WO2021088849A1 - Ultrasonic imaging method and apparatus, readable storage medium, and terminal device - Google Patents

Abstract

An ultrasonic imaging method and apparatus, a computer-readable storage medium, and a terminal device. The method comprises: collecting an ultrasonic transmission signal, wherein the ultrasonic transmission signal is a signal formed after an ultrasonic signal passes through a target biological tissue; performing image reconstruction according to the ultrasonic transmission signal to obtain a first image, wherein the first image is an image of the target biological tissue; and processing the first image by using a preset image processing model to obtain a second image, wherein the second image is the image formed after noise and artifacts are removed from the first image, and the image processing model is a neural network model obtained after training using a preset training sample set. According to the method, the image quality restoration process consumes a very short amount of time, such that a good imaging quality can be achieved, while ensuring a higher imaging speed.

Description

Ultrasound imaging method, device, readable storage medium and terminal equipment Technical field

This application belongs to the field of computer technology, and in particular relates to an ultrasonic imaging method, device, computer-readable storage medium, and terminal equipment.

Background technique

The ultrasound imaging methods in the prior art mainly fall into the following categories: The first category is the ultrasound CT reconstruction algorithm based on the straight line model. This algorithm does not require recalculation of the acoustic ray path with the correction of the sound velocity, so the calculation amount is small. The reconstruction speed is faster, but because the model is relatively simple and cannot accurately describe the propagation process of sound waves in biological tissues, the reconstructed images have obvious noise and artifacts; the second type is the ultrasound CT reconstruction algorithm based on the curve model. Compared with the linear model algorithm, this algorithm takes into account the refraction effect of sound waves. The theoretical model is more accurate, so the reconstructed image quality is better. However, because the reconstruction process involves multiple alternating forward and backward propagation processes, Every time the sound velocity distribution of the sound field is updated, the acoustic ray path needs to be recalculated. The amount of calculation is significantly increased, so the reconstruction time is also very long; the third type is the full-wave inversion algorithm, the theory of the algorithm is more complete, and the reconstruction process is the same It requires multiple forward and inversion processes, so the reconstruction quality of the image is also better, and the corresponding cost is a huge amount of calculation and calculation time. In summary, it is difficult for the existing ultrasound imaging methods to take into account both imaging speed and imaging quality.

Summary of the invention

In view of this, the embodiments of the present application provide an ultrasound imaging method, device, computer-readable storage medium, and terminal equipment to solve the problem that the existing ultrasound imaging method is difficult to balance imaging speed and imaging quality.

The first aspect of the embodiments of the present application provides an ultrasound imaging method, which may include:

Collecting ultrasound transmission signals, the ultrasound transmission signals being signals formed after the ultrasound signals pass through the target biological tissue;

Performing image reconstruction according to the ultrasound transmission signal to obtain a first image, where the first image is an imaging of the target biological tissue;

Use a preset image processing model to process the first image to obtain a second image, where the second image is an image formed by removing noise and artifacts from the first image, and the image processing model is A neural network model obtained after training with a preset training sample set. The training sample set includes N training samples, and each training sample includes an input image containing noise and artifacts and a noise and artifact removal Output image, N is a positive integer.

Further, the construction process of any training sample in the training sample set includes:

Construct the original sound velocity distribution image;

Generating a simulated transmission signal corresponding to the original sound velocity distribution image through a simulation experiment;

Performing image reconstruction according to the simulated transmission signal to obtain a reconstructed sound velocity distribution image;

The training sample is constructed according to the original sound velocity distribution image and the reconstructed sound velocity distribution image, wherein the reconstructed sound velocity distribution image is an input image in the training sample, and the original sound velocity distribution image Is the output image in the training sample.

Further, the performing image reconstruction according to the simulated transmission signal to obtain the reconstructed sound velocity distribution image includes:

Calculating the transit time of each acoustic ray according to the simulated transmission signal;

Calculate the distance of each acoustic ray in each pixel grid according to the positions of the transmitting and receiving array elements corresponding to each acoustic ray, and the preset linear model;

Calculate the slowness of each pixel grid according to the transit time of each acoustic ray and the distance that each acoustic ray travels in each pixel grid;

Perform gray value mapping on the slowness of each pixel grid to obtain the reconstructed sound velocity distribution image.

