An ultrasonic speckle tracking blood flow velocity measurement method and system with adaptive kernel block width

Abstract

This invention discloses an adaptive block width ultrasound speckle tracking blood flow velocity measurement method and system, relating to the field of ultrasound blood flow velocity measurement. The method includes: determining a reference frame image and a comparison frame image from a B-mode ultrasound image; dividing the pixels in the reference frame image corresponding to the blood vessel into multiple blocks; searching for the best matching block of each block in the comparison frame image; calculating the displacement of each block relative to its corresponding best matching block; calculating the displacement gradient and velocity of each block; when the displacement gradient of a block meets a set threshold range, the velocity of the block is taken as the blood flow velocity at that block; otherwise, the size of the block is scaled. When the displacement gradient of the scaled block meets a set threshold range corresponding to the scaled block, the velocity of the scaled block is taken as the blood flow velocity at the block's location. Based on the velocities of each block, a blood flow velocity profile is determined. This invention can improve the accuracy of ultrasound imaging blood flow velocity measurement results.

Description

Technical Field

[0001] This invention relates to the field of ultrasonic blood flow velocity measurement, and in particular to an ultrasonic speckle tracking blood flow velocity measurement method and system with adaptive core width. Background Technology

[0002] With rapid socio-economic development and the continuous improvement of people's living standards, people's lifestyles have become increasingly unhealthy, leading to a year-on-year increase in the incidence of cardiovascular and cerebrovascular diseases. Cardiovascular and cerebrovascular diseases are known as the number one killer of modern people due to their high incidence, high mortality, high disability rate, and high recurrence rate. Studies have shown that atherosclerosis is the main cause of cardiovascular and cerebrovascular diseases. Therefore, accurately monitoring the development of atherosclerosis can effectively assess the risk of developing cardiovascular and cerebrovascular diseases, which is of great significance for the prevention and early diagnosis of these diseases.

[0003] Currently, commonly used clinical techniques for detecting atherosclerosis include CT scans, magnetic resonance imaging (MRI), digital subtraction angiography (DSA), and ultrasound. These techniques can all visualize vascular structure and blood flow, and assess the risk of atherosclerosis based on changes in vascular structure and hemodynamic parameters. However, the first three techniques are expensive and involve radiation exposure, making them unsuitable for long-term monitoring of atherosclerosis. In contrast, ultrasound offers advantages such as being radiation-free, inexpensive, and real-time. It can detect the progression of atherosclerosis based on measured hemodynamic information, enabling early diagnosis of cardiovascular and cerebrovascular diseases.

[0004] Hemodynamics is based on blood flow velocity distribution. By analyzing blood flow velocity distribution, hemodynamic parameters such as shear stress, velocity shear rate, and wall shear rate can be calculated. In healthy blood vessels, the blood flow velocity distribution follows a parabolic shape, with the highest velocity located at the radial center of the lumen and gradually decreasing towards the vessel wall. However, in vessels with atherosclerosis, the blood flow velocity distribution is altered by plaque, no longer exhibiting a parabolic shape, and may instead show flattened or skewed velocity profiles. Therefore, accurate detection of blood flow velocity distribution is crucial for the prevention and diagnosis of cardiovascular and cerebrovascular diseases.

[0005] Ultrasound diagnostic techniques can be divided into ultrasound Doppler and speckle tracking techniques, both of which can acquire information on blood flow velocity distribution. Ultrasound Doppler primarily relies on extracting frequency shifts or phase shifts in the blood flow signal to calculate velocity. However, limited by the Nyquist law, it has a limited maximum detectable velocity and a high angle dependence, hindering accurate velocity measurement. Therefore, speckle tracking technology was proposed. This technology tracks blood flow speckle image blocks between two ultrasound frames to calculate velocity, overcoming the shortcomings of ultrasound Doppler. The performance of speckle tracking technology is closely related to the size of the speckle block. Traditional speckle tracking uses a fixed block size. When the displacement gradient is large, the block deforms significantly, leading to a larger block size and thus a larger deviation in the calculated blood flow velocity. Conversely, when the displacement gradient is small, the block does not deform significantly, and a small block size may not capture enough blood flow information, affecting the accuracy of the blood flow velocity measurement. Summary of the Invention

[0006] The purpose of this invention is to provide an adaptive block width ultrasonic speckle tracking blood flow velocity measurement method and system, which can improve the accuracy and reliability of ultrasonic imaging blood flow velocity measurement results.

[0007] To achieve the above objectives, the present invention provides the following solution:

[0008] An adaptive block width ultrasonic speckle tracking blood flow velocity measurement method, the velocity measurement method comprising:

[0009] Acquire multiple consecutive ultrasound images of the target region; the target region is the area where the blood vessel whose blood flow velocity is to be measured is located.

[0010] One frame of all the ultrasound images is designated as the reference frame image, and the next frame of the reference frame image is used as the comparison frame image.

[0011] Based on the pixel coordinates of the blood vessel in the reference frame image, the pixels corresponding to the blood vessel are divided into multiple kernel blocks according to a preset kernel block size;

[0012] In the comparison frame image, a search region is set for each kernel block, and the corresponding best matching block is searched within the search region;

[0013] The displacement between each kernel block and the corresponding best matching block is calculated based on the pixel coordinates of each kernel block in the reference frame image and the pixel coordinates of the corresponding best matching block in the comparison frame image.

