Matching layer material parameter prediction method and device, electronic equipment and storage medium

Abstract

The invention discloses a matching layer material parameter prediction method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining sound pressure distribution information in a working domain based on a matching layer sound velocity matrix, a matching layer density matrix and preset sound source information, inputting the sound pressure distribution information into an initial matching layer material parameter prediction model, and predicting the matching layer material parameter. A training material parameter matrix is obtained, the initial matching layer material parameter prediction model is updated based on the training material parameter matrix and the packing ratio matrix, and a trained matching layer material parameter prediction model is obtained; and inputting the target skull sound pressure information into the trained matching layer material parameter prediction model to obtain a target material parameter matrix. The target material parameter matrix of the matching layer can be obtained based on the target skull sound pressure information, the matching layer manufactured based on the target material parameter matrix can enable the sound pressure information in the skull to meet the target skull sound pressure information, manual calculation by experience is not needed, and the efficiency is high.

Description

Technical Field

[0001] The present application relates to the technical field of acoustic impedance matching materials, and in particular to a matching layer material parameter prediction method, device, electronic device and storage medium. Background Art

[0002] Transcranial ultrasound is a method that uses ultrasonic energy to penetrate the skull and directly act on the lesion area in the brain. It mainly uses high-intensity focused ultrasound (HIFU) technology to emit low-energy ultrasound through an external transducer, focusing on the target area in the body, generating instantaneous high temperature, thereby inducing thermal effects, mechanical effects and cavitation effects.

[0003] In the related art, in order to make the sound pressure information inside the skull meet the sound pressure requirements, the ultrasound transcranial is optimized through a matching layer. For example, the matching layer parameters are calculated manually based on experience according to the impedance data on the transducer surface, and then the matching layer is made based on the matching layer parameters. However, in the related art, when the matching layer is made by manually calculating the matching layer parameters, the sound pressure inside the skull is difficult to meet the sound pressure requirements. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a matching layer material parameter prediction method, device, electronic device and storage medium, which can obtain a target material parameter matrix of the matching layer based on the target skull sound pressure information, and the matching layer manufactured based on the target material parameter matrix can make the sound pressure information in the skull meet the target skull sound pressure information.

[0005] To achieve the above-mentioned purpose, the first aspect of the present application provides a matching layer material parameter prediction method, including:

[0006] constructing a skull model, and constructing a working domain including the skull model;

[0007] Based on the skull model, determining a matching layer region in the working domain, dividing the matching layer region into a plurality of grids of equal size and connected to each other, configuring a filling ratio for each grid, and obtaining a filling ratio matrix; wherein the filling ratio represents a material parameter within the grid;

[0008] Based on the filling ratio, respectively calculating the sound velocity and density of each grid, and obtaining a matching layer sound velocity matrix and a matching layer density matrix;

[0009] Based on the matching layer sound velocity matrix, the matching layer density matrix and preset sound source information, obtaining sound pressure distribution information in the working domain;

[0010] Inputting the sound pressure distribution information into an initial matching layer material parameter prediction model to obtain a training material parameter matrix, and updating the initial matching layer material parameter prediction model based on the training material parameter matrix and the filling ratio matrix to obtain a trained matching layer material parameter prediction model;

[0011] Target skull sound pressure information is obtained, and the target skull sound pressure information is input into the trained matching layer material parameter prediction model to obtain a target material parameter matrix.

[0012] According to some embodiments of the present application, constructing a skull model includes:

[0013] Obtaining computed tomography images of multiple skull slices of different thicknesses scanned at different angles;

[0014] determining Hounsfield unit values ​​at multiple locations of the skull based on the computed tomography image;

[0015] Based on the Hounsfield unit values ​​at multiple positions of the skull, an average value of the Hounsfield units is calculated, and the skull model is constructed based on the average value of the Hounsfield units; wherein the Hounsfield unit values ​​at each position of the skull model are all the average value of the Hounsfield units.

[0016] According to some embodiments of the present application, configuring a fill ratio for each grid to obtain a fill ratio matrix includes:

[0017] Setting a value range of the filling ratio;

[0018] Determine a starting column grid among multiple grids, randomly select a value from the value range as the filling ratio of the starting column grid, so that the filling ratio of each column grid increases or decreases column by column within the value range; wherein the filling ratios of grids located in the same column are the same.

[0019] According to some embodiments of the present application, the number of columns of the grid in the matching layer region is N;

[0020] The randomly selecting a value from the value range as the filling ratio of the starting column grid, so that the filling ratio of each column grid increases or decreases column by column within the value range, comprises:

[0021] A value X1 is randomly selected from the value range as the filling ratio of the starting column grid, and the filling ratios of the second column grid to the Nth column grid are calculated according to a first calculation formula, wherein the first calculation formula is:

[0022]

[0023] Among them, X iCharacterizes the fill ratio of the i-th column grid, X i-1 Characterizes the filling ratio of the i-1th column of grids, i is 2, 3..., N; c is a preset attenuation factor.

[0024] According to some embodiments of the present application, calculating the filling ratio of the second column of grids to the Nth column of grids according to the first calculation formula includes:

[0025] Detection X i Whether it exceeds the value range, when X is detected i If the value exceeds the range, the preset attenuation factor in the first calculation formula is increased, and X is recalculated. i , so that X i Within the value range; X i Characterizes the filling ratio of the i-th column of the grid, where i is 2, 3, ..., N.

[0026] According to some embodiments of the present application, inputting the sound pressure distribution information into an initial matching layer material parameter prediction model to obtain a training material parameter matrix includes:

[0027] Extracting the sound pressure information around the skull model from the sound pressure distribution information;

[0028] The sound pressure information around the skull is input into the initial matching layer material parameter prediction model to obtain the training material parameter matrix.

