Reflection ultrasound imaging using full waveform inversion

Abstract

A medical ultrasound (US) imaging system (10) includes a US probe (30) and a processor (48). The US probe includes an array (50) of transducers (401, 402) arranged in a reflection geometry, the probe configured to transmit US waves and receive reflected US waves reflected from a body part of a patient. The processor is configured to generate an image of the body part of the patient by applying an inversion model to the transmitted US waves and the reflected US waves.

Description

Invention Field

[0001] This invention relates generally to medical imaging, and more particularly to quantitative ultrasound (US) imaging. Background of the Invention

[0002] Various methods of medical ultrasound imaging have been considered in patent literature and scientific publications. For example, U.S. Patent Application Publication 2020 / 0008779 describes a medical ultrasound system comprising an ultrasound transducer and a processor for transmitting and receiving ultrasound. The ultrasound transducer is electrically connected to the processor, and the processor is configured to determine an ultrasound-based tomographic image dominated by ultrasound waves received by the ultrasound transducer in response to ultrasound waves emitted by the ultrasound transducer and scattered and / or reflected by the tissue under study.

[0003] As another example, R. Pratt's paper, "Medical ultrasound tomography: lessons from exploration geophysics," published in the Proceedings of the International Workshop on Medical Ultrasound Tomography, November 1-3, 2017, in Speyer, Germany, describes potential improvements in medical ultrasound transmission tomography based on full-waveform inversion (FWI). Invention Overview

[0004] The embodiments of the invention described below provide a medical ultrasound (US) imaging system including a US probe and a processor. The US probe includes an array of transducers arranged in a reflection geometry, the probe being configured to emit US waves and receive reflected ultrasound waves reflected from a patient's body part. The processor is configured to generate an image of the patient's body part by applying an inversion model to the emitted US waves and the reflected US waves.

[0005] In some embodiments, when generating images using an inversion model, the processor is configured to estimate and indicate one or more of (i) physiological tissue parameters and (ii) tissue boundaries in the images.

[0006] In some embodiments, physiological tissue parameters include one of local density, local sound velocity, and local energy decay within a body part.

[0007] In one embodiment, the inversion model is full waveform inversion (FWI). In another embodiment, the inversion model is reverse time migration (RTM).

[0008] In some embodiments, the processor is configured to jointly apply the inversion model to US waves emitted and reflected at multiple different locations of the probe relative to the body portion.

[0009] In some embodiments, the processor is configured to control the probe to emit and receive ultrasonic waves in a series of acquisitions, wherein in each acquisition, a corresponding subset of one or more transducers emits ultrasonic waves, and one or more other transducers receive reflected ultrasonic waves.

[0010] In one embodiment, a subset of transducers is selected to produce a signal-to-noise ratio (SNR) value higher than a predefined threshold.

[0011] In another embodiment, a subset of transducers selected from a series of acquisitions forms a two-dimensional multi-static basis.

[0012] In yet another embodiment, a subset of transducers selected from a series of acquisitions forms a two-dimensional Hadamard basis.

[0013] In some embodiments, the processor is configured to apply a low-pass filter to the reflected US wave, use the low-pass filtered reflected US wave to generate an initial image of the patient's body part, and use the initial image in subsequent inversion model calculations to generate the image.

[0014] In some embodiments, the array of transducers is two-dimensional.

[0015] In some embodiments, the processor is located away from the probe.

[0016] Additionally, another embodiment of the present invention provides a medical ultrasound (US) imaging method, which includes emitting US waves using a US probe and receiving reflected ultrasound waves reflected from a patient's body part, the US probe comprising an array of transducers arranged in a reflective geometry. An image of the patient's body part is generated by applying an inversion model to the emitted US waves and the reflected US waves.

