in a certain area and measuring the filling time, it is possible according to known rules to obtain numerical values of blood circulation in the tissue. Imaging of a contrast agent of high sensitivity and high resolution is also useful to track the allocation of lymph to find signal lymph nodes during surgical removal of tumors.

Since equations (27) and (30) give image parameters that are independent of absorption in the object, these parameters of the object can be used to characterize the object (for example, the content of fat or connective tissue in soft tissue, the content of oil, gas or water in a porous mountain breed or number and size of fish or marine animals). In addition, it is possible to determine local changes in the propagation velocity with temperature, based on experiments, to monitor changes in tissue temperature during thermal treatment of diseased tissue, for example, through focused high-power ultrasound (FUVM) (NGEI), radiofrequency ablation or cryosurgery. The temperature can be monitored by changes in the numerical parameters, as well as by changes in the propagation velocity, which cause the delay time between backscattered signals from image to image as the temperature changes. The radial gradients of this delay time determine the local temperature.

Since the low-frequency pulse does not affect the local linear scattering of the high-frequency pulse, it is understood that changes in the amplitude and / or phase and / or frequency of the low-frequency components in the emitted pulses, but not those shown in FIG. 3 can lead to a similar result in suppressing reverberations of a pulse from a near source to produce a linearly scattered signal with reduced reverberation noise and in suppressing linear backscatter to produce a nonlinear scattered signal, as noted above. For example, the low frequency component of the 2nd pulse in FIG. 3b may be absent or high-frequency pulses do not have to travel exactly on the positive crest or negative cavity of the low-frequency pulses. This versatility is important, because the phase between two frequency pulses can occur due to diffraction or mismatch of the radiating surfaces of the transducer gratings, since the low-frequency and high-frequency components change with the propagation distance along the beam.

The low-frequency and high-frequency bands of emitted pulses are often so widely spaced from each other that they sometimes prefer to use separate transducer arrays to emit two pulse bands. Such gratings can be made in the form of concentric rings with different resonant frequencies, in which the beams from the gratings automatically overlap, or the gratings can be installed side by side with the intersection of the beams at an acute angle.

When two different gratings are used for low-frequency and high-frequency components with beams that overlap at an acute angle, the phase relations between low-frequency and high-frequency pulses can have a strong spatial dependence, depending on the geometrical location and surface sizes of two radiating gratings. An exemplary arrangement of separate low-frequency and high-frequency transducer arrays emitting along the ζ axis in accordance with the invention is shown in FIG. 8a, in which reference numeral 801 denotes a high frequency grating, and reference numeral 802 denotes a low frequency grating consisting of two parts on each side of the high frequency grating. The drawing may, for example, show a cross section of a linear or curvilinear arrangement of the lattice, where the y-axis is the vertical axis perpendicular to the azimuthal scanning plane, i.e. χ-ζ-plane perpendicular to the ν-ζ-plane. The drawing may also show a cross-section of the arrangement of the annular array with the ζ axis as the radiating axis, where reference numeral 801 denotes the cross section of the high-frequency annular array and reference numeral 802 indicates the cross-section of the low-frequency annular element. Both linear and circular arrangements exhibit different propagation delays for low-frequency and high-frequency gratings, which should be duly taken into account when designing the grating and signal processing in order to take full advantage of the basic physical effects underlying the invention.

The boundaries of the illustrative low-frequency beam are shown by the position number 803, while the boundaries of the high-frequency beam are shown by the position number 804. It should be noted that there is a near zone 805 in front of the grating, where there is limited overlap between the low-frequency and high-frequency beams, thereby leading to particularly strong suppression of the reverberation by the wall of the body outgoing pulse, using the methods described in connection with equations (14) and (16) and FIG. 7.

- 26 014167 (37)

We also note that the propagation distance from the low-frequency grating to the scattering center located near the ζ axis is much larger for a low-frequency pulse than for a high-frequency pulse, depending on the geometric dimensions of the low-frequency and high-frequency gratings. The axial low-frequency field at a point 806 on the ζ axis and with an angular frequency ω ₀ = ο1 < ₀ will be equal to

О- / Л η „ηΐ (Ζ) Ειι - —-- = ее ^-1 * ⁰ (“ ι. ^^ + ^ ν ^) / ²⁵¹¹¹ ^ ⁰ ) ^ ”^ ^{+ до} ( ^г )) / ² рр _and where Rc is the radiated pressure on the surface of the grating;

Κΐο (ζ) is the distance 807 from the outer edge of the low-frequency grating to the point 806 on the ζ axis; but

Κ.ι, (ζ) is the distance 808 from the inner edge of the low-frequency grating to point 806.

Phase term represents the average propagation delay

where τ ₁ (ζ) represents the propagation delay from the low-frequency grating to the point 806;

τι, (ζ) represents the propagation delay from the high-frequency grating to the point 806; Κ _{| ο} (ζ) is the distance 809 from the outer edge of the high-frequency grating to point 806.

In addition to the delay in the propagation phase, there will also be a change in the sign of the term with the sine in equation (37), as part of the phase of the term H ₁ (g), which changes in steps of ± π when zero passes through this term with the sine. By reference numeral 810 in FIG. 8b shows the difference in the phase delay between the low-frequency field and the propagation delay of the high-frequency field, which can be expressed as

Δθ / ζ} = ί ~ ^ oL (g) (39) as a function of ζ for a typical geometry with a high-frequency aperture It = 7 mm and with internal and external parts of the low-frequency aperture Ό ₁₁ = 10 mm and _{Ό 1ο} = 15 mm, _respectively .

A sharp drop in the phase delay by π at 811 (ζ ~ 6.6 mm) occurs when the difference propagation phase Δθ ₁ (ζ) = ω ₀ (Κ ₁₁ (ζ) + Κ _1ο (ζ)) / 2ο ₀ between the outer and the inner edges of the low-frequency grating passes through zero, which leads to a change in the sign of the term with sine in equation (37) from -1 to +1. The amplitude of the low-frequency field Η ₁ (ζ) is shown not to scale in logarithmic units by the position number 812, and it should be noted that the amplitude zero coincides with the shift by π in phase. Zeros of the field can be found when the difference in the delay of the propagation phase from the outer and inner edges of the lattice to the point of the field is an odd number π, which gives a jump by π in the expression for Δθ1 (ζ).

For comparison, the position number 813 shows the difference in the propagation phase between the low-frequency and high-frequency gratings, equal to

Δφιίζ) = шо (т _г (г) - -тф)) (40)

It should be noted that Δθ ₁ (ζ) repeats Δφ ₁ (ζ) until it jumps to -π at 811 and repeats this delay with the difference π after that. Due to the large and ζ-dependent delay of the propagation phase between low-frequency and high-frequency gratings, we obtain a ζ-dependent relative position between high-frequency and low-frequency pulses. So, for example, a high-frequency pulse, which usually begins on the upper crest of a low-frequency pulse, in the situation shown by the position number 814 slides to the lower trough of the low-frequency pulse, in the situation shown by the number of the position 815, in which the phase delay Δθ ₁ (ζ) has changed to π, etc.

Substituting pressure

P0 (α) = somDb ¹ ^) (41) into equation (10), whence together with β _η = 5, _0a = 400 · 10 ^-12-1 , s _0a = 1540 m / s and P _{1 £} = 50 kPa, we obtain the nonlinear propagation delay shown by the position number 816 in FIG. 8s

Since the phase delay Δθ1 (ζ) (810) changes by several π along the ζ axis, as shown in FIG. 8b, we obtain a strongly oscillating delay of nonlinear propagation, shown by the position number 816, with a maximum delay equal to m _max = 2 ns. If the delay for this delay of nonlinear propagation is not corrected correctly, we obtain the maximum suppression of the linear backscattered signal equal to 28 dB (<E1T _max ) —18 dB for Г ₁ = 10 MHz.

To construct an image of a contrast medium, a higher low-frequency pressure p _{0 is} usually used, for example, p ₀ ~ 200 kPa, which will give _{max max} = 8 ns in the example above and reduce the suppression of the linearly scattered high-frequency signal by 28 psi (<E1T _max ) —6 dB for 1 = 10 MHz even with a large slip of the phase relationship between the high-frequency and low-frequency components, as is obtained in this example. Consequently, the amplitude of 50 kPa for low-frequency pressure is too small, but in special cases it can provide interesting images of contrast medium and other microbubbles of gas. Thus, it is possible to some extent to amplify the signal from the contrast medium without adjusting for the delay of nonlinear propagation, provided that the gratings are made

- 27 014167 so that there is an oscillating change in the delays of nonlinear propagation, as in the case of 816. It should be noted that when the high-frequency pulse is near zero in the low-frequency oscillation, the effects of nonlinear scattering and direct propagation are small. This effect can be avoided by using different phases between the low-frequency and high-frequency components in successive emitted pulses, which shifts the spatial position in which the high-frequency pulse is near zero low-frequency oscillations. Zeros in amplitude | Ηι (δ) | low-frequency fields can be shifted in space by changing the low-frequency center between successive radiated pulses. In addition, due to the width of the high-frequency beam, the signal is also received outside the axis, where | Η ₁ (§) | ^ 0, and the pulse consists of a frequency band that averages the zero points for many frequencies. Zeros of amplitude can be avoided by reducing the width of the low-frequency grating, which, however, will also reduce the ratio of the pressure amplitude to the control voltage of the grating.

However, in order to distinguish nonlinear scattering from other objects, such as microcalcifications in soft tissue and gas in porous rock, and to obtain the numerical parameters of objects, it is necessary to use higher low-frequency amplitudes, at which correction of nonlinear propagation delays becomes necessary to suppress linear scattering and separation of nonlinear scattering. This is also the case for using the pulse reverb noise reduction method, as described with respect to FIG. 7. In addition, with a weaker oscillating change in the delays of nonlinear propagation, there is a strong advantage of delay adjustment for detecting gas microbubbles even with a small amplitude of a low-frequency pulse.

