A single-element super-resolution ultrasonic imaging method based on acoustic metamaterials

Abstract

The application provides a single-element super-resolution ultrasonic imaging method based on acoustic metamaterials, which comprises the following steps: first, an acoustic field coding board is made of acoustic metamaterials to construct a single-element super-resolution ultrasonic imaging system; then, the system is calibrated in terms of acoustic field, and the system matrix in the imaging area is obtained by means of angular spectrum method and automatic convolution; secondly, the flowing microbubbles in the tissue are imaged by the single-element super-resolution ultrasonic imaging system to obtain a group of simulated original ultrasonic data, the acoustic coding board is rotated to different angles, the multi-angle data of the flowing microbubbles are collected, and multi-angle original ultrasonic echo data are obtained; then, based on the system matrix and the obtained original ultrasonic echo data, the ultrasonic microbubble image is reconstructed by means of linear inversion algorithm; finally, the center of the ultrasonic microbubble image is positioned by a single-molecule positioning method, the microbubble position is obtained, the microbubble positioning results of all frames are superimposed, and the single-element super-resolution ultrasonic imaging is realized.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of ultrasonic imaging, and particularly relates to a single-element super-resolution ultrasonic imaging method based on acoustic metamaterials. BACKGROUND

[0002] Super-resolution ultrasound technology is proposed and widely used in preclinical experiments, showing super-high spatial resolution. Compared with traditional ultrasonic imaging, with the help of ultrasonic contrast agent, the super-resolution ultrasonic imaging method performs single-molecule positioning on flowing ultrasonic microbubbles, realizes ultrasonic imaging with super-high resolution by accumulating microbubble positioning information, breaks through the acoustic diffraction limit, and realizes subwavelength scale observation of in-vivo structure information.

[0003] At present, super-resolution ultrasonic imaging mostly uses a linear array transducer to realize two-dimensional imaging, and uses a two-dimensional surface array to realize three-dimensional super-resolution ultrasonic imaging by moving and rotating the linear array. Three-dimensional ultrasonic images have been proven to have great auxiliary effect on the evaluation of many medical problems in the clinic, and are an imaging technology with great clinical application value. However, due to the limitation of the Nyquist theorem, the spatial resolution and imaging speed of three-dimensional ultrasonic imaging are highly dependent on high-quality surface array ultrasonic transducers, high data transmission rates and integrated electronic devices. The ultrasonic image quality directly affects the accuracy of single-molecule positioning, and therefore limits the application range of three-dimensional super-resolution ultrasonic imaging.

[0004] In order to solve this problem, in recent years, compressed sensing technology (CS) has been widely used in various imaging devices. This technology attempts to use the original natural structure of the signal to sample the signal at a much lower Nyquist frequency, thereby compressing the signal into a much smaller signal than the original signal, and then accurately reconstructing the original signal through numerical optimization algorithm.

[0005] Considering that super-resolution ultrasonic imaging is usually based on a linear array and is imaged by translation to a three-dimensional imaging space, the quality of three-dimensional ultrasonic images is also limited by the number and quality of ultrasonic elements. The present application attempts to realize super-resolution ultrasonic imaging by a single element with the help of compressed sensing technology and acoustic field coding board, thereby reducing the hardware requirements of the imaging system for super-resolution imaging and realizing efficient super-resolution ultrasonic imaging. Considering that real-time performance has always been a major advantage of ultrasonic imaging, the image reconstruction quality needs to be improved under the premise of ensuring imaging speed. SUMMARY

[0006] The present application aims at the deficiencies existing in the prior art, and provides a single-element super-resolution ultrasound imaging method based on acoustic metamaterials, which reduces the requirements of the imaging system hardware for super-resolution ultrasound imaging and realizes single-element super-resolution imaging while ensuring the imaging quality.