[0017] The invention will be more fully understood from the following detailed description of embodiments thereof, taken in conjunction with the accompanying drawings, in which: Brief description of the attached diagram

[0019] Figure 1This is a schematic illustration of a medical ultrasound (US) reflective computed tomography system according to an embodiment of the present invention, the system including a US probe comprising a flat detector array;

[0020] Figure 2 This is an embodiment of the present invention. Figure 1 A schematic illustration of the reflection geometry used in the US reflection tomography imaging system;

[0021] Figure 3 According to an embodiment of the present invention, the use of [material name] is illustrated schematically. Figure 1 The system collects data and according to Figure 2 A flowchart illustrating the iterative process of FWI reconstruction of US images using the FWI reconstruction algorithm applied to the reflection geometry defined in the diagram.

[0022] Figure 4A and Figure 4B The embodiments of the present invention are shown respectively by Figure 1 The graph shows the spectrum of the driving signal and the analog spectrum of the US wave transmitted by the system.

[0023] Figure 5 It is according to an embodiment of the present invention and Figure 1 A schematic top view of the multi-static acquisition mode used with the US probe; and

[0024] Figure 6 This is a flowchart illustrating an embodiment of the present invention, which schematically shows the use of... Figure 1 The system collects data and Figure 3 This paper describes a method for FWI reconstruction of US images using the FWI reconstruction algorithm.

[0025] Detailed description of the embodiments

[0026] Overview

[0027] Unlike computed tomographic imaging modalities such as CT and MRI, ultrasound (US) images are typically only indicative in nature. The clinical usefulness of a US imaging session depends heavily on the user’s expertise to apply the US system well enough to acquire meaningful images, and also on the ability of a trained radiologist to interpret the acquired US images.

[0028] The fundamental reason for this quality in US images stems from the acquisition and reconstruction methods used, which are similar to those used for sonar images. Therefore, unlike protocol-based CT and MRI examinations, which largely represent the actual anatomical structures being imaged, US images are acquired ad-hoc and manually, and are primarily suggestive. Only in very specific cases do US images achieve the quality of modalities such as CT and MRI, such as in breast US images produced by US transmission tomography imaging systems. However, despite acquiring US waves that carry important information, transmission geometry is cumbersome and often impractical for imaging larger body parts, such as the abdomen or trunk.

[0029] The embodiments of the invention described below provide systems, methods, and algorithms for tomographic US imaging of the human body using reflective geometry. Body parts capable of being imaged in this manner include, for example, the abdomen, pelvis, and heart. In some of the disclosed embodiments, a handheld US probe is optimized to generate and detect US waves for use in the disclosed reconstruction algorithms and for use with protocol-based image acquisition methods.

[0030] In some embodiments, a US reflection tomography imaging system is provided, including a US probe comprising an array of transducers arranged in a reflective geometry, the probe being configured to emit ultrasound waves and receive reflected ultrasound waves reflected from a body part of a patient. The system also includes a processor configured to generate images of the patient's body part by applying an inversion model to the emitted and reflected ultrasound waves.

[0031] In one embodiment, the image includes at least one physiological tissue parameter generated by applying full waveform inversion (FWI) to the emitted and reflected ultrasound waves. In another embodiment, the processor is configured to generate an image of a body part by applying a reverse time migration (RTM) model to the emitted and reflected ultrasound waves.

[0032] In this context, the term "reflection geometry" refers only to the geometry of a US wave that changes direction by at least 90 degrees relative to the emitted US wave during its propagation within the imaged body, as shown below. Figure 2 As shown. Any received beam with such characteristics is considered in this paper to belong to a “reflecting hemisphere.” The two-dimensional transducer array of a conventional US probe typically captures only a small portion of such a reflecting hemisphere (e.g., covering a narrow solid angle, much less than 2). Therefore, the reflection geometry allows for the use of probes that capture US waves from a fairly large portion of the reflecting hemisphere. The disclosed reflection tomography imaging system can use this reflected US wave to acquire US data and reconstruct images from the acquired data using FWI.

[0033] As a rule of thumb for probe design, the imaging depth using reflection inversion methods is proportional to the aperture of the probe's transducer array. The scaling factor typically varies between 2 and 0.5, thus requiring a large aperture in at least one dimension of a two-dimensional array, typically ranging from 5 cm to 15 cm, depending on the clinical application. If the array is longer in one dimension, then the two-dimensional transducer array is referred to as having both a major and minor axis (e.g., as in a rectangular array).