[0029] According to some embodiments of the present application, before obtaining the sound pressure distribution information in the working domain based on the matching layer sound velocity matrix, the matching layer density matrix and the preset sound source information, the method further includes:

[0030] Setting the focal position of the sound source information around the skull model;

[0031] The step of extracting the sound pressure information around the skull model from the sound pressure distribution information includes:

[0032] The focal point periphery sound pressure information around the focal position is extracted from the sound pressure distribution information as the skull periphery sound pressure information.

[0033] To achieve the above-mentioned purpose, a second aspect of the present application provides a matching layer material parameter prediction device, including:

[0034] A construction module, used for constructing a skull model and a working domain including the skull model;

[0035] A configuration module, for determining a matching layer region in the working domain based on the skull model, dividing the matching layer region into a plurality of grids of equal size and connected to each other, configuring a filling ratio for each grid, and obtaining a filling ratio matrix; the filling ratio represents a material parameter in the grid;

[0036] A calculation module, used to calculate the sound velocity and density of each grid based on the filling ratio, and obtain a matching layer sound velocity matrix and a matching layer density matrix;

[0037] A sound pressure determination module, configured to obtain sound pressure distribution information in the working domain based on the matching layer sound velocity matrix, the matching layer density matrix and preset sound source information;

[0038] A training module, used for inputting the sound pressure distribution information into an initial matching layer material parameter prediction model to obtain a training material parameter matrix, and updating the initial matching layer material parameter prediction model based on the training material parameter matrix and the filling ratio matrix to obtain a trained matching layer material parameter prediction model;

[0039] The application module is used to obtain target skull sound pressure information, input the target skull sound pressure information into the trained matching layer material parameter prediction model, and obtain a target material parameter matrix.

[0040] To achieve the above-mentioned purpose, an embodiment of the third aspect of the present application provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the matching layer material parameter prediction method described in any one of the embodiments of the first aspect when executing the computer program.

[0041] To achieve the above-mentioned purpose, a fourth aspect of the present application provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the matching layer material parameter prediction method described in any one of the first aspect embodiments is implemented.

[0042] The matching layer material parameter prediction method, device, electronic device and storage medium of the embodiment of the present application first prepare a training sample of the matching layer material parameter prediction model. Specifically, a skull model and a working domain including the skull model are constructed, and then a matching layer area is determined in the working domain, and the matching layer area is divided into a plurality of grids of equal size and connected to each other, and a filling ratio is configured for each grid to obtain a filling ratio matrix; since the filling ratio characterizes the material parameters in the grid, the filling ratio is configured for each grid, which can be used to simulate the characteristics of a real matching layer; then the sound velocity and density of each grid are calculated respectively to obtain a matching layer sound velocity matrix and a matching layer density matrix. Since the transmission of ultrasonic waves in a medium is mainly related to the sound velocity and density of the medium, the sound pressure distribution information in the working domain can be obtained based on the matching layer sound velocity matrix, the matching layer density matrix and the preset sound source information, and the sound pressure distribution information is used as a training sample of the matching layer material parameter prediction model. Input the sound pressure distribution information into the initial matching layer material parameter prediction model to obtain the training material parameter matrix, update the initial matching layer material parameter prediction model based on the training material parameter matrix and the filling ratio matrix, and obtain the trained matching layer material parameter prediction model; obtain the target skull sound pressure information, input the target skull sound pressure information into the trained matching layer material parameter prediction model, and obtain the target material parameter matrix. In this way, the present application can obtain the target material parameter matrix of the matching layer based on the target skull sound pressure information, and the matching layer made based on the target material parameter matrix can make the sound pressure information in the skull meet the target skull sound pressure information, and there is no need to manually calculate using experience, which is highly efficient.

[0043] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The present application is further described below with reference to the accompanying drawings and embodiments, wherein:

[0045] Figure 1 A schematic diagram of a flow chart of a matching layer material parameter prediction method according to an embodiment of the present application;

[0046] Figure 2 A schematic diagram of a partial structure of a working domain according to an embodiment of the present application;

[0047] Figure 3 A schematic diagram of simulation using a crystal cell and an aluminum cell according to an embodiment of the present application;

[0048] Figure 4 It is a schematic diagram of the relationship between filling ratio, equivalent sound velocity and equivalent density;

[0049] Figure 5 for Figure 1A specific flow chart of step S110;

[0050] Figure 6 for Figure 1 A specific flow chart of step S120 in FIG. 1 ;

[0051] Figure 7 for Figure 1 A specific flow chart of step S150;

[0052] Figure 8 A schematic diagram of target skull sound pressure information according to an embodiment of the present application;

[0053] Fig. 9 It is a schematic diagram of the first sound pressure information of a skull without a matching layer under the action of a preset focused sound field;

[0054] Fig.10 for Fig. 9 Schematic diagram of the sound pressure at the focus of the section;

[0055] Fig.11 It is a schematic diagram of the second sound pressure information of a skull provided with a target matching layer under the action of a preset focused sound field;

[0056] Fig.12 for Fig.11 Schematic diagram of the sound pressure at the focus of the section;

[0057] Fig.13 This is a schematic diagram of the structure of a matching layer material parameter prediction device according to an embodiment of the present application;

[0058] Fig.14 A schematic diagram of the hardware structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0059] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0060] In the description of the present application, it should be understood that descriptions involving orientation, such as up, down, front, back, left, right, etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0061] In the description of this application, "several" means more than one, "more" means more than two, "greater than", "less than", "exceed", etc. are understood to exclude the number itself, and "above", "below", "within", etc. are understood to include the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0062] In the description of this application, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in this application based on the specific content of the technical solution.