[0014] The displacement gradient of each core block is calculated based on the displacement of each core block and the corresponding best matching block.

[0015] The velocity of each kernel block is calculated based on the time interval between the reference frame image and the comparison frame image and the displacement of each kernel block relative to the corresponding best matching block;

[0016] When the displacement gradient of the core block meets the corresponding set threshold range, the velocity of the core block is taken as the blood flow velocity of the blood vessel corresponding to the location of the core block; otherwise, the preset core block size is scaled according to the set rules. When the displacement gradient corresponding to the scaled core block meets the set threshold range corresponding to the scaled core block, the velocity corresponding to the scaled core block is taken as the blood flow velocity of the blood vessel corresponding to the location of the core block. Based on the velocity of each core block, the blood flow velocity profile of the blood vessel is determined.

[0017] Optionally, acquiring multiple consecutive ultrasound images of the target region specifically includes:

[0018] An ultrasound probe with set ultrasound emission parameters is placed in the target area of ​​the blood flow velocity to be measured and then emits an ultrasound signal.

[0019] The echo signal generated after the interaction between the ultrasound signal and the blood is acquired, and the echo signal is beamformed to obtain a radio frequency signal.

[0020] The radio frequency signal is processed to obtain an ultrasound image.

[0021] Optionally, the radio frequency signal is sequentially subjected to envelope detection, dynamic compression, longitudinal downsampling, and lateral interpolation to obtain an ultrasound image.

[0022] Optionally, searching for the corresponding best matching block within the search area specifically includes:

[0023] Search for matching blocks within the search area;

[0024] The sum of squared errors is calculated by applying the sum of squared errors function to the kernel block and the corresponding matching block, and an error sum of squared error matrix is ​​established.

[0025] Calculate the value of the sum of squared errors matrix, and take the matching block corresponding to the minimum value of the sum of squared errors matrix as the best matching block corresponding to the kernel block.

[0026] Optionally, the displacement of each core block to its corresponding best-matching block is calculated using the following formula:

[0027]

[0028] Among them, D i Let x be the displacement of the i-th kernel block in the reference frame image and the best matching block corresponding to the i-th kernel block in the comparison frame image. startThe x-axis component of the starting pixel coordinates of the ultrasound image, y start S represents the y-axis component of the starting pixel coordinates of the ultrasound image. i (x start S' represents the x-axis component of the pixel coordinates of the i-th kernel block in the reference frame image. i-optimum (x start S represents the x-axis component of the pixel coordinates of the best-matching block corresponding to the i-th kernel block in the comparison frame image. i (y start S' represents the y-axis component of the pixel coordinates of the i-th kernel block in the reference frame image. i-optimum (y start ) represents the y-axis component of the pixel coordinates of the best matching block corresponding to the i-th kernel block in the comparison frame image, where i is a positive integer.

[0029] Optionally, the displacement gradient of each of the core blocks is calculated using the following formula:

[0030]

[0031] in, Let D be the displacement gradient of the i-th core block. i+1 Let D be the displacement of the (i+1)th kernel block in the reference frame image and the best matching block corresponding to the (i+1)th kernel block in the comparison frame image. i Let be the displacement of the i-th kernel block in the reference frame image and the best matching block corresponding to the i-th kernel block in the comparison frame image. Let x be the x-axis component of the displacement gradient. This represents the y-axis component of the displacement gradient.

[0032] Optionally, the velocity of each kernel block is calculated based on the time interval between the reference frame image and the comparison frame image, and the displacement of each kernel block relative to its corresponding best-matching block, using the following formula:

[0033]

[0034] Among them, v i D represents the actual velocity corresponding to the i-th core block from the upper wall to the lower wall of the blood vessel. i Let t be the displacement of the i-th kernel block in the reference frame image and the best matching block corresponding to the i-th kernel block in the comparison frame image, where t is the time interval between the reference frame image and the comparison frame image.

[0035] Optionally, when the displacement gradient of the core block meets the corresponding set threshold range, the velocity of the core block is used as the blood flow velocity at the location of the core block; otherwise, the size of the core block is scaled according to a set rule. When the displacement gradient corresponding to the scaled core block meets the set threshold range corresponding to the scaled core block, the velocity corresponding to the scaled core block is used as the blood flow velocity at the location of the core block. Specifically, this includes:

[0036] When (1-M)×THD i ≤|▽D i |≤(1+M)×THD i At that time, the velocity of the core block is taken as the blood flow velocity at the location of the core block;

[0037] When |▽D i |>(1+M)×THD i When the i-th kernel block is reduced to C times its original size, return to the step "set the search area for each kernel block in the comparison frame image and search for the best matching block within the search area" to continue execution; 0 < C < 1;

[0038] When |▽D i |<(1-M)×THD i When the size of the i-th kernel block is increased to E times its original size, return to the step "set the search area for each kernel block in the comparison frame image and search for the best matching block in the search area" to continue execution; 1 < E.

