Ultrasonic system and method for generating morphing functions and elastic properties of moving media using full waveform inversion - Patents.com

Abstract

A system and method for producing an ultrasound image of a moving medium is provided, the method including the steps of transmitting a plurality of ultrasound waves by a portion of a plurality of emitters, the plurality of ultrasound waves being emitted toward the moving medium, until a motion of the moving medium below a predetermined threshold is determined, receiving a first plurality of received signals corresponding to the plurality of ultrasound waves by a portion of a plurality of sensors, generating a model of a morphing function when the motion of the moving medium is below a predetermined threshold, the motion of the moving medium being determined based on at least one received signal of the first plurality of received signals, the morphing function defining a morphing of the moving medium from the initial plurality of received signals, and performing a waveform inversion on the model of the morphing function.

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of U.S. Provisional Application No. 63 / 281,981, filed November 22, 2021, the contents of which are incorporated herein by reference.

[0002] The present disclosure relates generally to ultrasound imaging systems, and more particularly to full waveform inversion processing, and even more particularly to their use in conjunction with ultrasound on body tissue. [Background technology]

[0003] Inversion problems aim to compute, from a set of observed signals, the causal factors that caused them. It is the inverse of the forward problem, which starts with the causes and computes the effect. Inversion tries to find the best model that fits the acquired data. In the case of subsurface exploration of the Earth and medical ultrasound, (ultrasonic) waves are emitted from an emitter and sensed by sensors placed on the surface of the medium. Inversion algorithms can be used to generate high-resolution models of the elastic properties and attenuation of the medium. This accuracy and resolution is obtained by utilizing waveform data acquired by multiple simple sensors spread over a large area, rather than relying solely on the returning echoes of a highly localized sensor array.

[0004] Seismic inversion typically refers to a set of algorithms applied to construct models of the Earth's interior elastic properties by processing measured seismic wave data records. Most algorithms are based on simulating the seismic pressure wavefield at the receiver locations. The simulated data is matched against the measured data and data misfits are inverted to update the model of the elastic properties. Algorithms differ in the way they simulate the wavefield, the wave modes selected for inversion, and how they modify the model of the elastic properties. Algorithms for wave propagation simulation generally refer to ray tracing theory (or ray bundles), Kirchhoff or Born modeling approximations, and one- or two-way wave simulation.

[0005] Current waveform inversion algorithms rely on the medium being static in time during data collection. However, for various applications, this assumption does not hold and there are some time dependencies introduced by short-scale motion that cannot and should not be ignored. For example, a patient undergoing a medical examination may continue to breathe, move slightly, or twist, which can potentially cause inaccuracies in the readings. In the field of photogrammetry and x-ray correction, algorithms exist that correct for motion by image processing techniques. Such correction is performed by finding features in the image, constructing a morphing function that describes the motion, and finally correcting the image. Current solutions are based on image processing techniques that require an image initially (or initially). Note that the raw data is not an image for the inversion problem. A morphing function is needed to construct the image, but such a morphing function does not exist at the time.

[0006] In one field of endeavor, a certain type of motion during ultrasound imaging is proposed. In such a case, for example, a first ultrasound image is taken, followed by a second ultrasound image. The data of the two images are processed to determine the overlap region. Preferably, image registration of the images with each other is performed using rigid registration. Then, a non-rigid deformation between the two images is performed, for example, using an elastic registration algorithm. A clear image in motion can be achieved by applying alpha-morphing to each overlap region of each image using non-rigid deformation, and obtaining a blend region of the overlap region by performing alpha-blending. The alpha-blending uses the result of the alpha-morphing to blend the two images by changing the relative contribution of the two image volumes. However, this limits ultrasound motion modes that require scanning, i.e., the motion of an emitter / sensor on a surface, and further requires the initial presence of an image. Such a problem remains unsolved, since the practical effects of the motion may be detrimental to the generation of a usable ultrasound image in general, specifically a first reference ultrasound image.

[0007] Another area of ​​endeavor is the use of ultrasound signals to provide images from body tissues and internal organs. The use of ultrasound transmitters and receivers requires a certain degree of technical ability by the user of the transponder to transmit and receive ultrasound. The covered area is small, and when a larger body area needs to be covered, the transponder needs to be moved back and forth on the surface of the body. Such a moving operation allows for the scanning of a larger area and also takes into account the movement of the transponder as well as the change in the position of the emitter and the sensor.

[0008] However, challenges remain because conventional full waveform inversion (FWI) is not adapted to motion at speeds that are on the order of motion of bodily organs or tissues. Although the analysis may be acceptable when the object (or medium) is stationary or nearly stationary, for various applications this assumption of stationary or nearly stationary does not hold. Time-dependence of motion on short scales that cannot be ignored is detected. For example, a patient undergoing a medical examination is still breathing or twisting slightly, and such bodily changes cannot be properly taken into account. DISCLOSURE OF THEINVENTION [Problem to be solved by the invention]

[0009] It would therefore be advantageous to provide a solution that overcomes the shortcomings of conventional FWI that does not rely on processing of images.

[0010] <Summary of the Invention> A summary of some exemplary embodiments of the present disclosure is provided below. This summary is provided for the convenience of the reader to provide a basic understanding of such embodiments, but does not completely define the breadth of the present disclosure. This summary is not a comprehensive overview of all contemplated embodiments, nor does it identify key or critical elements of all embodiments, nor delineate the scope of any or all aspects. Its sole purpose is to present some concepts of one or more embodiments in a simplified form as a prelude to the more detailed description that is presented later. For convenience, the terms "some embodiments" or "specific embodiments" may be used herein to refer to a single embodiment or multiple embodiments of the present disclosure. [Means for solving the problem]

[0011] Certain embodiments disclosed herein include a method for generating an ultrasound image of a moving medium by generating a morphing function, the method including the steps of emitting a plurality of ultrasonic waves by a portion of a plurality of emitters, the plurality of ultrasonic waves being emitted toward the moving medium, until a motion of the moving medium is determined to be below a predetermined threshold, receiving a first plurality of received signals corresponding to the plurality of ultrasonic waves by a portion of a plurality of sensors, generating a model of a morphing function when the motion of the moving medium is below a predetermined threshold, the motion of the moving medium being determined based on at least one received signal of the first plurality of received signals, the morphing function defining a morphing of the moving medium from an initial plurality of received signals, and performing a waveform inversion on the model of the morphing function.

[0012] Certain embodiments disclosed herein also include a non-transitory computer-readable medium stored on a computer that causes a processing circuit to perform a process including emitting a plurality of ultrasonic waves by a portion of a plurality of emitters until a motion of the moving medium below a predetermined threshold is determined, where the plurality of ultrasonic waves are emitted toward the moving medium; receiving a first plurality of received signals corresponding to the plurality of ultrasonic waves by a portion of the plurality of sensors; generating a model of a morphing function when the motion of the moving medium is below a predetermined threshold, where the motion of the moving medium is determined based on at least one received signal of the first plurality of received signals, the morphing function defining a morphing of the moving medium from the initial plurality of received signals; and performing waveform inversion on the model of the morphing function.

