Parallel computing accelerated ultrasound speckle tracking based velocity imaging method

Abstract

The embodiment of the application discloses a speed imaging method based on ultrasonic plaque tracking with parallel computing acceleration, and specifically comprises the following steps: acquiring a to-be-processed image, wherein the to-be-processed image comprises continuous multiple two-dimensional or three-dimensional images; performing plaque tracking processing on each two continuous images in the to-be-processed image; and imaging based on the method of ultrasonic plaque tracking, using parallel computing acceleration to improve the speed of post-processing, so that real-time myocardial and blood flow motion tracking can be realized, and the method has the advantages of fast imaging, large imaging field of view and high resolution; meanwhile, the measurement precision of the method is irrelevant to the imaging angle, and the method is suitable for high-precision measurement of complex myocardial motion and cardiac blood flow.

Description

Technical Field

[0001] This invention relates to the field of ultrasound technology, and more particularly to a velocity imaging method based on ultrasound patch tracking that is accelerated by parallel computing. Background Technology

[0002] Cardiovascular disease is the leading cause of death worldwide, with 15 million people dying from it annually, causing enormous clinical and social problems. The vast majority of cardiovascular diseases are closely related to abnormal cardiac function, including valvular stenosis, valvular regurgitation, heart failure, and arrhythmias. Imaging and functional diagnosis of the heart are of significant clinical value for clinical diagnosis and intraoperative surgical guidance. For example, the rapid and complex movements of the heart and blood flow within its chambers allow for myocardial motion tracking and blood flow imaging, which are crucial for diagnosing heart failure and providing real-time guidance for valve replacement / repair surgery, thus improving surgical precision.

[0003] Currently, blood flow and myocardial tissue motion imaging techniques typically use transthoracic ultrasound for Doppler imaging. However, the accuracy of the Doppler method is affected by the imaging beam angle, and the imaging frame rate is low with a small field of view, making it difficult to capture complex and rapid cardiac and cardiac chamber blood flow motion. Summary of the Invention

[0004] This application provides a parallel computing-accelerated velocity imaging method, apparatus, ultrasound equipment, and storage medium based on ultrasound patch tracking.

[0005] Firstly, a parallel computing-accelerated velocity imaging method based on ultrasound patch tracking is provided, the method comprising:

[0006] Step S1: Acquire ultrasound images to be processed, which include multiple consecutive frames of two-dimensional or three-dimensional images;

[0007] Step S2: Perform patch tracking processing on every two consecutive frames of the image to be processed, wherein the two consecutive frames include a first image and a second image; the patch tracking processing includes:

[0008] Step 1: Divide the first image and the second image into several non-overlapping window sub-regions of the same size;

[0009] Step 2: Perform Fourier transform on the corresponding window sub-regions in the first image and the second image to obtain the first data and the second data;

[0010] Step 3: Perform a Fourier transform-based cross-convolution using the first data and the second data mentioned above to obtain the convolution calculation result;

[0011] Step 4: Based on the above convolution calculation results, calculate the displacement information of the second sub-region relative to the first sub-region. The first sub-region is the k-th sub-region in the first image, and the second sub-region is the k-th sub-region in the second image, where k is a positive integer.

[0012] Step 5: Based on the displacement information above, perform deformation processing on the second image above;

[0013] Step 6: Reduce the size of the window sub-region in Step 2 above, and repeat Step 2 to Step 5 above until the iteration accuracy requirement is met, and obtain the target velocity vector distribution image.

[0014] Secondly, a velocity imaging device is provided, comprising:

[0015] The acquisition module is used to acquire the image to be processed, which includes multiple frames of two-dimensional or three-dimensional images;

[0016] The processing module is used to perform patch tracking processing on every two consecutive frames of the image to be processed, wherein the two consecutive frames include a first image and a second image;

[0017] The processing module is specifically used for:

[0018] Step 1: Divide the first image and the second image into several non-overlapping window sub-regions;

[0019] Step 2: Perform Fourier transform on the corresponding window sub-regions in the first image and the second image to obtain the first data and the second data;

[0020] Step 3: Perform a Fourier transform-based cross-convolution using the first data and the second data to obtain the convolution calculation result;

[0021] Step 4: Based on the convolution calculation result, calculate the displacement information of the second sub-region relative to the first sub-region, where the first sub-region is the k-th sub-region in the first image, and the second sub-region is the k-th sub-region in the second image, where k is a positive integer;

[0022] Step 5: Based on the displacement information, perform deformation processing on the second image;

[0023] Step 6: Reduce the size of the window sub-region in Step 2, and repeat Step 2-Step 5 until the iteration accuracy requirement is met to obtain the target velocity vector distribution image.