[0039] in, M is a preset percentage value, 0 < M < 100%; Let THD be the displacement gradient of the i-th core block; i V is the displacement gradient threshold for the i-th kernel block; i V represents the theoretical velocity corresponding to the i-th core block from the upper tube wall to the lower tube wall; i+1 t represents the theoretical velocity corresponding to the (i+1)th core block from the upper tube wall to the lower tube wall; t is the time interval between the reference frame image and the comparison frame image.

[0040] An adaptive block width ultrasonic speckle tracking blood flow velocity measurement system, applying the aforementioned adaptive block width ultrasonic speckle tracking blood flow velocity measurement method, the velocity measurement system comprising:

[0041] The acquisition module is used to acquire multiple consecutive B-mode ultrasound images of the target area; the target area is the region where the blood vessel whose blood flow velocity is to be measured is located.

[0042] The setting module is used to determine one frame of all the ultrasound images as a reference frame image and use the next frame of the reference frame image as a comparison frame image;

[0043] The segmentation module is used to divide the pixels corresponding to the blood vessel into multiple kernel blocks according to the pixel coordinates of the blood vessel in the reference frame image and a preset kernel block size;

[0044] The search module is used to set the search area for each kernel block in the comparison frame image and search for the corresponding best matching block within the search area;

[0045] The first calculation module is used to calculate the displacement between each kernel block and the corresponding best matching block based on the pixel coordinates of each kernel block in the reference frame image and the pixel coordinates of the corresponding best matching block in the comparison frame image.

[0046] The second calculation module is used to calculate the displacement gradient of each core block based on the displacement of each core block and the corresponding best matching block.

[0047] The third calculation module is used to calculate the velocity of each kernel block based on the time interval between the reference frame image and the comparison frame image and the displacement of each kernel block and the corresponding best matching block.

[0048] The blood flow velocity determination module is used to determine the blood flow velocity of the blood vessel corresponding to the location of the core block when the displacement gradient of the core block meets the corresponding set threshold range; otherwise, the preset core block size is scaled according to a set rule. When the displacement gradient corresponding to the scaled core block meets the set threshold range corresponding to the scaled core block, the velocity corresponding to the scaled core block is used as the blood flow velocity of the blood vessel corresponding to the location of the core block. Based on the velocities of each core block, the blood flow velocity profile of the blood vessel is determined.

[0049] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0050] The adaptive block width ultrasound speckle tracking blood flow velocity measurement method provided by this invention first acquires multiple consecutive B-mode ultrasound images of a target region. The target region is the area where the blood vessel whose blood flow velocity is to be measured is located. A reference frame image is set for the B-mode ultrasound images, and the next frame of the reference frame image is used as a comparison frame image. Based on the pixel coordinates of the blood vessel position in the reference frame image, the pixels corresponding to the blood vessel are divided into multiple blocks according to a preset block size. Second, a search area is set for each block in the comparison frame image, and the best matching block is searched within the search area. The displacement between each block and the corresponding best matching block is calculated based on the pixel coordinates of the multiple blocks and the pixel coordinates of the corresponding best matching block. Based on the displacement of each nucleus block relative to its corresponding best-matching block, the displacement gradient of each nucleus block is calculated. Furthermore, based on the time interval between the reference frame image and the comparison frame image, and the displacement of each nucleus block relative to its corresponding best-matching block, the velocity of each nucleus block is calculated. Finally, when the displacement gradient of a nucleus block meets a set threshold range, the velocity of the nucleus block is taken as the blood flow velocity at the location of the nucleus block. Otherwise, the size of the nucleus block is scaled according to a set rule. When the displacement gradient of the scaled nucleus block meets the set threshold range corresponding to the scaled nucleus block, the velocity of the scaled nucleus block is taken as the blood flow velocity at the location of the nucleus block. Based on the velocities of each nucleus block, a blood flow velocity profile is determined. This invention proposes a method for adaptive speckle nucleus block size based on blood flow displacement gradients, avoiding the situation where the nucleus block is severely deformed due to a large displacement gradient during blood flow velocity measurement, thus preventing accurate measurement of blood flow velocity. This helps improve the accuracy and reliability of ultrasound imaging blood flow velocity measurement results. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart of the ultrasonic speckle tracking blood flow velocity measurement method with adaptive core block width according to the present invention;

[0053] Figure 2 This is a flowchart illustrating the practical application of the present invention;

[0054] Figure 3 This is a schematic diagram of the speckle tracking part of the present invention;

[0055] Figure 4 This is a schematic diagram illustrating the principle of the present invention;

[0056] Figure 5 This is a schematic diagram of the blood flow model of the present invention;

[0057] Figure 6 This is a schematic diagram comparing the blood flow velocity profile of the present invention, the blood flow velocity profile of the traditional method, and the theoretical velocity profile. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] The purpose of this invention is to provide an adaptive block width ultrasonic speckle tracking blood flow velocity measurement method and system, which can improve the accuracy and reliability of ultrasonic imaging blood flow velocity measurement results.