Often such a design of low-frequency and high-frequency emitting apertures is chosen in order to obtain a minimum delay glide between low-frequency and high-frequency pulses based on maximizing the nonlinear control of the scattering and propagation parameters for the high-frequency pulse along the entire image range. This is necessary to obtain a monotonic increase in the delay of nonlinear propagation, as shown in FIG. 4. To avoid a phase shift of π-810 (and the corresponding zero amplitude), use of low-frequency elements of a smaller width (i.e. a smaller difference between I and I _{11 1 °),} and to reduce the phase slip between the low and high frequency pulses, should pull the low-frequency radiating surface as close as possible and may even be overlapping the high-frequency radiating surface. However, for many applications, one can come to terms with the oscillating behavior of non-linear propagation delays and even use this fact to obtain low non-linear propagation delays, as in FIG. 7c, or change the compression of the pulse length for the near and middle fields to the extension of the pulse length for the deep ranges in order to reduce the frequency and improve the sensitivity, as described in relation to FIG. 2.

In order to reduce the effect of pulse reverberation noise on the estimation of corrections for non-linear propagation delays, as well as in a non-linearly scattered signal, it is possible in the third method according to the invention to emit more than two pulses with more than two different amplitudes and / or phases and / or low-frequency frequencies momentum. As a first example according to the invention, in which there is no movement between the scattering centers and the transducer array (i.e., zero Doppler shift), low-frequency pulses with amplitudes p ₀ , 0, -p ₀ are emitted, in which the high-frequency pulse follows close to the crest or in the cavity of a low-frequency pulse, as, for example, shown in FIG. 3a and 3b. Received high-frequency signals from three pulses in the interval T ₁ can be written as

2 / ι (ί) = χι (ί + η) + _xn (1 + τ.) + Τ (ί) y ₂ (1) = χι (ί) + τ (ί) (42) = Χι (ί- h) - χ _η (ί - Τί) + ί · (ΐ) where χ ₁ (ΐ) is a linearly scattered signal;

χ _Π (ΐ) is a nonlinearly scattered signal with an amplitude p _{0 of the} low-frequency pulse;

t (1) is the reverberation signal for a high-frequency pulse and

-τ ₁ , is the delay of nonlinear propagation for a high-frequency pulse, when it is transmitted on a positive ridge with a pressure p0 of a low-frequency pulse.

According to one embodiment of the invention, we form a combination of these three signals in the form (ί) = Yi (9-Y1 (Y = Yi (£, h) - + Τ;) ζ ₂ (ΐ) = Uz (1) - th (* ) = ~ H, h) ““ h) where x _b (1, τ ₁ ) = χ ₁ (ΐ) -χ ₁ (ΐ + τ ₁ ) is a combination of linearly scattered signals.

- 28 014167

These combinations excluded pulse reverberation noise from ζι (ΐ) and ζ ₂ (ΐ), and since the amplitude of the nonlinearly scattered signal is much lower than the amplitude of the linearly scattered signal, we have approximately ζ ₂ (ΐ) = ζι (ΐ-τ ₁ ), and, therefore, for the interval T [1, it is possible to determine the delay correction τ ₁ by maximizing the following functional:

A = [(// ^ (/ - rC + r ₂ (/) | ² (44) / m, with respect to τ ₁ .

When τ _{1 is} determined, it is possible to calculate the components of the linear and nonlinear scattering signals from equation (43), for example, by the Fourier transform of equation (43) in fast time, which leads to a system of linear equations in Χι. (Ω) and Χ _η (ω )

The essence of the above operation is that three or more pulses with three or more levels of non-linear propagation delay control allow you to combine measurements to obtain at least two signals in which the pulse reverb noise is significantly suppressed, and these new signals can be used to determine delay corrections. In the described operation, the difference in the amplitude of adjacent low-frequency pulses should be constant, providing a non-linear propagation delay for each signal, which is an integer multiple of one delay. With a constant time-constant radial movement between the transducer array and scattering centers, at which there is a constant Doppler delay between the received signals for each pulse, there is also a fourth unknown quantity, which must be determined, which, in turn, requires the emission of at least four pulses with different amplitudes of the low-frequency pulse, or 5 pulses with 3 amplitudes of the low-frequency pulse can be used, as in equation (42), in which by maximizing the power signal in the same manner as in equations (21) and (44) define a combined delay nonlinear propagation and Doppler delays. Non-linear propagation delay and Doppler delay can be separated by operations, as in equations (24) and (25), in which the details of these and similar operations can be worked out in accordance with the invention by any person skilled in the art.

Nevertheless, the quantity τ _{1 itself} must be determined from combinations like equation (44), or from their equivalents in the form of the Fourier transform, and since ζ1 and ζ2 contain both linearly scattered and nonlinearly scattered signal, the latter introduces, although and a very insignificant error in determining the delay correction, which, in turn, introduces an error in the estimation of a nonlinearly scattered signal in the same way as the estimates given by equations (17) - (22).

The main reason for this error is that for stationary objects there are four unknowns:

linear scattering χ / ΐ);

nonlinear scattering χ _η (ΐ);

pulse reverberation noise g (1);

delay τ (ΐ) of nonlinear direct propagation (i.e., stationary objects in the absence of Doppler delay).

Since τ (ΐ) slowly changes with ΐ, this value can be considered approximately constant in the time interval T ₁ , as was shown above. To more accurately determine the four unknowns need to perform at least four measurements from four pressure levels + p0 + p0 / 3, p ₀ / p ₀ 3 and the low frequency pulse, to obtain

2/1 (0 = ДД / + _{г g} ) + „„ (/ + 7,) + g (/)

1/2 (0 = Χιίί + Τ, / 3) + | m ((/ + 7, / 3) + r (0 (45)

Uz (0 = ^ / (/ - l / Z) - \ x _n (1 - g / Z) + g (0

2/4 (0 = x _g (/ - 7,) - Χ _η (ί - 7,) + 7 (/)

Then we can eliminate r (1), as in equation (43), to obtain three signals ζ ₂ (ΐ), ζ ₂ (ΐ) and ζ ₃ (ΐ) and determine τ ₁ to synchronize these signals with maximizing a functional similar to functional in equation (44). The influence of χ _η (ΐ) on the estimate of τ ₁ will be less than in the case of three measurements, but the error introduced by the measurements will still not be completely excluded.

The best approach in accordance with the fourth method of the invention, which eliminates the influence of χ _η (ΐ) on the determination of τ ₁ , is obtained by the Fourier transform of equation (45) on the interval T ₁ , where τ (ΐ) can be approximated by the constant τ ₁ , which provides an overdetermined a system of linear equations in the form

Υ (ω) = Λ (ωτ,) Χ (ω) (46) where Υ (ω) ^Τ = {Υ1 (ω), Υ ₂ (ω), Υ3 (ω), Υ4 (ω)};

Χ (ω) ^τ = {Χι (ω), Χ _π (ω), Κ (ω)}, and

Α (ωτ ₁ ) is a 4x3 matrix obtained from the Fourier transform of equation (43) in accordance with

- 29 014167 with known methods.

Using such an overdetermined system, one can find X (w) that provides a better approximation from the point of view of the least squares method using the pseudoinverse matrix of the matrix Ά (ωτ ₁ ) as

where Α ^Η (ωτ ₁ ) is the Hermitian transposed matrix of the matrix Ά (ωτ ₁ ).

Now τ ₁ can be determined from the error minimization condition when applying the least squares method ίΚ (ω) = Υ - Α (ωτ,) Χ ⁷ »= - A (ogg.) (A ^I (pcs,) A (pcs _g )) 'a ^ unit ^ UY (48) with respect to τ ₁ , which is often called the full least squares method.

This method provides a systematic procedure for using at least four measurements with at least four low-frequency pulse levels to determine all four unknowns, especially non-linear propagation delays with greatly reduced effects of pulse and non-linear signal reverberations, although in many cases the methods described in relation to equations (17) - (22) and (42) - (45) provide adequate results. It should also be noted that the amplitude of low-frequency pulses can be set arbitrarily, so that nonlinear delays for different pulses can be set in the form τ; = ^ τ, where k ₁ is the ratio of low-frequency amplitudes to the control value, and τ is the nonlinear propagation delay with this the reference value, which is determined according to the procedure described above.

We also emphasize that the methods in accordance with this structure provide an estimate of linearly and nonlinearly scattered signals (as well as pulse reverberation noise) with minimal effect on the noise of reverberation of the pulse and, in principle, a complete separation of the definition of linearly and nonlinearly scattered signals. In this method, it is therefore necessary to use a specific signal x1 (1), which is the first linear scattering image signal with a strong suppression of pulse reverb noise, as in equation (14) in the second method, and the third linear scattering image signal, as in equation (29), while the specific signal {x _n 1) is the second image signal from the nonlinear scattering as in equations (19) and (28), both these signals are used subsequently in the calculation of the second numerical parameter nonlin ynogo scattering in equation (30), wherein certain non-linear delay τ is used to determine the first numerical nonlinear parameter in equation (27).

It is possible, as in the previous methods, to add a constant Doppler delay τ, ι between the received signals for each emitted group of pulses, which also gives 5 unknowns, namely Χ ₁ (ω), Χ _η (ω), Κ (ω), τ and τ ₄ . These unknowns can be determined from the received signals from five emitted groups of pulses, for example, with amplitudes (+ p0, + p0 / 2, 0, p0 / 2, p0) of low-frequency pulses and the same amplitude of the high-frequency pulse, by minimizing δΥ in equation (48).

The above discussion outlines how to deal with even more complex cases when it is necessary to determine a larger number of unknowns. For example, there may be uncertainty in the amplitudes of the low frequency pulses emitted, for example, as + p0 + ₀ A1P / 3, s ₂ r _0/3, s ₃ r ₀ for equation (45), where the vector a = _(ax, and _2a , ₃ ) represents unknown changes in the amplitudes. These uncertainties affect both the amplitude of the nonlinear scattering and the delays in equation (45); it should be noted that in this case we get seven unknowns, namely Χ ₁ (ω), Χ _η (ω), Κ (ω), τ, И1, and ₂ and a ₃ , which means that we do not have enough equations obtained from measurements to adequately determine τ while minimizing δΥ in equation (48). Therefore, it is necessary to add new, more thorough measurements, so that the number of uncertainties in the amplitude is minimized. For example, by careful measurements can receive only one indefinite amplitude and with five pulses emitted at low frequency pulse amplitudes equal to (p + ₀ + ₀ ap / 2, 0, ₀ ap / 2, p ₀₎ which gives five unknown Χ ₁ (ω), Χ _η (ω), Κ (ω), τ and a, i.e. the same number as the number of equations, while minimizing δΥ in equation (48) will give all five unknowns in the absence of Doppler delay.