[0024] Thirdly, an ultrasound device is provided, including an ultrasound probe and an ultrasound host, and also including a velocity imaging device as described in the second aspect and any possible implementation thereof.

[0025] Optionally, the ultrasonic probe is selected according to the array element structure, including but not limited to ultrasonic linear / ring array or convex array probes, ultrasonic surface array probes, and rotating ultrasonic linear / ring array or convex array probes. The ultrasonic probe is selected according to the material, including but not limited to PZT ceramics, composite materials, single crystal materials, CMUT or PMUT.

[0026] Optionally, the transducer array of the above-mentioned ultrasonic probe emits sound waves in the following ways: by sequentially emitting multiple focused ultrasonic beams for phased array or linear array planar scanning, or by sequentially emitting multiple non-focused plane waves or divergent waves for planar scanning; its transmission and reception adopt fundamental wave mode or harmonic imaging mode.

[0027] Fourthly, a computer storage medium is provided, the computer storage medium storing one or more instructions, the one or more instructions being adapted to be loaded by a processor and executed as described in the first aspect and any possible implementation thereof.

[0028] This application provides a parallel computing-accelerated velocity imaging method based on ultrasound patch tracking. The method includes: Step S1: acquiring an image to be processed, which includes multiple consecutive frames of two-dimensional or three-dimensional images; Step S2: performing patch tracking processing on every two consecutive frames of the image to be processed, where each pair of consecutive frames includes a first image and a second image; The patch tracking processing includes: Step 1: dividing the first image and the second image into several non-overlapping sub-regions; Step 2: performing Fourier transform on the corresponding window sub-regions of the first image and the second image to obtain first data and second data; Step 3: making... Step 4: Perform a Fourier transform-based cross-convolution on the first data and the second data to obtain the convolution calculation result; Step 5: Calculate the displacement information of the second sub-region relative to the first sub-region based on the convolution calculation result. The first sub-region is the k-th sub-region in the first image, and the second sub-region is the k-th sub-region in the second image, where k is a positive integer; Step 6: Perform deformation processing on the second image based on the displacement information; Step 7: Reduce the size of the window sub-region in Step 2, and repeat Step 2-Step 5 until the iteration accuracy requirement is met to obtain the target velocity vector distribution image. The beneficial effects of the technical solution of this application are: imaging based on ultrasound patch tracking, using parallel computing to accelerate and improve the post-processing speed, can realize real-time myocardial and blood flow motion tracking. Compared with the Doppler method, it has the advantages of fast imaging (real-time imaging), large imaging field of view, and high resolution. At the same time, the measurement accuracy of this method is independent of the imaging angle, and it is suitable for high-precision measurement of complex myocardial motion and cardiac blood flow. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of this application or the background art, the accompanying drawings used in the embodiments of this application or the background art will be described below.

[0030] Figure 1 A schematic flowchart illustrating a parallel computing-accelerated velocity imaging method based on ultrasound patch tracking, provided in an embodiment of this application;

[0031] Figure 2 This is a schematic diagram of the structure of a velocity imaging device provided in an embodiment of this application;

[0032] Figure 3 A schematic diagram illustrating the principle of a linear array / convex array / ring array ultrasonic probe provided for embodiments of this application;

[0033] Figure 4 A schematic diagram illustrating the principle of an area array ultrasonic probe provided in an embodiment of this application;

[0034] Figure 5 This is a schematic diagram of the structure of a spiral array / convex array / ring array ultrasonic probe provided in an embodiment of this application. Detailed Implementation

[0035] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0036] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0037] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0038] The ultrasound involved in this application embodiment is a mechanical wave that propagates in an elastic medium (biological tissue). It possesses complex acoustic effects, including wave effect, thermal effect, and mechanical effect, and has advantages such as deep penetration, good spatial directivity, and dynamic focusing scanning, making it widely used in the medical field. In traditional biomedical ultrasound, ultrasound diagnostic imaging technology mainly utilizes its wave effect, while high-intensity focused ultrasound therapy technology mainly utilizes its thermal effect.