Further, the calculation of the slowness of each pixel grid according to the transit time of each acoustic ray and the distance that each acoustic ray traverses in each pixel grid includes:

An equation set is constructed, where the transit time of each acoustic ray and the distance that each acoustic ray travels in each pixel grid are known quantities in the equation set, and the slowness of each pixel grid is the equation Unknown quantity in the group;

The synchronous algebraic iterative algorithm is used to solve the equations to obtain the slowness of each pixel grid.

Further, the image processing model is a convolutional neural network model based on reaction diffusion equation;

The processing procedure of the image processing model includes:

Identifying the local structural details of the input image through a preset two-dimensional convolution filter, the two-dimensional convolution filter being parameterized by a discrete cosine transform basis;

The local structural details are anisotropically smoothed by a preset influence function to obtain an output image, and the influence function is parameterized by a Gaussian radial basis function.

The second aspect of the embodiments of the present application provides an ultrasonic imaging device, which may include:

A signal acquisition module for acquiring ultrasound transmission signals, the ultrasound transmission signals being signals formed after the ultrasound signals pass through the target biological tissue;

An image reconstruction module, configured to perform image reconstruction according to the ultrasound transmission signal to obtain a first image, where the first image is an imaging of the target biological tissue;

The model processing module is used to process the first image using a preset image processing model to obtain a second image, and the second image is an image formed after noise and artifacts are removed from the first image, The image processing model is a neural network model obtained after training with a preset training sample set. The training sample set includes N training samples, and each training sample includes an input image containing noise and artifacts and a The output image with noise and artifacts removed, N is a positive integer.

A third aspect of the embodiments of the present application provides a computer-readable storage medium that stores a computer program that, when executed by a processor, implements the steps of any of the above-mentioned ultrasound imaging methods.

The fourth aspect of the embodiments of the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, Realize the steps of any of the above-mentioned ultrasound imaging methods.

The fifth aspect of the embodiments of the present application provides a computer program product, which when the computer program product runs on the terminal device, causes the terminal device to execute the steps of any of the above-mentioned ultrasound imaging methods.

Compared with the prior art, the embodiment of the present application has the beneficial effect that: the embodiment of the present application first collects the ultrasound transmission signal, which is the signal formed after the ultrasound signal passes through the target biological tissue, and then according to the ultrasound transmission The signal is image-reconstructed to obtain a first image, the first image is an imaging of the target biological tissue, which will have obvious noise and artifacts, and then the first image is processed using a preset image processing model , Get the second image. Because the image processing model is a neural network model obtained after training with a preset training sample set, and each training sample includes an input image containing noise and artifacts and an output image that removes noise and artifacts , So that the trained image processing model can remove noise and artifacts from the first image, so as to obtain the second image that does not contain noise and artifacts. And because the model has been pre-trained, the process of restoring the image quality takes a very short time, so that while ensuring a faster imaging speed, better imaging quality can be obtained.

Description of the drawings

In order to more clearly describe the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only of the present application. For some embodiments, those of ordinary skill in the art can obtain other drawings based on these drawings without creative labor.

Figure 1 is a schematic diagram of the construction process of any training sample in the training sample set;

Figure 2 is a schematic diagram of image reconstruction based on simulated transmission signals;

Figure 3 is a schematic diagram of part of the training samples in the training sample set;

Figure 4 is a schematic diagram of the overall structure of the image processing model;

FIG. 5 is a flowchart of an embodiment of an ultrasound imaging method in an embodiment of the application;

Figure 6 is a schematic diagram of part of the test results on the test sample set;

FIG. 7 is a schematic diagram of the recovery result of the sound velocity image in the test sample set;

Fig. 8 is a schematic diagram of the sound velocity value distribution along the dotted line in Fig. 7;

FIG. 9 is a structural diagram of an embodiment of an ultrasound imaging device in an embodiment of the application;

FIG. 10 is a schematic block diagram of a terminal device in an embodiment of this application.

Detailed ways

In order to make the purposes, features, and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the following The described embodiments are only a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

It should be understood that when used in this specification and the appended claims, the term "comprising" indicates the existence of the described features, wholes, steps, operations, elements and/or components, but does not exclude one or more other features The existence or addition of, whole, step, operation, element, component and/or its collection.

It should also be understood that the terms used in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in the specification of this application and the appended claims, unless the context clearly indicates other circumstances, the singular forms "a", "an" and "the" are intended to include plural forms.

It should be further understood that the term "and/or" used in the specification and appended claims of this application refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations .

As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "once" or "in response to determination" or "in response to detection" depending on the context . Similarly, the phrase "if determined" or "if detected [described condition or event]" can be interpreted as meaning "once determined" or "in response to determination" or "once detected [described condition or event]" depending on the context ]" or "in response to detection of [condition or event described]".

In addition, in the description of the present application, the terms "first", "second", "third", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

The core of the embodiments of the present application is to use a preset image processing model to process the reconstructed image after the image is reconstructed according to the ultrasound transmission signal to remove noise and artifacts therein, thereby obtaining a high-quality image.

In order to ensure that the image processing model can achieve the function of removing noise and artifacts, it needs to be trained in advance through a large number of training samples. In the embodiment of the present application, a training sample set including N training samples (N is a positive integer) can be constructed in advance to train the image processing model, wherein each training sample includes an input containing noise and artifacts. An image (as input to the image processing model) and an output image with noise and artifacts removed (as the expected output of the image processing model).

As shown in Figure 1, the process of constructing any training sample in the training sample set may include:

Step S101: Construct an original sound velocity distribution image.

Taking into account the sound velocity distribution of different tissue structures in organisms in a normal environment, the constructed sound velocity range can be 1300-1700 meters per second. The relatively simple sound velocity distribution image is mainly composed of some regular geometric figures, which divide the entire imaging area into different areas, and the sound velocity value in each area is set to a fixed value. In order to improve the ability of the image processing model to process various sound velocity distribution images, it is necessary to increase the complexity of the sound velocity distribution image. In the embodiments of this application, the complexity of the sound velocity distribution image is improved from two aspects: on the one hand, the regular geometric figures are twisted and stretched to simulate the complex and irregular boundaries between different tissues in the organism; on the other hand, in the A smooth and continuously changing sound velocity field is generated in the divided areas.

Step S102: Generate a simulated transmission signal corresponding to the original sound velocity distribution image through a simulation experiment.

In the embodiments of this application, the k-Wave ultrasonic simulation toolbox of the MATLAB platform can be used to perform simulation experiments. The k-Wave ultrasonic simulation toolbox can set the position of the ultrasonic transducer elements, signal waveforms, and signal waveforms in the calculation grid. Frequency, launch time, sound velocity, density and attenuation coefficient of the propagation medium, and a perfect matching layer can also be set at the boundary of the calculation grid. In addition, parameters such as the number and size of the calculation grid and the acquisition time of the simulation signal can be set. Using the k-Wave ultrasonic simulation toolbox, an ultrasonic ring array transducer with an internal radius of 9.9 cm, a total of 512 array elements, and a probe center frequency of 1Mhz is set in a 400×400 calculation grid. The previously constructed original sound velocity distribution image is also input into the calculation grid, and each array element is controlled in turn to emit signals and received by all other array elements. Through such a simulation experiment, the original sound velocity distribution image corresponding to the original sound velocity distribution image is generated. Simulate the transmitted signal.

It should be noted that the above process is based on a ring array ultrasonic transducer, but this method can also be transplanted to other types of ultrasonic transducers, such as linear arrays, sector arrays, or triangular arrays.

Step S103: Perform image reconstruction according to the simulated transmission signal to obtain a reconstructed sound velocity distribution image.

As shown in FIG. 2, step S103 may specifically include the following process:

Step S1031: Calculate the transit time of each acoustic ray according to the simulated transmission signal.

After the simulated transmission signal is obtained, the threshold method, the maximum value method or the correlation function method may be used to calculate the transit time of each acoustic ray of the simulated transmission signal.

Step S1032, according to the positions of the transmitting array element and the receiving array element corresponding to each acoustic ray, and the preset linear model, calculate the distance that each acoustic ray passes in each pixel grid.

In order to reconstruct the sound velocity distribution image, first determine the size of the reconstructed image to be 180×180, and then calculate the passage of each sound ray according to the positions of the transmitting and receiving elements corresponding to each sound ray and the preset linear model. All pixel grids and the distance of each sound ray in each pixel grid.

Step S1033: Calculate the slowness of each pixel grid according to the transit time of each acoustic ray and the distance that each acoustic ray passes in each pixel grid.

First, construct a system of equations, where the transit time of each acoustic ray and the distance that each acoustic ray travels in each pixel grid are known quantities in the equations, and the slowness of each pixel grid is determined by The unknown quantity in the above equations.