[0007] The present application provides a single-element super-resolution ultrasound imaging method based on acoustic metamaterials, which comprises the following steps: S1, an acoustic field coding board is made of acoustic metamaterials, and a single-element super-resolution ultrasound imaging system is constructed based on the acoustic field coding board; S2, the single-element super-resolution ultrasound imaging system is calibrated for the acoustic field, and a system matrix in an imaging area is obtained; S3, with the aid of ultrasound contrast agents, the single-element super-resolution ultrasound imaging system is used to perform ultrasound imaging on randomly activated flow microbubbles in the tissue to be imaged in the imaging area, so as to obtain a group of simulated original ultrasound data; S4, the acoustic coding board is rotated to different angles, and the single-element super-resolution ultrasound imaging system is used to collect multi-angle imaging data of the flow microbubbles, so as to obtain multi-angle original ultrasound echo data; S5, based on the system matrix, a linear inversion algorithm is used to reconstruct ultrasound microbubble images according to the multi-angle original ultrasound echo data; S6, a single molecule positioning method is used to position the center of each ultrasound microbubble image, so as to obtain the microbubble positioning results in the corresponding images; and S7, all the microbubble positioning results are superimposed, so as to realize single-element-based super-resolution ultrasound imaging.

[0008] The single-element super-resolution ultrasound imaging method based on acoustic metamaterials provided by the present application can also have the following technical features: the single-element super-resolution ultrasound imaging system at least comprises a single-element ultrasound transducer, an acoustic field coding board and an imaging area, the acoustic field coding board is arranged in front of the single-element ultrasound transducer, the surface of the acoustic field coding board is covered with randomly distributed circular holes, the diameter of the circular holes is 1mm, and the depth is 0.1mm-1mm.

[0009] The single-element super-resolution ultrasound imaging method based on acoustic metamaterials provided by the present application can also have the following technical features: the system matrix is obtained by the following process: in the imaging space, a certain plane acoustic field in the acoustic propagation direction at different distances from the acoustic field coding board is measured by means of a three-dimensional translation table and a hydrophone, the plane acoustic fields at different distances are calculated and obtained by an angular spectrum method, and then the echo signals received by the single-element ultrasound transducer are calculated by convolution, so as to obtain the system matrix at a single phase plate angle; after the acoustic field is measured by measuring the four angle coding boards, the system matrix at the corresponding four angles is obtained by the convolution calculation, so as to form the system matrix.

[0010] The single-element super-resolution ultrasonic imaging method based on acoustic metamaterials provided by the application can also have the following technical features: in step S5, the linear inversion algorithm is: assuming that the received echo signal is linearly related to the image v, the echo signal u can be expressed as:

[0011] u=Hv+n

[0012] In the formula, u is an M-dimensional echo signal, M represents the product of the number of time sampling points and the number of rotations; n is an M-dimensional additive noise; H is a system matrix, which is determined by an ultrasonic measurement device, and H is a two-dimensional matrix with a size of M*N, N represents the number of pixels in an imaging plane; the image reconstruction solving problem is expressed based on the linear inversion algorithm as:

[0013]

[0014] The linear solving is performed on the pre-processed conjugate gradient algorithm.

[0015] The single-element super-resolution ultrasonic imaging method based on acoustic metamaterials provided by the application can also have the following technical features: step S6 includes the following sub-steps: step S6-1, pre-processing is performed on the image, and the pre-processing includes denoising and microbubble extraction; step S6-2, a single molecule is positioned by using a Gaussian fitting method: assuming that the point spread function (PSF) of the ultrasonic microbubble conforms to a Gaussian distribution, that is, the PSF of the microbubble can be written as a 2D Gaussian function with different characteristics, and therefore the ultrasonic microbubble image can be described as:

[0016]

[0017] In the formula, (x0, y0) represents the actual position of the microbubble, A represents the amplitude, and sigma x and sigma y are standard deviations in each direction of the image plane; step S6-3, a least square method is used to require that the residual sum of squares is minimum to solve the nonlinear optimization problem in the formula in step S6-2, and a corresponding microbubble positioning point is obtained by fitting.

[0018] Invention action and effect

[0019] ​According to the single-element super-resolution ultrasonic imaging method based on the acoustic metamaterial, first, an acoustic field coding plate is made of the acoustic metamaterial to construct a single-element super-resolution ultrasonic imaging system; then, the system is calibrated in the acoustic field, and the system matrix in an imaging area is obtained by means of the angular spectrum method and automatic convolution; secondly, the single-element super-resolution ultrasonic imaging system is used to perform ultrasonic imaging on the flowing microbubbles in the tissue to obtain a group of simulated original ultrasonic data, and the acoustic coding plate is rotated to different angles to collect multi-angle data of the flowing microbubbles to obtain multi-angle original ultrasonic echo data; then, based on the system matrix and the obtained original ultrasonic echo data, the ultrasonic microbubble image is reconstructed by means of the linear inversion algorithm; finally, the center of the ultrasonic microbubble image is positioned by the single molecule positioning method to obtain the microbubble position, and the microbubble positioning results of all frames are superimposed, so that the single-element-based super-resolution ultrasonic imaging is realized.

