A frequency domain bone ultrasound imaging method based on virtual source synthetic aperture

Abstract

The present invention relates to a frequency-domain bone ultrasound imaging method based on virtual source synthetic aperture. The method comprises the following steps: S1, transmitting ultrasonic signals to the bone N times and collecting echo signals; S2, extracting the time delay of the echo signals at the upper and lower interfaces of the bone; S3, establishing a sound velocity model; S4, obtaining the echo frequency-domain wavefield and the transmission frequency-domain wavefield; S5, sequentially processing the echo signals collected from the N transmissions to obtain subframe images for each transmission; and S6, superimposing multiple subframe images to obtain a final bone image. Compared with existing technologies, the present invention has the advantages of improving the clarity of the image of the lower surface of the intermediate bone while correcting phase distortion caused by the sound velocity of soft tissue and bone.

Description

Technical Field

[0001] The present invention relates to the field of ultrasonic imaging, and in particular to a frequency domain bone ultrasonic imaging method based on virtual source synthetic aperture. Background Art

[0002] Ultrasound imaging technology is non-destructive, portable, and rapid, and has important applications in the imaging and diagnosis of biological soft tissues (such as muscles, blood, and organs). However, bone hard tissue presents difficulties and challenges in ultrasound imaging. On the one hand, bone is an inhomogeneous tissue with high acoustic impedance characteristics. Ultrasound propagation in bone is greatly attenuated, and the depth of bone ultrasound imaging is limited. Traditional methods have difficulty reconstructing the true shape of the bone. On the other hand, due to the large difference in sound velocity between bone and soft tissue (muscles and bone marrow, etc.), the phase distortion of ultrasound propagation is severe, making it difficult for traditional beamforming to obtain accurate results, resulting in distorted bone ultrasound images. Summary of the Invention

[0003] The purpose of the present invention is to provide a frequency domain bone ultrasound imaging method based on virtual source synthetic aperture in order to improve the clarity of the image of the lower surface of the middle bone, correct the phase distortion caused by the sound velocity of soft tissue and bone, provide accurate bone thickness, and reduce the computational complexity of the bone image.

[0004] The purpose of the present invention can be achieved by the following technical solutions:

[0005] A frequency domain bone ultrasound imaging method based on virtual source synthetic aperture comprises the following steps:

[0006] S1. Use a linear probe with N array elements and a virtual source synthetic aperture transmission method to transmit N ultrasonic signals to the bone and collect echo signals;

[0007] S2. Based on the echo signal, extract the echo signal delay of the upper and lower interfaces of the bone in combination with wavelet transform;

[0008] S3. Obtain the depth of the upper and lower interfaces of the bone based on the echo signal delay and the prior sound velocity of the bone, thereby establishing a sound velocity model;

[0009] S4, the entire imaging space is divided into grids, and the echo signal p at the depth z=0 obtained in step S1 is r (t,x,z0) and the transmitted signal p T (t,x,z0) is converted to the frequency domain, and the echo frequency domain wave field P at z=0 is obtained. r (ω,k x ,z0) and transmit frequency domain wave field P T (ω,k x ,z0), where t represents the time, (x,z) represents the coordinates of the imaging area, ω represents the frequency, and kx represents the wave number component of the wave number vector in the X direction, z0 represents the depth z = 0;

[0010] S5, successively process the echo signals acquired by N transmissions, and obtain the frequency domain wave field P of the echo signal through two-dimensional Fourier transform r (ω,k x ,z0), the echo frequency domain wave field P r (ω,k x ,z0) and transmit frequency domain wave field P T (ω,k x ,z0) gradually extrapolate to obtain the echo frequency domain wave field P at different depths r (ω,k x ,z) and transmit frequency domain wave field P T (ω,k x ,z), obtain image information of different depths, and combine the image information of different depths to obtain the subframe image of each emission;

[0011] S6. Superimpose multiple sub-frame images to obtain a final bone image.

[0012] Furthermore, the specific steps of S1 are:

[0013] The probe uses a linear probe with N array elements and adopts a virtual source synthetic aperture transmission method. The transmission parameters are set. Each transmission is centered on a different array element and uses a multi-element delayed transmission method to transmit ultrasound into the bone. All array elements receive the echo signal at the same time, and a total of N transmissions are performed.