It should be noted that the third and fourth methods, as well as the first and second methods, use the second harmonic band of the received signal to process and generate the image signal, while they have the advantage of even better suppression of pulse reverb noise in the image signals, but due to the smaller range images for the same frequencies. The devices, as described below, are therefore configured to select between the first and second harmonic bands of the received signals for processing in accordance with the present invention.

Nonlinear control of the low-frequency pulse by the parameters of scattering and propagation of the high-frequency pulse can give favorable results when reconstructing an image in computed tomography based on measurements of transmission and angular scattering of an object, for example, by measurements shown in FIG. 9a. In FIG. 9a shows a cross section of an object enclosed in an annular array 902 of transducers with transducer elements 903 mounted around the entire object. The acoustic coupling medium 904 may be, for example, water or other fluid. A group of 905 elements is arbitrarily selected among the entire group for emitting an acoustic beam 906, the direction of which through the object indicated by the unit vector e, can be scanned in all directions through the object by selecting different groups of radiating elements from the entire group of elements. In the forward direction e, the pulse hits the annular lattice at point 907 with the delay of direct propagation equal to some modification of equation (10) ίο / (β,) = / _um

(49) where G Yes; represents the path of direct propagation along the axis of the beam through the entire diameter

2a lattice.

The propagation delay at zero amplitude of the low-frequency pulse is ί ₀ £, when the nonlinear delay caused by the low-frequency pulse is τ. The received signal at point 907 will be an emitted pulse, followed by a tail of multiple reflected pulses. However, with reference to FIG. Figure 7 shows that a multiple-reflected signal has a much shorter non-linear delay time τ and can be significantly suppressed by combining at least two received signals with oscillations in the amplitude and / or phase and / or frequency of the low-frequency radiated pulse along the lines described above . When the high frequency pulse is located on the negative spatial gradient of the low frequency pulse, as shown in FIG. 1, the co-propagation of these pulses creates a high-frequency pulse compression, as described in relation to this drawing, while the suppression of the reverberation noise of the pulse can then be performed by frequency filtering, preferably with slip, as described in relation to FIG. 2. For the deepest penetration, the high-frequency and low-frequency radiating surfaces can be made such that the position of the high-frequency pulse with respect to the low-frequency pulse slides to obtain an expansion of the pulse with a reduced frequency, as described with reference to FIG. 2. Examples of such radiating surfaces with an analysis of the effects that result from sliding are shown in FIG. 7 and in the reasoning corresponding to this drawing.

In accordance with the reconstruction methods using computed tomography, you can use the delay C in the absence of a low-frequency pulse to find the propagation speed c _0a (g), where r is the spatial coordinate in the plane. Adding a low-frequency pulse and measuring the delay τ of non-linear propagation provides reconstruction in _pa (g) to _0a (g). Using the amplitude of a directly propagating pulse, one can also reconstruct the spatial change in absorption in the object. Moving the scan plane in the vertical direction then allows you to get a 30-image of the object. All these reconstructed images will have a spatial resolution limited by the width of the emitted beam, since the delay in the momentum and amplitude is averaged over the entire beam, while the model described in equation (49) assumes that all delays are tied to the beam axis. A similar model is also used for absorption.

Improved resolution can also be obtained using an angularly scattered signal in methods called diffraction tomography, reflective tomography, backscatter tomography, and the like. On elements, for example, indicated by reference numeral 908 in FIG. 9a, b, in the direction at an angle to the direction of the radiated beam, denoted by a unit vector e ₈ passing from the center of the grating, signals scattered at an angle will be observed as a function of time from different depths along the radiated beam. A high-frequency pulse, which is scattered from a depth r at a point 909 along the emitted beam, will first propagate along with a low-frequency pulse of large amplitude along the path Г shown in FIG. 9b from point 910 to point 909, and then from point 909 to point 908 along path D along with a low-frequency pulse of small amplitude due to a decrease in amplitude during scattering. The delay time of the entire propagation from point 910 to point 908 takes the form

Pr, ^ e ^) = eC + m (r, e,) k (r, e, _b ) = / _{Γ / (Γ!} ,, _{| + Γ} , _(Ch . _B) ά (r + ^ 6 + 2 «(r + <·) (! + £, &)) (50) + £.) - Λ, where in the last expression for ίο an assumption was made of a constant propagation speed with _0a along both paths Γ and Γ ₈ .

As described with respect to FIG. 7, the low-frequency pulse has a smaller effect on the multiply scattered high-frequency signal, so that by combining two or more high-frequency signals with different amplitudes and / or phases and / or frequencies of the low-frequency pulse, it is possible to significantly

- 31 014167 degrees suppress the reverb noise of the pulse. Similarly, in order to suppress pulse reverberation noise, it is possible to use glide in frequency of a high frequency pulse by a low frequency pulse, as described with reference to FIG. 1 and 2, and it is also possible to use the second harmonic band of the received signals together with methods 1-4 described above to further suppress the pulse reverberation noise and determine the nonlinearly scattered signal and the numerical parameters of propagation and scattering as the basis for reconstruction of the tomographic image. It will also be apparent to those skilled in the art that the arrangement of the transducer elements is different from the particular case shown in FIG. 9 can be used to obtain the same result of computed tomography reconstructions.

Delay corrections that maximize the functional of C, such as shown in equation (21) and similar to equation (44), can be (for both the first and second harmonic bands of the received signals, as well as for the full received signals), for example, found by differentiating equations (21)

where the total number of delays is K-1, see the discussion in connection with equation (17). Due to the Hermitian symmetry K _k1 ', this equation can be transformed into

where Ke is the real part of the sum.

Note that 5χ ₁ (ΐ + τ _ιρ ) / 5τ _ιρ = 5χ ₁ (ΐ + τ _ιρ ) / 5ΐ, we can rewrite equation (52) in the form

where ⁾ ζ ₁₁ (ΐ) is the signal from equation (17) with a delay correction of t _1k .

Since there is no analytical expression for K _k1 ', delays that satisfy this system of equations must be found numerically, for example, according to the following iterative scheme:

where s. | indicates the iteration step number; and μ is the gain that is chosen to achieve convergence of the iterative process. Minimization of 1 _Ш in equation (22) with respect to Η _ιρ can be performed by equating the derivatives to zero, i.e.

Due to Hermitian symmetry ⁾ Ν this equation can be transformed into the following equation for eigenvectors:

The minimum of the functional 1 Hz is then given by the eigenvector with the smallest eigenvalue λ ₁ . Substituting this expression for the correlation matrix in equation (22), we can transform equation (56) to the form

^where) ζ _ηι ^(ΐ) - signal from the equation (19), adjusting the delay t _1k.

The amplitude correction is thus found from the correlation between the signals of the RF element and the delay-adjusted amplitude of the RF signal at the output of the high-pass filter, where _{1k is} present. Equation (57) can then be solved in an iteration procedure, for example, as

- 32 014167

* = 0

In the above analysis of the signal used in a complex analytical representation that can be obtained from the received RF signal by using the Hilbert transform, as in equation (11), approximations of the Hilbert transform are given in equation (74). It is also obvious to a person skilled in the art that the above analysis can be modified to act on a received radio frequency signal as a real part of the analytical signal.

From the procedures of equations (54) and (58) it follows that it is natural to start an estimate for the interval T ₀ at the least delay with a further estimate at successive intervals. Then, certain filter amplitudes Η _ιρ with delay corrections τ _ιρ for the previous interval are used as initial values in the iterative procedure for each new interval, starting with τ0ρ, 0 = 0 and Η _ιρ = 1, as initial values for the first interval. An additional advantage of using the values determined in the previous interval as initial values is that when the jump in the phase ω ₁ δτ _ίρ associated with the delay is less than ± π / ω ₁ between each interval, the delays can be tracked using with phases | ωιτ _ιρ | >> π in the entire depth range.

When the signal bandwidth is small enough and the delay is less than the center frequency period, the following approximations can be applied:

⁹ "w [x _k (ί)

α) (59)

δ) 'hell _{(. £ + τ (ί)} «+ τ.ξ °) where in equation (56) we divided delay correction as a = 8t, t _k +% (60) where τ ^ - score or approximation τ ^, for example, we will use τ ^ = далее below.

Equation (59b) therefore represents the combination of the phase delay with the phase ωηδτ, v .. and the actual delay τ _{1 |} % as an estimate of τ _{1 |} . Improved accuracy of this approximation is obtained by filtering the signals with a band pass filter near ω1 to reduce the bandwidth of the value ⁾ x _k (!). Substituting the approximations from equations (59) into equations (53) and (54), we obtain

Equating the last expression to zero, we can calculate δτ _ιρ as

where Ζ {} denotes the phase angle of the complex expression.

This leads to the following iterative equation for amplitude and delay corrections:

in which the iteration is usually started with the values τ _ιρ , ₀ = τ _1- ι, _ρ with т _0р , ₀ = 0 as the initial values for the first interval.

Using the approximation in equation (59), it is possible to convert the power maximization in equations (17) and (21) into the eigenvalue problem. In this case, we use the delay separation in equation (60), which gives

The functional in equation (21) is converted to

- 33 014167

Maximizing C with respect to the amplitude and phase of 8 _1k gives 8 _1k as a solution to the following eigenvalue problem:

(66) for the largest eigenvalue λ ₁ .

The maximum of functional C is found using the best possible synchronization of signal components, which ensures the finding of optimal delay corrections from the phases of the components of the eigenvectors. The components of the vectors s ₁ defined in equation (64) have a module equal to unity, while the components of the eigenvectors in equation (66) generally have modules different from unity, since the value of the functional C can be further increased by prescribing different weights different signal components. The phases of the components of the eigenvectors in this case give delay corrections that maximize the power in equations (17) and (64). It should be noted that the matrix ^ K ¹ is Hermitian, i.e. ⁾ К1т ¹ = ( ⁾ Кт1 ¹ ) *. It follows that the eigenvalues are real, and the eigenvectors form a complete orthogonal basis for C _k , a complex K-dimensional space.