[0020] The method provided by the application can realize high-quality ultrasonic imaging in three-dimensional space by using only a single element, greatly reduces the requirements of super-resolution ultrasonic imaging on hardware devices (including the number of sensors and the data transmission rate), transfers the complexity from the hardware system to the computer, greatly reduces the manufacturing cost of the device, and enables the ultrasonic imaging to be more widely used. In addition, the method only uses single-element ultrasonic signals to reconstruct the ultrasonic image, to a certain extent, improves the data acquisition speed, and enables the super-resolution ultrasonic imaging of moving tissues to be possible. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 is a flowchart of the single-element super-resolution ultrasonic imaging method based on the acoustic metamaterial in the embodiment of the application;

[0022] Figure 2 is a structural schematic diagram of the single-element super-resolution ultrasonic imaging system in the embodiment of the application;

[0023] Figure 3 is a result schematic diagram of reconstructing the image based on the PCG method and positioning the reconstructed image based on the single molecule positioning in the embodiment of the application;

[0024] Figure 4 is a super-resolution ultrasonic image obtained based on the method in the embodiment of the application. DETAILED DESCRIPTION

[0025] In order to make the technical means, creative features, purposes and effects realized by the application easy to understand, the single-element super-resolution ultrasonic imaging method based on the acoustic metamaterial is specifically described below in combination with the embodiments and the drawings.

[0026] <EMBODIMENT>

[0027] Figure 1is a flow chart of a single-element super-resolution ultrasonic imaging method based on acoustic metamaterial in an embodiment of the present application.

[0028] As shown in Figure 1 , the single-element super-resolution ultrasonic imaging method based on acoustic metamaterial in the embodiment includes the following steps:

[0029] Step S1, an acoustic field coding board is made of acoustic metamaterial, and a single-element super-resolution ultrasonic imaging system is constructed based on the acoustic field coding board.

[0030] Figure 2 is a structural schematic diagram of a single-element super-resolution ultrasonic imaging system in an embodiment of the present application.

[0031] In the embodiment, a piece of acoustic field coding board is made by means of 3D printing technology, and the size of the acoustic field coding board is consistent with the size of the single-element ultrasonic transducer. The surface of the acoustic field coding board is covered with randomly distributed circular holes with a diameter of 1 mm and a depth of 0.1-1 mm. The acoustic field coding board is placed in front of the ultrasonic transducer, which can realize local delay of acoustic waves and thus produce a time-varying ultrasonic field, playing a role of acoustic field coding. The finally constructed single-element super-resolution ultrasonic imaging system is shown in Figure 2 .

[0032] Step S2, before formal imaging, the acoustic field of the single-element super-resolution ultrasonic imaging system is calibrated, and the system matrix in the imaging area is obtained.

[0033] In the embodiment, the sound field of a certain plane at a certain distance (in the direction of sound propagation) from the acoustic field coding board in the imaging space is measured by means of a three-dimensional translation table and a hydrophone, and the sound field of each plane at different distances is calculated by means of angular spectrum method. Next, the echo signal received by the ultrasonic transducer is obtained by convolution calculation, and thus the system matrix at a single phase plate angle can be obtained.

[0034] Since the acoustic field coding board is not completely axisymmetric, and subsequent measurement of the coding board and reverse rotation is required, the coding board needs to be rotated four times during the calibration process, and the sound field at four angles (0°, 90°, 180° and 270°) is measured by means of the hydrophone. Similarly, the system matrix at each angle is obtained by automatic convolution calculation to form the overall system matrix for image reconstruction. Specifically, the size of the system matrix at a certain angle is KxN, K is the number of time sampling points, and N is the number of pixels in the imaging area. The overall system matrix at four angles is 4KxN. This process only needs to be performed once in theory, because for a specific imaging system and a specific angle coding board, the system matrix H is unique.

[0035] Step S3, the single-element super-resolution ultrasonic imaging system is used to perform ultrasonic imaging on the flowing microbubbles in the imaging area to obtain a group of simulated original ultrasonic data.

[0036] Specifically, in each imaging process, the microbubbles in 5 random positions of the phantom structure template are activated to simulate single-element ultrasonic imaging. The above process is repeated to obtain a series of simulated raw ultrasonic data.