Specifically, the transit time of the acoustic ray between the transmitting element i and the receiving element j can be expressed as:

Among them, u _k is the sound velocity value in the k-th pixel grid, l _i,j,k are the distances traveled by the acoustic ray between the transmitting element i and the receiving element j in the k-th pixel grid, for For a 512-element ultrasonic transducer, the number of acoustic rays is 512×512, that is, 512×512 equations with the same form as equation (1) can be obtained. These equations can be sorted into the following equation group:

T=L×S Formula (2)

Among them, L is ^{a matrix with a size of 512 2} rows and 180 ² columns, which represents the distance that each acoustic ray travels in each pixel grid. S is a column vector with a length of 180 ² and represents the reciprocal of the speed of sound in each pixel grid, also known as slowness. T is a column vector with a length of 512 ² and represents the transit time of each sound ray.

Then, the synchronous algebraic iterative algorithm can be used to solve the equations to obtain the slowness of each pixel grid.

Specifically, in order to solve the unknown slowness distribution S, the Simultaneous Algebraic Reconstruction Technique (SART) can be used to solve the equations. The iterative formula of the synchronous algebraic iterative algorithm is:

among them,

Is the slowness of the k-th pixel grid after the p-th iteration, the initial value of the slowness of the iteration is all 1/1500, and l _q,k is the distance of the q-th acoustic ray in the k-th pixel grid , Λ is the iterative relaxation coefficient, and its value is preferably set to 0.2.

Step S1034: Perform gray value mapping on the slowness of each pixel grid to obtain the reconstructed sound velocity distribution image.

After obtaining the iterative numerical solution of the equations, it is mapped to a gray value from 0 to 255 by linear compression, thereby obtaining an image of the sound velocity (or slowness) distribution, that is, the reconstructed sound velocity distribution image .

It should be noted that the above process uses an algebraic iterative reconstruction algorithm based on straight line assumptions. In practical applications, other ultrasound CT reconstruction algorithms can also be used, such as filtered back projection algorithm, curve model-based reconstruction algorithm, and full wave inversion Reconstruction algorithm and so on.

Step S104: Construct the training sample according to the original sound velocity distribution image and the reconstructed sound velocity distribution image.

Wherein, the reconstructed sound velocity distribution image is an input image in the training sample, and the original sound velocity distribution image is an output image in the training sample.

Since each simulation experiment takes a long time, it is difficult to generate a large number of training samples in a short time. In the embodiment of the present application, a part of the training samples can be constructed first, and then the data can be expanded to obtain more training samples. In a specific implementation, only 55 training samples can be constructed, and then 440 training samples can be obtained through data expansion, and these training samples are combined to form the training sample set. Figure 3 shows a part of the training sample set. For training samples, each column is a training sample. For any training sample, the upper picture is the original sound velocity distribution image, and the lower picture is the reconstructed sound velocity distribution image.

In the embodiment of the present application, the image processing model may be any machine learning or deep learning neural network model, for example, it may be an image semantic segmentation network model represented by U-Net. Preferably, the convolutional neural network model based on the reaction-diffusion equation is used in the embodiment of the present application, and the model first recognizes the local structure details of the input image through a preset two-dimensional convolution filter, and then, through the preset influence The function performs anisotropic smoothing on the local structural details to obtain an output image.

Specifically, in the method of solving the image restoration problem, nonlinear anisotropic diffusion defines a class of efficient image restoration methods. In each step of the diffusion process, some linear filters are used to convolve the image to identify the outliers in the image, and then the outliers in the image are corrected and smoothed through the spread function. This type of method is derived from the physical model of the free diffusion of matter. Suppose there is a density field ρ(x,y,z,t) in space, which represents the number of molecules of matter in a unit volume at any point in space at time t. Considering the conservation of matter, the change in concentration in a certain area of space must be due to the outflow or inflow of matter into the area. This law can be described by the continuity equation

Among them, F represents the flux field, and div(F) represents the divergence of the flux field F, that is, the external flux density of the material per unit volume per unit time at a certain point.

The flux field is equal to the product of the velocity field and the density field:

F(x,y,z)=u(x,y,z)ρ(x,y,z) Equation (5)

Substances usually move from high-concentration areas to low-concentration areas, and the greater the concentration difference, the more intense the movement. According to Fick's law, the flux can be expressed by a negative gradient of concentration:

Among them, K is the diffusion coefficient, which is used to adjust the relationship between the concentration difference and the diffusion direction.