[0063] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0064] First, some nouns involved in this application are analyzed:

[0065] Matching layer: The matching layer is an important concept in transcranial ultrasound. It is located between the transducer and the skull and is used to improve the propagation efficiency of ultrasound between two media (usually water and skull). Since the skull has strong attenuation and absorption of ultrasound, direct penetration of the skull will result in a large loss of ultrasound energy. By selecting materials with appropriate acoustic impedance and sound velocity, the matching layer can reduce the reflection and scattering of ultrasound when passing through the matching layer, thereby improving the transmittance and focusing effect of ultrasound.

[0066] Transducer: The transducer is a key component in the transcranial ultrasound system. It is responsible for converting electrical energy into ultrasonic energy and focusing the ultrasonic energy on the target area in the body through focusing technology. The performance of the transducer directly affects the focusing effect and treatment efficiency of the ultrasound. With the development of phased transducer technology, more precise ultrasonic focusing and more efficient energy transmission can be achieved.

[0067] Skull transmittance: Skull transmittance refers to the proportion of energy remaining after ultrasound penetrates the skull. Due to the attenuation and absorption of ultrasound by the skull, the energy of ultrasound will be significantly reduced after penetrating the skull. The level of skull transmittance directly affects the effect of ultrasound transcranial treatment. In order to improve the skull transmittance, a variety of measures can be taken, such as optimizing the design of the matching layer, selecting the appropriate ultrasound frequency and power, and improving the focusing technology of the transducer.

[0068] Sound pressure: Sound pressure refers to the change in atmospheric pressure caused by sound wave disturbance, that is, the residual pressure of atmospheric pressure, which is equivalent to superimposing a pressure change caused by sound wave disturbance on the atmospheric pressure. Specifically, when there is a sound field in a medium (such as air), the difference between the pressure at a certain point at a certain moment and the static pressure at that point when there is no sound wave is called the sound pressure at that point. The magnitude of sound pressure reflects the strength of the sound wave and is an important physical quantity used to describe sound waves in acoustics. The unit of sound pressure is Pascal (Pa), which is the basic unit of pressure in the International System of Units. In addition, sound pressure can also be expressed in other units, such as bar, where 1 bar is equal to 100KPa.

[0069] In the related art, ultrasound transcranial optimization technology is roughly divided into two types, improving the sound source and making a matching layer. For example, the time inversion method is used to obtain the sound source sequence information of the focused sound field, optimize the focused sound field, and overcome the phase mismatch. However, the impedance matching problem between the transducer, water, and skull has not been solved. Most of the energy focused by the ultrasound is still absorbed by the skull, resulting in low transmittance, which may not reach the focal energy intensity required for treatment. For example, a matching layer close to the transducer is made according to the impedance data on the surface of the transducer to overcome the impedance mismatch between the water and the transducer surface. However, the impedance mismatch at the water-skull interface has not been solved, and a large amount of energy will still accumulate in the skull, resulting in too low transmittance. And there is currently no better way to quickly calculate the material parameters of different matching layers based on different skulls.

[0070] Based on this, the embodiments of the present application provide a matching layer material parameter prediction method, device, electronic device and storage medium, which can obtain a target material parameter matrix of the matching layer based on the target skull sound pressure information, and the matching layer manufactured based on the target material parameter matrix can make the sound pressure information in the skull meet the target skull sound pressure information. The manufactured matching layer can improve the focusing ability of transcranial ultrasound and improve the skull transmittance.

[0071] The matching layer material parameter prediction method of the embodiment of the present application can be applied to a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server can be configured as an independent physical server, or a server cluster or a distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the matching layer material parameter prediction method, etc., but is not limited to the above forms.

[0072] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0073] Reference Figure 1 The first aspect of the present application provides a method for predicting matching layer material parameters. Figure 1 The illustrated method flow includes but is not limited to steps S110 to S160.

[0074] Step S110, constructing a skull model and a working domain including the skull model;

[0075] In one embodiment, in order to make the training samples produced subsequently conform to the actual situation, a skull model is constructed based on the real skull, and a working domain including the skull model is constructed. The working domain is used to simulate the distribution area of ​​ultrasound waves around the skull during transcranial ultrasound. Figure 2 , Figure 2 It is a schematic diagram of part of the structure of the working domain, and the skull model is located in the working domain.

[0076] Step S120, based on the skull model, determining a matching layer region in the working domain, dividing the matching layer region into a plurality of grids of equal size and connected to each other, configuring a filling ratio for each grid, and obtaining a filling ratio matrix; the filling ratio represents a material parameter in the grid;

[0077] It is worth noting that the matching layer area is the area for placing the matching layer. Generally, the transducer is used as an ultrasonic emission source, and the transducer is deployed on one side of the skull. The transducer transmits ultrasonic waves to the skull, forming ultrasonic distribution information around the skull. When performing transcranial ultrasound, the matching layer is usually located between the transducer and the skull. Therefore, in the embodiment of the present application, in the working domain, a part of the area between the position of the transducer and the position of the skull model is selected as the matching layer area. For example, referring to Figure 2 , a 36*10 grid is selected from the working domain as the matching layer area. The matching layer area is composed of multiple grids of equal size and connected to each other. Each grid is an independent unit. There are 360 ​​independent units in the matching layer area that can be used to place different materials. The sound velocity and impedance of each column of materials from left to right are larger than those of the previous column, showing a gradual increasing trend.