[0013] Certain embodiments disclosed herein also include a system for generating an ultrasound image of a moving medium by generating a morphing function, the system comprising a processing circuit and a memory, the memory including instructions communicatively connected to the processing circuit when executed by the processing circuit to configure the system to: emit a plurality of ultrasound waves toward the moving medium by a portion of the plurality of emitters until a motion of the moving medium below a predetermined threshold is determined, the plurality of ultrasound waves being emitted toward the moving medium, receive a first plurality of received signals by a portion of the plurality of sensors corresponding to the plurality of ultrasound waves, generate a model of the morphing function when the motion of the moving medium is below a predetermined threshold, the motion of the moving medium being determined based on at least one received signal of the first plurality of received signals, the morphing function defining a morphing of the moving medium from an initial plurality of received signals, and perform a waveform inversion on the model of the morphing function. [Brief description of the drawings]

[0014] The subject matter disclosed herein is particularly pointed out and distinctly claimed in the claims appended hereto. The foregoing and other objects, features, and advantages of the disclosed embodiments will become apparent from the following detailed description taken in conjunction with the accompanying drawings.

[0015] [Figure 1] FIG. 1 is a system for performing full waveform inversion in a moving medium, according to one embodiment.

[0016] [Diagram 2] FIG. 2 is a flow chart for performing acquisition, modeling, and waveform inversion to create an ultrasound image of a moving medium, according to one embodiment.

[0017] [Diagram 3] FIG. 3 is a flow chart for performing the acquisition phase of waveform inversion in a moving medium, according to one embodiment.

[0018] [Figure 4] FIG. 4 is a flow chart for generating models and morphing functions for waveform inversion in moving media according to one embodiment.

[0019] [Diagram 5] FIG. 5 is a flow chart for performing full waveform inversion (FWI) in a moving medium, according to one embodiment.

[0020] [Figure 6] FIG. 6 is a flow chart for performing reverse time migration (RTM) in moving media according to one embodiment.

[0021] [Figure 7] FIG. 7 is a flow chart for performing least-squares migration (LSM) in moving media according to one embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] <Detailed Description> It is important to note that the embodiments disclosed herein are merely examples of the many advantageous uses of the innovative teachings herein. In general, statements made in the specification of this application do not necessarily limit any of the various claimed embodiments. Moreover, some statements may apply to some inventive features but not to others. In general, unless otherwise stated, singular elements may be in the plural and vice versa without loss of generality. In the drawings, like numbers refer to like parts throughout the several views.

[0023] For the generation of an ultrasound image of a moving medium, an ultrasound system adapted to generate a morphing function is used. A plurality of emitters and further a plurality of sensors are both sparsely arranged on a surface applied around the moving medium. A group of emitters is repeatedly emitted to capture ultrasound waves captured by a group of sensors. An initial model of a morphing function and elastic properties is generated to account for the change in the motion of the medium over time. During the inversion process, the morphing function and elastic properties are continuously refined until such time that their accuracy exceeds a predefined accuracy threshold before the generation of the ultrasound image. The morphing function and elastic properties may be used in the generation process of the ultrasound image. To this end, the disclosed embodiments consider the motion in the medium and improve the ultrasound imaging. It is noted that the motion in the medium is a small motion, such as, but not limited to, bowel movement, blood vessel expansion, rib case protuberance, and any combination thereof. Furthermore, refinement of the model and morphing functions allows creation of the elastic property model prior to generation of the ultrasound image, eliminating the need for the ultrasound image to perform the morphing analysis.

[0024] According to the disclosed embodiment, some of the components, such as emitters, sensors, etc., are activated to emit and receive ultrasound waves in conjunction with a moving medium. By activating some of the components, measurements of the same area can be obtained for validation and improved accuracy. Furthermore, the process of generating the model and morphing function is performed upon detection of small motion below a predefined threshold. That is, to reduce the processing burden of the system, processing may not be performed until an acceptable range of data indicating small movements is received. It is noted that such partial and / or conditional activation of components provides a means for power and resource conservation in various components of the ultrasound system, such as, but not limited to, emitters, sensors, and controllers.

[0025] The disclosed embodiments described herein apply to various methods that employ full-wave propagation (e.g., FWI and its variants, least squares migration (LSM), one-way or two-way wave equation migration, reverse time migration (RTM), other seismic migration variants, etc.), but are primarily demonstrated with respect to FWI. Those skilled in the art will appreciate that migration variants include, but are not limited to, e.g., least squares migration (LSM), reverse time migration (RTM), etc., and inversion variants include, but are not limited to, e.g., FWI, adaptive wave equation (AWI), physically based deep neural network (DNN) inversion, and other similar variants. However, these examples should not be considered as limiting the scope of the disclosed embodiments.

[0026] FWI is a method for generating high-resolution elastic models by the application of ultrasound. Accuracy is obtained by using the full waveform acquired by multiple simple sensors spread over a large area, rather than only the return echo of a highly localized sensor array. Given an initial model estimate and emitter signal, the FWI method solves the wave equation to find the expected signal at multiple sensors for the model. The model is then iteratively updated to reduce the misfit between the calculated signal and the actually acquired signal. According to the disclosed embodiment, FWI is modified to adapt to non-static media over the time of data acquisition. That is, in one embodiment, a morphing function is generated and incorporated into the FWI to account for the motion in the medium. Furthermore, the morphing function is continuously refined by iterative modification of the model of the morphing function. It is noted that the modified FWI refines the model of the morphing function, the sound speed model, and the attenuation model prior to the generation of the ultrasound image for improved accuracy without the need for prior images.

[0027] FIG. 1 illustrates an exemplary system 100 for performing full waveform inversion (FWI) in a moving medium, according to one embodiment. The system 100 includes an ultrasound sensing garment (USG) 110 having embedded sensors 112 (sensors 112-1 through 112-i, where "i" is an integer greater than "1"), emitters 113 (emitters 113-1 through 113-j, where "j" is an integer greater than "1"), and markers 114 (markers 114-1 through 114-k, where "k" is an integer greater than "1"), according to one embodiment. Exemplary embodiments are described in further detail in International Application No. PCT / IB2021 / 061474, entitled "Wearable Garment Adapted for Ultrasound Detection and Method for Full Wave Inversion with Imprecise Sensor Location," which is assigned to a common assignee and is herein referred to as the "'474 Patent Application," and which is incorporated herein by reference in its entirety. The embedding of the sensors 112, emitters 113, and markers 114 may be accomplished by various techniques, such as, but not limited to, weaving, gluing, mechanical attachment, etc., and any combination thereof. The USG 110 is designed to provide an imaging solution that transmits and receives ultrasound signals and then processes the ultrasound signals to generate high resolution three-dimensional (3D) images of the scanned body part, and the connectors 115, e.g., 115-1 and 115-2, provide the ability to fixate the USG 110 around a moving medium, e.g., a body part.