Substituting formula (6) into formula (3), the following diffusion equation can be obtained:

The physical meaning of formula (4) is that in each small period of time, if the second derivative of the substance concentration at a certain point is greater than 0, then the concentration of that point is increased; conversely, if the second derivative of the substance concentration at a certain point is The derivative is less than 0, then the concentration at that point is reduced. The second derivative greater than 0 means that the concentration value of the point is concave, so the concentration of the point will increase with time; the second derivative less than 0 means that the concentration value of the point is convex, so the concentration of the point increases with time Changes will be reduced. Assuming a two-dimensional image is I(x,y), according to formula (4), the diffusion equation of the image can be obtained as follows:

However, formula (5) is a homogeneous diffusion equation with homogeneity, directly used to smooth the image will cause all the image details to be uniformly blurred. The classic diffusion equation used for image processing is the P-M equation as shown below:

Among them, the diffusion function c is a monotonically decreasing function, so when the absolute value of the gradient of a certain part of the image is large, the diffusion speed of the image at that position is low. Generally, the edge structure gradient value in the image is relatively large, so the P-M equation can smooth the image while protecting the edge structure of the image. By changing the form of the spread function and the directional derivative, the restoration and processing of different types of details and structures of the image can be achieved. However, the partial differential equation corresponding to the traditional anisotropic diffusion process usually has a fixed form. Therefore, different forms of partial differential equations need to be designed for different types of images and different types of processing tasks. In order for the computer to learn appropriate equation parameters for different training images through machine learning, it is necessary to build a learning network model based on the reaction-diffusion equation.

The discrete form of the P-M equation is:

among them:

g(x)=xc(x) Formula (11)

The function c is the spread function, and the function g is the influence function. In order to expand the capability of the diffusion network, the one-dimensional gradient filter in the P-M equation is replaced with a larger two-dimensional convolution filter, and the number of filters is increased to facilitate the extraction of more types of image features. In addition, a numerical fidelity item can be added to control the deviation of the image after the diffusion process from the original image. After the above expansion, the image processing model can be expressed as:

Among them, I ₀ is the input image.

Is the i-th two-dimensional convolution filter in the t-th diffusion process, which is used to extract the local structural features of the image,

For and

The corresponding influence function is based on

The extracted local structural features of the image are used to determine the diffusion speed of the image value at the location, N _t is the number of two-dimensional convolution filters used in _{the t-th diffusion process, μ t} is the relaxation of the t-th diffusion process The coefficient, Δt is the time difference between the two-step diffusion process, and I _t is the image obtained after the t-th diffusion process.

In order to use the training data to train a suitable diffusion network model, the core part of the network model, namely the influence function g and the two-dimensional convolution filter K, can be parameterized.

Among them, the influence function is parameterized by a set of Gaussian radial basis functions. In the embodiment of this application, a set of 63 Gaussian radial basis functions in total is preferably used:

α _n ＝-320+10n,σ＝0.1 formula (13)

The two-dimensional convolution filter is parameterized by a set of discrete cosine transform bases (removing the DC component). The size of the convolution kernel corresponding to the two-dimensional convolution filter is 5×5, and the number of filters is 24. Preferably, a total of 5 steps of diffusion process are set in the embodiment of the present application, that is, T=5. The overall structure of the network model is shown in Figure 4.

After the network model is constructed, the L-BFGS gradient descent algorithm can be used on the training sample set to minimize the loss function to train the parameters of the image processing model. The parameters that need to be trained are the influence function g, the two-dimensional convolution filter K, and the relaxation coefficient μ _t . The loss function to be minimized during the training process is:

Among them, N _s represents the number of training samples,

Represents the true value image of the sth training sample (that is, the expected output),

Represents the actual output of the sth training sample.

After the training of the image processing model is completed, it can be used for ultrasound imaging. Specifically, as shown in FIG. 5, an ultrasound imaging method provided in an embodiment of the present application may include the following processes:

Step S501: Acquire ultrasonic transmission signals.

The ultrasound transmission signal is a signal formed after the ultrasound signal passes through the target biological tissue.

Step S502: Perform image reconstruction according to the ultrasound transmission signal to obtain a first image.

The first image is an imaging of the target biological tissue, in which there will be obvious noise and artifacts. The image reconstruction process in step S502 is similar to the image reconstruction process in step S103. For details, reference may be made to the detailed description in step S103, which will not be repeated here.