[0060] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0061] Example 1

[0062] like Figure 1 , Figure 2 and Figure 4 As shown, this invention provides an adaptive kernel width ultrasonic speckle tracking blood flow velocity measurement method, the velocity measurement method comprising:

[0063] Step S1: Acquire multiple consecutive B-mode ultrasound images of the target area; the target area is the region where the blood vessel whose blood flow velocity is to be measured is located.

[0064] S1 specifically includes:

[0065] Step S11: Place the ultrasound probe with the set ultrasound emission parameters in the target area of ​​the blood flow velocity to be measured, and emit an ultrasound signal.

[0066] Step S12: Obtain the echo signal generated after the ultrasound signal interacts with the blood, and perform beamforming on the echo signal to obtain a radio frequency signal.

[0067] Step S13: Process the radio frequency signal to obtain an ultrasound image. Specifically, the radio frequency signal is sequentially subjected to envelope detection, dynamic compression, longitudinal downsampling, and lateral interpolation to obtain an ultrasound image.

[0068] In practical applications, the transducer type is selected. Transducers include phased array transducers, convex array transducers, and linear array transducers. The selection of the transducer type usually needs to be determined based on specific application requirements and the inspection site to obtain the best imaging effect and function. The number of array elements (m) and the element spacing (l) of the ultrasonic transducer are set; ultrasonic emission parameters are set: ultrasonic wave type, pulse type, center frequency (f), and sampling frequency (f). s The pulse repetition frequency (PRF) is used to determine the ultrasonic probe's position in the target area. The probe is positioned appropriately, and the piezoelectric ceramic effect of the transducer converts electrical energy into mechanical energy, which is then converted into acoustic energy to transmit an ultrasonic signal. The echo signal generated after the transmitted ultrasonic signal interacts with the blood is received, and beamforming is performed on these echo signals to obtain a radio frequency (RF) signal. The obtained RF signal undergoes envelope detection, dynamic compression, longitudinal downsampling, and lateral interpolation to generate a B-mode ultrasound image.

[0069] Specifically, a linear array transducer was selected as the transducer type, and the number of array elements m = 128 and the element spacing l = 0.02 μm were set. The ultrasonic transmission parameters were set as follows: the ultrasonic wave used in this example was a plane wave, the transmission pulse type was a continuous pulse, the center frequency f = 4.8 MHz, and the sampling frequency f... s =19.2mHZ, pulse repetition frequency prrf=10kHz.

[0070] Step S2: Determine one frame from all the ultrasound images as the reference frame image, and use the next frame of the reference frame image as the comparison frame image.

[0071] In practical applications, a reference frame image F and a comparison frame image F′ are determined. Specifically, an image with better quality, less noise, and fewer artifacts is selected from the generated B-ultrasound images as the reference frame image, and the next frame is used as the comparison frame image. The quality of the ultrasound image is comprehensively evaluated based on the normal structure of the human body, the clarity and uniformity of the ultrasound image, and the distribution of echo morphology, thereby selecting the reference frame image. The example in this invention is the carotid artery, and the criteria for determining the reference frame image are: (1) the muscle layer structure is clearly displayed and the trend is clear; (2) the carotid artery wall is thin, the intima is clearly displayed, the continuity is complete, and the echo in the lumen is strong. In this invention, since the comparison frame image is taken as the frame following the reference frame image, the time interval between the reference frame image and the comparison frame image is the reciprocal of the set pulse repetition frequency prrf.

[0072] Step S3: Based on the pixel coordinates of the blood vessel in the reference frame image, divide the pixels corresponding to the blood vessel into multiple kernel blocks according to a preset kernel block size.

[0073] The location of the blood vessel is determined in the reference frame F. Specifically, the pixel coordinates of the upper and lower walls of the blood vessel, as well as the leftmost and rightmost edges of the blood vessel, are determined in the reference frame image F using the measurement function of the ultrasound equipment. The blood vessel lumen is divided into N layers, with one core block set in each layer, resulting in N core blocks S.

[0074] The core block size is set to a×b (N>0):

[0075]

[0076]

[0077] Among them, v max x is the theoretical maximum blood flow velocity, xl is the leftmost pixel coordinate of the blood vessel in the ultrasound image, x r Let y be the coordinates of the rightmost pixel of the blood vessel in the ultrasound image. u Let y be the pixel coordinates of the upper tube wall in the ultrasound image. d represents the pixel coordinates of the lower tube wall in the ultrasound image, and prrf represents the pulse repetition frequency.

[0078] In practical applications, the location of the blood vessel is determined in the reference frame F, and the core blocks are divided according to the layers of the blood vessel lumen. Specifically, the blood vessel lumen is divided into N=43 layers from the upper wall to the lower wall, and a core block S is defined for each layer. The size of the core block is set to a×b=4×181 (N>0).

[0079]

[0080]

[0081] Among them, v max =0.73m / s is the theoretical maximum blood flow velocity, x l =65 is the leftmost pixel coordinate of the blood vessel in the ultrasound image, x r =554 is the rightmost pixel coordinate of the blood vessel in the ultrasound image, y u =335 is the pixel coordinate of the upper tube wall in the ultrasound image, y d =425 represents the pixel coordinates of the lower tube wall in the ultrasound image.