[0037] Step S4, rotate the acoustic coding plate to the above four angles, and perform the imaging operation of step S3 on the flowing microbubbles at the four angles respectively to obtain multi-angle raw ultrasonic echo data.

[0038] Step S5, based on the system matrix, the linear inversion algorithm is used to realize ultrasonic microbubble image reconstruction according to the multi-angle raw ultrasonic echo data.

[0039] Assuming that the received echo signal is linearly related to the image v, the echo signal u obtained under multi-angle can be expressed as:

[0040] u=Hv+n (1)

[0041] In the formula, u is an M-dimensional echo signal, M represents the product of the number of time sampling points and the number of rotations; n is an M-dimensional additive noise; H is a system matrix, which is determined by the ultrasonic measurement device. H is a two-dimensional matrix of M*N size, and N represents the number of pixels in the imaging plane.

[0042] In this embodiment, the preconditioned conjugate gradient method (PCG) is used to realize image reconstruction. Compared with the traditional conjugate gradient method, the PCG method optimizes the convergence mode by introducing a preconditioning matrix to realize fast convergence. The specific process is as follows:

[0043] When the linear model is effective, and n is additive white Gaussian noise, the image reconstruction solving problem can be expressed as:

[0044]

[0045] For this problem, the PCG method is used to solve First, according to the Jacobi method, the diagonal matrix of the system matrix H is taken as the preconditioning matrix F. Take the initial value as 0, and the precision index ε as 0.001, and set z0=F -1 r0,p1=z0, k=1. Start iteration calculation:

[0046] w=Hp k (3)

[0047]

[0048]

[0049] r k =r k-1 +a k w (6)

[0050] z k =F -1 r k (7)

[0051]

[0052] β k =ρ k / ρ k-1 (9)

[0053] p k+1 =z k +β k p k (10)

[0054] When ||r k ||<ε, then is the reconstructed ultrasound microbubble image, otherwise let k=k+1 and repeat the calculation of formula (3) to formula (10) until ||r k ||<ε is satisfied.

[0055] Step S6, center positioning is performed on each ultrasound microbubble image by a single molecule positioning method, and a microbubble positioning result in the corresponding image is obtained.

[0056] In this embodiment, before single molecule positioning is performed, the image sequence needs to be preprocessed, mainly including denoising and microbubble extraction, which helps to improve the positioning accuracy in the later stage.

[0057] Generally, it can be considered that the point spread function (PSF) of the ultrasound microbubble is consistent with the Gaussian distribution, that is, the PSF of the microbubble can be written as an anisotropic 2D Gaussian function, so the ultrasound microbubble image can be described as PSF:

[0058]

[0059] In formula (11), (x0, y0) represents the actual position of the microbubble, A represents the amplitude, σ x and σ y are the standard deviations in each direction of the image plane.

[0060] The nonlinear optimization problem in formula (11) can be solved by using the least square method and the least residual sum of squares, so that the corresponding microbubble positioning result of fitting can be obtained.

[0061] Step S7, superimpose the microbubble positioning results of all frames to realize single-element-based super-resolution ultrasound imaging, and the result is as shown in FIG. 8. Figure 4

[0062] Effects of the Embodiments

[0063] According to the single-element super-resolution ultrasound imaging method based on acoustic metamaterial provided in the embodiment, first, an acoustic field coding board is made of acoustic metamaterial to construct a single-element super-resolution ultrasound imaging system; then, the system is calibrated in acoustic field, and the system matrix in the imaging area is obtained by means of angular spectrum method and automatic convolution; secondly, the flowing microbubbles in the tissue are imaged by the single-element super-resolution ultrasound imaging system to obtain a group of simulated original ultrasound data, and the acoustic coding board is rotated to different angles to collect multi-angle data of the flowing microbubbles to obtain multi-angle original ultrasound echo data; then, based on the system matrix and the obtained original ultrasound echo data, the ultrasound microbubble image is reconstructed by means of linear inversion algorithm; finally, the center of the ultrasound microbubble image is positioned by a single-molecule positioning method to obtain the microbubble position, and the microbubble positioning results of all frames are superimposed to realize single-element-based super-resolution ultrasound imaging.

[0064] In summary, the method provided in the embodiment can realize high-quality ultrasound imaging in three-dimensional space only with a single element, greatly reducing the requirements of super-resolution ultrasound imaging on hardware devices (including the number of sensors and data transmission rate), transferring the complexity from the hardware system to the computer, greatly reducing the manufacturing cost of the device, and making the ultrasound imaging more widely used.