[0034] The term "inversion model" refers to a class of formal inverse models, such as FWI and RTM, which involve finding medium parameters and / or medium boundaries given the wave equation of the acoustic pressure wave and performing some partial measurements of the pressure wave field at the medium boundaries, for example, using the emitted and received waves (at the probe) as boundary conditions. Typically, FWI enables resolution up to half the wavelength of the US signal, which is not achievable with standard methods.

[0035] In some disclosed embodiments, the processor uses the FWI algorithm to calculate medium parameters (density, sound velocity, elasticity, etc.) by iteratively applying a procedure to solve the wave equations and compare the solved wavefield data with actual measured values ​​of the wavefield data. In each iteration, this comparison produces small corrections to the medium parameters, such that the residual error between the solved wavefield and the actually measured wavefield is close to zero. At this point, the found medium parameters well explain the measured wavefield data and, in the disclosed embodiments, can be used to generate quantitative US images or enhance the image quality of qualitative US images, such as the image quality of B-mode US images.

[0036] In various embodiments of the invention, the processor uses the reflected US signal to estimate various physiological tissue parameters and visualize them for the user. Examples include local tissue density, local sound velocity, local energy attenuation, elasticity, etc.

[0037] In one embodiment, the ultrasound probe is configured to generate and detect a broadband US wave with an average frequency at least an order of magnitude lower than the average frequency used in conventional modern US systems (e.g., 250 kHz vs. 2.5 MHz or higher). Using a low-pass filter, the low-frequency tail of the spectrum of the US signal acquired by the probe (e.g., a frequency tail below 100 kHz) is extracted and analyzed. The processor uses the low-frequency tail of the US signal with an FWI algorithm to generate an initial image of the patient's body portion. Based on the initial image, the processor then uses a higher US frequency with the FWI algorithm to achieve a full US image with the robustness and accurate convergence of the FWI algorithm.

[0038] In other disclosed embodiments, the processor applies an RTM model to solve for the source wavefield. Ps (This wave field is the pressure at every point in space and time, and the source means the pulse emitted from the source transmitter.) The RTM model involves acquiring measured sensor data (from an actual probe) and using a receiver as a transmitter to computationally propagate the received signal backwards in time; this wave field is called… Pr (Mathematically, this operation uses the adjoint operator of the wave equation to computationally "reverse-time" the signal.) Finally, the processor... Ps and Pr Cross-correlation is performed (along the time axis) and taken at time offset = 0. This image may not be quantitative and may primarily result in sharp edges of the medium (similar to the characteristics of b-mode US, but with considerably better accuracy). The RTM method produces particularly accurate results if there is a good estimate of the sound velocity in the medium. In this respect, the RTM method is also similar to the b-mode, except that it uses all the data and considers complex wave phenomena. Ultrasonic b-mode is more similar to Kirchhoff migration in geophysics, which is another algorithm that can be used with the published techniques.

[0039] In another embodiment, the disclosed probe operates in a frequency range (e.g., several MHz or higher) common in conventional modern US systems. Using algorithmic methods, the processor is still able to achieve robust and accurately converged full US images with the FWI algorithm.

[0040] In some embodiments, acquisitions from different locations of the probe (e.g., from locations where the probe is moved by the user) are combined. In this way, using the FWI reconstruction algorithm, the body depth, spatial resolution, and field of view (FOV) of the image can be increased.

[0041] Unlike conventional ultrasound (US) systems that use beamforming technology to direct incident US waves (i.e., the US beam emitted by the probe) to a specific body location, the disclosed technique emits US waves that simultaneously cover a wide field of view. To this end, in some embodiments, the disclosed technique uses an ultrasound probe array in a multi-static acquisition mode. In the simplest form of the multi-static acquisition mode, one transducer emits, while the remaining transducers in the array receive reflected signals. In alternative embodiments, the disclosed technique can be used over a set of beamformed signals, provided the probe illuminates the target volume at a sufficiently wide range of angles and distances.

