Rapid beam forming method for ultrasonic imaging and related equipment

Abstract

The embodiment of the invention discloses a fast beam forming method for ultrasonic imaging and related equipment, and the method comprises the steps: constructing a DAS matrix for each transmitting and receiving beam before imaging according to a preset ultrasonic imaging mode and the space angle, focusing position, imaging resolution and imaging depth of ultrasonic beam scanning, and pre-storing the DAS matrix in a memory; secondly, acquiring a radio frequency digital signal (RF signal) of the wave beam acquired by the ultrasonic analog front end in real time during imaging, and multiplying the constructed corresponding DAS matrix by the RF signal to generate a plurality of signals after wave beam synthesis; according to the method, linear operation such as matrix multiplication is carried out on a constructed DAS matrix set and a digital signal, so that only one time of complex delay and interpolation calculation needs to be carried out before imaging, and only linear matrix multiplication is carried out in imaging; the operand, the calculation complexity and the required storage space of processing after image synthesis of an ultrasonic system in imaging are remarkably reduced, and complex clinical applications such as 4D ultrasonic imaging are met.

Description

[Technical field]

[0001] The present invention relates to the field of ultrasonic imaging, and in particular to a fast beamforming method for ultrasonic imaging and related equipment. [Background technology]

[0002] Ultrasonic beam synthesis is an ultrasound image reconstruction algorithm. The current ultrasound imaging beam synthesis method generally implements the Delay and Sum (DAS) method in the FPGA, converting the sampling time delay of each imaging voxel into the distance from the ultrasound array element, and calculating the transmission and reception delays for imaging. During the calculation process, each voxel on the image needs to be subjected to multiple delay calculations and interpolation calculations according to the distance to each ultrasound array element of the ultrasound probe during each frame of the image. This calculation method is computationally intensive and requires huge storage space to store intermediate data. In some special ultrasound application scenarios, such as ultra-fast imaging and 4D ultrasound imaging, the amount of data calculation is huge and requires faster post-processing methods. Therefore, a more efficient and storage-saving imaging beam synthesis method is clinically needed. [Summary of the invention]

[0003] In view of this, the present invention provides a fast beamforming method for ultrasonic imaging and related equipment to reduce the amount of calculation of the beamforming method in imaging and reduce the required storage space, thereby simplifying the calculation method.

[0004] The specific technical scheme of the first embodiment of the present invention is: a method for rapid beam synthesis of ultrasonic imaging, the method comprising: according to multiple transmission and reception parameters of each frame of image preset by the ultrasonic system, before imaging, the spatial scanning direction, focus position, imaging depth and resolution of each transmitted and received beam are calculated, the delay of each voxel point on the beam is interpolated, a DAS matrix set is constructed, and the DAS matrix set is stored in a memory; the DAS matrix set includes the same number of DAS matrices as the number of beams; each beam transmitted and received by the ultrasonic system corresponds to a DAS matrix; during imaging, the preset angle of space in the ultrasonic system is scanned, and then the DAS matrix set is stored in a memory; The echo signal of the beam is received, and a digital signal of the echo signal is acquired in real time; the digital signal is a signal of the radio frequency digital signal emitted by the beam after gain, filtering and analog-to-digital conversion in the analog front end; the DAS matrix set and the digital signal are multiplied to obtain multiple beam-synthesized signals of multiple beams; the multiple beam-synthesized signals are Hilbert transformed to obtain the amplitude of the complex signals after the multiple beams are synthesized, and the amplitude is an image grayscale image; the image grayscale image is coordinate transformed and three-dimensionally reconstructed to obtain a frame of ultrasound image; and each frame of image is obtained by using the same DAS matrix set and different echo signals for calculation.

[0005] Preferably, before imaging, the delay of each voxel point on the beam is calculated and interpolated for the spatial scanning direction, focal point position, imaging depth and resolution of each transmitted and received beam, and a DAS matrix set is constructed, including: obtaining the distance between each voxel and all the ultrasonic array elements according to the coordinates of each ultrasonic array element in the ultrasonic system and the coordinates of each voxel in a beam; obtaining the transmission delay of each voxel relative to all the ultrasonic array elements according to the minimum distance between each voxel and all the ultrasonic array elements and divided by the speed of sound; obtaining the total delay of each voxel according to the transmission delay of each voxel relative to all the ultrasonic array elements and the receiving delay of each voxel relative to each ultrasonic array element; obtaining the DAS matrix of each beam according to the total delay of each voxel transmitted once by the beam, the position of each voxel and the sampling frequency; and constructing a DAS matrix set according to the DAS matrices of the multiple beams, and the spatial scanning direction, focal point position and resolution.