[0078] Step S130, based on the filling ratio, respectively calculating the sound velocity and density of each grid to obtain a matching layer sound velocity matrix and a matching layer density matrix;

[0079] In one embodiment, the present application is designed based on the equivalent medium theory of microspheres periodically arranged in water, so that one grid represents a material, and the material is simulated by a crystal cell and an aluminum cell, referring to Figure 3 , aluminum cells are filled in the crystal cells, and the aluminum cells are located in the center of the crystal cells. The filling ratio refers to the volume ratio of aluminum cells to crystal cells in a grid. The filling ratio is used to characterize the properties of the material. Figure 3 In the case of 2 / bb 2 The speed of sound refers to the speed at which sound propagates in a medium. The speed of sound of a crystal cell is 1500 m / s, and the speed of sound of an aluminum cell is 2800 m / s. Therefore, the speed of sound of the material corresponding to the filling ratio of each grid can be calculated based on the filling ratio, thereby obtaining the matching layer speed matrix. Each element in the matching layer speed matrix is ​​the speed of sound corresponding to a grid. The density of the crystal cell is 1000 kg / m 3 , the density of aluminum unit cell is 1800kg / m 3 Therefore, the density of the material corresponding to the filling ratio of each grid can be calculated based on the filling ratio, thereby obtaining the matching layer density matrix. Each element in the matching layer density matrix is ​​the density corresponding to a grid.

[0080] Step S140, obtaining sound pressure distribution information in the working domain based on the matching layer sound velocity matrix, the matching layer density matrix and preset sound source information;

[0081] It is worth noting that the sound source information refers to the relevant information of the ultrasonic source, including the frequency, amplitude, ultrasonic emission source position and focal position of the ultrasonic wave. In one embodiment, the frequency of the ultrasonic wave is set to 0.75 MHz, and the amplitude is set to 1*10 6 Pa, the focal position is set at a preset position in the working domain. The sound pressure distribution information refers to the distribution of the sound pressure of the ultrasonic wave in the working domain.

[0082] In one embodiment, the matching layer sound velocity matrix, the matching layer density matrix, the frequency of the ultrasonic wave, the amplitude of the ultrasonic wave, the focal position of the ultrasonic wave, and the position of the ultrasonic wave emission source are input into the comsol toolkit, and the coordinate information of the working domain and the position information of the skull in the working domain are input into the comsol toolkit, and the comsol toolkit is used to perform simulation to obtain the sound pressure distribution of the ultrasonic wave in the working domain, that is, to obtain the sound pressure distribution information of the working domain. It should be noted that the full name of the comsol toolkit is COMSOL Multiphysics, which is used to simulate the propagation of sound waves in non-uniform media, can output waveforms at any point in space, and effectively generate steady-state pressure distribution at the frequency of interest.

[0083] Step S150, inputting the sound pressure distribution information into the initial matching layer material parameter prediction model to obtain a training material parameter matrix, and updating the initial matching layer material parameter prediction model based on the training material parameter matrix and the filling ratio matrix to obtain a trained matching layer material parameter prediction model;

[0084] In one embodiment, the resnet50 model is imported from the ResNet library as a matching layer material parameter prediction model. ResNet (Residual Network) is a very important convolutional neural network structure in the field of deep learning. It provides an effective solution to the problem of gradient vanishing in the deep network training process. The core idea of ​​ResNet is to solve the problem of gradient vanishing in the deep network training process by introducing Skip Connection. In the forward propagation process of the traditional deep network, information needs to pass through multiple network layers in sequence, and in the reverse propagation, the gradient also needs to be propagated back layer by layer through multiple layers. When the number of network layers is deep, the layer-by-layer propagation of the gradient will lead to the problem of gradient vanishing or gradient explosion. ResNet solves this problem by introducing skip connections in the network. In the skip connection, the input can be directly passed to the output through the cross-layer connection, so that the gradient has a shorter path to propagate back, thereby avoiding the problem of gradient vanishing or gradient explosion. The basic module of ResNet is the Residual Block, which consists of two or three convolutional layers and skip connections. The output and input of the two convolutional layers are added together and then passed through a nonlinear activation function to form the final residual block output.

[0085] It is worth noting that after passing through step S110 to step S140, the filling ratio matrix and the sound pressure distribution information corresponding to the filling ratio matrix are obtained as training samples. The sound pressure distribution information is input into the initial matching layer material parameter prediction model to obtain a training material parameter matrix, which represents the parameters of the matching layer material predicted during the training process. Then, based on the training material parameter matrix and the filling ratio matrix, an error comparison is performed on the initial matching layer material parameters to obtain the corresponding loss value. By returning the gradient of the loss value, the parameters of the matching layer material parameter prediction model are readjusted, thereby updating and optimizing the matching layer material parameter prediction model. Among them, the loss value is calculated by L1loss, and the learning rate is set to 1e-5.

[0086] In one embodiment, it is necessary to repeatedly perform steps S110 to S140 to obtain more training samples. In this process, the sound pressure distribution information corresponds to the filling ratio matrix one by one, so a mapping relationship between the sound pressure distribution information and the filling ratio matrix can be established. After the sound pressure distribution information is input into the initial matching layer material parameter prediction model and the training material parameter matrix is ​​obtained, the filling ratio matrix corresponding to the sound pressure distribution information can be determined based on the mapping relationship, and then the initial matching layer material parameter prediction model can be updated based on the training material parameter matrix and the corresponding filling ratio matrix.