[0039] The embodiments of this application are described below with reference to the accompanying drawings.

[0040] Please see Figure 1 , Figure 1 This is a schematic flowchart of a parallel computing-accelerated velocity imaging method based on ultrasound patch tracking provided in an embodiment of this application. Figure 1 As shown, the method includes:

[0041] 101. Obtain the image to be processed, which includes multiple consecutive frames of two-dimensional or three-dimensional images.

[0042] The method in this application embodiment can be applied to electronic devices. In specific implementation, the electronic device is a device with ultrasonic detection function (ultrasonic device), including but not limited to any ultrasonic detector.

[0043] Optionally, the images to be processed can be transthoracic ultrasound images, transesophageal ultrasound images, or intracardiac ultrasound images, etc. Specifically, multiple frames can be captured continuously and rapidly; for example, the 0 frames can be denoted as I1, I2, ..., I... t ,…,I o .

[0044] 102. Perform patch tracking processing on every two consecutive frames of the above-mentioned image to be processed to obtain the target image. The above-mentioned two consecutive frames include the first image and the second image.

[0045] Specifically, for the acquired images to be processed, starting from the first frame, patch tracking processing is performed on two consecutive frames using multi-window iteration.

[0046] The above-mentioned patch tracking process includes:

[0047] Step 1: Divide the first image and the second image into several non-overlapping window sub-regions;

[0048] Step 2: Perform Fourier transform on the corresponding window sub-regions in the first image and the second image to obtain the first data and the second data;

[0049] Step 3: Perform a Fourier transform-based cross-convolution using the first data and the second data mentioned above to obtain the convolution calculation result;

[0050] Step 4: Based on the above convolution calculation results, calculate the displacement information of the second sub-region relative to the first sub-region. The first sub-region is the k-th sub-region in the first image, and the second sub-region is the k-th sub-region in the second image, where k is a positive integer.

[0051] Step 5: Based on the displacement information above, perform deformation processing on the second image above;

[0052] Step 6: Reduce the size of the window sub-region in Step 2 above, and repeat Step 2 to Step 5 above until the iteration accuracy requirement is met, and obtain the target velocity vector distribution image, i.e. the target image above.

[0053] Specifically, for the sake of simplicity, the first and second images mentioned above are represented as two consecutive frames of two-dimensional images I. t and I t+1 For example:

[0054] For step 1, I t and I t+1 (Assuming the number of pixels is M×N) Divide the image into several non-overlapping sub-regions, and assuming the size of each sub-region is m×n, then each frame of the image can be divided into... Each sub-region, among which This indicates rounding down. (The last part, "I", appears to be a typo and can be left as is.) t The k-th subregion on is denoted as

[0055] For step 2, for I t and I t+1 Perform a Fourier transform on the two corresponding window sub-regions, denoted as and

[0056] Optionally, for step 3, the following calculations can be performed:

[0057]

[0058] In the above formula, * denotes element-wise multiplication. The cross-convolution based on Fourier transform in this embodiment can be implemented through high-performance parallel computing, i.e., on a multi-core computing platform, allocating m×n threads, each thread performing... and Perform Fourier transforms separately, then perform... and The product of the corresponding pixels is then used to perform inverse Fourier transform and summation operations using high-performance parallel computing.

[0059] For step 4, the following formula can be used to calculate... Compared to Displacement information (also known as displacement field or deformation field):

[0060]

[0061]

[0062] For step 5, based on the displacement information obtained in step 4, image interpolation is used to perform I... t+1 Deformation is performed to eliminate the influence of x0 and y0 displacements.

[0063] The image interpolation involved in this application embodiment is a process of generating a high-resolution image from a low-resolution image under a model-based framework, in order to recover the information lost in the image. A specific image interpolation method can be selected as needed.