[0006] Preferably, the DAS matrix of each beam is obtained according to the total delay of each voxel emitted by the beam once, the position of each voxel and the sampling frequency, including: obtaining multiple sampling points of the digital signal according to the total delay of each voxel in the spatial scanning direction and the sampling frequency; wherein the number of sampling points is the same as the number of voxels on the beam; performing linear interpolation processing on the multiple sampling points to obtain multiple linear interpolation coefficients; and using the multiple linear interpolation coefficients to assign a preset sparse matrix to obtain the DAS matrix.

[0007] Preferably, the linear interpolation coefficient is obtained using the following formula:

[0008]

[0009] Among them, c k =ceil(τ i，j *fs)-τ i，j *fs,c k+1 =τ i，j *fs-floor(τ i，j *fs),s i,j is the sampling point after linear interpolation, N is the number of array elements, c k and c k+1 is the linear interpolation coefficient, τ i，j is the total delay, fs is the sampling frequency, ceil means rounding up, floor means rounding down, k is the value of the sampling point after rounding down, RF j (k) and RF j (k+1) are two adjacent radio frequency digital signals.

[0010] Preferably, the use of the multiple linear interpolation coefficients to assign values ​​to a preset sparse matrix to obtain the DAS matrix includes: obtaining all target columns of the multiple linear interpolation coefficients based on the values ​​of multiple sampling points rounded down; obtaining all target rows of the multiple linear interpolation coefficients based on the coordinates of each voxel; and using the linear interpolation coefficients to assign values ​​to elements of all target columns and all target rows in a preset sparse matrix to obtain the DAS matrix.

[0011] Preferably, performing Hilbert transform on the multiple beam-synthesized signals to obtain amplitudes of the multiple beam-synthesized complex signals, wherein the amplitudes are image grayscale images, includes: performing vectorization processing on the multiple beam-synthesized signals to obtain multiple one-dimensional vectors; performing Hilbert transform accelerated by multi-threaded parallel computing on the multiple one-dimensional vectors to obtain complex signals; and taking the amplitudes of the complex signals, which are image grayscale images corresponding to all beam-synthesized signals.

[0012] Preferably, the Hilbert transform of the multiple one-dimensional vectors with multi-threaded parallel computing acceleration obtains a complex signal, including: the signal after the Hilbert transform is a complex number, including a real part and an imaginary part; wherein the real part is the original signal; the imaginary part calculation requires phase shifting of different sampling points of the original signal, and then performing a discrete Fourier transform; the original signal is the multiple one-dimensional vectors; wherein, when performing the discrete Fourier transform, high-performance computing with multi-threaded acceleration is adopted, and each thread corresponds to a sampling point.

[0013] The specific technical scheme of the second embodiment of the present invention is: a fast beam synthesis system for ultrasonic imaging, characterized in that the system includes: a DAS matrix construction module, an analog front end, a beam synthesis signal acquisition module, a Hilbert transform module and an ultrasonic image acquisition module; the DAS matrix construction module is used to calculate the delay of each voxel point on the beam and interpolate the spatial scanning direction, focus position, imaging depth and resolution of each transmitted and received beam before imaging according to multiple transmission and reception parameters of each frame of the ultrasonic system preset, and construct a DAS matrix set, and store the DAS matrix set in the memory; the DAS matrix set includes the same number of DAS matrices as the number of beams; each beam transmitted and received by the ultrasonic system corresponds to a DAS matrix; the analog front end is used to scan the image during imaging. The preset angle of space in the ultrasound system is scanned, the echo signal of the beam is received, and the digital signal of the echo signal is acquired in real time; the digital signal is the signal of the radio frequency digital signal emitted by the beam after gain, filtering and analog-to-digital conversion in the analog front end; the beam synthesis signal acquisition module is used to multiply the DAS matrix set and the digital signal to obtain multiple beam synthesised signals of multiple beams; the Hilbert transform module is used to perform Hilbert transform on the multiple beam synthesised signals to obtain the amplitude of the complex signals after the multiple beam synthesis, and the amplitude is the image grayscale image; the ultrasound image acquisition module is used to perform coordinate transformation and three-dimensional reconstruction on the image grayscale image to obtain a frame of ultrasound image, and use the same DAS matrix set and different echo signals for calculation to obtain each frame of image.

[0014] The specific technical solution of the third embodiment of the present invention is: an ultrasonic imaging fast beamforming device, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method described in any one of the first embodiments of the present application.

[0015] The specific technical solution of the fourth embodiment of the present invention is: a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to perform the steps of the method described in any one of the first embodiments of the present application.