[0087] In one embodiment, steps S110 to S140 are repeated 2000 times in total, thereby obtaining 2000 sets of data, each set of data including a sound pressure distribution information and a filling ratio matrix. A part of the 2000 sets of data is used as a training set, and the other part is used as a verification set. Step S150 is repeatedly performed based on the training set to iteratively update the matching layer material parameter prediction model until the number of iterations reaches a preset threshold or the loss value converges, thereby obtaining a trained matching layer material parameter prediction model. Then, the trained matching layer material parameter prediction model is verified using the data of the verification set, and a parameter file with the best performance on the verification set among all parameter files of the matching layer material parameter prediction model is selected and saved, thereby obtaining a trained matching layer network prediction model.

[0088] In another embodiment, the sound pressure distribution information is a sound pressure distribution matrix. Since the matching layer material parameter prediction model has a weak perception of the characteristics of the sound pressure distribution information, for example, the difference between an excellent focused sound field and a distorted focused sound field in terms of acoustic angle is very large, but after conversion into a sound pressure distribution matrix, it is difficult for the model to perceive the difference between an excellent focused sound field and a distorted focused sound field. In order to improve the sensitivity of the model to the sound pressure distribution information, a weight matrix with the same size as the sound pressure distribution matrix is ​​preset. In the weight matrix, the weight value corresponding to the focal position is set to a maximum value of 10, the farther from the center, the lower the weight value, and the minimum value of 0 is set at the edge. Then, the sound pressure distribution matrix of each set of data in the 2000 sets of data is multiplied by the weight matrix to obtain an updated sound pressure distribution matrix, thereby updating the data, and then the matching layer material parameter prediction model is trained based on the updated 2000 sets of data.

[0089] Step S160, obtaining target skull sound pressure information, inputting the target skull sound pressure information into the trained matching layer material parameter prediction model, and obtaining a target material parameter matrix.

[0090] It is worth noting that the target skull sound pressure information refers to the sound pressure requirement that the ultrasound wave is expected to achieve around the skull. Normally, the sound pressure achieved by the ultrasound wave around the skull needs to meet certain requirements to achieve the therapeutic purpose. The target skull sound pressure information is input into the trained matching layer material parameter prediction model to obtain the target material parameter matrix. Relevant technicians can make a matching layer based on the target material parameter matrix, and then apply the matching layer to the ultrasound transcranial treatment. During the application process, the sound pressure of the ultrasound wave around the skull can meet the sound pressure requirements.

[0091] In the embodiment of the present application, through the above steps S110 to S160, a training sample of the matching layer material parameter prediction model is first prepared. Specifically, a skull model and a working domain including the skull model are constructed, and then a matching layer area is determined in the working domain, and the matching layer area is divided into a plurality of grids of equal size and connected to each other, and a filling ratio is configured for each grid to obtain a filling ratio matrix; since the filling ratio characterizes the material parameters in the grid, the filling ratio is configured for each grid, which can be used to simulate the characteristics of a real matching layer; then the sound velocity and density of each grid are calculated respectively to obtain a matching layer sound velocity matrix and a matching layer density matrix. Since the transmission of ultrasonic waves in a medium is mainly related to the sound velocity and density of the medium, the sound pressure distribution information in the working domain can be obtained based on the matching layer sound velocity matrix, the matching layer density matrix and the preset sound source information, and the sound pressure distribution information is used as a training sample of the matching layer material parameter prediction model. Input the sound pressure distribution information into the initial matching layer material parameter prediction model to obtain the training material parameter matrix, update the initial matching layer material parameter prediction model based on the training material parameter matrix and the filling ratio matrix, and obtain the trained matching layer material parameter prediction model; obtain the target skull sound pressure information, input the target skull sound pressure information into the trained matching layer material parameter prediction model, and obtain the target material parameter matrix. In this way, the present application can obtain the target material parameter matrix of the matching layer based on the target skull sound pressure information, and the matching layer made based on the target material parameter matrix can make the sound pressure information in the skull meet the target skull sound pressure information, and there is no need to manually calculate using experience, which is highly efficient.

[0092] In one embodiment, referring to Figure 5 , Figure 5 for Figure 1 A specific flow chart of step S110 in FIG. Figure 5 The illustrated process steps include but are not limited to step S510 to step S530.

[0093] Step S510, acquiring computerized tomography images of a plurality of skull slices of different thicknesses scanned at different angles;

[0094] It is worth noting that computed tomography (CT) is a medical imaging diagnostic technology that uses X-rays for tomographic imaging. CT uses a rotating X-ray scanner to perform continuous scans at multiple angles in the patient's body, and then uses a computer to process and reconstruct the scanned data to generate high-resolution tomographic images.

[0095] Step S520, determining the Hounsfield unit values ​​at multiple locations of the skull based on the computed tomography image;

[0096] It is worth noting that the Henry unit value, also known as the CT value, is a unit of measurement for determining the density of a certain local tissue or organ in the human body, medically known as the Henry unit (HU). It is a relative value converted from the X-ray attenuation coefficient measured by the detector through a certain mathematical model, and is used to indicate the density of body tissues.

[0097] Step S530, based on the Hounsfield unit values ​​at multiple locations of the skull, calculate the average value of the Hounsfield unit, and construct a skull model based on the average value of the Hounsfield unit; wherein the Hounsfield unit value at each location of the skull model is the average value of the Hounsfield unit.

[0098] The present application calculates the CT value of each position of the real skull through steps S510 to S530, and then calculates the average value of the Hounsfield unit based on multiple CT values ​​to construct a skull model, so that the CT value of each position in the skull model is the average value of the Hounsfield unit, so that the skull module is a uniform medium.

[0099] In one embodiment, referring to Figure 6 , Figure 6 for Figure 1 A specific flow chart of step S120 in FIG. Figure 6 The illustrated step flow includes but is not limited to step S610 to step S620.