[0028] In one embodiment, the USG 110 may further comprise electronic circuitry 116 adapted to provide power for consumption by the components of the USG 110 (sensor 112, emitter 113, and / or marker 114). Furthermore, the electronic circuitry 116 may include a combination of digital, analog, and optical components as may be required for proper operation of the USG 110. Signals received from the sensor 112 may be processed locally by the electronic circuitry 116 or, after initial or minimal processing, may be transmitted, by wire or wirelessly, to a processing circuit (not shown) that may further process the signals and display an image corresponding to the processed signals on a display device (not shown).

[0029] The USG 110 and its embedded components are communicatively connected (wired, wireless, or any acceptable combination thereof) to a controller 101 including a processing circuit (PC) 180, a memory 120, a sensor control interface (SCI), an emitter control interface (ECI) 140, a power control unit (PCU), a communication interface unit (CIU), and optionally a marker control interface (MCI). In an embodiment, the components of the controller 101 may be communicatively connected. The controller 101 is configured to control the USG 110 and its embedded components, such as, but not limited to, the emitter 113. In an exemplary embodiment, when the controller 101 determines operation below a predetermined threshold, it instructs the emitter 113 to stop emitting. In another exemplary embodiment, the controller 101 is adapted to activate a particular emitter 113 to emit ultrasound waves into a medium.

[0030] Processing circuitry (PC) 180 may be implemented as one or more hardware logic components and circuits. For example, but not limited to, exemplary types of hardware logic components that may be used include field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), graphics processing units (GPUs), tensor processing units (TPUs), general purpose microprocessors, microcontrollers, digital signal processors (DSPs), and the like, or any other hardware logic components capable of performing calculations or other manipulations of information.

[0031] A portion of the memory 120 may include code 125 and may be stored therein (or may be executed by the PC 180) when the code 125 is executed by the PC 180. The methods described herein are adapted to account for moving body tissues or organs and provide the benefits of performing FWI as described in more detail herein. The SCI 130 is adapted to at least receive signals from the sensors 112 and sensed by the sensors 112. The SCI 130 may receive signals in parallel from all, some, or only one of the sensors 112. The ECI 140 is adapted to send control signals to at least the emitters 113 to activate the emitters 113. The ECI 140 may transmit signals in parallel to all, some, or only one of the emitters 113.

[0032] An optional MCI 150 may be used for the active markers 114 and is adapted to at least activate the active markers 114. The MCI 150 may send control signals in parallel to all, some, or only one of the active markers 114. The PCU 160 is adapted to provide the necessary operating power to the elements of the USG 110 (e.g., the sensors 112, the emitters 113, the markers 114, etc.), which may be done in parallel to some of the USG 110 or to only a single element.

[0033] Additionally, the CIU 170 is adapted to provide communication between the USG 110. For example, without limitation, the CIU 170 may a) provide a communication means for activating the USG 110, b) receive signals from an external device (not shown) that controls the USG 110, and c) transmit processed or unprocessed signals captured by the sensor 112 according to any of the embodiments described herein. The PC 180 may include one or more processing circuits without departing from the scope of the present disclosure.

[0034] At the start of the inversion process, the subsurface sound speed model (also called simply velocity model) and attenuation model are unknown. The forward simulation takes the input signals (sent into the body in the case of medical ultrasound) and simulates their propagation on the initial virtual model. The output of this forward simulation represents the pressure measured at each sensor. The forward simulation results are compared to the actual measurements, and the error between them is used to modify the sound speed model and attenuation. This process is repeated iteratively until the convergence criteria of the simulation and the actual results are met. At this point, the model is assumed to be a valid replica of the subsurface structure. To improve the morphing model, the sound speed model needs to be updated and vice versa. The term sound speed model generally refers to the inversion of the 3D data of the sound speed at each point.

[0035] It should be noted that subsurface structures may vary depending on, for example, but not limited to, the portion or location on the moving medium, the type of medium, the age of the patient's body, the particular condition, etc., and any combination thereof. To this end, accurate models increase accuracy in producing ultrasound images and accommodate minor motions that may otherwise be ignored.

[0036] The moving medium in which the waves propagate is governed by a function that maps (morphs) each point x′ in the medium at time t to its original position x at time 0: The wave may be described by the following equation: TIFF0007659943000001.tif8169. In an exemplary embodiment, the medium is the object that moves and changes its shape during the measurement. The medium in which the wave travels is assumed to move slowly in time relative to the speed of sound in that medium.

[0037] The disclosed embodiments are described using the acoustic wave equation in a medium with slowly varying density. However, it should be understood that the disclosed embodiments apply equally to variable density, elastic, and viscoelastic cases by adding the appropriate equation terms. The wave equation in the moving medium reference frame is given by the Laplacian in a curvilinear system: For TIFF0007659943000002.tif7169, this is written:

[0038] Formula 1: TIFF0007659943000003.tif11169

[0039] where P is the pressure at point x' (not real space point x) in the medium, c is the speed of sound at point x' (not real space point x) in the medium, and f is the external force density acting at that point. That force will be ignored in the following discussion.

[0040] The changes in the medium through which the wave travels can be described as a combination of a spatial warping transformation and a change in elasticity. Typically the changes in density and elasticity (or stiffness) during the motion are small, and therefore the warping mainly affects the speed of sound in the medium. Specifically, the mapping function above Using TIFF0007659943000004.tif8169, the sound speed c(x',t) at point x' can be linearly approximated using the original velocity at point x and the change in sound speed due to elastic change, as shown in Equation 2.

[0041] Formula 2: TIFF0007659943000005.tif9169

[0042] In one embodiment, the time taken for exposure T f is assumed to be short enough that the change in the elastic constants is negligible. Furthermore, the transformation can be approximated by a linear function of time as the change from the initial position to the final position occurs. This is shown by Equation 3.

[0043] Formula 3: TIFF0007659943000006.tif11169

[0044] therefore, TIFF0007659943000007.tif7169, and as the final position of each unit volume of the medium TIFF0007659943000008.tif8169. In one embodiment, it is further assumed that the mapping function G(x) is doubly differentiable. It should therefore be appreciated that the disclosed embodiments relate morphing to the use of FWI in the field of ultrasound imaging, including biological tissue, for example.