[0065] In addition, the method only reconstructs the ultrasound image based on single-element ultrasound signals, to a certain extent, improves the data acquisition speed, and makes it possible to realize super-resolution ultrasound imaging of moving tissues.

[0066] The above embodiments are only used to illustrate the specific embodiments of the present application, and the present application is not limited to the description range of the above embodiments.​

Claims

1. A single-element super-resolution ultrasound imaging method based on acoustic metamaterials, characterized in that, The method comprises the following steps: Step S1, manufacturing an acoustic field coding board by using an acoustic metamaterial, and constructing a single-element super-resolution ultrasonic imaging system based on the acoustic field coding board; Step S2, calibrating the acoustic field of the single-element super-resolution ultrasonic imaging system, and obtaining a system matrix in an imaging area; Step S3, using the single-element super-resolution ultrasonic imaging system to perform ultrasonic imaging on randomly activated flow microbubbles in a tissue to be imaged in the imaging area by means of an ultrasonic contrast agent, so as to obtain a group of simulated original ultrasonic data; Step S4, rotating the acoustic coding board to different angles, and using the single-element super-resolution ultrasonic imaging system to perform multi-angle imaging data acquisition on the flow microbubbles, so as to obtain multi-angle original ultrasonic echo data; Step S5, based on the system matrix, using a linear inversion algorithm to realize ultrasonic microbubble image reconstruction according to the multi-angle original ultrasonic echo data; Step S6, performing center positioning on each of the ultrasonic microbubble images by using a single molecule positioning method, so as to obtain microbubble positioning results in the corresponding images; Step S7, superimposing all the microbubble positioning results, so as to realize single-element-based super-resolution ultrasonic imaging. The single-element super-resolution ultrasonic imaging system at least comprises a single-element ultrasonic transducer, an acoustic field coding board and an imaging area, the acoustic field coding board is arranged in front of the single-element ultrasonic transducer, the surface of the acoustic field coding board is covered with randomly distributed circular holes, the diameter of the circular holes is 1 mm, and the depth is 0.1 mm to 1 mm.

2. The single-element super-resolution ultrasonic imaging method based on the acoustic metamaterial according to claim 1, wherein: wherein, The system matrix is obtained in the following manner: In the imaging space, the sound propagation direction of a certain plane sound field at different distances from the acoustic field coding board is measured by means of a three-dimensional translation table and a hydrophone, the sound field of each plane at different distances is calculated and obtained by an angular spectrum method, and then the received echo signal of the single-element ultrasonic transducer is calculated by convolution, so as to obtain the system matrix at a single phase plate angle. After the sound field of the four-angle coding board is measured, the system matrix at the corresponding four angles is obtained by the convolution calculation, so as to form the system matrix.

3. The single-element super-resolution ultrasonic imaging method based on the acoustic metamaterial according to claim 1, wherein: wherein The linear inversion algorithm is that the received echo signal is linearly related to the image v, and the echo signal u can be represented as: ， In the formula, u is an M-dimensional echo signal, M represents the product of the number of time sampling points and the number of rotations; n is an M-dimensional additive noise; H is a system matrix, which is determined by an ultrasonic measurement device, and H is a two-dimensional matrix with a size of M × N, and N represents the number of imaging plane pixels; Based on the linear inversion algorithm, the image reconstruction solving problem is represented as: ， The pre-conditioned conjugate gradient algorithm is used to solve the linear system. linearly.

4. The single-element super-resolution ultrasound imaging method based on acoustic metamaterials of claim 1, The step S6 comprises the following substeps: Step S6-1, pre-processing the image, and the pre-processing comprises denoising and microbubble extraction. ​ Step S6-2, single molecule localization is realized by using Gaussian fitting method: setting the point spread function PSF of the ultrasound microbubble to be consistent with Gaussian distribution, i.e. the PSF of the microbubble can be written as a 2D Gaussian function with different parameters, and therefore the ultrasound microbubble image can be described as PSF: ， wherein represents the actual position of the microbubbles, represents the amplitude, and is the standard deviation in each direction of the image plane; Step S6-3, a least square method is used to require the residual sum of squares to be minimum to solve the nonlinear optimization problem in step S6-2, and the corresponding microbubble localization point is fitted.

Images