[0014] Furthermore, the transmit parameters include transmit aperture size, center frequency, sampling frequency, and transmit delay. The transmit aperture size and the number of transmit elements are determined by the virtual source focal depth and F number, specifically:

[0015]

[0016] Where L is the excitation aperture size, D is the virtual source focal depth, and F is the F number.

[0017] Furthermore, the specific steps of S3 are:

[0018] S3-1. Determine the spatial position of the upper surface of the bone using the ranging principle and obtain a double-layer sound velocity model;

[0019] S3-2. The spatial position of the lower surface is determined by combining the echo delay of the lower bone interface and the prior sound velocity of the bone, thereby obtaining a three-layer sound velocity model.

[0020] Furthermore, in the sound velocity model, the medium below the lower surface and above the upper surface is water.

[0021] Furthermore, the sound velocity model and imaging space in S4 are meshed with the same mesh size.

[0022] Furthermore, the steps of S5 are:

[0023] S5-1, the echo frequency domain wave field P r (ω,k x ,z0) and transmit frequency domain wave field P T (ω,k x ,z0) is gradually extrapolated to obtain the echo frequency domain wave field P at different depths r (ω,k x ,z) and transmit frequency domain wave field P T (ω,k x ,z), where the frequency domain wavefield at depth z is calculated by the frequency domain wavefield at depth z-1;

[0024] S5-2. Calculation of echo frequency domain wave field P at different depths r (ω,k x ,z) and transmit frequency domain wave field P T (ω,k x ,z) is used to calculate the image information at different depths, and the image information at different depths is combined to obtain the subframe image of each emission.

[0025] Furthermore, the frequency domain wavefield at depth z is:

[0026]

[0027] Where i is the complex component, Δz is the depth step, and P represents the echo frequency domain wavefield or the transmission frequency domain wavefield.

[0028] Furthermore, the subframe image is:

[0029]

[0030] Among them, I ui Indicates the uth i The sub-frame image transmitted is x, and x represents the x direction.

[0031] Furthermore, in S4, the echo signal p at the depth z=0 obtained in step S1 is transformed based on Fourier transform. r (t,x,z0) and the transmitted signal p T Convert (t,x,z0) to the frequency domain.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The present invention adopts a virtual source synthetic aperture emission mode, which has higher incident energy and can more clearly see the lower surface of the middle bone; at the same time, the phase shift algorithm combined with the constructed prior sound velocity model can correct the phase distortion caused by the sound velocity of soft tissue and bone to provide accurate bone thickness; and the phase shift algorithm has lower computational complexity, and the matrix-based operation form is suitable for parallel operation, which has the potential to achieve real-time imaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 Schematic diagram of the relationship between the virtual source aperture size, the virtual source focal depth, and the F number of the present invention;

[0035] Figure 2 1 is a schematic diagram of the time delay of the echo of the upper and lower interfaces of the bone extracted by wavelet transform in an embodiment of the present invention;

[0036] Figure 3 Schematic diagram of a three-layer sound velocity model constructed based on echo delay and transmission parameters in an embodiment of the present invention;

[0037] Figure 4 is a flow chart of the present invention;

[0038] Figure 5 is a schematic diagram of the experimental results in the embodiment of the present invention, wherein Figure 5 (a) is the result of traditional synthetic aperture without combining the sound velocity model. Figure 5 (b) is the result of traditional synthetic aperture combined with three-layer sound velocity model. Figure 5 (c) is the result of virtual source synthetic aperture without combining the sound velocity model. Figure 5 (d) is the result of virtual source synthetic aperture combined with three-layer sound velocity model. DETAILED DESCRIPTION

[0039] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0040] In order to achieve accurate imaging of the upper and lower interfaces of bones, the present invention proposes a frequency domain bone ultrasound imaging method based on virtual source synthetic aperture. The flow chart of the method is as follows: Figure 4 As shown, the present invention includes the following steps:

[0041] S1. Use a linear probe with N array elements and a virtual source synthetic aperture transmission method to transmit N ultrasonic signals to the bone and collect echo signals;

[0042] S2. Based on the echo signal, extract the echo signal delay of the upper and lower interfaces of the bone in combination with wavelet transform;

[0043] S3. Obtain the depth of the upper and lower interfaces of the bone based on the echo signal delay and the prior sound velocity of the bone, thereby establishing a sound velocity model;