[0016] Implementing the embodiments of the present invention will have the following beneficial effects:

[0017] The present invention calculates the transmission delay, reception delay and interpolation coefficient before imaging according to the imaging parameters of the ultrasound system, namely, the spatial scanning direction, focus position, imaging depth and resolution of the beams transmitted and received each time, so as to pre-construct a DAS matrix set, and store the DAS matrix set in a preset memory; the DAS matrix set includes DAS matrices with the same number of beams; during the imaging process, each beam transmitted and received by the ultrasound system corresponds to a DAS matrix; according to the spatial angle of the beam scanned by the preset ultrasound system, the echo signal of the beam is received, and the digital signal of the echo signal is obtained in real time; the digital signal is the signal of the radio frequency digital signal transmitted by the beam after analog-to-digital conversion in the analog front end; the DAS matrix set and the corresponding digital signal are multiplied to obtain a signal after multiple beams of multiple beams are synthesized; the signal after multiple beams are synthesized is Hilbert transformed to obtain the amplitude of the signal after beam synthesis, and the amplitude is the image grayscale image; the image grayscale image is coordinate transformed and three-dimensionally reconstructed to obtain a frame of ultrasound image. Since ultrasound imaging is composed of multiple frames of images, the acquisition parameters of multiple transmissions and receptions of each frame of image are exactly the same, so the calculation of the complex DAS matrix only needs to be calculated once before imaging. By using the constructed DAS matrix set to multiply the digital signal obtained by each transmission and reception, there is no need to perform real-time delay calculation and interpolation calculation on the digital signal collected for each frame during the imaging process, thereby reducing the amount of calculation in the imaging process of the ultrasound system. Secondly, during the imaging process, only simple linear operations such as matrix multiplication need to be performed in computing units such as CPU, FPGA or GPU, which greatly reduces the storage space required for image processing during the image acquisition process of the ultrasound system.

Brief Description of the Drawings

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flowchart of the steps of an ultrasound imaging method based on fast beam synthesis;

[0020] Figure 2 A flowchart of the steps for constructing a DAS matrix set;

[0021] Figure 3 A flowchart of the steps for constructing a DAS matrix;

[0022] Figure 4 A flowchart of the steps for assigning values ​​to a preset sparse matrix;

[0023] Figure 5 A flow chart of the steps for obtaining a grayscale image;

[0024] Figure 6 A flow chart of the steps for obtaining a complex signal;

[0025] Figure 7 It is a structural diagram of an ultrasound imaging system based on fast beam synthesis;

[0026] Among them, 701 is a DAS matrix construction module; 702 is an analog front end; 703 is a beam synthesis signal acquisition module; 704 is a Hilbert transform module; 705 is an ultrasound image acquisition module. [Specific implementation method]

[0027] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0028] See also Figure 1 , is a flowchart of a method for rapid beamforming of ultrasound imaging in the first embodiment of the present application, which does not require real-time delay calculation and interpolation calculation of digital signals, thereby reducing the amount of calculation of the ultrasound system and the storage space required by the ultrasound system. The method includes:

[0029] Step 101: According to a plurality of transmission and reception parameters of each frame of an image preset by an ultrasound system, before imaging, the spatial scanning direction, focus position, imaging depth and resolution of each transmitted and received beam are calculated, the delay of each voxel point on the beam is interpolated, a DAS matrix set is constructed, and the DAS matrix set is stored in a memory; the DAS matrix set includes the same number of DAS matrices as the number of beams; each beam transmitted and received by the ultrasound system corresponds to one DAS matrix;

[0030] Step 102: During imaging, a preset angle of space in the ultrasound system is scanned, an echo signal of the beam is received, and a digital signal of the echo signal is acquired in real time; the digital signal is a signal of the radio frequency digital signal emitted by the beam after gain, filtering, and analog-to-digital conversion in an analog front end;

[0031] Step 103: multiply the DAS matrix set and the digital signal to obtain a plurality of beam-synthesized signals of a plurality of beams;

[0032] Step 104: perform Hilbert transform on the signals after the plurality of beamformations are synthesized to obtain amplitudes of the complex signals after the plurality of beamformations are synthesized, wherein the amplitude is an image grayscale image;

[0033] Step 105, performing coordinate transformation and three-dimensional reconstruction on the image grayscale image to obtain a frame of ultrasound image;

[0034] Step 106: Use the same DAS matrix set and different echo signals to perform calculations to obtain each frame of image.

[0035] Specifically, the ultrasound system transmits l beams, each beam has M voxels, and a DAS matrix set is constructed according to the spatial scanning direction, focus position, imaging depth and resolution of the beams transmitted and received by the ultrasound system each time, and the DAS matrix set is stored in a preset memory; wherein a DAS matrix is ​​constructed according to the coordinates of the M voxels in a beam, l DAS matrices are constructed from the voxels in l beams, and l DAS matrices constitute a DAS matrix set; the digital signal RF emitted by l beams is acquired in real time i , RF i ∈R H×1 , where R is the RF data transmitted by the beam, H is the number of samples of the RF digital signal, and H depends on the sampling frequency and imaging depth of the ultrasound system; matrix multiplication is performed on the DAS matrix set and the digital signal to obtain the beam-synthesized signal of l beams; the beam-synthesized signal of l beams is subjected to Hilbert transform to obtain the image grayscale image of l beams, and the image grayscale image is subjected to coordinate transformation (such as three-dimensional reconstruction in 4D imaging) to obtain a frame of ultrasound image.