[0082] Step S4: Set the search area for each kernel block in the comparison frame image, and search for the corresponding best matching block within the search area.

[0083] S4 specifically includes:

[0084] Step S41: Search for matching blocks within the search area.

[0085] In practical applications, a search region R is defined for each core block in the comparison frame F′, and the matching block S' is searched by traversing within the search region R.

[0086] Specifically, the principle for defining the search area is: the search area size is larger than the kernel block size, and all matching kernel blocks are contained within the search area; in this embodiment, the search area size is: a1×b1=4×301;

[0087] The method for finding matching blocks in the search area is to traverse the search, searching from top to bottom and from left to right. Each time, a block with the same size as the core block is found in the search area as a matching block; the number of matching blocks found in the search area is always greater than 1.

[0088] Step S42: Apply the error sum of squares function to calculate the error sum of squares between the kernel block and the corresponding matching block, and establish the error sum of squares matrix.

[0089] In practical applications, the sum of squared errors (SSD) between the kernel block S and the matching block S' is calculated, and the sum of squared errors (MSSD) matrix is ​​established.

[0090]

[0091] Where a and b are the kernel block sizes, i and j are the row and column indices of the error sum of squares function matrix (i≤N (N is the number of vascular lumen layers); j≤b1-b+1), SSD(i,j) represents the value at the i-th row and j-th column position in the error sum of squares function matrix, where the rows of the error sum of squares function matrix represent the number of vascular lumen layers, and the columns represent the number of matching blocks that can be found in each layer; S(i+k-1,j+m-1) is the information of the pixel at the x-coordinate i+k-1 and y-coordinate j+m-1 in the kernel block, and S'(k,m) is the information of the pixel at the x-coordinate k and y-coordinate m in the matching block. S and S' have the same size and are a×b.

[0092] The essence of the error sum of squares formula in this invention is as follows: for each piece of information within the core block size, the error is calculated and squared for each piece of information in its corresponding matching block. Finally, all the error squares are summed. In this embodiment, the error sum of squares function matrix has 43 rows and 121 columns.

[0093] Specifically, in the comparison frame F′, a search area R corresponding to each core block is defined, and a matching block S' is searched within the search area R. The size of the matching block is set to a×b=4×181.

[0094] Calculate the sum of squared errors (SSD) between the kernel block S and the matching block S', and construct the sum of squared errors (MSSD) matrix, as shown in the appendix. Figure 3 As shown:

[0095]

[0096] Where a = 4 and b = 181 are the core block sizes.

[0097] Step S43: Calculate the value of the error sum of squares matrix, and take the matching block corresponding to the minimum value of the error sum of squares matrix as the best matching block corresponding to the core block. That is, the matching block corresponding to the minimum value of each row in the error sum of squares function matrix is ​​the best matching block S' corresponding to the core block determined by that layer of blood vessel lumen. i-optimum .

[0098] Step S5: Calculate the displacement of each kernel block and the corresponding best matching block based on the pixel coordinates of each kernel block in the reference frame image and the pixel coordinates of the corresponding best matching block in the comparison frame image.

[0099] The formula used is:

[0100]

[0101] Among them, D i Let x be the displacement of the i-th kernel block in the reference frame image and the best matching block corresponding to the i-th kernel block in the comparison frame image. start The x-axis component of the starting pixel coordinates of the ultrasound image, y start S represents the y-axis component of the starting pixel coordinates of the ultrasound image. i (x start S' represents the x-axis component of the pixel coordinates of the i-th kernel block in the reference frame image. i-optimum (x start S represents the x-axis component of the pixel coordinates of the best-matching block corresponding to the i-th kernel block in the comparison frame image. i (y start S' represents the y-axis component of the pixel coordinates of the i-th kernel block in the reference frame image. i-optimum (y start ) represents the y-axis component of the pixel coordinates of the best matching block corresponding to the i-th kernel block in the comparison frame image, where i is a positive integer, 0 < i < N, and N is the total number of kernel blocks.

[0102] In practical applications, each core block S is calculated. i Its best matching block S' i-optimum Displacement D between i :

[0103]

[0104] Where 1≤index≤b1-b+1, index is the kernel block number, b1 is the length of the search area, and b is the length of the kernel block.

[0105] Step S6: Calculate the displacement gradient of each core block based on the displacement of each core block and the corresponding best matching block.

[0106] The formula used is:

[0107]

[0108] in, Let D be the displacement gradient of the i-th core block. i+1 Let D be the displacement of the (i+1)th kernel block in the reference frame image and the best matching block corresponding to the (i+1)th kernel block in the comparison frame image. i Let be the displacement of the i-th kernel block in the reference frame image and the best matching block corresponding to the i-th kernel block in the comparison frame image. Let x be the x-axis component of the displacement gradient. This is the y-axis component of the displacement gradient. The displacement gradient is a vector with both magnitude and direction. The final displacement gradient direction and magnitude are calculated according to the mathematical principles of vector addition and subtraction. In This means that the derivative calculation is limited to the x-axis direction of the displacement. Similarly, In The derivative calculation is limited to the y-axis direction of the displacement.