[0100] Step S610, setting a value range of the filling ratio;

[0101] It is worth noting that the value range of the filling ratio is set to (0, 1).

[0102] Step S620, determining a starting column grid among the multiple grids, randomly selecting a value from a value range as the filling ratio of the starting column grid, so that the filling ratio of each column grid increases or decreases column by column within the value range; wherein the filling ratios of grids in the same column are the same.

[0103] In one embodiment, referring to Figure 2 and Figure 3 , in the multi-column grid, the grid closest to the line with the horizontal axis of 0 is used as the starting column grid, and then starting from the starting column grid, the filling ratio of each column grid increases from left to right within the value range. Since the filling ratio increases column by column, the sound speed and density of each column grid also increase column by column, refer to Figure 4 , Figure 4 It is a schematic diagram of the relationship between filling ratio, equivalent sound velocity and equivalent density. Figure 4In the figure, the solid line represents the relationship between the equivalent density and the filling ratio, and the dotted line represents the relationship between the equivalent sound speed and the filling ratio. As the filling ratio changes from 0 to 1, the equivalent sound speed of the corresponding material also changes from the water sound speed (1500m / s) to the average sound speed of the skull (2800m / s). Since the filling ratio produced in this application is increased column by column, the elements in the filling ratio matrix are also increased column by column, so that the target material parameter matrix predicted by the matching layer material parameter prediction model is also increased column by column. The material parameters of the matching layer are increased column by column, which can make the matching layer achieve a gradual effect, can alleviate the impedance matching problem between water and skull to a certain extent, and can enrich the training data set. In one embodiment, the filling ratio is randomly increased column by column.

[0104] In another embodiment, starting from the starting column grid, the filling ratio of each column grid decreases column by column from left to right within the value range. Since the filling ratio produced by the present application decreases column by column, the elements in the filling ratio matrix also decrease column by column, so that the target material parameter matrix predicted by the matching layer material parameter prediction model also decreases column by column. The decreasing material parameters of the matching layer can make the matching layer achieve a gradual effect, which can alleviate the impedance matching problem between water and skull to a certain extent, and can enrich the training data set.

[0105] In one embodiment, the number of columns of the grid in the matching layer region is N, and step S620 may include the following steps:

[0106] A value X1 is randomly selected from the value range as the filling ratio of the starting column grid, and the filling ratios of the grids from the second column to the Nth column are calculated according to the first calculation formula. The first calculation formula is:

[0107]

[0108] Among them, X i Characterizes the filling ratio of the i-th column grid, X i-1 Characterizes the filling ratio of the i-1th column grid, i is 2, 3..., N; c is a preset attenuation factor. It should be noted that c is a positive number with a large value. Those skilled in the art can set the value of c according to actual needs to avoid the filling ratio exceeding the value range. Through the first calculation formula, the filling ratio of each column grid can be increased column by column within the value range, and the increasing relationship is neither geometric nor arithmetic, which is more in line with the actual situation. Among them, X1 is the filling ratio of the starting column grid.

[0109] When calculating using the first calculation formula, detect X i Whether it exceeds the value range, when X is detected i If the value exceeds the range, the preset attenuation factor in the first calculation formula is increased and X is recalculated. i, so that X i Within the value range; X i Characterizes the fill ratio of the i-th column grid, i is 2, 3..., N. In one embodiment, N is 75. In addition, refer to Figure 3 Since the sound velocity is positively correlated with the filling ratio, and the density is positively correlated with the filling ratio, the sound velocity of the material also increases column by column. The impedance of the material is positively correlated with the sound velocity, so the impedance also increases column by column.

[0110] In another embodiment, step S620 may include the following steps:

[0111] A value X1 is randomly selected from the value range as the filling ratio of the starting column grid, and the filling ratios of the grids from the 2nd column to the Nth column are calculated according to the second calculation formula. The second calculation formula is:

[0112]

[0113] Among them, X i Characterizes the filling ratio of the i-th column grid, X i-1 Characterizes the filling ratio of the i-1th column grid, i is 2, 3..., N; c is a preset attenuation factor. It should be noted that c is a positive number with a large value. Those skilled in the art can set the value of c according to actual needs to avoid the filling ratio exceeding the value range. Through the second calculation formula, the filling ratio of each column grid can be reduced column by column within the value range. Among them, X1 is the filling ratio of the starting column grid.

[0114] When calculating using the second calculation formula, detect X i Whether it exceeds the value range, when X is detected i If the value exceeds the range, the preset attenuation factor in the second calculation formula is increased and X is recalculated. i , so that X i Within the value range; X i Characterizes the filling ratio of the i-th column of the grid, where i is 2, 3, ..., N.

[0115] In one embodiment, referring to Figure 7 , Figure 7 for Figure 1 A specific flow chart of step S150 in FIG. Figure 8 The illustrated process steps include but are not limited to step S810 to step S820.

[0116] Step S810, extracting the sound pressure information around the skull model from the sound pressure distribution information;

[0117] Step S820: Input the sound pressure information around the skull into the initial matching layer material parameter prediction model to obtain a training material parameter matrix.

[0118] It is worth noting that the sound pressure distribution information includes the sound pressure distribution of the entire working domain, and the present application intercepts the sound pressure information around the skull model from the sound pressure distribution information through steps S810 to S820, and inputs the sound pressure information around the skull into the initial matching layer material parameter prediction model to obtain a training material parameter matrix, which can reduce the size of the data input into the initial matching layer material parameter prediction model, thereby improving calculation efficiency.