The approximation accuracy in equation (59) and, therefore, finding the solution of eigenvectors in equation (66) are improved by reducing the signal bandwidth. The Fourier transform of the partial summed signals into ω ₁ provides such a strong bandpass filtering that the differentiation in equation (59a) and phase delay in equation (59b) become accurate. However, in order to avoid the case where the phase delay exceeds the fundamental range ± π, it is useful to separate the delays, as in equation (60), to obtain% (ω) = (67)

After that, you can determine the delay by maximizing the functionality

A, = pr _and (w)] ² - λ, | i _1k | ² = - λ, 1.5, + ²

Maximization leads to the problem of eigenvalues for each frequency, similar to equation (66), in which delay corrections are found from the phases of 8 _1k values. Since the phase delay for the Fourier transform is an accurate representation of the true delay, equation (67) is accurate, in contrast to equation (64), in which the phase delay is taken only at the center frequency.

The main part of the analysis performed above is based on the availability of complex analytical signals from radio frequency elements and on the further processing of these signals to obtain adjusted received signals. The calculations can be simplified further, and it should be noted that the phases in equation (62) can be calculated as

(69) / _n άί (He {\ (/ + d-v, ρ)) 1 ^ (4 (/)} + 1m {^ _p (/ + 7 -, _ 1. _P )} 1m {\ r ₂ , (one)})

Integration over T ₁ is an estimate of the following mathematical expectations:

where \ 7 denotes ensemble averaging; and A ₁ takes the following forms:

Α, (ί) = + 7 -, _ _1ιΡ )} 1t {^ ι, (ί)} Α, (ί) = 1t { ^} a: p (/ + τ, _1ιΡ)} Ηθ { ⁾ ζ1, (ί )}

Α (ί) = NeTs.Gr) '/ _r + 7- _{_]; P)} VE} {';!, (/; )} Α, (ί) = + τ, _ 1ιΡ)} 1t {\ , (£)} (71)

The approximation in equation (70) becomes equality as T ^ is yes, whereas for finite values of T ₁ there is a random error in estimating ensemble mean values, which is the reason for the sign of the approximate equality in equation (70). In signal processing textbooks, you can find that

For the final values of T _1, random estimation errors lead to the following approximations:

(73)

- 34 014167

It should be noted further that

Ηβ ^ ζ / ί)} = x _p (1)

Not {^, (2)} = ζ ^ ί)

Ιιη { ⁵ ίΓρ (ί)} - # {; Ερ (ί)} «χ _ρ (ί - π / 2ωι)

1m { ⁾ r ₁₁ (/)} = "2 _{£ ι} ( _ί - π / 2ωι) (74) where the approximation of the Hilbert transform by means of signals delayed by π / 2ω _£ is satisfactory for narrow-band signals centered at ω1.

By combining equations (73) and (74) we obtain simplified expressions, _7m . ~ ¹ 4. _о „-1 Д ^ Лх _p ^ + г _г - _1р ) Н { _г1 Д)} _ 1; _{N <& χ ρ (ί + τ} , _ι, ρ) 2 _{£ ι (ΐ - π / 2ωι} ) ~~ E0L1 'μ ^-' · ^- ~ g ~ · * 4ζ · 'E £ 1C p. . E2 * -Ά αΤϊ ^ ΐ— (ί) I

In this expression, the Hilbert transform or its approximation of the delay acts only on the adjusted output of the high-pass filter and, therefore, should be applied to only one signal, simplifying the whole procedure. A similar expression can be obtained from the last part of equation (73), but here the Hilbert transform or its delay approximation should be applied to the received signals χ _ρ (ΐ), which requires much more processing. Some reduction in the variance of the estimate can be obtained by combining the first and last expressions in equation (73) in accordance with equation (69).

Also note the following ratio:

^ Be {\ p (T)} 1t {C (C}) = (Ke {^ (r)} 1t {x ₍ ^% (0) ^ = - <7t {t ^)} Not {r, ^% (0 } ^ which allows you to replace the above actions on analytical signals with actions on complex envelopes of signals, where their real and imaginary parts are found as active and reactive components of the output signals of a standard quadrature demodulator, as described above.

With constant mechanical scanning of the acoustic beam, it is also of interest to use low-pass filters and high-pass filters with an infinite impulse response (IIR), in which an illustrative first-order low-pass filter can be described as

where T is the filter time constant with a cutoff frequency < _c = 1 / T;

the subscript k denotes the pulse number, as before, and the image signal ζ (ΐ) is updated for each new emitted pulse.

The same structure of the equation is true both for the analytical signal and for its complex envelope.

Thus, a recurrence scheme for estimating delay corrections of t _1k for each new emitted beam has now been developed. We assume that m 1 _m are given for m up to k-1. With a new radiated beam with number k with the received signal x _k (1), then it is necessary to determine the delay correction t _1k . This estimate is usually performed by minimizing the functional, where the approximation of the analytical signal is used for simplicity

Differentiation with respect to m _1k gives « _{= 2 ЛКе} {^ lm ' _{ad (()} p _{0 (79)}

Substituting this approximation into Eq. (59), we obtain ~ -AcSm / _Τι + τ, -ιλ) + ф1х _к (1 + т _11к )} (80)

Since the last integrand is real, equality in the above expression holds when

Higher-order high-pass filters can be described in a similar way with a spatial representation of the filter vector states known to any person skilled in the art. Also, any person skilled in the art can perform a recursive determination

- 35 014167 division t _1k in accordance with the procedure described above.

For proper suppression of a linearly scattered signal, it is necessary to have a delay accuracy that is much smaller than the sampling interval of the received RF signal in most acoustic imaging systems. This accuracy can be obtained by interpolating the signal between samples in accordance with known methods. Often, but less appropriate accuracy can be obtained by combining delay correction with the accuracy of the closest sample 1+ ⁾ t _1k and phase correction ω ^ τ ^ by analogy with equation (59) + т _гк ) т ^) (82) ^where § ^т ] k = ^T 1k ^-T 1k.

The radio frequency signal is a real part of equation (82).

The first method in accordance with the invention described in connection with FIG. 1 and 2 can be implemented as a variant of the receiving filter, which is present in most modern digital scanners, or filtering at the reception can be performed in the way of processing the radio frequency signal. The block diagram of the processing node for determining the correction of the amplitude and / or delay in accordance with the second or third methods according to the invention, together with the image parameters in accordance with this invention, as defined in equations (14) - (30), (42) - ( 44), shown in FIG. 10. The received radio frequency signals x _to (1), indicated by the position number 1001, from the successive emitted pulse groups are fed to the high-speed filter 1002 of the slow time to give a linear image signal 1003 with suppressed reverb, in accordance with equation (14), as well as to the memory unit 1004 to provide greater versatility in signal processing. The received signals are then sent to the amplitude and / or delay corrector determiner 1005, which operates in accordance with one of the methods presented above or similar in order to obtain accurate estimates of the amplitude and / or nonlinear propagation delay correction 1006. The obtained corrections of the amplitude and / or delay of nonlinear propagation are received in the correction block 1007, which receives signals from the memory block and generates corrections of the amplitude and / or delay of nonlinear propagation. The corrected signals then go to the block.

1008, which extracts the corrected linear signal 1009, for example, in accordance with equations (17) and (18), or directly solves equation (42), or the like, and to block 1010, which extracts the corrected non-linearly scattered signal 1011, for example, in accordance with equations (19) and (20), either a direct solution to equation (42), or similar. It should be noted that when methods similar to Eqs. (42) - (44) are used for estimation, the effect of pulse reverberation on the determination of delay correction is minimal, as it is also minimal on the determination of linear and nonlinear scattering signals. According to the reasoning above, you can also use the second harmonic band of the received signals to determine the delay of nonlinear propagation, as well as for image signals. The corresponding blocks on the block diagram should then contain filters to isolate the second harmonic band, or the input signals themselves should be signals of the second harmonic band.

Linear 1009 and non-linear 1011 delay-corrected signals along with certain delay corrections 1006 are supplied to a numerical parameter determination unit 1012 in which one or more numerical non-linear propagation parameter 1015 is calculated in accordance with equation (27) or a similar and one or more numerical parameter 1016 nonlinear scattering in accordance with equation (30) or similar, as well as the coefficient a (d) 1017 of local acoustic absorption in accordance with equation (36). The delay correction block 1005 may also represent Doppler phases 1013, for example, according to equation (24), which enter the Doppler block 1014, which calculates the radial displacement 1018 of the scattering centers, the radial velocity 1019 of the scattering centers, the radial deformation of the displacement 1020 of the scattering centers and the radial speed 1021 deformation of the displacement of the scattering centers or calculates other parameters. Signals 1003,

1009, 1011, 1015, 1016, 1017, 1018, 1019, 1020 and 1021 are then typically served for further processing and displayed to create complete acoustic images in accordance with known methods.

The fourth method in accordance with the invention, described in connection with equations (45) to (48), operates more in batch mode to evaluate corrections for non-linear propagation delays and linearly and nonlinearly scattered signals. The procedure for such an assessment can be represented by the block diagram in FIG. 11, where the position number 1101 denotes the incoming measured signals x _to (1), which enter the block 1102 correction and determination of the amplitude and / or delay, the output of which is a linearly scattered signal χ ₁ (обознач), indicated by the position number 1103, nonlinear scattered signal x _p (1). indicated by the position number 1104, and the corrections found for the delays τ (нелиней) of the nonlinear propagation and amplitude oscillations a (1), indicated by the position number 1105. In connection with this method, it should be noted that the pulse reverberations and nonlinear scattering signals have a minimal effect on the determination of delay corrections although there are certain advantages to using the second harmonic band of the received signals, as discussed above in connection with equations (42) - (44). The found values of χ ₁ (ΐ), χ _η (ΐ), τ (ΐ) and a (1) then go to block 1106 to determine

- 36 014167 numerical parameters, which gives one or more nonlinear propagation parameters 1015, nonlinear scattering parameter 1016, local absorption parameter 1017, radial displacement 1018 of scattering centers, radial velocity 1019 of scattering centers, radial strain 1020 of scattering center displacement and radial strain rate 1021 of deformation displacement of scattering centers, or calculates other parameters.