[0036] Specifically, ultrasound imaging is spliced ​​in time by acquiring multiple frames of ultrasound images continuously. Each frame of ultrasound image is formed by multiple transmission and reception beams, and the imaging parameters of each frame of imaging are the same. Before imaging, a DAS matrix set is pre-constructed and stored in memory according to the imaging depth, resolution, focus position of multiple beam transmissions and spatial scanning angle of the ultrasound system. Among them, each ultrasound transmission and reception corresponds to a DAS matrix. During imaging, the received beam undergoes a series of analog operations in the analog front end, and is converted into a digital signal to obtain the digital signal of multiple beam transmissions in real time. For each transmission and reception, the corresponding DAS matrix and digital signal are multiplied to obtain the beam synthesized signal; each beam synthesized signal is Hilbert transformed, and then the amplitude is calculated to obtain the image grayscale image; for each transmission and reception, different beam angles are used to repeatedly obtain the digital signals of different beam transmissions, and the corresponding DAS matrix set is multiplied to obtain the beam synthesized signal, and the Hilbert transform and amplitude calculation are performed to obtain the image grayscale image, until the scanning of a frame of image is completed using different transmission angles, that is, the beam synthesis of an ultrasound image is completed. Repeating this process multiple times can achieve continuous multi-frame ultrasound image acquisition.

[0037] This embodiment pre-constructs a DAS matrix set according to the spatial scanning direction, focus position, imaging depth and resolution of the beam transmitted and received by the ultrasound system each time, and stores the DAS matrix set in a preset memory; the DAS matrix set includes the same number of DAS matrices as the number of beams; each beam transmitted and received by the ultrasound system corresponds to a DAS matrix; according to the spatial angle of the beam scanned by the preset ultrasound system, the echo signal of the beam is received, and the digital signal of the echo signal is obtained in real time; the digital signal is the signal of the radio frequency digital signal transmitted by the beam after analog-to-digital conversion in the analog front end; the DAS matrix set and the digital signal are multiplied to obtain a signal after multiple beams of multiple beams are synthesized; the signal after the multiple beams are synthesized is Hilbert transformed to obtain the amplitude of the multiple beams synthesized signal, and the amplitude is the image grayscale image; the image grayscale image is coordinate transformed (for example, 4D imaging requires three-dimensional reconstruction) to obtain a frame of ultrasound image. Since ultrasound imaging is composed of multiple frames of images, the acquisition parameters of multiple transmissions and receptions of each frame of image are exactly the same, so the DAS matrix set calculation only needs to be performed once and repeatedly calculated for different frames. Each DAS matrix is ​​multiplied by the digital signal received in the corresponding beam direction. It only needs to calculate the delay once before imaging, and there is no need to perform delay calculation and interpolation calculation on the digital signal in real time during the imaging process, which significantly reduces the amount of calculation of the ultrasound system. Secondly, beam synthesis is performed in computing units such as CPU, FPGA or GPU, and only simple linear operations such as matrix multiplication are required, which greatly reduces the storage space required for image processing during the image acquisition process of the ultrasound system.

[0038] In the specific embodiments, see Figure 2 In step 101, according to the multiple transmission and reception parameters of each frame of the ultrasound system preset, the spatial scanning direction, focus position, imaging depth and resolution of each transmitted and received beam are calculated and interpolated before imaging, and a DAS matrix set is constructed, including:

[0039] Step 201: Obtain the distance between each voxel and all the ultrasound array elements according to the coordinates of each ultrasound array element in the ultrasound system and the coordinates of each voxel in a beam;

[0040] Step 202: Obtain the transmission delay of each voxel relative to all the ultrasound array elements according to the minimum distance between each voxel and all the ultrasound array elements and divide it by the speed of sound;

[0041] Step 203: Obtain the total delay of each voxel according to the transmission delay of each voxel relative to all ultrasound array elements and the reception delay of each voxel relative to each ultrasound array element; wherein, the total delay = transmission delay + reception delay;

[0042] Step 204, obtaining a DAS matrix of each beam according to the total delay of each voxel transmitted once by the beam, the position of each voxel and the sampling frequency;

[0043] Step 205: construct a DAS matrix set according to the DAS matrices of the multiple beams, and the spatial scanning direction, focus position and resolution.