[0064] For step 6, reduce the size of the window sub-region in step 2 according to the set rules, and then repeat steps 2-5 above; iterate the above steps p times until the tracking resolution meets the requirements (the number of iterations p can be set as needed, for example, p=3), and obtain multiple deformation fields denoted as x0,…x p and y0,…y p .

[0065] The final obtained I t+1 Compared to I t The velocity field of this window sub-region is x0 + ... + x p and y0+…+y p .

[0066] Optionally, the image to be processed is a two-dimensional or three-dimensional image obtained based on an ultrasound probe;

[0067] When the image to be processed is a two-dimensional image, the displacement information includes displacement information in two directions; when the image to be processed is a three-dimensional image, the displacement information includes displacement information in three directions.

[0068] In one optional implementation, obtaining the target velocity vector distribution image until the iteration accuracy requirement is met includes:

[0069] The target velocity vector distribution image is obtained after the number of iterations reaches a preset threshold; or,

[0070] The process continues until the obtained image resolution reaches a preset resolution threshold, thus obtaining the aforementioned velocity vector distribution image that meets the accuracy requirements.

[0071] The aforementioned preset number threshold, i.e., the number of iterations p, can be set as needed, for example, p = 3; or the aforementioned preset resolution threshold can be set, and the iteration will stop when the obtained image reaches the preset resolution threshold, thus obtaining the target image, i.e., the target velocity vector distribution image.

[0072] In an optional implementation, reducing the size of the window sub-region in step 2 includes:

[0073] The size of the window sub-region mentioned in step 2 above is reduced to a certain proportion of the original size in each direction, for example, reduced to half of the original size. This application embodiment does not limit this.

[0074] Optionally, the parallel computing proposed in this application embodiment can be an FPGA, CPU, or GPU architecture, based on, for example, OpenMP, CUDA, OpenCL, or other parallel computing platforms.

[0075] The patch tracking method proposed in this application embodiment can be based on ultrasound B-mode images. In practical applications, it can also be directly tracked based on ultrasound radio frequency digital (RF) signals, thereby further improving the accuracy and speed of post-processing.

[0076] To improve imaging accuracy, ultrasound contrast agents, such as microbubbles or droplets, may also be used in the embodiments of this application to improve imaging contrast.

[0077] To improve the contrast of blood, the ultrasound emission scheme in this embodiment can be harmonic imaging, thereby reducing the signal of myocardial tissue.

[0078] Ultrasonic Doppler is a commonly used method for tracking cardiac tissue and blood flow. Since blood flow and myocardial motion are vectors containing three velocity components, it is typically necessary to rapidly emit at least three beams in different directions, alternatingly scanning the same pixel from different emission positions and directions, based on the Doppler principle. For each beam direction, the velocity component located in that beam direction is obtained using the following Doppler formula:

[0079]

[0080] Among them, f d The Doppler frequency is proportional to the tissue or blood flow velocity; θ is the angle between the ultrasound beam and the tissue or blood flow velocity; c is the ultrasound velocity, typically 1540 m / s; and f0 is the center frequency of the ultrasound. Repeating this scanning process at least three times will obtain the three velocity components of the myocardium and blood flow.

[0081] The imaging accuracy of Doppler imaging depends on the angle between the sound beam and the tissue or blood flow. The larger the angle, the lower the accuracy. Doppler imaging is relatively slow. Therefore, in order to ensure real-time imaging, it is often necessary to scan the imaging window area at a lower sampling frequency. At the same time, the imaging window area is generally small, which results in Doppler imaging often having low resolution.

[0082] The parallel computing-accelerated velocity imaging method based on ultrasound plaque tracking in this application can be used for tracking myocardial and blood flow motion, wherein parallel computing is used for accelerated post-processing. Compared with the Doppler method, it has the advantages of fast imaging (real-time imaging), large imaging field of view, and high resolution. Furthermore, the measurement accuracy of the parallel computing-accelerated velocity imaging method based on ultrasound plaque tracking in this application is independent of the imaging angle; therefore, its accuracy is higher than that of the Doppler method in complex myocardial motion and cardiac blood flow measurements.