The signal processing units can then be applied both to tomographic reconstruction methods based on transmission and angular scatter measurements, as shown in FIG. 9, and for backscatter imaging devices. A block diagram of a backscatter imaging device in the broadest sense in accordance with the invention is shown in FIG. 12, where reference number 1201 denotes an acoustic transducer array comprising high-frequency (HF) and low-frequency (LF) sections. In this widest application of methods, the lattice has a two-dimensional distribution of elements, which allows for full electronic 3B control of high-frequency and low-frequency beams, which is hereinafter referred to as the 2B lattice, and the device is also capable of both determining and correcting wavefront aberrations. However, it is understood that the methods can also be used with less complex lattices, as described below.

The high-frequency part of the grating in applications of full 3B images can have a large number of elements, for example, from 3000 to 10000, and the number of channels for receiving and emitting is usually reduced in subaperture block 1202, where in the reception mode the signals from several neighboring grating elements are delayed and summed with the signals 1203 subapertures for further processing. To correct aberration, the widths of subaperture groups on the surface of the lattice are less than the correlation length of wavefront aberrations, where the usual number of subapertures and signals can be from 100 to 1000.

To emit groups of pulses, the RF emitted beam former 1204 supplies pulses to the subaperture block 1202, which delays and distributes the signals to all elements or subgroups of the RF grating elements, while the low frequency emitted beam former 1205 simultaneously delivers pulses to the low-frequency array elements. Pulse group radiation is triggered by a device controller 1206, which handles subunits via the device bus 1207.

The received subaperture signals 1203 arrive at block 1208, in which the subaperture signals are delayed to control the direction of the received beam and focus under the assumption of a homogeneous medium with a constant average propagation velocity, which is hereinafter referred to as uniform delays. 3B control and focusing can also be performed for sparse gratings, with subaperture block 1202 usually being excluded. For 1.75B gratings, the number of HF grating elements can be so reduced that subaperture blocks can be eliminated. In the future, therefore, we will use the concepts of element signals and subaperture signals as synonyms.

Signals 1209 of the elements, which are corrected by uniform delays, are fed to block 1210, where corrections for wavefront aberrations are also applied, for example, determined as described in equations (33) and (34), or in accordance with the methods described in US patents No. 6485423, 6905465 and in the application for US patent No. 10/89438, before the signals of the elements are summed into the final signal of the received beam. To construct 3B images, numerous received beams with small angular deviations, which parallel cover a wide emitted beam, are usually used. Aberration corrections for beams with angular deviations can be laterally offset correction cases for the central beam, which are added together with uniform delays for angular deviation in block 1210.

The output 1211 of block 1210 is therefore one or more radio frequency signals for one or more parallel directions of the received beam, which enters the processing block 1212 in accordance with the present invention, which, in turn, performs one or more operations in accordance with FIG. 2, 10 and 11. It should be noted that for the operation in accordance with FIG. 2, the high-frequency pulse over the vast majority of the propagation distance is located on the negative spatial gradient of the low-frequency pressure oscillations, whereas for the methods described in connection with FIG. 10 and 11, a high-frequency pulse on the vast majority of the propagation distance is located near the crest or trough of low-frequency pressure fluctuations.

Aberration corrections are determined in block 1213, for example, according to the methods described in connection with the cited patents and patent applications, and possibly also using methods based on equations (33) and (34). Block 1213 takes as input the signals 1209 with homogeneous delay corrections, and possibly also the final beam signals 1214 with noise reverberation noise reduction in accordance with the invention. Delay correction element signals 1209 are usually first processed using the methods of the invention, typically using the method described in connection with FIG. 2 or equation (14) to suppress the pulse reverb noise before determining the delay corrections. It should be noted that the use of a signal from moving scattering centers, such as, for example, takes place for blood or myocardium and as proposed in US Pat. No. 6,485,423, improves the effect of methods for suppressing pulse reverberation noise, such as

- 37 014167 described in relation to equation (14). Estimates based on the nonlinear propagation delays for individual element / subaperture signals given by equations (33) and (34) also represent interesting estimates in themselves, and also serve as a starting point for further estimates in accordance with the cited patents, as for focusing 1 of the emitted beam, as well as in the iterative scheme.

When the estimate of the corrections for the wavefront aberration is based on the correlations of the signal with the output signal of the beam former 1214 with strongly suppressed reverberation noise, the reverberation noise in the element signals is uncorrelated with respect to the output of the beam former. If slow updates to aberration correction estimates are acceptable, long correlation times can be used such that the effect of the reverberation noise in the element signals on the correction estimates can be negligible. However, the correlation time is usually so short that it is preferable to also suppress the reverberation noise in the element signals before performing aberration correction estimates.

The output signals of block 1212 are linearly and nonlinearly scattered signals, two numerical nonlinear parameters, and also Doppler phase and frequency data, as described with respect to FIG. 10 and 11. This data can be fed directly to the imaging unit 1216 and to a scan converter, which presents images in compressed and colored form of amplitudes of linearly and nonlinearly scattered signals, numerical nonlinear parameters / signals, radial displacements of the object, velocities, displacement deformations and displacement strain rates based on the output signals given in FIG. 10 and 11. However, to measure the radial velocities of blood or gas bubbles or other fluids, it is necessary to further process linearly or nonlinearly scattered signals in the slow time domain to suppress the spurious echo from the object due to interfering reflections in order to isolate the signals of the fluids for Doppler processing in in accordance with known methods, which is done by block 1215. The output signals from this block are sent to block 1216 imaging to be selected and superimposed on images with other information iey. Block 1216 provides its output signals to a display device 1217.

It should be apparent to any person skilled in the art that the numerous simplifications of the device illustrated in FIG. 12, although using essential aspects of the invention in a device. For example, it is possible to have a rough separation of the elements in the vertical direction, which limits the electronic control of the beam direction in this direction, although corrections for wavefront aberrations and dynamic focusing with depth in the vertical direction can still be obtained. Such gratings are often called 1.750 gratings, which have a smaller number of total number of grating elements than 2O gratings for full 3O beam control, as a result of which the subaperture block can be eliminated. Sparse gratings are another way of eliminating a number of elements, so it becomes practical to exclude subaperture block 1202. However, there is a gain from using a subaperture block if the size of the subaperture group along the lattice surface is less than the correlation length of the wavefront aberrations.

You can also exclude estimates and corrections for wavefront aberrations, i.e., blocks 1210 and 1213, and nevertheless be able to perform processing in block 1212 to obtain both linearly and nonlinearly scattered signals, etc., as described above. The lattice can be further simplified when elements symmetrical with respect to the scanning axis of the beam (azimuth axis) are galvanically combined to further halve the number of independent channels, and such a lattice is often called a 1.5 O-lattice. Similarly, one-dimensional (III) gratings can be used, as well as ring gratings with mechanical scanning of the beam direction, with the only change in the block diagram shown in FIG. 12 will be the exclusion of subaperture block 1212, aberration correction block 1210, and aberration correction estimation block 1213.

Thus, since it is desirable to have devices with different complexity or with a choice of complexity, for best efficiency, depending on the measuring situation, it is desirable to have devices that are capable of choosing between different processing methods described above. The numbering of the methods described above corresponds to the increasing complexity of the methods with the consequent increase in the number of pulses required for a single radial line of the image, and, consequently, an increase in the time taken to build one image, which is the reciprocal of the frame rate.

The advantages and disadvantages of these methods are as follows.

1. In a first method in accordance with the invention described in relation to FIG. 1 and 2, the result is obtained using a single group of emitted pulses and suppresses the noise of the reverberation of the pulse with the sensitivity of the first harmonic by means of radio-frequency filtering of the received signal in fast time in block 1212 shown in FIG. 12. Signals with suppressed reverb are further processed according to known methods for obtaining a structural image of an object and constructing a Doppler image of moving scattering centers and fluids, as well as radial strain of displacement and strain rates of relative displacement

- 38 014167 of the scattering center in blocks 1215 and 1216. This method does not provide non-linear scattering parameters or non-linear propagation delay parameters, but gives the highest frame rate of all methods.

2. In a second method in accordance with the invention described in relation to FIG. 3-7 and equations (10) - (41), the result is obtained using two or more groups of emitted pulses for each radial image line with changes in the frequency and / or phase and / or amplitude of the low-frequency pulse for each emitted group of pulses. By combining the received signals from several pulses, the first image signal with the suppression of pulse reverberation noise with the sensitivity of the first harmonic is obtained, an estimate of the nonlinear propagation delay, which gives a second image signal giving information about nonlinear scattering from the object, microcalcifications and microbubbles, as well as the first and second numerical nonlinear image parameters. When three or more pulses are emitted, Doppler information is also obtained according to equation (24), which is extremely useful in studying the radial displacement and velocity, as well as the radial deformation of displacement and strain rates of object structures, such as myocardium. The processed first and second image signals can be used to construct an image of the amplitude of the structures of the object in block 1216 and to construct a Doppler image of moving scattering centers with interference noise filtering in block 1215. For amplitude imaging of a tissue, the method provides a lower frame rate than the first method, described above, since it is necessary to radiate two or more groups of pulses for each radial line of the image, whereas when building a Doppler image, images of deformations ation displacement and deformation rate frame rate of the two methods are comparable. When the first harmonic signals are used to determine non-linear propagation delays, delay estimates will show errors caused by pulse reverberation noise, while with less sensitivity, non-linear propagation delays can be determined from the second harmonic component of the received signals, which has suppressed reverb noise, or it can be reduced reverb noise in another way to get estimates of non-linear propagation delays with less errors caused by roar noise rberatsii. Nonlinear scattering also produces small errors in the delay estimates, which affect the accuracy of the determination of the second image signal and numerical nonlinear image parameters. However, these errors only minimally reduce the suppression of the linearly scattered signal during the formation of the second image signal, which is an important result. The present invention also notes the basic principles of the design of two-way acoustic transducer arrays, which create oscillatory changes with depth in the phase of the emitted low-frequency pulse with respect to the high-frequency pulse, to minimize the maximum non-linear propagation delays, so that for low amplitudes (~ 50 kPa) of the low-frequency pulse, approximately determine the second non-linear image signals without corrections for the delay of non-linear propagation.