Step S503: Use a preset image processing model to process the first image to obtain a second image.

The second image is an image formed after noise and artifacts are removed from the first image.

The trained image processing model was tested on a test sample set different from the training sample set. Some of the test results are shown in Figure 6. The original image is compared with the algebraic iterative method to reconstruct the image and after the image processing model is restored It can be seen that the image processing model effectively removes noise and streak artifacts in the reconstructed image. Although there are obvious image distortions in the reconstructed image, the image processing model recovers a part of the boundary information of the tissue on the basis of these distortions to a certain extent. The mean square error of the sound velocity distribution reconstructed on the test set by the two sound velocity reconstruction methods, the peak signal-to-noise ratio of the sound velocity image, and the mean and standard deviation of the structural similarity of the sound velocity image are shown in the following table:

The three quantitative indicators all show that after the restoration process of the image processing model, the accuracy of the sound velocity reconstruction and the quality of the sound velocity image have been significantly improved, indicating the effectiveness of the diffusion network reconstruction method.

In addition, in order to more intuitively investigate the correction effect of the diffusion network on the reconstructed sound velocity value, Fig. 7 shows the restoration results of the sound velocity images of the three test sample sets, and the sound velocity value distribution along the dotted line is shown in Fig. 8. It can be seen from Figure 8 that the sound velocity distribution reconstructed by the traditional algebraic iterative method has strong noise interference, and large errors will occur at the position where the sound velocity value jumps. After the diffusion network is restored, the noise is effectively transplanted, and the error in the sound velocity jump position is also better corrected, thereby improving the accuracy of the sound velocity distribution reconstruction result.

To sum up, the embodiment of the present application first collects ultrasound transmission signals, which are signals formed after the ultrasound signals pass through the target biological tissue, and then perform image reconstruction based on the ultrasound transmission signals to obtain the first image. The first image is an imaging of the target biological tissue, in which there will be obvious noise and artifacts, and then a preset image processing model is used to process the first image to obtain a second image. Because the image processing model is a neural network model obtained after training with a preset training sample set, and each training sample includes an input image containing noise and artifacts and an output image that removes noise and artifacts , So that the trained image processing model can remove noise and artifacts from the first image, so as to obtain the second image that does not contain noise and artifacts. And because the model has been pre-trained, the process of restoring the image quality takes a very short time, so that while ensuring a faster imaging speed, better imaging quality can be obtained.

It should be understood that the size of the sequence number of each step in the foregoing embodiment does not mean the order of execution. The execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

Corresponding to the ultrasound imaging method described in the above embodiment, FIG. 9 shows a structural diagram of an embodiment of an ultrasound imaging apparatus provided in an embodiment of the present application.

In this embodiment, an ultrasonic imaging apparatus may include:

The signal acquisition module 901 is configured to acquire ultrasound transmission signals, which are signals formed after the ultrasound signals pass through the target biological tissue;

The image reconstruction module 902 is configured to perform image reconstruction according to the ultrasound transmission signal to obtain a first image, where the first image is an imaging of the target biological tissue;

The model processing module 903 is configured to process the first image using a preset image processing model to obtain a second image, and the second image is an image formed after noise and artifacts are removed from the first image The image processing model is a neural network model obtained after training with a preset training sample set, the training sample set includes N training samples, and each training sample includes an input image containing noise and artifacts and An output image with noise and artifacts removed, N is a positive integer.

Further, the ultrasound imaging device may further include:

Sound velocity distribution construction module, used to construct the original sound velocity distribution image;

A transmission signal generating module, which is used to generate a simulated transmission signal corresponding to the original sound velocity distribution image through a simulation experiment;

A simulation reconstruction module for performing image reconstruction according to the simulated transmission signal to obtain a reconstructed sound velocity distribution image;

A training sample construction module, configured to construct the training sample according to the original sound velocity distribution image and the reconstructed sound velocity distribution image, wherein the reconstructed sound velocity distribution image is the input image in the training sample, The original sound velocity distribution image is an output image in the training sample.

Further, the simulation reconstruction module may include:

The transit time calculation sub-module is used to calculate the transit time of each acoustic ray according to the simulated transmission signal;

The distance calculation sub-module is used to calculate the distance of each acoustic ray in each pixel grid according to the positions of the transmitting and receiving array elements corresponding to each acoustic ray, and the preset straight line model;

The slowness calculation sub-module is used to calculate the slowness of each pixel grid according to the transit time of each sound ray and the distance that each sound ray passes in each pixel grid;

The image reconstruction sub-module is used to perform gray value mapping on the slowness of each pixel grid to obtain the reconstructed sound velocity distribution image.