[0083] The parallel computing-accelerated velocity imaging method based on ultrasound patch tracking in this embodiment includes the following steps: Step S1: Acquiring an image to be processed, which includes multiple frames of two-dimensional or three-dimensional images; Step S2: Performing patch tracking processing on every two consecutive frames of the image to be processed, where the two consecutive frames include a first image and a second image; The patch tracking processing includes: Step 1: Dividing the first image and the second image into several non-overlapping sub-regions; Step 2: Performing Fourier transform on the corresponding window sub-regions of the first image and the second image to obtain first data and second data; Step 3: Using the above... Step 4: Perform a Fourier transform-based cross-convolution on the first data and the second data to obtain the convolution calculation result; Step 5: Calculate the displacement information of the second sub-region relative to the first sub-region based on the convolution calculation result. The first sub-region is the k-th sub-region in the first image, and the second sub-region is the k-th sub-region in the second image, where k is a positive integer; Step 6: Perform deformation processing on the second image based on the displacement information; Step 7: Reduce the size of the window sub-region in Step 2, and repeat Steps 2-5 until the iteration accuracy requirement is met to obtain the target velocity vector distribution image. Imaging can be performed based on ultrasound patch tracking, using parallel computing to accelerate post-processing and achieve real-time myocardial and blood flow motion tracking. Compared with the Doppler method, it has the advantages of fast imaging (real-time imaging), large imaging field of view, and high resolution. At the same time, the measurement accuracy of this method is independent of the imaging angle, making it suitable for high-precision measurement of complex myocardial motion and cardiac blood flow.

[0084] Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of a velocity imaging device provided in an embodiment of this application. Figure 2 As shown, the velocity imaging device 200 includes:

[0085] The acquisition module 210 is used to acquire the image to be processed, which includes multiple frames of two-dimensional or three-dimensional images;

[0086] The processing module 220 is used to perform patch tracking processing on every two consecutive frames of the image to be processed, wherein the two consecutive frames include a first image and a second image.

[0087] Specifically, the aforementioned processing modules can be used to perform, for example... Figure 1 The patch tracking process in steps 1-6 of the illustrated embodiment will not be described again here.

[0088] Based on the description of the above method embodiments, this application also provides an ultrasonic device. This ultrasonic device includes an ultrasonic probe and an ultrasonic main unit, and further includes, for example, an ultrasonic transducer. Figure 2 The velocity imaging device described in the illustrated embodiment can perform the following: Figure 1 Any step of the method in the illustrated embodiment.

[0089] In one optional embodiment, the array element arrangement of the ultrasound probe is an ultrasound linear / ring array or convex array probe 1, an ultrasound planar array probe 2, and a rotating ultrasound linear / ring array or convex array probe 3. Probe 1 is primarily used for two-dimensional plaque tracking, obtaining two velocity components of the myocardium and blood flow in the imaging plane; probes 2 and 3 can be used for three-dimensional plaque tracking, obtaining three velocity components of the myocardium and blood flow in space.

[0090] See also Figure 3 The diagram shown illustrates the principle of a linear array / convex array / ring array ultrasonic probe. Figure 3 As shown, from left to right, the crystal arrangement of the linear array probe, the crystal arrangement of the convex array probe, and the crystal arrangement of the ring array probe are shown. Figure 4 A schematic diagram illustrating the principle of an area array ultrasonic probe provided in this application embodiment is shown below. Figure 4 As shown, the left and right images are a two-dimensional array ultrasonic probe and a 1.5D array ultrasonic probe, respectively.

[0091] Linear phased array probes come in single-line and dual-line array types. In a linear phased array probe, the crystals are arranged one-dimensionally in a straight line, allowing only beam deflection along the crystal arrangement direction. Dual-line phased array probes offer better near-field detection. Linear array probes have a planar probe surface, resulting in a large contact area, a large near-field field of view, and a small far-field field of view, forming a rectangular imaging field of view, making them suitable for blood vessels and small superficial organs. Convex array probes, on the other hand, have a convex probe surface, a small contact area, a small near-field field of view, and a large far-field field of view, forming a fan-shaped imaging field of view, making them widely used, particularly in the abdominal and lung areas.