3. In the third method described in relation to equations (42) to (44), three or more emitted pulse groups with at least three levels of frequencies and / or phases and / or amplitudes of the low-frequency pulse are used to obtain one processed signal. The method removes the pulse reverberation noise before further estimating the non-linear propagation delays and obtains estimates of the linearly and non-linearly scattered signals with strong suppression of the pulse reverberation noise. The first non-linear numerical parameters / image signals, equation (27), and the second non-linear numerical parameters / image signals, equation (30), are obtained, as in the second method. When using four or more pulses, it is also possible to estimate the Doppler delay between signals from successive pulses, which is constant for each estimated interval T ₁ , as in equation (24) for the second method. This Doppler delay is extremely useful for studying radial displacement and velocity, radial deformation of displacement, and strain rates of an object such as myocardium. To build a Doppler image of the velocities of the scattering centers, it is necessary to have a filter to suppress interfering reflections to remove interference from the object, and processing is performed in block 1215, as in other methods. Nevertheless, the nonlinearly scattered signal introduces small errors in the determination of the nonlinear delay, which affect the accuracy of determining the second nonlinearly scattered image signal, but compared to the second method, these errors do not reduce the suppression of the linearly scattered signal during the formation of the second nonlinear image signal, which is important result. This method gives a lower frame rate compared to the second method.

4. In the fourth method described in relation to equations (45) to (48), four or more emitted pulse groups with four or more frequency levels and / or phases and / or amplitudes of the low-frequency pulse are used to obtain one processed linear signal 1st order scattering, 1st order nonlinear scattering and nonlinear propagation delays. The first non-linear numerical parameters / image signals, equation (27), and the second non-linear numerical parameters / image signals, equation (30), are obtained, as in the second method. When using five or more pulses with five changes in frequency, and / or phase, and / or amplitude, low

- 39 014167 frequency pulses, it is also possible to evaluate errors in the phases and / or amplitudes of the low-frequency pulse and / or Doppler delays between the five emitted pulses. This Doppler delay is extremely useful for studying radial displacement and velocity, radial deformation of displacement, and strain rates of an object such as myocardium. To build a Doppler image of the velocities of the scattering centers, it is necessary to have a filter to suppress interfering reflections to remove interference from the object, and processing is performed in block 1215, as in other methods. In this case, the pulse reverberation noise and nonlinear scattering only minimally affect the determination of nonlinear propagation delays, thus providing the most accurate determination of nonlinear propagation delays, linearly and nonlinearly scattered signals due to the lowest frame rate of all four methods.

The invention provides a device that can work with at least two of these methods, with the possibility of choosing the best method depending on the need, and the choice can be made with direct control of the operator or the operator can set restrictions, while the device automatically selects methods for the best work efficiency in accordance with these restrictions for different working conditions.

An example of a restriction set by the operator can be the minimum frame rate, where for small depth ranges at which a high pulse repetition rate can be used, the method with the highest numbering from those listed above can be applied, which nonetheless satisfies the frame rate limit for obtaining better work efficiency with the required frame rate. For large depth ranges, at which the pulse repetition rate should be reduced, the device selects one of the remaining methods, which still satisfies the frame rate limit, although with a worse quality of determination. Another example of a limitation is the combination of the frame rate and the quality of determination, in which, as the range increases for medium ranges, the quality decreases at a constant frame rate, and for large depth ranges, the frame rate drops, while the quality is maintained.

The choice of the method can also automatically depend on the means of imaging, and for constructing linear object images of the heart, the first method is used with image signals with suppressed reverberation for the highest frame rate, while when examining movements in the myocardium, the device can switch in the second method from 2-4 groups of emitted pulses to each radial image line using equation (24) for myocardial movement. In order to construct an image of the velocities of scattering centers, the device can switch in the second method from 8-16 groups of emitted pulses to each radial image line using the processing in block 1215. For immovable objects like the prostate gland, mammary glands, liver, etc., usually you can choose the fourth method for the best possible determination of linearly and nonlinearly scattered signals of the first order, non-linear propagation delays and numerical image parameters.

For the purpose of tomographic reconstruction, the signal processing in accordance with the present invention is usually performed on separate received elemental signals before the signals are processed according to different types of reconstruction algorithms, moreover, a block diagram of a typical tomographic image reconstruction device in accordance with the invention is shown in FIG. 13. The drawing shows measurements using an annular array 1301, and it will be apparent to those skilled in the art that other array designs and transducer arrays that use mechanical scanning in whole or in part to collect data can also be used without departing from the spirit of the invention. The lattice surrounds the object 1302. Block 1303 from all elements arbitrarily selects a group of emitting elements and creates a group of emitted pulses consisting of a low-frequency and a high-frequency pulse overlapping in time, which, for example, is depicted in FIG. 1 and 3. The radiation of the pulse groups is triggered by the control unit 1307 via the control bus 1308. The unit 1304 selects the receiving elements from the entire group of elements in series or in parallel or in a series-parallel combination, and amplifies and digitizes the signals of the elements for further processing according to the invention in block 1305 This unit operates in accordance with the principles of the invention, for example, as described in connection with FIG. 1 and 2 for a single group of pulses per processed signal or in connection with FIG. 10 or 11 for multiple emitted pulses per processed signal. Signal processing in block 1305 provides one or more linearly scattered and radiated signals with significant suppression of pulse reverberation noise (multiple scattering), nonlinearly scattered signals, and numerical parameters of nonlinear propagation and scattering, which are transmitted to block 1206 that produces computer tomographic images of 2E layers of the object . By mechanically moving the grating relative to the object in a direction perpendicular to the drawing, a 3Ό reconstructed image of the object can be obtained.

- 40 014167

An example of a device for using methods of obtaining an acoustic image of geological structures around an oil well is shown in FIG. 14. Reference numeral 1401 shows a shot oil production casing with surrounding geological structures 1402, which typically consist of porous rock filled with oil, gas, water, or a mixture thereof, although hard rock areas can be found. Reference numeral 1403 shows an acoustic array for emitting and receiving groups of acoustic pulses according to the invention in sectors of beams 1404 that are selectable around a production casing. The received acoustic signals are processed in accordance with the methods described above with the illustrative devices shown in FIG. 10-12. When the oil wells are located close enough to each other, transmission measurements between the oil wells can be used and images from the transmission measurements and the angle measurements can be reconstructed as described above, in particular with respect to FIG. 9 and 13.

Thus, as the main new features of the invention have been shown, described and noted, as applied to the preferred embodiments, it should be understood that various exceptions, replacements and changes in the shape and details of the devices shown can be made by those skilled in the art as well as in principles of their work without departing from the essence of the invention.

It is also intentionally assumed that all combinations of those elements and steps of the method that perform essentially the same function, essentially the same way to achieve the same results, fall within the scope of the invention. In addition, it should be understood that the structures and / or elements and / or steps of the method shown and / or described in connection with any disclosed form or embodiment of the invention can be included in any other disclosed, or described, or proposed form or an embodiment as a basic principle when choosing a design. Thus, the invention is intentionally limited by the gist of the claims below.

In the claims we will use the following concepts.

The received signals are xSC signals). first introduced before equation (11), or their analytical form defined in equation (11), or their complex envelope defined in equations (12) and (13).

Fast and slow time or pulse coordinate number are determined with respect to FIG. 5.

Slow time filtering or filtering along the pulse coordinate number is determined with respect to FIG. 6.

The pulse reverb noise is determined with respect to FIG. 7.

The construction of tomographic reconstructed images is determined with respect to FIG. nine.

A nonlinearly scattered signal is a nonlinearly scattered signal from a high-frequency pulse with strongly suppressed linear scattering components defined in equations (9), (11) (13), (19), (42) - (48), or its analytical form, or its complex envelope defined similarly to x _k (!).

A linearly scattered signal is a received signal from linear scattering of a high-frequency pulse in an object, defined in equations (9), (11) - (13), (17), (42) - (48), or its analytical form, or its complex envelope defined similarly to x _k (1).

The image signal with suppressed reverb, or the first image signal defined in equations (14) and (42) to (48) or in relation to FIG. 2.

Non-linear propagation delays are defined with respect to equation (10).

The total propagation delays are the sum of the non-linear propagation delays and the Doppler shift delays (Doppler delays) defined in relation to equation (23).

Delay or correction corrections for non-linear propagation delays or total propagation delays are defined in equations (10), (17) - (25) and (49), (50).

The second image signal is defined in equations (19), (28) and (42) - (48) as a nonlinearly scattered signal.

The third image signal is defined in equations (17), (29) and (42) - (48) as a linearly scattered signal.

Amplitude corrections are defined for equations (19) and (22).

Estimated intervals T ₁ defined in relation to equations (21) and (44).

The first numerical nonlinear image parameter, or nonlinear propagation parameter, is defined in equation (27).

A second numerical non-linear image parameter, or non-linear scattering parameter, is defined in equation (30).

A recurrent procedure is a computational procedure that repeats several stages and in which parameter values are updated in the procedure at each stage; defined in relation to equations (54), (58), (63) and (81).

An iterative procedure is the same as a recursive procedure.

- 41 014167

References [1] Björn A., J. Angelsen. Building ultrasound images. - Waves, Signals, and Signal Processing, Volume 2, Emantek, Trondheim, Norway, Dec. 2000.