[0045] Data collection for accurate FWI is done in a series of exposures. In each exposure, a subset of ultrasound emitters produces a pulse or series of pulses, and multiple sensors record the sensed pressure over a period of time. Although an exposure can be relatively fast, the entire acquisition process, i.e., a set of many exposures and their processing, can be much slower.

[0046] Given the mapping function G(x), the model parameters are calculated using a slightly modified FWI scheme. First, a forward simulation is performed using the time-dependent model parameters. Second, the misfit gradient calculation for the model at each FWI iteration is changed to reflect the morphing changes. This is further explained herein. The modified forward simulation reflects the changed equation of motion. Specifically, a discrete version of Equation 1 is shown in Equation 4.

[0047] Formula 4: TIFF0007659943000009.tif11169

[0048] where Δt and Δh are the time and space steps, respectively. TIFF0007659943000010.tif8169 is the pressure at grid cell i,j at time t=nΔt, TIFF0007659943000011.tif7169 is a finite-difference constant for integration in curvilinear coordinates, which is a function of G(x) and time.

[0049] The actual measured pressure at the sensor is matched with the simulated pressure to estimate the misfit error function. This error is back-injected into the medium and back-propagated in time by using the adjoint wave equation to calculate the effect of every single point in the medium on the misfit error. The misfit error gradient with respect to the model parameters c(x) can then be calculated, leading to Equation 6:

[0050] Formula 6: TIFF0007659943000012.tif12169

[0051] J is This is the Jacobian of TIFF0007659943000013.tif7169, where TIFF0007659943000014.tif7169 is the adjoint wavefield and c0(x) is the speed of sound at time t = 0 (see Equation 2). Equation 6 has a similar form to the conventional gradient with an additional term.

[0052] The disclosed embodiments described herein utilize a transformation function that is continuous in time and changes throughout the data acquisition session. TIFF0007659943000015.tif7169 is the focus of this paper. However, in many practical applications and embodiments thereof, this constraint may be further relaxed. As described herein, data acquisition for FWI involves many "shot experiments", where during each shot time a subset of emitters are activated and measured. The duration of a single shot is relatively short, and therefore the transfer function can be considered to be constant during each shot time and to change only between shots. That is, within a particular shot experiment, the transfer function TIFF0007659943000016.tif7169 is constant as shown in Equation 7.

[0053] Formula 7: TIFF0007659943000017.tif13169

[0054] According to one embodiment, the changes are seen as a simple morphing from the initial image (i.e., the starting model) to the final image (i.e., the ending model). Thus, x′=G(x) maps each point x in space at time t=0 to another point x′ in space at time T. The morphing function is assumed to be linear in time; that is, the mapped point progresses in a straight line from the starting position to the ending position. Another assumption is that the changes are less than the wavelength for the time it takes to travel inside the body. Note that the entire FWI sequence can be made longer and processed according to the present disclosure.

[0055] G(x) is constructed in two steps: first, an initial estimation is generated from the data; second, the model is improved during the iterative FWI phase. The initial G(x) (or initial G(x) / initial G(x)) is estimated from given constraints on specific points at the start and end times of the motion. The points are calculated from specific signals and specific emissions mixed in the FWI data collection.

[0056] There are several sources of predefined constraints that may be used to initially generate G(x): a) the relative motion of the emitter and sensor to their original positions, keeping in mind that absolute positions contain large errors, but changes due to changes in the time of flight of the signal have fairly small errors; b) certain surfaces within the imaged body, e.g., muscle-bone interfaces, generate large echoes, and therefore points on those surfaces may be interpreted structurally using traditional imaging techniques such as reverse time migration; and c) accumulated knowledge from tracking signals over time, which may allow for the estimation of certain movements, such as respiration, by using machine learning (ML) of the signal's patterns.

[0057] G(x) can be reconstructed or otherwise interpolated from these constraints by various methods of elastic registration and shape deformation adapted for use as described herein. In one embodiment, the constraints are considered to have greater uncertainty (not precise) that needs to be incorporated into the building G(x) and refined only later during FWI (or the constraints that need to be incorporated into the building G(x) are considered to have greater uncertainty (not precise) and refined only later during FWI). Furthermore, under the assumption that the overall acquisition time is short enough for G(x) to describe small motion changes, G(x) can be considered to be locally linear in the vicinity of the constraint points. Thus, G(x) can be expressed as Equation 8.

[0058] Formula 8: TIFF0007659943000018.tif12169

[0059] Here, p i is the i-th constraint point, and g i =(xp i )·A i +B i is the linear transformation of that point. i ,B iThe vector can be found by minimization.

[0060] FIG. 2 is an exemplary flow chart 200 for performing data acquisition, modeling, and full waveform inversion (FWI) in a moving medium according to one embodiment. In one embodiment, the inputs to the FWI iterations are a morphing function and a model of elastic properties, and the output is an updated morphing function and a model of elastic properties. According to one embodiment, there are three distinct parts to find the morphing function G(x) and model: a) the system uses the raw signal to find some initial approximation (or initial approximation) of G(x), b) using an inversion technique such as FWI to calculate an approximate model of elastic properties given the initial morphing function, and c) the system improves G(x) by solving the inverse problem of G(x) given the model of elastic properties. The last two steps above are repeated until both G(x) and the model of elastic properties converge. At the system level, if a large movement is recorded during signal acquisition, code is used to find a first approximation of G(x) from the raw signal and restart the acquisition process. (Or use code to find a first approximation of G(x) from the raw signal, and restart the acquisition process if significant movement is recorded.)

[0061] In S210, an ultrasonic signal is acquired, which is described in more detail in FIG. 3. Here, sensor data is collected and the motion of the moving medium is estimated. If the collected data is within a predetermined range for use by the next processing step, execution continues to S220. In one embodiment, the predetermined range is defined as the deviation of the estimated motion.

[0062] In S220, an initial elastic property model (or initial elastic property model) and G(x) (i.e., a morphing function) are generated. The model and morphing function are later used to perform FWI in accordance with the principles of the disclosed embodiments. In one embodiment, the model of the morphing function describes the elastic properties and the morphing function. This generation is further described in FIG. 4. Once G(x) is available, execution proceeds to S230.

[0063] In S230, FWI processing is performed. FWI considers the motion of moving media, such as, but not limited to, body tissues, organs, etc. In one embodiment, model parameters may be updated as further described with respect to FIG. 5. FWI may be iterated as necessary, and once the iterations are completed, execution proceeds to S240. In one embodiment, the FWI process is performed on the generated model of elastic properties and G(x) to refine the generated model and morphing function.