[0044] Specifically, there are N channels in the ultrasound system, each channel corresponds to an ultrasound array element, and the coordinates of each ultrasound array element are calculated and recorded as (xe j ,ye j ,ze j ), where j≤N; record the coordinates of each voxel (x i ,y i , z i ), according to the coordinates of the voxel and the coordinates of the ultrasonic array element, the distance between the M voxels and the N ultrasonic array elements is obtained, the minimum distance between a voxel and the N ultrasonic array elements is selected, and the time for the wave velocity to reach the nearest voxel front is obtained according to the minimum distance, which is the transmission delay of the voxel relative to all ultrasonic array elements; the time for each voxel to reach each ultrasonic array element is the receiving delay of each voxel; it can be understood that one voxel has N receiving delays corresponding to N ultrasonic array elements, and the total delay value τ of one voxel relative to one channel i，j The sum of 1 transmission delay and the N receiving delays corresponding to the channel, the total number of delay values ​​corresponding to N channels of M voxels is M*N; the DAS matrix of a beam is obtained according to the total delay and sampling frequency of each voxel; the DAS matrix of l beams, and the spatial scanning direction, focus position and resolution constitute a DAS matrix set. Construct a DAS matrix D according to the sum of 1 transmission delay and the N receiving delays corresponding to the channel l , D l ∈R M×L When adjusting the ultrasonic array element distribution scheme, it is only necessary to obtain the total delay value of l beam transmissions to obtain the DAS matrix set. The number of beams transmitted when constructing the DAS matrix set can be set according to actual conditions. Therefore, when the ultrasonic system is working, it is only necessary to preset the DAS matrix set in advance according to the imaging parameters. During the imaging process, the RF signal is collected in real time. There is no need to calculate the total delay of each beam in real time. Only linear calculations such as matrix multiplication are required, which greatly reduces the amount of calculation, calculation complexity and storage space required by the ultrasonic system.

[0045] In the specific embodiments, see Figure 3 In step 204, the DAS matrix of each beam is obtained according to the total delay of each voxel transmitted by the beam once, the position of each voxel and the sampling frequency, including:

[0046] Step 301, obtaining a plurality of sampling points of the digital signal according to the total delay of each voxel in the spatial scanning direction and the sampling frequency; wherein the number of sampling points is the same as the number of voxels on the beam;

[0047] Step 302: Perform linear interpolation processing on the plurality of sampling points to obtain a plurality of linear interpolation coefficients;

[0048] Step 303: Use the multiple linear interpolation coefficients to assign values ​​to a preset sparse matrix to obtain the DAS matrix.

[0049] Specifically, the total delay τ of each voxel is i，j Multiply the sampling frequency by the sampling point τ to obtain the digital signal. i，j *fs; perform linear interpolation on multiple sampling points to obtain multiple linear interpolation coefficients c k and c k+1 ; Among them, interpolation processing methods such as spline interpolation, radial basis function interpolation or polynomial interpolation can be used to interpolate multiple sampling points. The DAS matrix is ​​a sparse matrix, and the sparse matrix storage method can be used to store the DAS matrix, that is, only the coordinate positions and matrix values ​​of non-zero elements are saved, thereby significantly reducing the storage space.

[0050] In a specific embodiment, the linear interpolation coefficient is obtained using the following formula:

[0051]

[0052] Among them, c k =ceil(τ i,j *fs)-τ i,j *fs,c k+1 =τ i,j *fs-floor(τ i,j *fs),s i,j is the sampling point after linear interpolation, N is the number of array elements, c k and c k+1 is the linear interpolation coefficient, τ i，j is the total delay, fs is the sampling frequency, ceil means rounding up, floor means rounding down, k is the value of the sampling point after rounding down, RF j (k) and RF j (k+1) are two adjacent RF digital signals. The linear interpolation processing method has high precision and can obtain more accurate linear interpolation coefficients.

[0053] For details, please refer to Figure 4In step 303, the preset sparse matrix is ​​assigned with the plurality of linear interpolation coefficients to obtain the DAS matrix, including:

[0054] Step 401, obtaining all target columns of the plurality of linear interpolation coefficients according to the values ​​of the plurality of sampling points rounded down;

[0055] Step 402: Obtain all target rows of the plurality of linear interpolation coefficients according to the coordinates of each voxel;

[0056] Step 403: Use linear interpolation coefficients to assign values ​​to all target columns and all target rows in a preset sparse matrix to obtain a DAS matrix.

[0057] Specifically, the preset sparse matrix is ​​assigned values, and the elements of its i-th row, k-th column, and (k+1)-th column are c k and c k+1 , all elements except the i-th row are 0, and a DAS matrix (i.e., a DAS sparse matrix) is obtained; wherein the selection of i is related to the coordinates of the voxel. More than 99% of the elements in the DAS matrix are 0, so when storing the DAS matrix, the DAS matrix requires less storage space.

[0058] For specific embodiments, please refer to Figure 5 In step 104, Hilbert transform is performed on the signals after the plurality of beam synthesis to obtain the amplitudes of the signals after the plurality of beam synthesis, where the amplitude is an image grayscale image, including:

[0059] Step 501: performing vectorization processing on the signals after the plurality of beamformations to obtain a plurality of one-dimensional vectors;

[0060] Step 502: Perform Hilbert transform accelerated by multi-threaded parallel computing on the multiple one-dimensional vectors to obtain complex signals; take amplitudes of the complex signals, which are grayscale images corresponding to the signals after all beam synthesis.