[0042] Alternatively, an equivalent mode of the multi-static acquisition mode can be used, which can produce a higher signal-to-noise ratio (SNR), such as running acquisition with a Hadamard sequence (i.e., applying US emission with a sequence consisting of Hadamard bases of the transmitting transducer). Using algorithms, the processor can map the Hadamard sequence (or any other suitable sequence) into a simpler multi-static sequence. In an embodiment, the sequence is selected to produce an SNR value above a predefined threshold, which is considered sufficient for generating clinical images.

[0043] By providing US systems and methods capable of generating quantitative US images with reflective geometry, high-quality and reliable medical images can be achieved with limited resources, with image quality comparable to that achieved through far more cumbersome imaging modalities (such as CT and MRI) and associated workflows.

[0044] System Description

[0045] Figure 1 This is a schematic illustration of a medical ultrasound (US) reflective computed tomography system 10 according to an embodiment of the present invention. The system 10 includes a handheld US probe 30, which includes a flat detector array 50.

[0046] The US imaging system 10 includes a US console 40, which includes an interface 46 to which a US probe 30 is connected via a cable 44. The console 40 also includes a system processor 48.

[0047] Interface 46 is configured to transfer electrical power via cable 44 to an array 50 of piezoelectric US transducers 100, which are included in probe 30 and are visible in illustration 25. The US transducers 100 are sized to generate and detect US signals with low-frequency components (e.g., <250 kHz) to achieve the aforementioned reflection FWI reconstruction of the US image. However, other transducer designs can be used, for example, including transducers sized to generate and detect US signals with a center frequency of 500 kHz or 1 MHz or higher.

[0048] The high-frequency components of the signal can also be used, for example, to increase spatial resolution. Furthermore, interface 46 can be controlled by a processor to deliver electrical energy to one or more transducers of the probe in the aforementioned multi-static acquisition mode or in other acquisition modes equivalent to multi-static (same forming base) but with higher SNR (e.g., hadamard).

[0049] Array 50 is configured to generate a US beam in response to driving energy and detect the resulting US echoes, then transmit the resulting electrical signals to processor 48 via cable 44 and interface 46. Processor 48 is configured to generate a quantitative US image by applying FWI reconstruction and display it on monitor 49.

[0050] In the illustrated embodiment, array 50 has a square shape with its corners cut off and includes, by way of example only, 376 transducers 100, each configured to transmit and detect a US wave with a center frequency of 250 kHz, including transmitting and detecting significant US energy in frequencies within the 50 kHz–100 kHz band. As can be seen, probe 30 has a substantially uniform array aperture in all lateral directions.

[0051] Considering the challenge of obtaining quantitative images using only data from a single-sided probe (with the aforementioned reflective geometry), the disclosed iterative reconstruction starts at the lowest possible frequency to aid in the convergence of the iterative reconstruction.

[0052] Typically, processor 48 includes a general-purpose computer programmed with software to perform the functions described herein. In some embodiments, processor 48 includes additional processing resources, such as a set of GPUs. Processor 48 is configured to upload data from memory 49 to perform a US image FWI reconstruction algorithm (e.g., ...). Figure 2 The software (the algorithm described in the text).

[0053] The configuration of array 50 is depicted by way of example. Other configurations of array 50 are possible. In this example, array 50 is planar, that is, transducers 100 are located in a single plane. Transducers 100 are arranged in an 11×20 rectangular layout. In one example, the dimensions of each transducer 100 are 6mm×6mm. The distance between the centers of adjacent transducers (horizontally or vertically) is 7mm. In another example, each element of the 376 transducers consists of multiple sub-transducers (e.g., 2×2 or 3×3), for example, to facilitate the production of such elements, wherein the sub-transducers of a given transducer are connected to the same driver using common wiring.

[0054] Therefore, the total aperture size of the array is 14 cm. In alternative embodiments, the array 50 may be slightly curved or made of a flexible material to conform to the body surface. The array 50 may have any other suitable shape (e.g., rectangular, circular, or elliptical) and any suitable number of transducers in any suitable layout.