[0064] In S240, it is checked whether convergence has been achieved, i.e., whether the change is below a predefined threshold; if so, the execution is completed; if not, the execution continues to S250. Convergence is determined at least one between the generated model and the predecessor model and between the generated G(x) and the predecessor G(x). In one embodiment, a threshold is predefined for each of the model and G(x). A small convergence indicates a high similarity between the compared models. In one embodiment, the generated model is designated as a predecessor model before the execution proceeds to S250. The "predecessor" model is the model generated in the previous round of generation and inversion, which is similarly applied to the "predecessor" morphing function. In one embodiment, the predecessor model is updated or replaced by the generated model from the iteration of steps S220-S240.

[0065] In S250 it is checked whether additional acquisition of the ultrasound signal is required, and if so, execution continues to S210, otherwise execution continues to S220. In an exemplary embodiment, additional acquisition is performed when the change in S240 is substantially large, e.g., above a predefined normal value.

[0066] In one embodiment, the generated elastic property model and morphing function are utilized to generate an initial ultrasound image (or initial ultrasound image) that compensates for small movements in the moving medium. In a further embodiment, subsequent ultrasound images are generated based on the initial ultrasound image that was created to include morphing considerations. It should be appreciated that the morphing model is generated from the onset, prior to having an initial ultrasound image. Thus, it is not necessary to establish an image prior to ultrasound signal acquisition. In a further embodiment, the morphing model may be continuously updated to improve the accuracy of the initial ultrasound image being generated.

[0067] 3 is an exemplary flow chart S210 for performing an acquisition stage of full-wave inversion (FWI) in a moving medium according to one embodiment. The method further describes S210 of FIG. 2, which shows a method for acquiring an ultrasound signal.

[0068] In S210-10, one or more emitters are activated. In one embodiment, emitters (e.g., emitter 113, FIG. 1) are activated using both lower and higher range ultrasonic frequencies. In one embodiment, a selected portion of the emitters are activated. In further embodiments, the selected portion of the emitters may vary between successive rounds of generating morphing functions and models. For example, emitters 1-10 are activated to emit ultrasonic waves to obtain signals for a first G(x) and model generation, and emitters 11-20 are activated to emit ultrasonic waves for a second G(x) and model generation.

[0069] In S210-20, ultrasonic data is collected. The ultrasonic data is reflected and / or refracted ultrasonic signals collected by a sensor (e.g., sensor 112 of FIG. 1). In one embodiment, a selected portion of the sensor may be utilized to collect signals emitted by a selected portion of the emitter. In a further embodiment, a selected portion of the sensor may be predefined for successive cycles of G(x) and model generation.

[0070] At S210-30, a motion estimation is calculated to determine the actual motion that may have occurred, which may be, for example, but is not limited to, tissue, bone, muscle, organ, and the like, and any combination thereof.

[0071] In S210-40 it is checked whether the motion is small enough. In one embodiment, a small enough motion is a motion less than a predetermined displacement. If so, execution continues to S210-50; otherwise, execution ends and continues to S220 of FIG. 2.

[0072] In S210-50, FWI is performed to allow for sufficiently small motions. Execution ends and continues to S220 of FIG. 2 as described herein.

[0073] According to one embodiment, during FWI acquisition, calibration shots are interleaved, for example, but not limited to, about every 100 msec. In an exemplary embodiment, the received sensed signal is recorded as being similar to the initial sensed signal when it is determined based on the calibration shot that the body has not moved. The emitter selected for the calibration process is at a predefined location on the surface of the USG 110 that provides a clear reflected signal, is more sensitive to motion, and is in the vicinity of the bone margin. In one embodiment, the collected ultrasound data (i.e., signals) are discarded when the motion detected during calibration exceeds a threshold. In a further embodiment, the acquisition cycle continues after discarding the ultrasound data. Thus, the activation of the emitters in S210-10 should be understood as an activation that may also include a calibration signal from the designated emitter.

[0074] FIG. 4 is an exemplary flow chart S220 illustrating a method for generating an initial model and a morphing function G(x) for FWI in a moving medium, according to one embodiment.

[0075] In S220-10, x on the outer periphery i (or x i on perimeter) of G(x i ) is determined based on the signals collected by the relative positions of the sensors (eg, sensor 112 in FIG. 1).

[0076] In S220-20, G(x s ) is determined based on echoes (eg, ultrasound signals) based on conventional imaging techniques.

[0077] In S220-30, optionally, Doppler signal components may be analyzed to provide greater imaging accuracy.

[0078] In S220-40, G(x) is determined. In one embodiment, G(x) is interpolated from given constraints. In another embodiment, G(x) is determined by applying an algorithm, for example a machine learning algorithm, based on previous G(x) estimates and received signal data. The method of determining G(x) may be selected automatically or manually, without combination.

[0079] 5, 6, and 7 show examples of different embodiments of S230 of FIG. 2 for performing FWI. FIG. 5 is an exemplary flow chart 500 for performing full waveform inversion (FWI) in a moving medium, according to one embodiment. In S230-510, a forward wave simulation (FWI) of the emitted signal is performed. In S230-520, the simulated signal at the sensor location is read. In S230-530, a backward wave simulation of the error is performed. In S230-540, the gradient of the model is calculated. In S230-550, the model parameters and G(x) are updated. In S230-560, it is checked whether the mini-cycle is complete, if not, execution continues to S230-510, otherwise execution ends and continues to S240 of FIG. 2 as described herein. In an exemplary embodiment, the end of a mini-cycle (or a complete mini-cycle) is identified by determining that the convergence of the input and output models of the FWI process falls below a predetermined standard value.

[0080] FIG. 6 is an exemplary flow chart 600 for performing reverse time migration (RTM) in a moving medium, according to another embodiment. In S230-610, a forward wave simulation of the emitted signal is performed. This forward simulation is done in curvilinear coordinates according to Equation 1, using G(x). In S230-620, a backward wave simulation of the received signal is performed. This backward simulation also uses G(x) to account for the motion. In S230-630, the gradient of the model is calculated as in Equation 6, where: (TIFF0007659943000019.tif7169 is now the back propagated received signal rather than the error). In S230-640, the perturbation of the model parameters is updated. Thereafter, execution of S230 ends and S240 of FIG. 2 is performed as described herein.

[0081] FIG. 7 is an exemplary flow chart 700 for performing least squares migration (LSM) in a moving medium, according to yet another embodiment. In S230-710, a forward wave simulation of the emitted signal is performed. This forward simulation is done in curvilinear coordinates according to Equation 1, using G(x). In S230-720, a backward wave simulation of the received signal is performed. This backward simulation also uses G(x) to compensate for motion. In S230-730, the gradient of the model is calculated as in Equation 6 ( (TIFF0007659943000020.tif7169 is now the back-propagated received signal rather than the error). In S230-640, the model parameter perturbations are updated. In S230-750, it is checked whether the mini-cycle is complete, and if not, execution continues with S230-710, otherwise execution ends and S240 of FIG. 2 is performed as described herein.