Claims (59)

CLAIM 1. A method of measuring or constructing an image of at least one of the acoustic properties of scattering and propagation in an object using a plurality of radial measurement or image lines, and for each radial measurement or image line in the method: a) emit beams of at least one group of acoustic pulses towards the object, wherein said at least one group of pulses consists of at least one high-frequency (HF) pulse and at least one low-frequency (LF) pulse and with the same or overlapping beam directions; b) receive the RF signals using the array of RF converters from at least one acoustic wave obtained by scattering of at least one RF pulse that has passed through the object; and c) generating measurement signals or images from the received RF signals in the process, using at least one LF pulse to create non-linear control of the elastic properties of the object observed by the at least one HF pulse in order to suppress the worsening effect on the measurement or signal construction due to pulse reverberation due to the multiple scattering of said at least one RF pulse. 2. The method according to claim 1, in which one or both of the following are performed: a) the specified RF pulse at least in part of the range of image depths propagate on a negative spatial gradient of the waveform of the specified RF pulse and b) the beams of the indicated HF and LF pulses are positioned so that the phase relationship between the HF and LF pulses changes so much with depth that in the same range the specified HF pulse propagates along the zero or positive spatial gradient of the waveform of the indicated LF pulse, as a result which results in improved resolution of the indicated RF pulse at deeper ranges and improved frequency separation of the noise of the reverberation of the pulse and the scattered signal of the first order of the RF pulse. 3. The method according to claim 2, in which the received RF signals are filtered in the fast-time space (depth time) by means of a filter that suppresses at least lower frequencies and in which at least the lower cut-off frequency of the filter changes from depth to obtain the first signal, which is a high-frequency signal linearly scattered from the object with a significant suppression of pulse reverberation noise at each image depth for further processing to form measurement or image signals. 4. The method according to claim 1, in which at least two groups of pulses emit towards the specified object with the same or overlapping beam directions for each image line and in which the frequency, and / or phase, and / or amplitude of the LF pulse change for each emitted group of pulses in the process of forming the measurement signals or image. 5. The method according to claim 4, in which at the stage of forming the measurement signals or image determine at least one of the following signals from the RF received signals: a first signal, which is a received signal from linearly scattered RF pulses with significant suppression of pulse reverb noise; a second signal, which is a received signal from non-linearly scattered RF pulses with significant suppression of the linearly scattered signal; and a third signal, which is a received signal from linearly scattered RF pulses with the same gain, varying with depth and acoustic absorption as the 2nd signal. 6. The method according to claim 5, in which the first signal is isolated from the HF received signals in a process that includes filtering steps along the pulse coordinate number (slow time) to suppress the low-frequency components of the slow time and pass the high-frequency components of the slow time. 7. A method of measuring or constructing an image of nonlinear acoustic scattering from a region in an object in which: a) emit at least two groups of acoustic pulses in the direction of the region of the object, each of the at least two groups of pulses consisting of at least one high-frequency (HF) pulse and at least one low-frequency (LF) pulse and with those the same or overlapping directions of the beams, while in each of the indicated at least two groups of pulses at least one of the parameters: the frequency, amplitude or phase of the specified at least one LF pulse varies; b) receive RF signals using an array of RF converters, which picks up at least - 42 014167 at least one of the acoustic waves generated as a result of scattering at least one RF pulse from the object and acoustic waves of said at least one RF pulse that passed through the object; c) evaluate the difference in the total propagation delay between the RF pulses based on the RF received signals, and adjust the received signals from at least two RF pulses with the estimated differences in the total propagation delay taking into account the RF delay and g) form a second signal, which is nonlinear scattering at the object with strong suppression of the linearly scattered signal at the object by combining the RF received signals, adjusted for the delay, and by using the specified at least one LF pulse to create a nonlinear control of the elastic properties of the object observed by the specified at least one RF pulse. 8. The method according to claim 1 or 7, in which high-frequency signals that were scattered at an angle from the object and / or passed through the object are used to reconstruct the tomographic image of the acoustic properties of the object. 9. The method according to claim 2 or 5, in which high-frequency signals that were scattered at an angle from the object and / or passed through the object are processed to significantly suppress the noise of the reverberation of the pulse before they are used for tomographic reconstructions of the image of the acoustic properties of the object. 10. The method according to claim 4 or 7, in which the variable total propagation delays with pulse-by-pulse accumulation, as the sum of the Doppler delays between groups of pulses and non-linear propagation delays caused by non-linear control of the low-frequency pulse by the frequency of propagation of the high-frequency pulse, is determined from at least the received signals from two groups of pulses, while certain total propagation delays are used in the process of generating measurement or image signals. 11. The method according to claim 10, in which the delay of nonlinear propagation is determined directly from one of the following parameters: a) Doppler delays that are zero due to the lack of movement between the scattering centers and the array of transducers; and b) at least three emitted groups of pulses with different amplitudes of the low-frequency pulse, and both non-linear propagation delays and Doppler delays are separately determined from the RF received signals, while certain Doppler delays and / or non-linear propagation delays are used in the process of generating measurement signals or Images. 12. The method of claim 10, in which the differences in the total propagation delay are estimated based on minimizing the power in the measurement or image signals, which is nonlinear scattering at the object. 13. The method according to claims 5, 9 or 10, in which the delay of said HF received signals is corrected with one of said specific total propagation delays and non-linear propagation delays to generate HF received signals with delay adjustments, and said second signal, which is a non-linear estimate the scattered signal, which is the local parameters of the nonlinear scattering of the object, is isolated in the process, which includes the steps of combining these HF received signals with delay adjustments along the the ordinates of the pulse number (slow time) to suppress the low-frequency components of the slow time of the specified RF received signals with delay adjustments. 14. The method according to claim 7 or 13, in which, in addition to the specified RF delay corrections, the received signals are also subjected to amplitude corrections for the maximum suppression of the linearly scattered signal during the formation of the specified second signal. 15. The method of claim 14, wherein said amplitude corrections are determined from minimizing power in said second signal with a restriction of a fixed norm of the amplitude correction vector. 16. The method according to item 12 or 15, which additionally uses visual observation of power in the measurement or image signals as feedback for manually adjusting at least one of: estimates of amplitude corrections and estimates of differences in propagation delay at different depths along the beam for minimize power in the measurement signals or image, which is a nonlinear scattering on the object. 17. The method according to claim 7 or 11, in which the indicated differences in the delay of nonlinear propagation are estimated from a calculation for several depths using a pressure-dependent variation of the propagation velocity determined by the formula c (s) = co (1 + βηΚΡίρ) and their dependence _depending on the pressure p LF of the LF pulse at the location of the HF pulse as a function of depth and the estimate of the pressure of the LF pulse at the location of the HF pulse for several depths with the selected material parameters β ,, κ and c ₀ at several depths are additionally adjusted - 43 014167 manually to improve the suppression of the linearly scattered signal. 18. The method of claim 11, wherein the non-linear parameter or image signal, which is a local parameter or non-linear direct propagation image signal defining local non-linear propagation parameters in an object, is formed from the difference along the fast time of the estimated non-linear propagation delays. 19. The method according to p, in which the first numerical non-linear parameter or image signal, which is a local parameter or image signal of non-linear direct propagation, which is the non-linear propagation parameters of the object, is formed as a combination of the derivative of these specific non-linear propagation delays along a fast time and how an estimate of the local pressure amplitude of said at least one LF pulse at the location of said at least one HF pulse. 20. The method according to claims 5 or 7 and 10 or 11, in which the delay of said HF received signals is corrected with one of said specific total propagation delays and non-linear propagation delays for generating HF received signals with delay adjustments, and said third signal, which is by evaluating a linearly scattered RF signal, with the same gain varying with depth and acoustic absorption as the second signal, which is a nonlinearly scattered signal, is isolated in a process that includes the steps of said signals of the connections of delay as adjusted along the pulse number coordinate (slow time) for passing slow time frequency components around zero and suppressing other frequency slow time components. 21. The method according to claims 5 or 7 and 20, wherein the non-linear parameter or image signal, which is a local parameter or non-linear scattering image signal, which determines the local non-linear scattering parameters of an object, is formed by combining the envelope of the second signal and the envelope of the third signal. 22. The method according to item 21, in which the second numerical non-linear parameter or image signal, which is a numerical parameter or non-linear scattering image signal, representing the local non-linear scattering parameters of the object, is formed by combining the envelope of the specified second signal, the envelope of the specified third signal and estimating the local the pressure amplitudes of the at least one LF pulse at a jointly propagating at least one HF pulse. 23. The method according to claim 11, in which of these specific Doppler delays evaluate one or more of the following parameters: radial displacement of the object along the beam direction as a function of depth along the beam; the radial velocity of the displacement of the object along the direction of the beam as a function of depth along the beam; the radial mechanical deformation of the object along the direction of the beam is estimated from the derivative of the estimated estimated displacement along the depth range and the radial speed of the mechanical deformation of the object along the direction of the beam is estimated from the derivative of the estimated estimated displacement along the depth range. 24. The method according to PP.19 and 22, in which one or both of the first and second numerical non-linear parameters / image signals are used to evaluate either the relative volume of microvessels in the object, or perfusion of the fluid through the object, or both. 25. The method according to claim 3 or 5, in which the first signal with the suppression of reverberation noise is used in the process of evaluating the correction of wavefront aberrations. 26. The method according A.25, in which the array of RF converters has a two-dimensional distribution of elements and in which at least one of the signals: signals received from elements that are RF received signals from elements of the array; and signals received from subapertures, which are a combination of RF received signals from groups of subapertures of adjacent lattice elements, are processed according to claim 3 or 5 to obtain new RF signals of the element or subaperture with significant suppression of pulse reverberation noise, said new high-frequency signals of the element or subapertures are used in the estimation of corrections for wavefront aberrations. 27. The method according to claims 26 and 11, wherein the nonlinear propagation delays are determined for the indicated signals of the element or subaperture and are used in the process of evaluating corrections for wavefront aberrations. 28. The method according to claim 3, or 5, or 7, in which wide HF and LF beams are emitted, covering numerous parallel HF received beams to increase the frame rate of the image when constructing 2Ό and 30 images, and in which at least at least one of the indicated 1-, 2- and 3rd signals is obtained for each of the indicated parallel received beams. 29. The method according A.25 or 28, in which the correction of aberration for one direction of the emitted beam is estimated using a highly focused RF beam, after which a wide RF and - 44 014167 LF beams with multiple parallel HF received beams to increase the frame rate of 2Ό and 30 images using aberration corrections for each HF received beam, calculated from the estimated aberration corrections obtained using focused HF radiated beams. 