[0061] In one embodiment, the signal after l beam synthesis is vectorized on the CPU to obtain l*M one-dimensional vectors; the l*M one-dimensional vectors are subjected to multi-threaded parallel computing accelerated Hilbert transform on the GPU to obtain a complex signal, the amplitude of the complex signal is the image grayscale image, and the image grayscale image is subjected to coordinate transformation (such as three-dimensional reconstruction of 4D ultrasound imaging) to obtain an ultrasound image. The multi-threaded parallel computing accelerated Hilbert transform can perform parallel calculation of vectors, that is, all one-dimensional vectors are calculated once to obtain the final image grayscale image, simplifying the calculation process.

[0062] In a specific embodiment, Hilbert transform may be performed on the multiple beamformed signals to obtain multiple image grayscale images, and coordinate transformation may be performed on the multiple image grayscale images (such as three-dimensional reconstruction of 4D ultrasound) to obtain ultrasound images.

[0063] In the specific embodiments, see Figure 6 In step 502, performing multi-threaded parallel computing accelerated Hilbert transform on the multiple one-dimensional vectors to obtain complex signals includes:

[0064] Step 601: The signal after Hilbert transformation is a complex number, including a real part and an imaginary part; wherein the real part is the original signal;

[0065] Step 602, the imaginary part calculation requires phase shifting of different sampling points of the original signal, followed by discrete Fourier transform; the original signal is the multiple one-dimensional vectors;

[0066] Step 603: When performing discrete Fourier transform, high-performance computing with multi-thread acceleration is adopted, and each thread corresponds to one sampling point.

[0067] Specifically, the real part of the signal after Hilbert transformation is the original signal; the multiple one-dimensional vectors are subjected to discrete Fourier transformation accelerated by multi-threaded parallel computing to obtain multiple frequency domain signals; different sampling points in the original signal are phase shifted, and then discrete Fourier inverse transformation is performed to obtain a modulated signal; the signal after Hilbert transformation is a complex number, that is, a complex signal, and its real part and imaginary part are respectively the real signal after phase shift and the modulated signal after discrete Fourier transformation; wherein, when performing discrete Fourier transformation, multi-threading is used for calculation, and each thread corresponds to one sampling point.

[0068] Among them, a plurality of one-dimensional vectors are subjected to discrete Fourier transform accelerated by CPU parallel computing to obtain a plurality of frequency domain signals F; different sampling points in the frequency domain signal F are subjected to different phase shift operations to obtain the real signal P. Specifically, if M is an even number, the phase shift is performed according to formula (1); if M is an odd number, the phase shift is performed according to formula (2), wherein i is an imaginary unit, and formulas (1) and (2) are as follows:

[0069]

[0070]

[0071] Where n is the number of channels, i.e. the number of ultrasound array elements; M is the number of voxels on each beam; P is a real signal, and F is a frequency domain signal; the real signal P is subjected to an inverse discrete Fourier transform to obtain the modulated signal I, and the modulated signal I is subjected to a Hilbert transform to obtain a complex signal S, S = R + i * abs (I), where abs represents the modulus of the complex number, and R is the real part of the complex signal. Through the Hilbert transform, phase shift operation, inverse discrete Fourier transform, and Hilbert transform accelerated by multi-threaded parallel computing, multiple one-dimensional vectors are converted into complex signals for representing the grayscale image at one time, simplifying the processing steps of one-dimensional vectors, and the Hilbert transform, phase shift operation, inverse discrete Fourier transform, and Hilbert transform accelerated by multi-threaded parallel computing are all conventional data processing methods, with simple calculation methods, which are convenient for improving calculation efficiency. When performing multi-threaded parallel computing, the number of threads is equal to the number of RF sampling points.

[0072] In the specific embodiments, see Figure 7 , is a structural diagram of an ultrasound imaging fast beamforming system in the second embodiment of the present application, the system comprising: a DAS matrix building module 701, an analog front end 702, a beamforming signal acquisition module 703, a Hilbert transform module 704 and an ultrasound image acquisition module 705;

[0073] The DAS matrix construction module 701 is used to calculate the delay of each voxel point on the beam and interpolate the spatial scanning direction, focus position, imaging depth and resolution of each transmitted and received beam before imaging according to multiple transmission and reception parameters of each frame of the ultrasound system, construct a DAS matrix set, and store the DAS matrix set in the memory; the DAS matrix set includes the same number of DAS matrices as the number of beams; each beam transmitted and received by the ultrasound system corresponds to a DAS matrix;

[0074] The analog front end 702 is used to scan a preset angle of space in the ultrasound system during imaging, receive the echo signal of the beam, and obtain a digital signal of the echo signal in real time; the digital signal is a signal of the RF digital signal emitted by the beam after gain, filtering and analog-to-digital conversion in the analog front end;

[0075] The beam synthesis signal acquisition module 703 is used to multiply the DAS matrix set and the digital signal to obtain a plurality of beam synthesised signals of a plurality of beams;

[0076] The Hilbert transform module 704 is used to perform Hilbert transform on the multiple beamformed signals to obtain the amplitude of the complex signals after the multiple beamformed signals are synthesized, and the amplitude is an image grayscale image;

[0077] The ultrasound image acquisition module 705 is used to perform coordinate transformation and three-dimensional reconstruction (such as three-dimensional reconstruction in 4D ultrasound imaging) on ​​the image grayscale image to obtain a frame of ultrasound image, and use the same DAS matrix set and different echo signals for calculation to obtain each frame of image.