[0092] Phased array probes can be further categorized into matrix and ring array types. Ring array phased array probes have their chips arranged in concentric rings, primarily for achieving focusing at different depths.

[0093] In addition, the probe material can be PZT ceramic, composite or single crystal ceramic, CMUT or PMUT material, and this application embodiment does not limit this.

[0094] Figure 5 This is a schematic diagram of the structure of a spiral array / convex array / ring array ultrasonic probe provided in an embodiment of this application. Figure 5 The ultrasonic probe shown includes a rotary transducer, a three-dimensional positioning sensor, a torque transmission device, a transducer lead-out harness, and a conduit. The rotary transducer is a type of rotating ultrasonic transducer; its function is to convert input electrical power into mechanical power (i.e., ultrasonic waves) and then transmit it.

[0095] In this embodiment, the ultrasound device can use transthoracic ultrasound, transesophageal ultrasound, or intracardiac ultrasound. Transesophageal or intracardiac ultrasound, due to the absence of rib obstruction and generally shallower imaging depth, typically provides a larger field of view and better imaging results compared to transthoracic ultrasound.

[0096] Optionally, the parameters of the linear array ultrasonic probe are not limited, and different arrays with different frequencies, number of array elements, array element spacing, array element distribution schemes, etc. can be selected according to various application environments.

[0097] Optionally, the transducer array can emit sound waves by sequentially emitting multiple focused ultrasonic beams for planar scanning, or by sequentially emitting multiple non-focused or divergent planar waves for planar scanning.

[0098] Optionally, the parallel computing proposed in this application embodiment can be an FPGA, CPU, or GPU architecture, based on, for example, OpenMP, CUDA, OpenCL, or other parallel computing platforms.

[0099] The patch tracking method proposed in this application embodiment can be based on ultrasound B-mode images. In practical applications, it can also be directly tracked based on ultrasound radio frequency digital (RF) signals, thereby further improving the accuracy and speed of post-processing.

[0100] To improve imaging accuracy, ultrasound contrast agents, such as microbubbles or droplets, may also be used in the embodiments of this application to improve imaging contrast.

[0101] To improve the contrast of blood, the ultrasound emission scheme in this embodiment can be harmonic imaging, thereby reducing the signal of myocardial tissue.

[0102] Currently, the clinically used methods for myocardial and cardiac chamber blood flow motion imaging are generally Doppler methods. This application proposes a parallel computationally accelerated method for myocardial and cardiac chamber blood flow motion imaging based on two-dimensional or three-dimensional plaque tracking. The key innovation of this application lies in:

[0103] 1. This paper proposes a method for imaging based on ultrasound plaque tracking and uses parallel computing to accelerate post-processing, thereby enabling real-time tracking of myocardial and blood flow motion.

[0104] 2. This study is the first to propose a method for motion tracking imaging using transesophageal or intracardiac ultrasound to track blood flow plaques in the myocardium and cardiac chambers. Compared to transthoracic ultrasound, transesophageal or intracardiac ultrasound provides superior imaging results while offering an unobstructed field of view, enabling motion tracking of the entire heart.

[0105] 3. This study first proposed using a rotating ultrasound probe for motion tracking of the myocardium and blood flow, such as... Figure 5 As shown. Compared to area array ultrasound, this approach is less expensive.

[0106] This application also provides a computer storage medium (memory), which is a memory device in an electronic device used to store programs and data. It is understood that the computer storage medium here can include both built-in storage media in the electronic device and extended storage media supported by the electronic device. The computer storage medium provides storage space that stores the operating system of the electronic device. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device; optionally, it can also be at least one computer storage medium located remotely from the aforementioned processor.

[0107] In one embodiment, a processor may load and execute one or more instructions stored in a computer storage medium to implement the corresponding steps in the above embodiments; specifically, one or more instructions in the computer storage medium may be loaded and executed by a processor. Figure 1 Any steps of the method are not described here.

[0108] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0109] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the division of modules is merely a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. The coupling, direct coupling, or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, apparatuses, or modules, and may be electrical, mechanical, or other forms.