30. The method according to claim 10, in which the total time interval T for signal reception is divided into sub-intervals T ₁ so short that the total propagation delays can be approximated by a constant in each sub-interval, and these full propagation delays are determined in the process of maximizing the power in each the sub-interval of the signal, which is obtained by correcting the delay of the specified RF received signals with the specified specific total propagation delays and by low-pass filtering of the RF received signal a delay correction in the slow time coordinate. 31. The method according to PP.5, 10 or 11, in which the RF received signals from a number of groups of emitted pulses are first combined to form a series of new RF signals with noise reduction of the reverberation of the pulse, and this series of new RF signals is used to estimate these propagation delays and one or all of a linearly scattered high-frequency signal and a non-linearly scattered RF signal with strong suppression of pulse reverberation noise, wherein said first and third signals are set equal to said estimated linearly scattered yannomu RF signal and said second signal is set equal to said linearly scattered HF signal. 32. The method according to claims 5 and 11, wherein at least five groups of pulses with different amplitudes of the low-frequency pulse are emitted, and the measured high-frequency received signals are obtained for each emitted group of pulses, wherein the linearly scattered signal, the nonlinearly scattered signal and the indicated delays non-linear propagation and Doppler delays are determined from the indicated measured HF received signals in a procedure in which: a) the measured RF received signals are approximated by a signal model, which is a combination of linearly scattered signals, non-linearly scattered signals and pulse reverberation noise, the specified signal model being determined by the delay parameters, which are non-linear propagation delays and Doppler delays of the measured RF received signals; and b) estimates of linearly scattered signals, non-linearly scattered signals and pulse reverberation noise are determined as RF signals ensuring continuity, in the sense of determining the signal model to the measured signals, for c) estimating the delay of nonlinear propagation and Doppler delays obtained as delay parameters, which minimize the error between the indicated signal model and the indicated measured HF received signals, and the first and third signals are set equal to the specified estimated linearly eyannomu signal and said second signal is set equal to the estimated said nonlinearly scattered signal. 33. The method according to p, in which the estimates of linearly scattered signals, non-linearly scattered signals and noise reverberation of the pulse are found as estimates that provide the best continuity of the specified signal model to the specified measured signals in the sense of least squares. 34. The method according to p, in which for known or negligible Doppler delays between the emitted groups of pulses emit one less, i.e. at least four groups of acoustic pulses with predetermined characteristics, and in which only the delay of nonlinear propagation is evaluated. 35. The method according to claim 19 or 22, in which at least one of the first and second numerical non-linear parameters / image signals is used in the process of tracking the local temperature of the object during its heat treatment. 36. The method according to claim 5 or 7, in which at least one of the 1st, 2nd and 3rd signals is used to detect or image objects with high compliance, including at least one of: fat or lipid particles, gas bubbles, fish or marine animals or gas in geological structures. 37. The method according to claim 1 or 7, in which in the formation of image signals using the second harmonic band of the received signal. 38. The method of claim 10, wherein a second harmonic band of received signals is used to estimate the total propagation delays. 39. The method according to claim 7, or 13, or 31, or 32, or 34, in which the specified non-linearly scattered signal is used to detect and / or image objects with high compliance, including at least one of: fat or lipid particles , gas bubbles, fish or marine animals or gas in geological structures. 40. The method according to claim 7, or 13, or 31, or 32, or 34, wherein said nonlinearly scattered signal is used to detect and / or image structures in softer objects with low compliance, including at least one of: microcalcifications, high density connective tissue in soft tissue or deposits in geological structures. 41. The method according to claim 1 or 7, in which the specified object is a neighborhood of an oil well. 42. The method according to claim 1 or 7, in which the specified object is a fish or sea animal. - 45 014167 43. The method according to claim 7, or 13, or 31, or 32, or 34, wherein said nonlinearly scattered signal is used to differentiate between rocks containing oil, gas or water near an oil well. 44. The method according to claim 19 or 22, in which at least one of the first and second numerical non-linear parameters / image signals is used to obtain numerical values of one of the following: the relative volume of gas or oil in the porous rock; changes in the volume of gas or oil in the porous rock and the size and number of fish with a swimming bladder or marine animals with lungs in the water. 45. The method according to claim 1 or 7, in which the harmonic components of the LF pulse in the received HF band are suppressed by emitting the LF pulse without an HF pulse, the received signal received from this LF pulse, the received signal is subtracted, and the indicated stored HF received signal is subtracted from the HF received signals from groups of emitted pulses that contain both low and high frequency pulses. 46. The method according to claim 11, in which the local absorption coefficient in the object is determined from the specified specific delay of non-linear propagation, the gradient in depth from the specified specific delay of non-linear propagation, and the estimate of the upper center frequency in the RF received signal and the gradient in depth from the specified estimate of the central frequency, with the HF pulse, a significant part of the depth range propagates along the negative spatial gradient of the LF pulse. 47. The method according to claim 2, in which the RF pulses for observation are used to observe the deformation of the object due to the emitting force of the acoustic pulses that are pressed and in which the indicated pulses for monitoring a significant part of the depth range propagate along the negative spatial gradient of the LF pulse to set the conversion up in frequency from the compression of the pulse of these RF pulses for observation; these emitting force pressing pulses emit as: a) an RF push pulse without a low frequency pulse or b) as a sequence of HF pushing pulses along with LF pulses, where a significant part of the depth range of said HF pushing pulses propagate along the positive spatial gradient of said LF pulses to convert downstream of said HF pushing pulses so that the received signals overlap from these pulses for observation and these pressure impulses can be separated by filtering by time depth ( three times). 48. The method according to claim 1 or 7, in which to increase the harmonic distortion of the RF pulse, the specified RF pulse, at least in the near depth range, is located near the positive pressure peak of the LF pulse in the emitted pulse group, so that the maximum amplitude of the RF pulse can be increased as much as possible due to the mechanical index of the group. 49. The method according to claim 1 or 7, in which the processing includes the steps of suppressing the components of the LF pulse in the HF received signal to extract the HF received signal for further processing and in which the suppression of these components of the LF pulse is performed in the filter. 50. A method of constructing an image of the acoustic properties of nonlinear scattering in the region of an object in which a sequence of at least two groups of acoustic pulses is emitted to a specified region, said groups of pulses consisting of a high-frequency (HF) pulse and a low-frequency (LF) pulse, overlapping in time , and with the same or overlapping beam directions; and between the emitted groups of pulses, the frequency and / or phase and / or amplitude of the emitted LF pulses are changed relative to the indicated HF pulses to provide non-linear control of the acoustic elasticity of the indicated object, as observed by the indicated HF pulses, which vary from pulse to pulse; and said LF and HF pulses are generated by separate arrays of acoustic transducers with spaced apart radiating surfaces, so that the phase of these LF pulses changes relative to the phase of these HF pulses in the entire actual image range to create a delay in the nonlinear propagation of these high-frequency pulses by these LF -pulses that exhibit a nonmonotonic change along the axis of the HF beam, which limits the maximum delay of nonlinear propagation so that for low amplitudes of the LF pulse (~ 50 kPa), the HF signal linearly scattered from the tissue can be suppressed and the nonlinearly scattered HF signal can be estimated through the combination of the coordinate number of the pulse of the HF received signals from at least two pulses without corrections for the delay of nonlinear propagation in quick time. 51. The method according to item 50, in which the specified array of converters for the RF pulse is - 46 014167 is a linear or curved grating, and the specified lattice of the converters for the low frequency pulse represents one grating of a linear or curved grating installed on one side of the specified RF grating, and two linear or curved gratings, each of which is installed on each side of the specified RF trellis. 52. An acoustic device for constructing an image of an object region, comprising: a) an emitting transducer configured to emit at least one acoustic pulse, consisting of a high-frequency (HF) and low-frequency (LF) pulse with overlapping beam directions; b) a receiving transducer configured to receive the RF received signals obtained by scattering in the object of the RF pulse and / or acoustic waves of the RF pulse that has passed through the object; and c) processing means for processing the RF received signal, said processing providing measurement signals or images from said RF received signals in processes in which the noise of the reverberation of the pulse due to multiple scattering of said RF pulse is suppressed by non-linear manipulation of the RF pulse by the properties the elasticity of the object, as observed by the RF pulse. 53. A device for measuring or imaging the acoustic properties of nonlinear scattering and / or propagation in an object, comprising: a) means for emitting groups of acoustic pulses consisting of at least one high-frequency (HF) pulse and at least one low-frequency (LF) pulse with the same or overlapping beam directions; b) an RF converter and reception means configured to receive RF received signals from at least the scattered waves received from said at least one RF pulse; c) processing means for processing RF received signals from at least two emitted pulse groups, and for each of said at least two groups of pulses, at least one of the parameters: frequency, amplitude and phase of said at least one LF- momentum is changing; and j) processing means, further comprising means for estimating non-linear propagation delays between the RF received signal, for adjusting for the delays of the RF received signals and for combining the received signals from at least two pulse groups corrected for the delays of the RF to obtain measurement or image signals, which are nonlinear scattering of the emitted RF pulses from the object, by using the specified at least one LF pulse to obtain a nonlinear control , the elastic properties of the object as observed by said at least one rf pulse. 54. The device according to paragraph 52 or 53, in which the processing means evaluates the measurement signals or images, which are at least one of the following parameters: linear scattering from the object; nonlinear scattering from the object; local parameter of nonlinear propagation in the object; local parameter of nonlinear scattering from the object; a numerical local parameter of nonlinear propagation in the object and a local parameter of nonlinear scattering from the object. 55. The device according to paragraph 52 or 53, in which the processing method is selected by the controller of the device for the best performance under the restrictions that are established in the manufacture of the device or by the operator. 56. The device according to paragraph 52 or 53, in which: a) the emitting transducer is designed to emit a wide beam; b) a receiving transducer is configured to receive an RF signal from multiple parallel RF received beams that are contained in a wide radiated beam; and c) the processing means is arranged for parallel processing of the HF received signals of their indicated parallel HF received signals so that an increased image frame rate is obtained for constructing 2Ό- and 3B-images. 57. The device according to paragraph 52 or 53, in which: a) said radiating transducer and receiving transducer comprise a transducer array with a two-dimensional distribution of the array elements; and b) the specified receiving transducer is designed to redirect received from the individual elements of the lattice or subapertures of the elements of the lattice of the RF signal to the specified processing means; and c) the specified processing means is designed to evaluate the wavefront aberration corrections from the indicated RF signals of the element or subaperture and contains j) the means for applying the specified wavefront aberration corrections to the indicated RF signals of the elements or subapertures before they are summed to form the received RF beam signals; and e) means for applying the indicated wavefront aberration corrections to the signals emitted from the specified radiating - 47 014167 converters. 58. The device according to paragraph 52 or 53, in which the processing means creates a reconstruction of the tomographic image from the RF received signals. 59. The device according to § 58, in which the processing means bases the reconstruction of the tomographic image on the estimates of the signals, which are at least one of the following parameters: linear scattering from the object; nonlinear scattering from the object; local parameter of nonlinear propagation in the object; local parameter of nonlinear scattering from the object; a numerical local parameter of nonlinear propagation in the object and a local parameter of nonlinear scattering from the object.