[0119] In one embodiment, before step S140, the focal position of the sound source information needs to be set around the skull model, and in step S810, specifically: the focal periphery sound pressure information around the focal position is intercepted from the sound pressure distribution information as the skull periphery sound pressure information. In this way, the matching layer material parameter prediction model of the embodiment of the present application only focuses on the sound pressure distribution at the focal point, which can improve the calculation efficiency.

[0120] It is worth noting that referring to Figure 8 , Figure 8 : is a schematic diagram of the target skull sound pressure information of the embodiment of the present application. The target skull sound pressure information is the sound pressure information obtained in the preset focused sound field without a skull. The target skull sound pressure information is input into the trained matching layer material parameter prediction model to obtain the target material parameter matrix, and the target matching layer is prepared through the target material parameter matrix. Fig. 9 , Fig. 9 is a schematic diagram of the first sound pressure information of a skull without a matching layer under the action of a preset focused sound field; Fig.10 , Fig.10 for Fig. 9 Schematic diagram of the sound pressure at the focus of the cross section. Fig.11 , Fig.11 Schematic diagram of the second sound pressure information of a skull with a target matching layer under the action of a preset focused sound field. Fig.12 , Fig.12 for Fig.11 Schematic diagram of the sound pressure at the focus in the cross section. Figures 9 to 12 It can be clearly seen that after applying the target matching layer, the position and shape of the focus have been significantly improved, the energy at the focus has also been significantly improved, and the sound pressure value at the focus has been increased by 60% compared to the sound pressure value of the skull alone without a matching layer.

[0121] Reference Fig.13 The second aspect of the present application provides a matching layer material parameter prediction device, including:

[0122] A construction module 1410 is used to construct a skull model and a working domain including the skull model;

[0123] A configuration module 1420 is used to determine a matching layer region in the working domain based on the skull model, divide the matching layer region into a plurality of grids of equal size and connected to each other, and configure a filling ratio for each grid to obtain a filling ratio matrix; the filling ratio represents a material parameter in the grid;

[0124] A calculation module 1430 is used to calculate the sound velocity and density of each grid based on the filling ratio to obtain a matching layer sound velocity matrix and a matching layer density matrix;

[0125] A sound pressure determination module 1440 is used to obtain sound pressure distribution information in the working domain based on the matching layer sound velocity matrix, the matching layer density matrix and preset sound source information;

[0126] A training module 1450 is used to input the sound pressure distribution information into an initial matching layer material parameter prediction model to obtain a training material parameter matrix, and to update the initial matching layer material parameter prediction model based on the training material parameter matrix and the filling ratio matrix to obtain a trained matching layer material parameter prediction model;

[0127] The application module 1460 is used to obtain target skull sound pressure information, and input the target skull sound pressure information into the trained matching layer material parameter prediction model to obtain a target material parameter matrix.

[0128] The matching layer material parameter prediction device can execute the matching layer material parameter prediction method of the embodiment of the present application. When executing the method, a training sample of the matching layer material parameter prediction model is first prepared. Specifically, a skull model and a working domain including the skull model are constructed, and then a matching layer area is determined in the working domain, and the matching layer area is divided into a plurality of grids of equal size and connected to each other, and a filling ratio is configured for each grid to obtain a filling ratio matrix; since the filling ratio characterizes the material parameters in the grid, the filling ratio is configured for each grid, which can be used to simulate the characteristics of a real matching layer; then the sound velocity and density of each grid are calculated respectively to obtain a matching layer sound velocity matrix and a matching layer density matrix. Since the transmission of ultrasonic waves in a medium is mainly related to the sound velocity and density of the medium, the sound pressure distribution information in the working domain can be obtained based on the matching layer sound velocity matrix, the matching layer density matrix and the preset sound source information, and the sound pressure distribution information is used as a training sample of the matching layer material parameter prediction model. Input the sound pressure distribution information into the initial matching layer material parameter prediction model to obtain the training material parameter matrix, update the initial matching layer material parameter prediction model based on the training material parameter matrix and the filling ratio matrix, and obtain the trained matching layer material parameter prediction model; obtain the target skull sound pressure information, input the target skull sound pressure information into the trained matching layer material parameter prediction model, and obtain the target material parameter matrix. In this way, the present application can obtain the target material parameter matrix of the matching layer based on the target skull sound pressure information, and the matching layer made based on the target material parameter matrix can make the sound pressure information in the skull meet the target skull sound pressure information, and there is no need to manually calculate using experience, which is highly efficient.

[0129] The specific implementation of the matching layer material parameter prediction device is basically the same as the specific implementation of the matching layer material parameter prediction method in the above-mentioned embodiment, and will not be repeated here. On the premise of meeting the requirements of the embodiment of the present application, the matching layer material parameter prediction device can also be provided with other functional modules to implement the matching layer material parameter prediction method in the above-mentioned embodiment.

[0130] To achieve the above purpose, the third aspect of the present application provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the matching layer material parameter prediction method of the above embodiment when executing the computer program. The electronic device can be any smart terminal including a tablet computer, a car computer, etc.

[0131] In one embodiment, referring to Fig.14 , Fig.14 The hardware structure of the electronic device of the embodiment of the present application is illustrated, and the electronic device includes:

[0132] The processor 151 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0133] The memory 152 may be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 152 may store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 152, and the processor 151 calls and executes the matching layer material parameter prediction method of the embodiment of this application;

[0134] Input / output interface 153, used to implement information input and output;

[0135] The communication interface 154 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0136] A bus 155 that transmits information between the various components of the device (e.g., the processor 151, the memory 152, the input / output interface 153, and the communication interface 154);

[0137] The processor 151 , the memory 152 , the input / output interface 153 and the communication interface 154 are connected to each other in communication within the device via the bus 155 .