Further, the slowness calculation submodule may include:

The equation group construction unit is used to construct the equation group, where the transit time of each acoustic ray and the distance that each acoustic ray travels in each pixel grid are the known quantities in the equation group, and each pixel grid The slowness of is the unknown quantity in the equations;

The iterative solving unit is used to solve the equation system using a synchronous algebraic iterative algorithm to obtain the slowness of each pixel grid.

Those skilled in the art can clearly understand that, for the convenience and conciseness of the description, the specific working processes of the above described devices, modules and units can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in an embodiment, reference may be made to related descriptions of other embodiments.

FIG. 10 shows a schematic block diagram of a terminal device provided by an embodiment of the present application. For ease of description, only parts related to the embodiment of the present application are shown.

As shown in FIG. 10, the terminal device 10 of this embodiment includes: a processor 100, a memory 101, and a computer program 102 stored in the memory 101 and running on the processor 100. When the processor 100 executes the computer program 102, the steps in the above embodiments of the ultrasound imaging method are implemented, for example, step S501 to step S503 shown in FIG. 5. Alternatively, when the processor 100 executes the computer program 102, the functions of the modules/units in the foregoing device embodiments, for example, the functions of the modules 901 to 903 shown in FIG. 9 are realized.

Exemplarily, the computer program 102 may be divided into one or more modules/units, and the one or more modules/units are stored in the memory 101 and executed by the processor 100 to complete This application. The one or more modules/units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program 102 in the terminal device 10.

The terminal device 10 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server. Those skilled in the art can understand that FIG. 10 is only an example of the terminal device 10, and does not constitute a limitation on the terminal device 10. It may include more or less components than shown in the figure, or a combination of certain components, or different components For example, the terminal device 10 may also include an input/output device, a network access device, a bus, and the like.

The processor 100 may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.

The memory 101 may be an internal storage unit of the terminal device 10, such as a hard disk or a memory of the terminal device 10. The memory 101 may also be an external storage device of the terminal device 10, such as a plug-in hard disk equipped on the terminal device 10, a smart memory card (Smart Media Card, SMC), and a Secure Digital (SD) Card, Flash Card, etc. Further, the memory 101 may also include both an internal storage unit of the terminal device 10 and an external storage device. The memory 101 is used to store the computer program and other programs and data required by the terminal device 10. The memory 101 can also be used to temporarily store data that has been output or will be output.

Those skilled in the art can clearly understand that, for the convenience and conciseness of description, only the division of the above functional units and modules is used as an example. In practical applications, the above functions can be allocated to different functional units and modules as needed. Module completion, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist alone physically, or two or more units can be integrated into one unit. The above-mentioned integrated units can be hardware-based Formal realization can also be realized in the form of software functional units. In addition, the specific names of the functional units and modules are only used to facilitate distinguishing from each other, and are not used to limit the protection scope of the present application. For the specific working process of the units and modules in the foregoing system, reference may be made to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail or recorded in an embodiment, reference may be made to related descriptions of other embodiments.

A person of ordinary skill in the art may realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.

In the embodiments provided in this application, it should be understood that the disclosed device/terminal device and method may be implemented in other ways. For example, the device/terminal device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units. Or components can be combined or integrated into another system, or some features can be omitted or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.

In addition, the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.

If the integrated module/unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiments and methods, and can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the program is executed by the processor, it can implement the steps of the foregoing method embodiments. Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file, or some intermediate forms. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electrical carrier signal, telecommunications signal, and software distribution media, etc. It should be noted that the content contained in the computer-readable medium can be appropriately added or deleted according to the requirements of the legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to the legislation and patent practice, the computer-readable medium Does not include electrical carrier signals and telecommunication signals.

The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, a person of ordinary skill in the art should understand that it can still implement the foregoing The technical solutions recorded in the examples are modified, or some of the technical features are equivalently replaced; these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the application, and should be included in Within the scope of protection of this application.