[0055] In an embodiment, the size of the individual transducer 100 may be varied on the array 50 to optimize the detection capability of the array 50, such as adding the ability to transmit and detect less US power, which includes very low frequencies (e.g., <50kHz), i.e. increasing probe sensitivity, and / or increasing the bandwidth of the transmitted US signal to higher frequencies (e.g., >1MHz).

[0056] Although Figure 1 An example system layout is shown, but embodiments of the disclosed invention can be implemented in other ways. For example, in one embodiment, the entire system (e.g., probe, cables, and processor) is housed in a handheld device. In other embodiments, processor 48 is located (e.g., via a network) remotely from the probe. The processor may be implemented in a cloud computing network or some off-premise computing resource that performs the reconstruction and sends the reconstructed image to a predefined receiver (e.g., back to display 49 for viewing at the location of the US program).

[0057] Reflection geometry for US computed tomography imaging

[0058] Figure 2 It is based on an embodiment of the present invention. Figure 1 A schematic illustration of the reflection geometry used at least in part by the US reflection tomography imaging system 10. As described above, FWI processing based on reflection geometry allows the use of a probe that captures a considerable portion of the reflection hemisphere 68, which the disclosed reflection tomography imaging system can apply to acquire US signals.

[0059] Figure 2 A beam emitted by a probe in direction 62 (which may define the x-axis) is illustrated by way of example. The disclosed embodiments, and in particular the FWI algorithm used, only consider beams that belong to the reflecting hemisphere and are received by the probe. Such beams (e.g., beam 64) form a complementary angle 66 not exceeding 90 degrees. This means that the reflected US wave will have its direction changed by at least 90 degrees relative to the incident direction. Otherwise, such a beam direction would fall within the transmission hemisphere 70. Disclosed FWI algorithms typically do not consider beams falling within the transmission hemisphere 70.

[0060] As mentioned above, the two-dimensional transducer array of a conventional US probe typically captures only a small portion of the reflecting hemisphere. Therefore, Figure 2 The definitions in the document cover a wider range of possible probe designs for use with the published FWI-based US reflection tomography method.

[0061] Reflection US tomography images reconstructed using FWI

[0062] The disclosed portion provides an iterative FWI algorithm for reconstructing tomographic images of a portion of the human body from US reflection data. To accurately reconstruct such an image, the body is modeled as a lossy medium in which acoustic absorption follows the form... The frequency power law, where It is the absorption ratio coefficient. y is the time frequency, and y is the power law exponent. In some cases, the power law exponent can be assumed to be a constant and equal to 1.5.

[0063] Limited absorption is added to the FWI model's ability to reconstruct images based solely on the acoustic contrast characteristics of generated reflected and scattered signals (e.g., differences in acoustic impedance between different tissue types). However, in some cases, by assuming... In the model, medium absorption can still be ignored.

[0064] The published model follows a paper entitled “Time domain reconstruction of sound speed and attenuation inultrasound computed tomography using full wave inversion” published by M. Perez-Liva et al. in March 2017 in The Journal of the Acoustical Society of America, Volume 141 (Issue 3), page 1595, which discusses modeling of transmission US tomography.

[0065] The linear propagation of sound waves in this medium can be achieved by using position... and time Sound pressure wave of function The fractional Laplace wave equation describes it as follows:

[0066] Equation 1

[0067] in, It is the source term (i.e., the US emission from the probe), and This is the speed of sound in the medium, typically averaging 1530 m / sec. The last two terms describe acoustic absorption and dispersion. and Depend on Given. In the preceding equations, and Based on spatial location And change.

[0068] Figure 3 This is a block diagram illustrating an embodiment of the present invention, which schematically describes the use of... Figure 1 The system 10 collects data and according to Figure 2 The FWI reconstruction algorithm 200, applied to the reflection geometry defined in the standard, performs an iterative process for FWI reconstruction of the US image. Typically, processor 48 uploads algorithm 200 from memory 49 and runs algorithm 200 during the US procedure to iteratively solve equation 1 to generate a quantitative tomographic image of a part of the body from the US reflection data (including scattering). The quantitative tomographic image may be one or more of local density, local sound velocity, and local energy attenuation images.