[0109] In practical applications, based on the calculation of the i-th core block S i Its best matching block S' i-optimum Displacement D between i and the (i+1)th core block S i+1 Its best matching block S' i+1-optimum Displacement D between i+1 Then calculate the displacement gradient of the i-th core block. At this point, 1≤i≤N-1.

[0110] Step S7: Calculate the velocity of each kernel block based on the time interval between the reference frame image and the comparison frame image and the displacement of each kernel block and the corresponding best matching block.

[0111] The formula used is:

[0112]

[0113] Among them, v i D represents the actual velocity corresponding to the i-th core block from the upper wall to the lower wall of the blood vessel. i Let t be the displacement of the i-th kernel block in the reference frame image and the best matching block corresponding to the i-th kernel block in the comparison frame image, where t is the time interval between the reference frame image and the comparison frame image.

[0114] Step S8: When the displacement gradient of the core block meets the corresponding set threshold range, the velocity of the core block is taken as the blood flow velocity of the blood vessel corresponding to the location of the core block; otherwise, the preset core block size is scaled according to the set rules. When the displacement gradient corresponding to the scaled core block meets the set threshold range corresponding to the scaled core block, the velocity corresponding to the scaled core block is taken as the blood flow velocity of the blood vessel corresponding to the location of the core block. Based on the velocity of each core block, the blood flow velocity profile of the blood vessel is determined.

[0115] Specifically:

[0116] when At that time, the velocity of the core block is taken as the blood flow velocity at the location of the core block.

[0117] when When the size of the i-th kernel block is reduced to C times its original size, return to the step "set the search area for each kernel block in the comparison frame image and search for the best matching block in the search area" to continue execution; 0 < C < 1.

[0118] when When the size of the i-th kernel block is increased to E times its original size, return to the step "set the search area for each kernel block in the comparison frame image and search for the best matching block in the search area" to continue execution; 1 < E.

[0119] Where M is a preset percentage value, 0 < M < 100%; Let THD be the displacement gradient of the i-th core block; i V is the displacement gradient threshold for the i-th kernel block; i V represents the theoretical velocity corresponding to the i-th core block from the upper tube wall to the lower tube wall; i+1 The theoretical velocity corresponds to the (i+1)th core block from the upper pipe wall to the lower pipe wall; t is the time interval between the reference frame image and the comparison frame image. Here, the displacement gradient... Since modulo operations have been performed, we only consider the magnitude of the displacement gradient and the displacement gradient threshold, and only discuss their magnitudes. It should be noted that this part of the judgment only discusses the magnitude of the displacement gradient and its threshold.

[0120] Displacement gradient threshold THD i Setting it too high can lead to severe kernel deformation and inaccurate velocity calculations; the displacement gradient threshold THD i Setting the threshold too small will increase the computational load and reduce computational efficiency. Therefore, the optimal gradient threshold is calculated using theoretical values.

[0121]

[0122]

[0123] Among them, V i Let v be the theoretical velocity corresponding to the i-th core block from the upper tube wall to the lower tube wall, R be the radius of the tube cavity, and v be the velocity of the i-th core block from the upper tube wall to the lower tube wall. max The theoretical maximum blood flow velocity is given by r, where r is the distance from the center of the i-th nucleus block in the radial direction of the blood vessel to the center of the blood vessel. The x-axis component represents the displacement gradient threshold. This represents the y-axis component of the displacement gradient threshold. The displacement gradient threshold is a vector with both magnitude and direction. The final displacement gradient direction and magnitude are calculated using the mathematical principles of vector addition and subtraction. Here, V... i and V i+1 They are all vectors, with both magnitude and direction. middle This involves differentiating the x-axis components of the displacement between two theoretical layers (layer i and layer i+1), but... middle This involves differentiating its y-axis component. This formula only calculates the theoretical value of the velocity; the specific velocity direction is obtained from the results of color Doppler ultrasound imaging.

[0124] In practical applications, such as Figure 5 As shown, the cylindrical blood vessel is located 25 mm subcutaneously, with a lumen radius of R = 3.5 mm. The vessel is not tilted, and its maximum central velocity is v. max =0.73 m / s, then the formula for calculating the theoretical blood flow velocity of each core block in the blood flow velocity profile from the upper wall to the lower wall is:

[0125]

[0126] Where 1≤r≤R, R=3.5mm, and r is the distance from the center of the nucleus block to the center of the blood vessel in the radial position of the blood vessel.

[0127] In practical applications, the displacement gradient is determined based on a predetermined displacement gradient threshold. With displacement gradient threshold THD i The size relationship between them:

[0128] when When the i-th core block is in the i-th core block, the velocity corresponding to it is calculated directly.

[0129] when If the size of the core block is doubled (the enlarged core block size is 2a×2b), then steps S4-S8 are repeated.

[0130] when When this happens, the core block size is reduced to half of its original size (the reduced core block size is...). Repeat steps S4-S8.

[0131] Finally, as Figure 6 As shown, the velocities obtained from N core blocks are combined to draw a blood flow velocity profile. Specifically, the actual blood flow velocity of each preset core block in the lumen can be calculated using the above method. The calculated blood flow velocity values ​​of each core block are plotted in a coordinate system, and then connected to obtain the blood flow velocity profile.