Claims (10)

1. An ultrasound imaging method, characterized in that it comprises:

   Collecting ultrasound transmission signals, the ultrasound transmission signals being signals formed after the ultrasound signals pass through the target biological tissue;

   Performing image reconstruction according to the ultrasound transmission signal to obtain a first image, where the first image is an imaging of the target biological tissue;

   Use a preset image processing model to process the first image to obtain a second image, where the second image is an image formed by removing noise and artifacts from the first image, and the image processing model is A neural network model obtained after training with a preset training sample set. The training sample set includes N training samples, and each training sample includes an input image containing noise and artifacts and a noise and artifact removal Output image, N is a positive integer.
2. The ultrasound imaging method according to claim 1, wherein the process of constructing any training sample in the training sample set comprises:

   Construct the original sound velocity distribution image;

   Generating a simulated transmission signal corresponding to the original sound velocity distribution image through a simulation experiment;

   Performing image reconstruction according to the simulated transmission signal to obtain a reconstructed sound velocity distribution image;

   The training sample is constructed according to the original sound velocity distribution image and the reconstructed sound velocity distribution image, wherein the reconstructed sound velocity distribution image is an input image in the training sample, and the original sound velocity distribution image Is the output image in the training sample.
3. The ultrasonic imaging method according to claim 2, wherein the image reconstruction according to the simulated transmission signal to obtain the reconstructed sound velocity distribution image comprises:

   Calculating the transit time of each acoustic ray according to the simulated transmission signal;

   Calculate the distance of each acoustic ray in each pixel grid according to the positions of the transmitting and receiving array elements corresponding to each acoustic ray, and the preset linear model;

   Calculate the slowness of each pixel grid according to the transit time of each acoustic ray and the distance that each acoustic ray travels in each pixel grid;

   Perform gray value mapping on the slowness of each pixel grid to obtain the reconstructed sound velocity distribution image.
4. The ultrasonic imaging method according to claim 3, wherein the calculation of the slowness of each pixel grid according to the transit time of each acoustic ray and the distance that each acoustic ray passes in each pixel grid comprises:

   An equation set is constructed, where the transit time of each acoustic ray and the distance that each acoustic ray travels in each pixel grid are known quantities in the equation set, and the slowness of each pixel grid is the equation Unknown quantity in the group;

   The synchronous algebraic iterative algorithm is used to solve the equations to obtain the slowness of each pixel grid.
5. The ultrasound imaging method according to claim 4, wherein the iterative formula of the synchronous algebraic iterative algorithm is:

   

   among them,

   

   Is the slowness of the k-th pixel grid after the p-th iteration, l _q,k is the distance traveled by the q-th acoustic ray in the k-th pixel grid, and λ is the iterative relaxation coefficient.
6. The ultrasound imaging method according to any one of claims 1 to 5, wherein the image processing model is a convolutional neural network model based on a reaction diffusion equation;

   The processing procedure of the image processing model includes:

   Identifying the local structural details of the input image through a preset two-dimensional convolution filter, the two-dimensional convolution filter being parameterized by a discrete cosine transform basis;

   The local structural details are anisotropically smoothed by a preset influence function to obtain an output image, and the influence function is parameterized by a Gaussian radial basis function.
7. The ultrasonic imaging method according to claim 6, wherein the image processing model is expressed as:

   

   Among them, I ₀ is the input image,

   

   Is the i-th two-dimensional convolution filter in the t-th diffusion process,

   

   For and

   

   Corresponding influence function, N _t is the number of two-dimensional convolution filters used in the t-th diffusion process, μ _t is the relaxation coefficient of the t-th diffusion process, Δt is the time difference between the two diffusion processes , I _t is the image obtained after the diffusion process in the t-th step.
8. An ultrasonic imaging device, characterized in that it comprises:

   A signal acquisition module for acquiring ultrasound transmission signals, the ultrasound transmission signals being signals formed after the ultrasound signals pass through the target biological tissue;

   An image reconstruction module, configured to perform image reconstruction according to the ultrasound transmission signal to obtain a first image, where the first image is an imaging of the target biological tissue;

   The model processing module is used to process the first image using a preset image processing model to obtain a second image, and the second image is an image formed after noise and artifacts are removed from the first image, The image processing model is a neural network model obtained after training with a preset training sample set. The training sample set includes N training samples, and each training sample includes an input image containing noise and artifacts and a The output image with noise and artifacts removed, N is a positive integer.
9. A computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program implements the ultrasound imaging method according to any one of claims 1 to 7 when the computer program is executed by a processor A step of.
10. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program as claimed in claims 1 to 7. The steps of the ultrasound imaging method described in any one of 7.

Images