[0069] As in Figure 3 As seen in the image, the sensor data 202 measured from the reflections entering the US probe array (e.g., array 50) is optionally preprocessed (201) and input into the FWI iterative algorithm 200.

[0070] The processor running the algorithm compares the measured sensor data (202) with the modeled sensor data 206 (204). If (at step 205) a stopping criterion is met, such as the difference between the data being below a given threshold, for example, a threshold defined by L2 or by the level of mismatch in arrival time or by any other suitable metric, the process stops, and the current image 214 becomes the final output image 216. If the stopping criterion is not met, the iterative computation continues with the back-projection step 212 of the difference between the data to update the modeled medium parameters (e.g., attenuation, sound velocity) and generate a new image 214. The acoustic model 208 uses Equation 1 to generate a new set of modeled data 206 from this new image 214.

[0071] like Figure 3 As shown in the specific embodiment, at the start of the iterative calculation, trivial image data (210) is fed to model 208, such as uniform values ​​resulting from a homogeneous medium (e.g., water).

[0072] choose Figure 3 The example illustrations shown are merely for clarity of concept. Figure 3Only the parts relevant to embodiments of the invention are shown. For example, detailed steps such as calibration have been omitted for simplicity.

[0073] In some cases, Algorithm 200 runs offline, for example, by a remote reconstruction and image processing console.

[0074] In other embodiments, in addition to presenting quantitative images, the disclosed techniques are also used to generate qualitative images (such as ultrasound b-mode) of higher quality and resolution based on the characteristics of media-based FWI calculations.

[0075] Figure 4A and Figure 4B The embodiments of the present invention are shown respectively by Figure 1 The graph shows the spectrum of the drive signal 65 and the analog spectrum 75 of the US wave emitted by the system.

[0076] like Figure 4A What we see here, due to Figure 3 The iterative FWI model converges better by using data from lower frequencies and gradually increasing the data frequency during reconstruction iterations, thus driving a wider signal spectrum (e.g., a range greater than 0.5 MHz). Specifically, the lower frequencies allow the FWI model to converge quickly to near the global minimum. The higher frequencies increase the spatial resolution of the reconstructed image.

[0077] Figure 4B Furthermore, Illustration 45 specifically shows that the peak spectral density of the emitted US wave is at 250 kHz, with the majority of the wave energy located below 200 kHz and above 300 kHz in the tail region. The transducer 100 of the probe 30 is designed to achieve such a US emission and detection profile. In particular, the thickness of the transducer elements and the composite material of the matching layer surrounding them are optimized for this purpose.

[0078] Multi-static data acquisition

[0079] Figure 5 It is according to an embodiment of the present invention and Figure 1A schematic top view of a multi-static acquisition mode used with a US probe. In the illustrated embodiment, at any given acquisition time window, one transducer of array 50 (e.g., transducer 401) transmits, while the remaining transducers (e.g., transducer 402) receive the resulting signal. This process is repeated, with each transducer of array 50 acting as a transmitter. In this example, where array 50 has 376 transducers, the received signals can be organized into tensors. The size of the received signal tensor is typically S×R×TS, where S is the number of source transmitters, R is the number of receivers, and TS is the number of A / D sampling time steps. If, for each transducer, transmission is performed multiple times, i.e., N times, for example to increase the SNR, a set of N tensors is obtained. For example, if the number of channels in the acquisition device is limited, scanning can be repeated, i.e., the same transmission is performed multiple times (e.g., N times), where each time the processor changes the receive channel multiplexer settings to collect the received signals from all elements.

[0080] Generally speaking, it is possible to have with Figure 5 The acquisition modes described herein are equivalent to the multi-static acquisition modes. For example, in a given acquisition, one or more transducers (e.g., an entire row or column, or any other suitable subset of transducers) can be used to transmit a US wave. Simultaneously, one or more other transducers (e.g., all remaining transducers) are in receive mode. However, typically, all transducers are used, as there is no restriction on making the transmitting element a receiver immediately after it has completed transmitting its short pulse. Any other arrangement involving the simultaneous use of more than a single transducer can be considered, such as the Hadamakki method applying the aforementioned transducers.