[0138] To achieve the above-mentioned purpose, a fourth aspect of the present application provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the matching layer material parameter prediction method of the first aspect of the present application is implemented.

[0139] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0140] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0141] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0142] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0143] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0144] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0145] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0146] In the several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0147] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0148] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0149] If the integrated 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 technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0150] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A matching layer material parameter prediction method, characterized in that: include: constructing a skull model, and constructing a working domain including the skull model; Based on the skull model, determining a matching layer region in the working domain, dividing the matching layer region into a plurality of grids of equal size and connected to each other, configuring a filling ratio for each grid, and obtaining a filling ratio matrix; wherein the filling ratio represents a material parameter within the grid; Based on the filling ratio, respectively calculating the sound velocity and density of each grid to obtain a matching layer sound velocity matrix and a matching layer density matrix; Based on the matching layer sound velocity matrix, the matching layer density matrix and preset sound source information, obtaining sound pressure distribution information in the working domain; Inputting the sound pressure distribution information into an initial matching layer material parameter prediction model to obtain a training material parameter matrix, and updating the initial matching layer material parameter prediction model based on the training material parameter matrix and the filling ratio matrix to obtain a trained matching layer material parameter prediction model; Target skull sound pressure information is obtained, and the target skull sound pressure information is input into the trained matching layer material parameter prediction model to obtain a target material parameter matrix.

2. The matching layer material parameter prediction method according to claim 1, characterized in that: The method of constructing a skull model comprises: Obtaining computed tomography images of multiple skull slices of different thicknesses scanned at different angles; determining Hounsfield unit values ​​at multiple locations of the skull based on the computed tomography image; Based on the Hounsfield unit values ​​at multiple positions of the skull, an average value of the Hounsfield units is calculated, and the skull model is constructed based on the average value of the Hounsfield units; wherein the Hounsfield unit values ​​at each position of the skull model are all the average value of the Hounsfield units.

3. The matching layer material parameter prediction method according to claim 1, characterized in that: The filling ratio is configured for each grid to obtain a filling ratio matrix, including: Setting a value range of the filling ratio; Determine a starting column grid among multiple grids, randomly select a value from the value range as the filling ratio of the starting column grid, so that the filling ratio of each column grid increases or decreases column by column within the value range; wherein the filling ratios of grids located in the same column are the same.

4. The matching layer material parameter prediction method according to claim 3, characterized in that: The number of columns of the grid in the matching layer region is N; The randomly selecting a value from the value range as the filling ratio of the starting column grid, so that the filling ratio of each column grid increases or decreases column by column within the value range, comprises: A value X1 is randomly selected from the value range as the filling ratio of the starting column grid; the filling ratios of the second column grid to the Nth column grid are calculated according to a first calculation formula, and the first calculation formula is: Among them, X i Characterizes the fill ratio of the i-th column grid, X i-1 Characterizes the filling ratio of the i-1th column of grids, i is 2, 3..., N; c is a preset attenuation factor.

5. The matching layer material parameter prediction method according to claim 4, characterized in that: The step of calculating the filling ratio of the second column of grids to the Nth column of grids according to the first calculation formula includes: Detection X i Whether it exceeds the value range, when X is detected i If the value exceeds the range, the preset attenuation factor in the first calculation formula is increased, and X is recalculated. i , so that X i Within the value range; X i Characterizes the filling ratio of the i-th column of the grid, where i is 2, 3, ..., N.

6. The matching layer material parameter prediction method according to claim 1, characterized in that: The step of inputting the sound pressure distribution information into an initial matching layer material parameter prediction model to obtain a training material parameter matrix includes: Extracting the sound pressure information around the skull model from the sound pressure distribution information; The sound pressure information around the skull is input into the initial matching layer material parameter prediction model to obtain the training material parameter matrix.

7. The matching layer material parameter prediction method according to claim 4, characterized in that: Before obtaining the sound pressure distribution information in the working domain based on the matching layer sound velocity matrix, the matching layer density matrix and the preset sound source information, the method further includes: Setting the focal position of the sound source information around the skull model; The step of extracting the sound pressure information around the skull model from the sound pressure distribution information includes: The focal point periphery sound pressure information around the focal position is extracted from the sound pressure distribution information as the skull periphery sound pressure information.

8. A matching layer material parameter prediction device, characterized in that: include: A construction module, used for constructing a skull model and a working domain including the skull model; A configuration module, for determining a matching layer region in the working domain based on the skull model, dividing the matching layer region into a plurality of grids of equal size and connected to each other, configuring a filling ratio for each grid, and obtaining a filling ratio matrix; the filling ratio represents a material parameter in the grid; A calculation module, used to calculate the sound velocity and density of each grid based on the filling ratio, and obtain a matching layer sound velocity matrix and a matching layer density matrix; A sound pressure determination module, configured to obtain sound pressure distribution information in the working domain based on the matching layer sound velocity matrix, the matching layer density matrix and preset sound source information; A training module, used for inputting the sound pressure distribution information into an initial matching layer material parameter prediction model to obtain a training material parameter matrix, and updating the initial matching layer material parameter prediction model based on the training material parameter matrix and the filling ratio matrix to obtain a trained matching layer material parameter prediction model; The application module is used to obtain target skull sound pressure information, input the target skull sound pressure information into the trained matching layer material parameter prediction model, and obtain a target material parameter matrix.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the matching layer material parameter prediction method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the matching layer material parameter prediction method according to any one of claims 1 to 7 is implemented.

Images