[0132] Example 2

[0133] To implement the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, an adaptive block width ultrasonic speckle tracking blood flow velocity measurement system is provided below. The velocity measurement system includes:

[0134] The acquisition module is used to acquire multiple consecutive B-mode ultrasound images of the target area; the target area is the region where the blood vessel whose blood flow velocity is to be measured is located.

[0135] The setting module is used to determine one frame of all the ultrasound images as a reference frame image and use the next frame of the reference frame image as a comparison frame image;

[0136] The segmentation module is used to divide the pixels corresponding to the blood vessel into multiple kernel blocks according to the pixel coordinates of the blood vessel in the reference frame image and a preset kernel block size;

[0137] The search module is used to set the search area for each kernel block in the comparison frame image and search for the corresponding best matching block within the search area;

[0138] The first calculation module is used to calculate the displacement between each kernel block and the corresponding best matching block based on the pixel coordinates of each kernel block in the reference frame image and the pixel coordinates of the corresponding best matching block in the comparison frame image.

[0139] The second calculation module is used to calculate the displacement gradient of each core block based on the displacement of each core block and the corresponding best matching block.

[0140] The third calculation module is used to calculate the velocity of each kernel block based on the time interval between the reference frame image and the comparison frame image and the displacement of each kernel block and the corresponding best matching block.

[0141] The blood flow velocity determination module is used to determine the blood flow velocity of the blood vessel corresponding to the location of the core block when the displacement gradient of the core block meets the corresponding set threshold range; otherwise, the preset core block size is scaled according to a set rule. When the displacement gradient corresponding to the scaled core block meets the set threshold range corresponding to the scaled core block, the velocity corresponding to the scaled core block is used as the blood flow velocity of the blood vessel corresponding to the location of the core block. Based on the velocities of each core block, the blood flow velocity profile of the blood vessel is determined.

[0142] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0143] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for ultrasound speckle tracking blood flow velocity measurement with adaptive kernel width, characterized in that, The speed measurement method includes: Acquire multiple consecutive ultrasound images of the target region; the target region is the area where the blood vessel whose blood flow velocity is to be measured is located. One frame of all the ultrasound images is designated as the reference frame image, and the next frame of the reference frame image is used as the comparison frame image. Based on the pixel coordinates of the blood vessel in the reference frame image, the pixels corresponding to the blood vessel are divided into multiple kernel blocks according to a preset kernel block size; In the comparison frame image, a search region is set for each kernel block, and the corresponding best matching block is searched within the search region; The displacement between each kernel block and the corresponding best matching block is calculated based on the pixel coordinates of each kernel block in the reference frame image and the pixel coordinates of the corresponding best matching block in the comparison frame image. The displacement gradient of each core block is calculated based on the displacement of each core block and the corresponding best matching block. The velocity of each kernel block is calculated based on the time interval between the reference frame image and the comparison frame image and the displacement of each kernel block relative to the corresponding best matching block; When the displacement gradient of the core block meets the corresponding set threshold range, the velocity of the core block is taken as the blood flow velocity of the blood vessel corresponding to the location of the core block; otherwise, the preset core block size is scaled according to the set rules. When the displacement gradient corresponding to the scaled core block meets the set threshold range corresponding to the scaled core block, the velocity corresponding to the scaled core block is taken as the blood flow velocity of the blood vessel corresponding to the location of the core block. Based on the velocity of each core block, the blood flow velocity profile of the blood vessel is determined.

2. The method for adaptive kernel width ultrasound speckle tracking blood flow velocity measurement according to claim 1, characterized in that, The acquisition of multiple consecutive ultrasound images of the target region specifically includes: An ultrasound probe with set ultrasound emission parameters is placed in the target area of ​​the blood flow velocity to be measured and then emits an ultrasound signal. The echo signal generated after the interaction between the ultrasound signal and the blood is acquired, and the echo signal is beamformed to obtain a radio frequency signal. The radio frequency signal is processed to obtain an ultrasound image.

3. The method for adaptive kernel width ultrasound speckle tracking blood flow velocity measurement according to claim 2, characterized in that, The radio frequency signal is sequentially subjected to envelope detection, dynamic compression, longitudinal downsampling, and lateral interpolation to obtain an ultrasound image.

4. The method for adaptive block width ultrasound speckle tracking blood flow velocity measurement according to claim 1, characterized in that, Searching for the best matching block within the search area specifically includes: Search for matching blocks within the search area; The sum of squared errors is calculated by applying the sum of squared errors function to the kernel block and the corresponding matching block, and an error sum of squared error matrix is ​​established. Calculate the value of the sum of squared errors matrix, and take the matching block corresponding to the minimum value of the sum of squared errors matrix as the best matching block corresponding to the kernel block.