[0081] In some embodiments, the transducers of array 50 are excited in a manner that allows array 50 to deliver more power into the medium in a wide range of directions. For example, the array may be excited in a multi-static configuration. As another example, the array may be excited using a two-dimensional Hadamard base or another two-dimensional base (e.g., Haar), ultimately achieving multi-static equivalent power delivery with a higher SNR. However, as stated above, there are no limitations to using FWI over a set of beamformed signals using the disclosed techniques, as long as the probe illuminates the target volume across sufficiently different angles and distances.

[0082] For example, the processor of system 10 can be configured to control probe 30 to operate with a given emission and respective acquisition sequence, wherein the processor is also configured to convert the acquisition sequence into a multi-static acquisition mode using a given sequence of emitted and reflected US waves, wherein one or more transducers are modeled as emitting US waves, and all transducers are modeled as acquiring reflected US waves (typically, the transducer emitting the wave can become a receiver (immediately after the emission pulse is completed)). Using the multi-static acquisition mode, the processor generates images of the patient's body parts. In particular, the given emission and respective acquisition sequence can be defined by a two-dimensional Hadamazed model.

[0083] The disclosed acquisition modes are particularly useful for FWI reconstruction and can provide extended FOV and imaging depth.

[0084] Methods for FWI reconstruction of US images in reflection

[0085] Figure 6 This is a flowchart illustrating an embodiment of the present invention, which schematically shows the use of... Figure 1 The system collects data and Figure 3 The method for performing FWI reconstruction of US images using an FWI reconstruction algorithm. The process begins at measurement data receiving step 302, where processor 48 receives reflected ultrasound waves from ultrasound probe 30, which are reflected from the patient's body parts.

[0086] Next, at the US image generation step 304, the processor 48 generates an image of the patient's body portion by applying full waveform inversion to the emitted and reflected ultrasound waves. This image includes at least one physiological tissue parameter. Step 304 may include preprocessing steps such as calibration.

[0087] Finally, at image presentation step 306, the processor presents one or more quantitative US images of at least one physiological tissue parameter. Such images may be images of local density, local sound velocity, and local energy attenuation within a body part.

[0088] choose Figure 6 The example flowchart shown is for conceptual clarity only. In particular, for simplicity, many sub-steps related to image acquisition and reconstruction have been omitted.

[0089] Although the embodiments described herein are primarily for reflective medical ultrasound, the methods and systems described herein can also be used in other applications, such as nondestructive testing.

[0090] Therefore, it will be understood that the embodiments described above are referenced by way of example, and the invention is not limited to what has been specifically shown and described above. Rather, the scope of the invention includes combinations and sub-combinations of the various features described above, as well as variations and modifications of the various features that would occur to those skilled in the art upon reading the foregoing description and which are not disclosed in the prior art. Documents incorporated herein by reference are considered part of this application, and in such cases, only the definitions in this specification should be considered, unless any term is defined in those incorporated documents in a manner that conflicts with the definitions expressly or implicitly made in this specification.

Claims

1. A medical ultrasound (US) imaging system, comprising: US probe, the US probe including an array of transducers, the probe being configured to emit US waves and receive reflected ultrasound waves reflected from parts of the patient's body; as well as A processor configured to generate images of a patient's body portion by applying a full waveform inversion FWI process, wherein the full waveform inversion FWI process calculates the medium parameters of the body portion by iteratively solving wave equations and comparing the solved wavefield data with actual measured values ​​of the wavefield data, wherein the FWI process is applied only to US waves emitted from the probe and the reflected US waves, and not to US waves transmitted through the body portion. The array of transducers is two-dimensional (2D) and arranged in a reflective geometry only.

2. The medical ultrasound imaging system according to claim 1, wherein, When generating the image using the FWI process, the processor is configured to estimate and indicate one or more of (i) physiological tissue parameters and (ii) tissue boundaries in the image.