[0082] Accordingly, an aspect of the disclosed embodiment is a computerized method for generating a morphing function for use in creating an ultrasound image of a moving medium, the method comprising the steps of emitting a first plurality of ultrasonic waves toward a patient's body by a first plurality of emitters; receiving a first plurality of received signals corresponding to the first plurality of ultrasonic waves by a first plurality of sensors; and determining, based on at least a first received signal of the first plurality of received signals, that motion of the medium is below a predetermined threshold and if so continuing or, otherwise, discarding the first plurality of received signals and repeating the emission by the first plurality of emitters and reception by the first plurality of sensors. and emitting a second plurality of ultrasonic waves toward the patient's body by a second plurality of emitters; receiving a second plurality of received signals corresponding to the second plurality of ultrasonic waves by a second plurality of sensors; determining, based on at least a second received signal of the second plurality of received signals, that motion of the medium is below a predetermined threshold, and if so continuing, or otherwise discarding the second plurality of received signals and repeating emission by the second plurality of emitters and receiving by the second plurality of sensors; and generating a model of a morphing function, the morphing function establishing a morphing of the moving medium between the first plurality of received signals and the second plurality of received signals.

[0083] Other aspects of the computerized method include, but are not limited to, repeating the steps of: emitting ultrasound from at least a first emitter of the plurality of emitters designated as a calibration emitter; the designation of the first emitter of the plurality of emitters being based on a determination that the first emitter is in proximity to a reflective element; the reflective element being a bone edge within the body; determining a motion based on reflections received from the reflective element; generating a sound speed model based on the generated morphing model; the plurality of received signals being one of a reflected signal, a refracted signal, and a combination thereof; generating at least an initial ultrasound image using the generated morphing model; emitting by a first plurality of emitters, receiving by a first plurality of sensors, determining based on at least the first received signals, emitting by a second plurality of emitters, receiving by a second plurality of sensors, determining based on at least the second received signals, and generating a model of the morphing function.

[0084] The teachings herein may further be embodied in a system including a processing element, a plurality of ultrasound emitters adapted to emit ultrasound waves under control of the processing element, the plurality of ultrasound emitters mounted within a wearable garment, a plurality of ultrasound sensors adapted to transmit readings corresponding to ultrasound waves received by the plurality of ultrasound sensors, the plurality of ultrasound sensors mounted within the wearable garment and disposed between the plurality of ultrasound emitters, and a memory communicatively coupled to the processing element and including a plurality of instructions therein that, when executed by the processing element, cause the system to: emit a first plurality of ultrasound waves toward a body of a patient with a first plurality of emitters of the plurality of ultrasound emitters; receive a first plurality of received signals corresponding to the first plurality of ultrasound waves with a first plurality of ultrasound sensors of the plurality of ultrasound sensors; and transmit at least a first plurality of received signals of the first plurality of received signals. determining, based on the received signals, that motion of the medium is below a predetermined threshold, and if so, continuing, otherwise discarding the first plurality of received signals, and repeating the emission by the first plurality of emitters and reception by the first plurality of sensors; emitting a second plurality of ultrasonic waves toward the patient's body by a second plurality of ultrasonic emitters of the plurality of ultrasonic emitters; receiving a second plurality of received signals corresponding to the second plurality of ultrasonic waves by a second plurality of ultrasonic sensors of the plurality of ultrasonic sensors; determining, based on at least a second received signal of the second plurality of received signals, that motion of the medium is below a predetermined threshold, and if so, continuing, otherwise discarding the second plurality of received signals, and repeating the emission by the second plurality of emitters and reception by the second plurality of sensors; and generating a model of a morphing function, the morphing function establishing a morphing of the moving medium between the first plurality of received signals and the second plurality of received signals.

[0085] Another aspect of the system includes a plurality of ultrasound emitters sparsely disposed on the wearable garment, a plurality of ultrasound sensors sparsely disposed on the wearable garment, at least one of the plurality of ultrasound emitters and the plurality of ultrasound sensors embodied within a single device, the plurality of ultrasound waves further including at least ultrasound emitted from a first emitter of the plurality of emitters designated as a calibration emitter, the designation of the first emitter of the plurality of emitters being based on a determination that the first emitter is in proximity to a reflective element, the reflective element being an edge of a bone within the body, the motion being determined based on reflections received from the reflective element, and a memory storing, when executed by the processing element, a morphing model generated by the morphing model. The memory further includes, but is not limited to, instructions for generating a sound speed model based on the morphing model, the plurality of received signals being one of reflected signals, refracted signals, and combinations thereof, the memory further includes instructions for generating at least an initial ultrasound image using the generated morphing model, the memory further includes instructions for, when executed by the processing element, repeating emission by a first plurality of emitters, receiving by a first plurality of sensors, determining based on at least the first received signals, emitting by a second plurality of emitters, receiving by a second plurality of sensors, determining based on at least the second received signals, and generating a model of the morphing function.

[0086] Further aspects of the inversion method for analyzing ultrasonic waves received from a moving medium include, but are not limited to, receiving ultrasonic data from a plurality of ultrasonic sensors, simulating ultrasonic waves in the moving medium using at least a morphing function, and updating at least an elastic property of the moving body based on at least the simulation. In yet another aspect, the simulation further includes at least one of forward simulation and backward simulation. In yet another aspect of the present invention, the simulation includes full waveform inversion (FWI), adaptive waveform inversion (AWI), physically based deep neural network (DNN) inversion, least squares migration (LSM), one-way wave equation migration, two-way wave equation migration, reverse time migration (RTM), and any combination thereof.

[0087] Various embodiments disclosed herein may be implemented as hardware, firmware, software, or any combination thereof. Furthermore, software is preferably implemented as an application program tangibly embodied on a program storage unit or computer-readable medium consisting of parts or specific devices and / or combinations of devices. The application program may be uploaded to and executed by a machine having any suitable architecture. Preferably, the machine is implemented on a computer platform having hardware such as one or more central processing units ("CPU"), general-purpose computation accelerators such as graphics processing units ("GPU"), memory, and input / output interfaces. The computer platform may also include an operating system and microinstruction code. Various processes and functions described herein may be either part of the microinstruction code or part of the application program, or any combination thereof, that may be executed by the CPU or GPU, whether or not such computers or processors are explicitly shown. In addition, various other peripheral units may be connected to the computer platform, such as additional data storage units and printing units. Furthermore, a non-transitory computer-readable medium is any computer-readable medium, except for a transitory propagating signal.

[0088] All examples and conditional language recited herein are for educational purposes to aid the reader in understanding the principles of the disclosed embodiments and concepts contributed by the inventors to advance the art, and should not be construed as being limited to such specifically recited examples and conditions. Moreover, all statements herein reciting principles, aspects, and embodiments of the disclosed embodiments, as well as specific examples thereof, are intended to encompass both structural and functional equivalents thereof. Moreover, such equivalents are intended to include both currently known equivalents and equivalents developed in the future, i.e., any elements developed that perform the same function, regardless of structure.