[0078] The ultrasound imaging fast beam synthesis system in this embodiment can be applied to low-cost ultrasound machines (such as FPGA, DSP and ARM architectures), and can meet the computing power and storage space requirements of real-time 4D imaging, greatly reducing the host cost of 4D ultrasound and real-time ultrasound guidance systems. By using the ultrasound imaging fast beam synthesis system in this embodiment, since the acquisition parameters of each frame of the image are the same, it is possible to use the constructed DAS matrix set to multiply the digital signal, so that each frame of the image acquisition does not need to perform delay calculations, etc., and only needs to update the RF signal. In the image acquisition process, only the pre-calculated DAS matrix and the real RF data need to be used for linear calculations such as matrix multiplication, and there is no need to perform delay calculations and interpolation calculations on the digital signal in real time, thereby reducing the amount of calculation and complexity of the ultrasound system and the storage space required for the ultrasound system.

[0079] In a specific embodiment, the transducer used in the matrix multiplication in this embodiment can be a two-dimensional array transducer, and can also be replaced with any form of transducer array, such as a linear array, a curved array, a circular array, etc.; the various parameters of the ultrasonic array element are not limited, and arrays with different frequencies, numbers of ultrasonic array elements, ultrasonic array element spacing, ultrasonic array element distribution schemes, etc. can be selected according to various application environments; the ultrasonic imaging scanning method is not limited to focused ultrasonic scanning, but can also be plane wave imaging, linear array imaging, and divergent wave imaging.

[0080] This embodiment can be applied to 2D ultrasound imaging, 3D ultrasound imaging and 4D ultrasound imaging. 3D and 4D ultrasound imaging can use different ultrasound probes, such as 2D array probes, 1.5D array probes, linear probes or convex array probes; the probe material can be PZT probes, composite material probes, CMUT or PMUT, etc.

[0081] In a specific embodiment, the third embodiment of the present application provides an ultrasound imaging fast beamforming device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described in any one of the first embodiments of the present application.

[0082] In a specific embodiment, the fourth embodiment of the present application is a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps of the method as described in any one of the first embodiments of the present application.

[0083] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0084] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0085] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A fast beamforming method for ultrasonic imaging, characterized in that: The method comprises: According to the multiple transmission and reception parameters of each frame of the ultrasound system preset, the spatial scanning direction, focus position, imaging depth and resolution of each transmitted and received beam are calculated and interpolated before imaging, a DAS matrix set is constructed, and the DAS matrix set is stored in the memory; the DAS matrix set includes the same number of DAS matrices as the number of beams; each beam transmitted and received by the ultrasound system corresponds to a DAS matrix; During imaging, a preset angle of space in the ultrasound system is scanned, an echo signal of the beam is received, and a digital signal of the echo signal is acquired in real time; the digital signal is a signal of the radio frequency digital signal emitted by the beam after gain, filtering and analog-to-digital conversion in the analog front end; Multiplying the DAS matrix set and the digital signal to obtain a plurality of beam-synthesized signals of a plurality of beams; Performing Hilbert transform on the signals synthesized by the multiple beams to obtain amplitudes of the complex signals synthesized by the multiple beams, wherein the amplitudes are image grayscale images; Performing coordinate transformation and three-dimensional reconstruction on the image grayscale image to obtain a frame of ultrasound image; Each frame of image is obtained by using the same DAS matrix set and different echo signals for calculation.

2. The fast beamforming method for ultrasonic imaging according to claim 1, characterized in that: Before imaging, the spatial scanning direction, focus position, imaging depth and resolution of each transmitted and received beam are calculated, the delay of each voxel point on the beam is interpolated, and a DAS matrix set is constructed, including: According to the coordinates of each ultrasonic array element in the ultrasonic system and the coordinates of each voxel in a beam, the distance between each voxel and all ultrasonic array elements is obtained; The transmission delay of each voxel relative to all the ultrasound array elements is obtained by dividing the minimum distance between each voxel and all the ultrasound array elements by the speed of sound; Obtaining a total delay of each voxel according to the transmission delay of each voxel relative to all ultrasound array elements and the reception delay of each voxel relative to each ultrasound array element; Obtaining a DAS matrix for each beam according to the total delay of each voxel transmitted once by the beam, the position of each voxel and the sampling frequency; A DAS matrix set is formed according to the DAS matrices of the multiple beams, and the spatial scanning direction, focus position and resolution.