[0044] S4, the entire imaging space is divided into grids, and the echo signal p at the depth z=0 obtained in step S1 is r (t,x,z0) and the transmitted signal p T (t,x,z0), converted to the frequency domain and wavenumber domain through two-dimensional Fourier transform, the echo frequency domain wave field P at z=0 is obtained r (ω,k x ,z0) and transmit frequency domain wave field P T (ω,k x ,z0), where t represents the time, (x,z) represents the coordinates of the imaging area, ω represents the frequency, and k x represents the wave number component of the wave number vector in the x direction, z0 represents the depth z = 0;

[0045] S5, successively process the echo signals acquired by N transmissions, and convert the echo frequency domain wave field P r (ω,k x ,z0) and transmit frequency domain wave field P T (ω,k x ,z0) gradually extrapolate to obtain the echo frequency domain wave field P at different depths r (ω,k x ,z) and transmit frequency domain wave field P T (ω,k x ,z), obtain image information of different depths, and combine the image information of different depths to obtain the subframe image of each emission;

[0046] S6. Superimpose multiple sub-frame images to obtain a final bone image.

[0047] In S1, the probe uses a linear probe with N array elements and adopts a virtual source synthetic aperture transmission method. The transmission parameters (including the transmission aperture size, center frequency, sampling frequency, and transmission delay, etc.) are set. Each transmission adopts a multi-element delayed transmission method to transmit ultrasound into the bone. All array elements receive the echo signal at the same time, and a total of N transmissions are performed.

[0048] The aperture size of the virtual source and the number of transmitting elements in step S1 are determined by the virtual source focal depth and F number (see Figure 1 ), whose expression is:

[0049]

[0050] L represents the excitation aperture size, D represents the virtual source focal depth, and F is the F number.

[0051] In S2, the echo signal is processed by combining the feature extraction property of wavelet transform to extract the echo signal delay of the upper and lower interfaces of the bone. Figure 2 shown.

[0052] In S3, the echo delay is extracted and combined with the sampling frequency and the prior sound velocity of the bone to obtain the approximate position of the upper and lower interfaces of the bone, and then a three-layer sound velocity model is established.

[0053] Under the premise of knowing the bone sound velocity, the sound velocity model is obtained by the following steps in step S3:

[0054] Step S3-1, using the distance measurement principle d=v 水 t / 2 determines the spatial position of the upper surface of the bone, where v 水 is the speed of sound in water;

[0055] Step S3-2, based on the premise of S3-1, combines the time delay of the lower surface of the bone and the known bone sound velocity v 骨 Determine the spatial position of the lower surface, and then obtain a double-layer sound velocity model. Assuming that the lower surface is a water medium, then obtain a three-layer sound velocity model (see Figure 3 ).

[0056] In S4, the entire imaging space is divided into grids, and the echo signal p at the depth z=0 obtained in step S1 is r (t,x,z0) and the transmitted signal p T (t,x,z0) is converted to the frequency domain P r (ω,k x ,z0) and P T (ω,k x ,z0), where t represents the time, (x,z) represents the coordinates of the imaging area, ω represents the frequency, k x Represents the wave number component of the wave number vector in the X direction. The grid size of the imaging area in S4 is the same as the grid size of the sound velocity model.

[0057] In S5, the subframe image is obtained by combining the phase shift method with the sound velocity model reconstruction, and the data is acquired by sequentially processing N transmission acquisitions. r (ω,k x ,z0) and P T (ω,k x ,z0) gradually extrapolate to obtain the frequency domain wave field P at different depths r (ω,k x ,z) and P T (ω,k x ,z), further calculate the correlation between the two to obtain the subframe image I i (x,z).

[0058] S5 includes the following steps:

[0059] In step S5-1, the frequency domain wavefield at depth z can be calculated by the frequency domain wavefield at depth z-1, and the expression is:

[0060]

[0061] Where i is the complex component and Δz is the depth step. Based on the above formula, we can know that the frequency domain wave field information P at different depths can be obtained by gradually calculating the frequency domain wave field at z = 0. r (ω,k x ,z) and P T (ω,k x ,z);

[0062] In step S5-2, the image information of the grids at different depths in the image can be obtained by calculating the correlation between the received sound field and the transmitted sound field, and the expression is:

[0063]

[0064] Combine image information of different depths to get the uth i Sub-frame image transmitted

[0065] In S6 , the reconstructed N sub-frame images are superimposed to obtain a final image I (x, z).