[0089] It should be understood that any reference to an element herein using a designation such as "first," "second," etc. does not generally limit the quantity or order of those elements. Rather, these designations are generally used herein as a convenient method of distinguishing between two or more elements or instances of an element. Thus, a reference to a first and a second element does not imply that only two elements may be used therein or that the first element must precede the second element in any way. Also, unless otherwise specified, a set of elements includes one or more elements.

[0090] As used herein, the phrase "at least one of" can be followed by a list of items meaning that any of the listed items can be utilized individually, or any combination of two or more of the listed items can be utilized. For example, if a system is described as including "at least one of A, B, and C," the system can include A alone, B alone, C alone, 2A, 2B, 2C, 3A, a combination of A and B, a combination of B and C, a combination of A, B, and C, a combination of 2A and C, a combination of A, 3B, and 2C, and the like. The following is an invention described at the beginning of the present application. <Claim 1> 1. A method for generating an ultrasound image of a moving medium by generating a morphing function, comprising: emitting a plurality of ultrasonic waves by a portion of a plurality of emitters toward the moving medium until a motion of the moving medium is determined to be below a predetermined threshold; receiving, by a portion of a plurality of sensors, a first plurality of received signals corresponding to the plurality of ultrasonic waves; generating a model of a morphing function when the motion of the moving medium falls below the predetermined threshold, the motion of the moving medium being determined based on at least one received signal of the first plurality of received signals, the morphing function defining a morphing of the moving medium from an initial plurality of received signals; and performing waveform inversion on the model of the morphing function. <Claim 2> comparing the model of the morphing function with a predecessor model, the predecessor model being determined prior to generation of the model of the morphing function; determining a convergence between the predecessor model and the model of the morphing function based on the comparison; The method of claim 1 , further comprising: updating the prior model with the model of the morphing function when the convergence is greater than a predetermined value. <Claim 3> the portion of the plurality of emitters is a first portion of the plurality of emitters, the portion of the plurality of sensors is a first portion of the plurality of sensors, and the convergence is greater than the predetermined value; repeating the emitting by a second portion of the plurality of emitters and the receiving by a second portion of the plurality of sensors, where a second plurality of receive signals is received; 3. The method of claim 2, further comprising repeating the steps of generating the model of the morphing function and performing waveform inversion, wherein the morphing function defines a morphing of the moving medium from the first plurality of received signals. <Claim 4> emitting an initial plurality of ultrasonic waves with an initial portion of the plurality of emitters until an initial motion of the moving medium below the predetermined threshold is determined; 2. The method of claim 1, further comprising: receiving, by an initial portion of a plurality of sensors, an initial plurality of received signals corresponding to an initial plurality of ultrasonic waves, and wherein the initial motion of the medium is determined based on at least the initial plurality of received signals. <Claim 5> The method of claim 2 , further comprising generating at least one ultrasound image using the model of the morphing function. <Claim 6> The method of claim 1 , further comprising generating a speed of sound model based on the generated model of the morphing function. <Claim 7> The method of claim 1 , wherein the plurality of received signals are one of reflected and refracted signals. <Claim 8> The step of performing waveform inversion includes: simulating the plurality of received signals using the morphing function, the morphing function being determined based on the model of the morphing function; The method of claim 1 , further comprising: iteratively updating at least elastic properties of the moving medium and the morphing function based at least on the simulation. <Claim 9> The method of claim 8 , wherein the simulation of the plurality of received signals includes at least one of a forward simulation and a backward simulation. <Claim 10> 2. The method of claim 1, wherein the waveform inversion comprises any one of full waveform inversion (FWI), adaptive waveform inversion (AWI), and physically based deep neural network (DNN) inversion. <Claim 11> The method of claim 1 , wherein the portion of the plurality of emitters includes at least one designated calibration emitter, the at least one calibration emitter being located proximate a reflective element of the moving medium. <Claim 12> 1. A system for ultrasonic imaging of a moving medium, the system comprising: Processing circuitry, a plurality of ultrasonic emitters adapted to emit ultrasonic waves under the control of a processing circuit; a plurality of ultrasonic sensors adapted to transmit readings corresponding to the ultrasonic waves received by the plurality of ultrasonic sensors; and a memory communicatively coupled to the processing circuitry, the memory being configured to, when executed by the processing circuitry, execute the system; emitting a plurality of ultrasonic waves with a portion of a plurality of emitters toward the moving medium until a motion of the moving medium below a predetermined threshold is determined; receiving, by a portion of a plurality of sensors, a first plurality of received signals corresponding to the plurality of ultrasonic waves; generating a model of a morphing function when the motion of the moving medium falls below the predetermined threshold, the motion of the moving medium being determined based on at least one received signal of the first plurality of received signals, the morphing function defining a morphing of the moving medium from an initial plurality of received signals; performing waveform inversion on the model of the morphing function; and said memory having a plurality of instructions therein configured to perform the steps. <Claim 13> The system further comprises: comparing the model of the morphing function with a predecessor model, the predecessor model being determined prior to generation of the model of the morphing function; determining a convergence between the predecessor model and the model of the morphing function based on the comparison; and updating the prior model with the model of the morphing function when the convergence is greater than a predetermined value. <Claim 14> the portion of the plurality of emitters is a first portion of the plurality of emitters, the portion of the plurality of sensors is a first portion of the plurality of sensors, the convergence is greater than the predetermined value, and the system further comprises: repeating the emitting by a second portion of the plurality of emitters and the receiving by a second portion of the plurality of sensors, where a second plurality of receive signals is received; 14. The system of claim 13, configured to repeat the steps of generating the model of the morphing function and performing waveform inversion, wherein the morphing function defines a morphing of the moving medium from the first plurality of received signals. <Claim 15> The system further comprises: emitting an initial plurality of ultrasonic waves with an initial portion of the plurality of emitters until an initial motion of the moving medium below the predetermined threshold is determined; receiving, by an initial portion of a plurality of sensors, an initial plurality of received signals corresponding to an initial plurality of ultrasonic waves, and wherein the initial motion of the medium is determined based at least on the initial plurality of received signals. <Claim 16> The system further comprises: The system of claim 13 , configured to perform the step of generating at least one ultrasound image using the model of the morphing function. <Claim 17> The system further comprises: The system of claim 12 , configured to perform the step of generating a speed of sound model based on the generated model of the morphing function. <Claim 18> The system of claim 12 , wherein the plurality of received signals are one of reflected and refracted signals. <Claim 19> The system further comprises: simulating the plurality of received signals using the morphing function, the morphing function being determined based on the model of the morphing function; and iteratively updating at least elastic properties of the moving medium and the morphing function based at least on the simulation. <Claim 20> 20. The system of claim 19, wherein the simulation of the plurality of received signals includes at least one of a forward simulation and a backward simulation. <Claim 21> 13. The system of claim 12, wherein the waveform inversion includes any one of full waveform inversion (FWI), adaptive waveform inversion (AWI), and physically based deep neural network (DNN) inversion. <Claim 22> 13. The system of claim 12, wherein the portion of the plurality of emitters includes at least one designated calibration emitter, the at least one calibration emitter located proximate a reflective element of the moving medium.