3. The medical ultrasound imaging system according to claim 2, wherein, The physiological tissue parameters include one of the local density, local sound velocity, and local energy decay within the body part.

4. The medical ultrasound imaging system according to any one of claims 1-3, wherein, The processor is configured to jointly apply an FWI model to the US waves emitted from the probe and the reflected US waves at multiple different locations of the probe relative to the body portion.

5. The medical ultrasound imaging system according to any one of claims 1-3, wherein, The processor is configured to control the probe to emit and receive the ultrasonic waves in a series of acquisitions, wherein in each acquisition, a corresponding subset of one or more transducers emits the ultrasonic waves, and one or more other transducers receive the reflected ultrasonic waves.

6. The medical ultrasound imaging system according to claim 5, wherein, The subset of transducers is selected to produce a signal-to-noise ratio (SNR) value higher than a predefined threshold.

7. The medical ultrasound imaging system according to claim 5, wherein, The subset of transducers selected from the series of acquisitions forms a two-dimensional multi-static basis.

8. The medical ultrasound imaging system according to claim 5, wherein, The subset of transducers selected from the series of acquisitions forms a two-dimensional hadamar base.

9. The medical ultrasound imaging system according to any one of claims 1-3, wherein, The processor is configured to apply a low-pass filter to the reflected US wave to generate an initial image of the patient's body portion using the low-pass filtered reflected US wave, and to use the initial image in a subsequent FWI calculation to generate the image.

10. The medical ultrasound imaging system according to any one of claims 1-3, wherein, The array of transducers is two-dimensional.

11. The medical ultrasound imaging system according to any one of claims 1-3, wherein, The processor is located away from the probe.

12. A medical ultrasound (US) imaging method, comprising: Using a US probe that includes an array of transducers, US waves are emitted and reflected ultrasound waves are received from parts of the patient's body, wherein the array of transducers is two-dimensional (2D) and arranged in a reflective geometry only. as well as Images of a patient's body are generated by applying a full waveform inversion FWI process, which calculates the medium parameters of the body by iteratively solving wave equations and comparing the solved wavefield data with actual measured values ​​of the wavefield data. The FWI process is applied only to US waves emitted from the probe and the reflected US waves, and not to US waves transmitted through the body.

13. The medical ultrasound imaging method according to claim 12, wherein, Generating the image includes estimating and indicating one or more of (i) physiological tissue parameters and (ii) tissue boundaries in the image.

14. The medical ultrasound imaging method according to claim 13, wherein, The physiological tissue parameters include one of the local density, local sound velocity, and local energy decay within the body part.

15. The medical ultrasound imaging method according to any one of claims 12-14, wherein, The application of the FWI process includes combining the FWI model with the US waves emitted from the probe and the reflected US waves at multiple different locations of the probe relative to the body portion.

16. The medical ultrasound imaging method according to any one of claims 12-14, wherein, Transmitting and receiving the ultrasonic waves includes performing a series of acquisitions, and in each acquisition, transmitting the ultrasonic waves through a corresponding subset of one or more of the transducers, and receiving the reflected ultrasonic waves through one or more of the other transducers.

17. The medical ultrasound imaging method of claim 16, further comprising selecting the subset of the transducers to generate a signal-to-noise ratio (SNR) value higher than a predefined threshold.

18. The medical ultrasound imaging method according to claim 16, wherein, The subset of transducers selected from the series of acquisitions forms a two-dimensional multi-static basis.

19. The medical ultrasound imaging method according to claim 16, wherein, The subset of transducers selected from the series of acquisitions forms a two-dimensional hadamar base.

20. The medical ultrasound imaging method according to any one of claims 12-14, wherein, The FWI process includes applying a low-pass filter to the reflected US wave, using the low-pass filtered reflected US wave to generate an initial image of the patient's body portion, and using the initial image to generate the image in subsequent FWI calculations.

21. The medical ultrasound imaging method according to claim 13, wherein, The array of transducers is two-dimensional.

22. The medical ultrasound imaging method according to claim 13, wherein, The image is generated at a location remote from the probe.

Images