[0066] The following is a practical experiment:

[0067] A bovine femur was used as the in vitro experimental subject, immersed in a water tank. Pulse-echo measurements revealed a sound velocity of 3219 m / s in the femur, compared to 1500 m / s in water, which is similar to the sound velocity of soft tissue. Signal acquisition was performed using a 128-element linear array at a sampling frequency of 25 MHz.

[0068] Figure 4 FIG. 1 is a flowchart of a bone ultrasound imaging method based on virtual source synthetic aperture according to an embodiment of the present invention, which mainly includes the following steps:

[0069] Execute step S1 and pre-set the virtual source transmission mode's focal depth to -20 mm and F-number to 1.5, resulting in a transmit aperture width of 13.3 mm and 39 excitation elements. Each virtual source transmission mode transmits from a different element, acquiring signals using a multi-element delayed transmission and simultaneous reception by all elements, for a total of 128 transmissions.

[0070] Executing step S2, combining wavelet transform with echo signal extraction, it is obtained that the echo delay of the upper surface of the bone is around 27.8 μs, and the echo delay of the lower surface of the bone is around 32.9 μs.

[0071] Execute step S3 to establish a water-bone-water three-layer medium sound velocity model based on the acquisition frequency and the echo delay of the upper and lower surfaces of the bone, such as Figure 3 shown.

[0072] Step S4 is executed to discretely grid the imaging area to a size of 400*128, and Fourier transform is performed on the received signal and the transmitted signal to obtain a frequency domain wave field.

[0073] Step S5 is executed to calculate the frequency domain wave fields at different depths using a phase shift algorithm, and calculate the correlation between the transmitted wave field and the received wave field to obtain image information at different depths, which is further combined to obtain subframe images.

[0074] Step S6 is executed to superimpose the different sub-frame images to obtain a final image.

[0075] Figure 5 Schematic diagram of the in vitro experimental results in the embodiment of the present invention, Figure 5 (a) is the result of traditional synthetic aperture without combining the sound velocity model. Figure 5 (b) is the result of traditional synthetic aperture combined with three-layer sound velocity model. Figure 5 (c) is the result of virtual source synthetic aperture without combining the sound velocity model. Figure 5 (d) is the result of combining virtual source synthetic aperture with three-layer sound velocity model. Figure 5 (a, b) and Figure 5 (c, d) It can be seen that compared with the results of traditional synthetic aperture, the lower surface of the bone is more clearly visible in the reconstruction results of virtual source synthetic aperture. Figure 5 (a, c) and Figure 5 (b, d), the results without using the sound velocity model ( Figure 5 (a, c) The femur appears "flat". In contrast, the results using the sound velocity model ( Figure 5 (b, d)) The femur is closer to the real femur. The actual bone thickness is about 7.5mm. The virtual source synthetic aperture is not combined with the sound velocity model result ( Figure 5 (c)) The thickness of the femur is 3.9 mm, and the relative error is 48.0%; the result of virtual source synthetic aperture combined with sound velocity model ( Figure 5 (d)) The femur thickness is 7.2 mm, and the relative error is 4.0%.

[0076] The present invention aims to provide an ultrasound imaging method that can accurately reconstruct bone morphology. The present invention features and advantages in that it utilizes a virtual source synthetic aperture emission mode, which provides higher incident energy and a clearer image of the underlying surface of the intermediate bone. Furthermore, a phase shift algorithm, combined with a constructed priori sound velocity model, can correct for phase distortion caused by the sound velocity of soft tissue and bone, providing accurate bone thickness. Furthermore, the phase shift algorithm has lower computational complexity, and its matrix-based operation is amenable to parallel computing, potentially enabling real-time imaging.

[0077] The present invention has the following advantages:

[0078] (1) The present invention adopts a virtual source synthetic aperture emission mode, which excites multiple array elements in a single emission, has higher incident energy, and can more clearly reconstruct deep image information;

[0079] (2) The present invention adopts a phase shift algorithm, combined with a sound velocity model, to correct the phase distortion caused by the large difference in sound velocity between soft tissue and bone, and accurately image the interior of the bone.