Claims

1. 1. A method for generating an ultrasound image of a moving medium by generating a morphing function, comprising: emitting a plurality of ultrasonic waves by a portion of a plurality of emitters toward the moving medium until a motion of the moving medium is determined to be below a predetermined threshold; receiving, by a portion of a plurality of sensors, a first plurality of received signals corresponding to the plurality of ultrasonic waves; generating a model of a morphing function when the motion of the moving medium falls below the predetermined threshold, the motion of the moving medium being determined based on at least one received signal of the first plurality of received signals, the morphing function defining a morphing of the moving medium from an initial plurality of received signals; and performing waveform inversion on the model of the morphing function.

2. comparing the model of the morphing function with a predecessor model, the predecessor model being determined prior to generation of the model of the morphing function; determining a convergence between the predecessor model and the model of the morphing function based on the comparison; The method of claim 1 , further comprising: updating the predecessor model with the model of the morphing function when the convergence is greater than a predetermined value.

3. the portion of the plurality of emitters is a first portion of the plurality of emitters, the portion of the plurality of sensors is a first portion of the plurality of sensors, and the convergence is greater than the predetermined value; repeating the emitting by a second portion of the plurality of emitters and the receiving by a second portion of the plurality of sensors, where a second plurality of receive signals is received; 3. The method of claim 2, further comprising repeating the steps of generating the model of the morphing function and performing waveform inversion, the morphing function defining a morphing of the moving medium from the first plurality of received signals.

4. emitting an initial plurality of ultrasonic waves with an initial portion of the plurality of emitters until an initial motion of the moving medium below the predetermined threshold is determined; 10. The method of claim 1, further comprising: receiving, by an initial portion of a plurality of sensors, an initial plurality of received signals corresponding to an initial plurality of ultrasonic waves, and wherein the initial motion of the medium is determined based at least on the initial plurality of received signals.

5. The method of claim 2 , further comprising generating at least one ultrasound image using the model of the morphing function.

6. The method of claim 1 , further comprising generating a speed of sound model based on the generated model of the morphing function.

7. The method of claim 1 , wherein the plurality of received signals are one of reflected and refracted signals.

8. The step of performing waveform inversion includes: performing a simulation of the plurality of received signals using the morphing function, the morphing function being determined based on the model of the morphing function; The method of claim 1 , further comprising: iteratively updating at least elastic properties of the moving medium and the morphing function based at least on the simulation.

9. The method of claim 8 , wherein the simulation of the plurality of received signals includes at least one of a forward simulation and a backward simulation.

10. The method of claim 1 , wherein the waveform inversion comprises any one of full waveform inversion (FWI), adaptive waveform inversion (AWI), and physically based deep neural network (DNN) inversion.

11. The method of claim 1 , wherein the portion of the plurality of emitters includes at least one designated calibration emitter, the at least one calibration emitter being located proximate a reflective element of the moving medium.

12. 1. A system for ultrasonic imaging of a moving medium, the system comprising: Processing circuitry, a plurality of ultrasound emitters adapted to emit ultrasound waves under the control of a processing circuit; a plurality of ultrasonic sensors adapted to transmit readings corresponding to the ultrasonic waves received by the plurality of ultrasonic sensors; and a memory communicatively coupled to the processing circuitry, the memory being operable when executed by the processing circuitry to: emitting a plurality of ultrasonic waves with a portion of a plurality of emitters toward the moving medium until a motion of the moving medium below a predetermined threshold is determined; receiving, by a portion of a plurality of sensors, a first plurality of received signals corresponding to the plurality of ultrasonic waves; generating a model of a morphing function when the motion of the moving medium falls below the predetermined threshold, the motion of the moving medium being determined based on at least one received signal of the first plurality of received signals, the morphing function defining a morphing of the moving medium from an initial plurality of received signals; performing waveform inversion on the model of the morphing function; and said memory having a plurality of instructions therein configured to perform the steps.

13. The system further comprises: comparing the model of the morphing function with a predecessor model, the predecessor model being determined prior to generation of the model of the morphing function; determining a convergence between the predecessor model and the model of the morphing function based on the comparison; and updating the predecessor model with the model of the morphing function when the convergence is greater than a predetermined value.

14. the portion of the plurality of emitters is a first portion of the plurality of emitters, the portion of the plurality of sensors is a first portion of the plurality of sensors, the convergence is greater than the predetermined value, and the system further comprises: repeating the emitting by a second portion of the plurality of emitters and the receiving by a second portion of the plurality of sensors, where a second plurality of receive signals is received; 14. The system of claim 13, configured to repeat the steps of generating the model of the morphing function and performing waveform inversion, wherein the morphing function defines a morphing of the moving medium from the first plurality of received signals.

15. The system further comprises: emitting an initial plurality of ultrasonic waves with an initial portion of the plurality of emitters until an initial motion of the moving medium below the predetermined threshold is determined; 13. The system of claim 12, configured to perform the steps of: receiving, by an initial portion of a plurality of sensors, an initial plurality of received signals corresponding to an initial plurality of ultrasonic waves, and wherein the initial motion of the medium is determined based at least on the initial plurality of received signals.

16. The system further comprises: The system of claim 13 , configured to perform the step of generating at least one ultrasound image using the model of the morphing function.

17. The system further comprises: The system of claim 12 , configured to perform the step of generating a speed of sound model based on the generated model of the morphing function.

18. The system of claim 12 , wherein the plurality of received signals are one of reflected and refracted signals.

19. The system further comprises: performing a simulation of the plurality of received signals using the morphing function, the morphing function being determined based on the model of the morphing function; and iteratively updating at least elastic properties of the moving medium and the morphing function based at least on the simulation.

20. 20. The system of claim 19, wherein the simulation of the plurality of received signals includes at least one of a forward simulation and a backward simulation.

21. The system of claim 12 , wherein the waveform inversion includes any one of full waveform inversion (FWI), adaptive waveform inversion (AWI), and physically based deep neural network (DNN) inversion.

22. The system of claim 12 , wherein the portion of the plurality of emitters includes at least one designated calibration emitter, the at least one calibration emitter located proximate a reflective element of the moving medium.

Images