5. The method for adaptive block width ultrasound speckle tracking blood flow velocity measurement according to claim 1, characterized in that, The displacement of each core block to its corresponding best-matching block is calculated using the following formula: Among them, D i Let x be the displacement of the i-th kernel block in the reference frame image and the best matching block corresponding to the i-th kernel block in the comparison frame image. start The x-axis component of the starting pixel coordinates of the ultrasound image, y start S represents the y-axis component of the starting pixel coordinates of the ultrasound image. i (x start S' is the x-axis component of the pixel coordinates of the i-th kernel block in the reference frame image. i-optimum (x start S represents the x-axis component of the pixel coordinates of the best-matching block corresponding to the i-th kernel block in the comparison frame image. i (y start S' is the y-axis component of the pixel coordinates of the i-th kernel block in the reference frame image. i-optimum (y start ) represents the y-axis component of the pixel coordinates of the best matching block corresponding to the i-th kernel block in the comparison frame image, where i is a positive integer.

6. The method for adaptive block width ultrasound speckle tracking blood flow velocity measurement according to claim 1, characterized in that, The displacement gradient of each core block is calculated using the following formula: in, Let D be the displacement gradient of the i-th core block. i+1 Let D be the displacement of the (i+1)th kernel block in the reference frame image and the best matching block corresponding to the (i+1)th kernel block in the comparison frame image. i Let be the displacement of the i-th kernel block in the reference frame image and the best matching block corresponding to the i-th kernel block in the comparison frame image. Let x be the x-axis component of the displacement gradient. This represents the y-axis component of the displacement gradient.

7. The method for adaptive block width ultrasound speckle tracking blood flow velocity measurement according to claim 1, characterized in that, The velocity of each kernel block is calculated based on the time interval between the reference frame image and the comparison frame image, and the displacement of each kernel block relative to its corresponding best-matching block. The applied formula is as follows: Among them, v i D represents the actual velocity corresponding to the i-th core block from the upper wall to the lower wall of the blood vessel. i Let t be the displacement of the i-th kernel block in the reference frame image and the best matching block corresponding to the i-th kernel block in the comparison frame image, where t is the time interval between the reference frame image and the comparison frame image.

8. The method for adaptive block width ultrasound speckle tracking blood flow velocity measurement according to claim 1, characterized in that, When the displacement gradient of the core block meets the corresponding set threshold range, the velocity of the core block is taken as the blood flow velocity at the location of the core block; otherwise, the size of the core block is scaled according to a set rule. When the displacement gradient corresponding to the scaled core block meets the set threshold range corresponding to the scaled core block, the velocity corresponding to the scaled core block is taken as the blood flow velocity at the location of the core block, specifically including: When (1-M)×THD i ≤|▽D i |≤(1+M)×THD i At that time, the velocity of the core block is taken as the blood flow velocity at the location of the core block; When |▽D i |>(1+M)×THD i When the i-th kernel block is reduced to C times its original size, return to the step "set the search area for each kernel block in the comparison frame image and search for the best matching block within the search area" to continue execution; 0 < C < 1; When |▽D i |<(1-M)×THD i When the i-th kernel block size is increased to E times its original size, return to step "Set the search area for each kernel block in the comparison frame image and search for the best matching block within the search area" to continue execution; 1 < E; in, M is a preset percentage value, 0 < M < 100%; Let THD be the displacement gradient of the i-th core block; i V is the displacement gradient threshold for the i-th kernel block; i V represents the theoretical velocity corresponding to the i-th core block from the upper tube wall to the lower tube wall; i+1 t represents the theoretical velocity corresponding to the (i+1)th core block from the upper tube wall to the lower tube wall; t is the time interval between the reference frame image and the comparison frame image.

9. An adaptive block width ultrasonic speckle tracking blood flow velocity measurement system, characterized in that, The ultrasonic speckle tracking blood flow velocity measurement method with adaptive core width according to any one of claims 1-8, wherein the velocity measurement system comprises: The acquisition module is used to acquire multiple consecutive B-mode ultrasound images of the target area; the target area is the region where the blood vessel whose blood flow velocity is to be measured is located. The setting module is used to determine one frame of all the ultrasound images as a reference frame image and use the next frame of the reference frame image as a comparison frame image; The segmentation module is used to divide the pixels corresponding to the blood vessel into multiple kernel blocks according to the pixel coordinates of the blood vessel in the reference frame image and a preset kernel block size; The search module is used to set the search area for each kernel block in the comparison frame image and search for the corresponding best matching block within the search area; The first calculation module is used to calculate the displacement between each kernel block and the corresponding best matching block based on the pixel coordinates of each kernel block in the reference frame image and the pixel coordinates of the corresponding best matching block in the comparison frame image. The second calculation module is used to calculate the displacement gradient of each core block based on the displacement of each core block and the corresponding best matching block. The third calculation module is used to calculate the velocity of each kernel block based on the time interval between the reference frame image and the comparison frame image and the displacement of each kernel block and the corresponding best matching block. The blood flow velocity determination module is used to determine the blood flow velocity of the blood vessel corresponding to the location of the core block when the displacement gradient of the core block meets the corresponding set threshold range; otherwise, the preset core block size is scaled according to a set rule. When the displacement gradient corresponding to the scaled core block meets the set threshold range corresponding to the scaled core block, the velocity corresponding to the scaled core block is used as the blood flow velocity of the blood vessel corresponding to the location of the core block. Based on the velocities of each core block, the blood flow velocity profile of the blood vessel is determined.

Images