[0080] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A frequency domain bone ultrasound imaging method based on virtual source synthetic aperture, characterized in that: The method comprises the following steps: S1. Use a linear probe with N array elements and a virtual source synthetic aperture transmission method to transmit N ultrasonic signals to the bone and collect echo signals; S2. Based on the echo signal, extract the echo signal delay of the upper and lower interfaces of the bone in combination with wavelet transform; S3. Obtain the depth of the upper and lower interfaces of the bone based on the echo signal delay and the prior sound velocity of the bone, thereby establishing a sound velocity model; S4, the entire imaging space is divided into grids, and the echo signal p at the depth z=0 obtained in step S1 is r (t,x,z0) and the transmitted signal p T (t,x,z0) is converted to the frequency domain, and the echo frequency domain wave field P at z=0 is obtained. r (ω,k x ,z0) and transmit frequency domain wave field P T (ω,k x ,z0), where t represents the time, (x,z) represents the coordinates of the imaging area, ω represents the frequency, and k x represents the wave number component of the wave number vector in the X direction, z0 represents the depth z = 0; S5, successively process the echo signals acquired by N transmissions, and obtain the frequency domain wave field P of the echo signal through two-dimensional Fourier transform r (ω,k x ,z0), the echo frequency domain wave field P r (ω,k x ,z0) and transmit frequency domain wave field P T (ω,k x ,z0) gradually extrapolate to obtain the echo frequency domain wave field P at different depths r (ω,k x ,z) and transmit frequency domain wave field P T (ω,k x ,z), obtain image information of different depths, and combine the image information of different depths to obtain the subframe image of each emission; S6. Superimpose multiple sub-frame images to obtain a final bone image.

2. The frequency domain bone ultrasound imaging method based on virtual source synthetic aperture according to claim 1, characterized in that: The specific steps of S1 are: The probe uses a linear probe with N array elements and adopts a virtual source synthetic aperture transmission method. The transmission parameters are set. Each transmission is centered on a different array element and uses a multi-element delayed transmission method to transmit ultrasound into the bone. All array elements receive the echo signal at the same time, and a total of N transmissions are performed.

3. The frequency domain bone ultrasound imaging method based on virtual source synthetic aperture according to claim 2, characterized in that: The transmit parameters include transmit aperture size, center frequency, sampling frequency, and transmit delay. The transmit aperture size and the number of transmit elements are determined by the virtual source focal depth and F number, specifically: Where L is the excitation aperture size, D is the virtual source focal depth, and F is the F number.

4. The frequency domain bone ultrasound imaging method based on virtual source synthetic aperture according to claim 3, characterized in that: The specific steps of S3 are: S3-1. Determine the spatial position of the upper surface of the bone using the ranging principle and obtain a double-layer sound velocity model; S3-2. The spatial position of the lower surface is determined by combining the echo delay of the lower bone interface and the prior sound velocity of the bone, thereby obtaining a three-layer sound velocity model.

5. The frequency domain bone ultrasound imaging method based on virtual source synthetic aperture according to claim 4, characterized in that: In the sound velocity model, the medium below the lower surface and above the upper surface is water.

6. The frequency domain bone ultrasound imaging method based on virtual source synthetic aperture according to claim 5, characterized in that: In S4, the sound velocity model and imaging space are meshed with the same grid size.

7. The frequency domain bone ultrasound imaging method based on virtual source synthetic aperture according to claim 6, characterized in that: The steps of S5 are: S5-1, the echo frequency domain wave field P r (ω,k x ,z0) and transmit frequency domain wave field P T (ω,k x ,z0) is gradually extrapolated to obtain the echo frequency domain wave field P at different depths r (ω,k x ,z) and transmit frequency domain wave field P T (ω,k x ,z), where the frequency domain wavefield at depth z is calculated by the frequency domain wavefield at depth z 1; S5-2. Calculation of echo frequency domain wave field P at different depths r (ω,k x ,z) and transmit frequency domain wave field P T (ω,k x ,z) is used to calculate the image information at different depths, and the image information at different depths is combined to obtain the subframe image of each emission.

8. The frequency domain bone ultrasound imaging method based on virtual source synthetic aperture according to claim 7, characterized in that: The frequency domain wave field at depth z is: Where i is the complex component, Δz is the depth step, and P represents the echo frequency domain wavefield or the transmission frequency domain wavefield.

9. The frequency domain bone ultrasound imaging method based on virtual source synthetic aperture according to claim 8, characterized in that: The subframe image is: in, Indicates the uth i The sub-frame image transmitted is x, and x represents the x direction.

10. The frequency domain bone ultrasound imaging method based on virtual source synthetic aperture according to claim 9, characterized in that: In S4, the echo signal p at the depth z=0 obtained in step S1 is transformed based on Fourier transform. r (t,x,z0) and the transmitted signal p T Convert (t,x,z0) to the frequency domain.

Images