3. The fast beamforming method for ultrasonic imaging according to claim 2, characterized in that: The step of obtaining the DAS matrix of each beam according to the total delay of each voxel transmitted once by the beam, the position of each voxel and the sampling frequency comprises: Obtaining a plurality of sampling points of the digital signal according to the total delay of each voxel in the spatial scanning direction and the sampling frequency; wherein the number of sampling points is the same as the number of voxels on the beam; Performing linear interpolation processing on the multiple sampling points to obtain multiple linear interpolation coefficients; The DAS matrix is ​​obtained by assigning values ​​to a preset sparse matrix using the multiple linear interpolation coefficients.

4. The fast beamforming method for ultrasonic imaging according to claim 3, characterized in that: The linear interpolation coefficient is obtained using the following formula: Among them, c k =ceil(τ i,j *fs)-τ i,j *fs,c k+1 =τ i,j *fs-floor(τ i,j *fs),s i,j is the sampling point after linear interpolation, N is the number of array elements, c k and c k+1 is the linear interpolation coefficient, τ i,j is the total delay, fs is the sampling frequency, ceil means rounding up, floor means rounding down, k is the value of the sampling point after rounding down, RF j (k) and RF j (k+1) are two adjacent radio frequency digital signals.

5. The fast beamforming method for ultrasonic imaging according to claim 3, characterized in that: The step of assigning a preset sparse matrix using the plurality of linear interpolation coefficients to obtain the DAS matrix includes: Obtaining all target columns of the plurality of linear interpolation coefficients according to the values ​​of the plurality of sampling points rounded down; Obtaining all target rows of the plurality of linear interpolation coefficients according to the coordinates of each voxel; The DAS matrix is ​​obtained by assigning values ​​to all target columns and all target rows in the preset sparse matrix using linear interpolation coefficients.

6. The rapid beamforming method for ultrasonic imaging according to claim 1, characterized in that: The step of performing Hilbert transform on the signals synthesized by the multiple beamforms to obtain amplitudes of the complex signals synthesized by the multiple beamforms, wherein the amplitude is an image grayscale image, comprises: Performing vectorization processing on the signals synthesized by the multiple beamformations to obtain multiple one-dimensional vectors; The multiple one-dimensional vectors are subjected to Hilbert transform accelerated by multi-threaded parallel computing to obtain complex signals; the amplitude of the complex signals is taken, which is the image grayscale image corresponding to the signals after all beam synthesis.

7. The fast beamforming method for ultrasonic imaging according to claim 6, characterized in that: The step of performing multi-threaded parallel computing accelerated Hilbert transform on the multiple one-dimensional vectors to obtain a complex signal comprises: The signal after Hilbert transformation is a complex number, including a real part and an imaginary part; the real part is the original signal; the imaginary part calculation requires phase shifting of different sampling points of the original signal, followed by discrete Fourier transformation; the original signal is the multiple one-dimensional vectors; Among them, when performing discrete Fourier transform, high-performance computing with multi-thread acceleration is adopted, and each thread corresponds to one sampling point.

8. A fast beamforming system for ultrasonic imaging, characterized in that: The system comprises: a DAS matrix building module, an analog front end, a beam synthesis signal acquisition module, a Hilbert transform module and an ultrasound image acquisition module; The DAS matrix construction module is used to calculate the delay of each voxel point on the beam and interpolate the spatial scanning direction, focus position, imaging depth and resolution of each transmitted and received beam before imaging according to multiple transmission and reception parameters of each frame of the ultrasound system preset, construct a DAS matrix set, and store the DAS matrix set in the memory; the DAS matrix set includes the same number of DAS matrices as the number of beams; each beam transmitted and received by the ultrasound system corresponds to a DAS matrix; The analog front end is used to scan a preset angle of space in the ultrasound system during imaging, receive the echo signal of the beam, and obtain a digital signal of the echo signal in real time; the digital signal is a signal of the radio frequency digital signal emitted by the beam after gain, filtering and analog-to-digital conversion in the analog front end; The beam synthesis signal acquisition module is used to multiply the DAS matrix set and the digital signal to obtain a plurality of beam synthesised signals of a plurality of beams; The Hilbert transform module is used to perform Hilbert transform on the signals after the plurality of beams are synthesized to obtain the amplitude of the complex signals after the plurality of beams are synthesized, and the amplitude is an image grayscale image; The ultrasonic image acquisition module is used to perform coordinate transformation and three-dimensional reconstruction on the image grayscale image to obtain a frame of ultrasonic image, and use the same DAS matrix set and different echo signals for calculation to obtain each frame of image.

9. A fast beamforming device for ultrasonic imaging, comprising a memory and a processor, characterized in that: The memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

Images