[0110] The modules described as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0111] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media can be read-only memory (ROM), random access memory (RAM), or magnetic media, such as floppy disks, hard disks, magnetic tapes, magnetic disks, or optical media, such as digital versatile discs (DVDs), or semiconductor media, such as solid state disks (SSDs).

Claims

1. A parallel computing-accelerated velocity imaging method based on ultrasound patch tracking, characterized in that, The method includes: Step S1: Obtain the image to be processed, which includes multiple consecutive frames of two-dimensional or three-dimensional images; Step S2: Perform patch tracking processing on every two consecutive frames of the image to be processed, wherein the two consecutive frames include a first image and a second image, and the patch tracking processing includes: Step 1: Divide the first image and the second image into several non-overlapping window sub-regions of the same size in space; Step 2: Perform Fourier transform on the corresponding window sub-regions in the first image and the second image to obtain the first data and the second data; Step 3: Perform a Fourier transform-based cross-convolution using the first data and the second data to obtain the convolution calculation result; the cross-convolution is implemented using a high-performance parallel computing architecture including but not limited to CUDA and OPENMP, and the implementation includes: allocating m×n threads, each thread performing a Fourier transform on the first data and the second data respectively, then multiplying the corresponding pixels on the first data and the second data, and then using high-performance parallel computing to perform inverse Fourier transform and summation operations; Step 4: Based on the convolution calculation result, calculate the displacement information of the second sub-region relative to the first sub-region, where the first sub-region is the k-th sub-region in the first image, and the second sub-region is the k-th sub-region in the second image, where k is a positive integer; Step 5: Based on the displacement information, perform deformation processing on the second image; Step 6: Reduce the size of the window sub-region in Step 2, and repeat Step 2-Step 5 until the iteration accuracy requirement is met to obtain the target velocity vector distribution image.

2. The parallel computing-accelerated velocity imaging method based on ultrasonic patch tracking according to claim 1, characterized in that, The process of obtaining the target velocity vector distribution image until the iteration accuracy requirement is met includes: The target velocity vector distribution image is obtained until the number of iterations reaches a preset threshold; or, The process continues until the obtained image resolution reaches a preset resolution threshold, thus obtaining the velocity vector distribution image that meets the accuracy requirements.

3. The parallel computing-accelerated velocity imaging method based on ultrasonic patch tracking according to claim 1, characterized in that, Reducing the size of the window sub-region in step 2 includes: The size of the window sub-region in step 2 is reduced in all directions.

4. The parallel computing-accelerated velocity imaging method based on ultrasonic patch tracking according to claim 1, characterized in that, The image to be processed is obtained based on an ultrasonic probe, and the ultrasonic imaging method used for the image to be processed includes, but is not limited to, plane wave, divergent wave, phased array imaging, linear array imaging, and harmonic imaging. When the image to be processed is a two-dimensional image, the displacement information includes displacement information in two directions; when the image to be processed is a three-dimensional image, the displacement information includes displacement information in three directions.

5. The parallel computing-accelerated velocity imaging method based on ultrasonic patch tracking according to any one of claims 1-4, characterized in that, The image to be processed is a blood flow or tissue image.

6. A velocity imaging device, characterized in that, include: The acquisition module is used to acquire the image to be processed, which includes multiple frames of two-dimensional or three-dimensional images; The processing module is used to perform patch tracking processing on every two consecutive frames of the image to be processed, wherein the two consecutive frames include a first image and a second image; The processing module is specifically used for: Step 1: Divide the first image and the second image into several non-overlapping window sub-regions; Step 2: Perform Fourier transform on the corresponding window sub-regions in the first image and the second image to obtain the first data and the second data; Step 3: Perform a Fourier transform-based cross-convolution using the first data and the second data to obtain the convolution calculation result; the cross-convolution is implemented using a high-performance parallel computing architecture including but not limited to CUDA and OPENMP, and the implementation includes: allocating m×n threads, each thread performing a Fourier transform on the first data and the second data respectively, then multiplying the corresponding pixels on the first data and the second data, and then using high-performance parallel computing to perform inverse Fourier transform and summation operations; Step 4: Based on the convolution calculation result, calculate the displacement information of the second sub-region relative to the first sub-region, where the first sub-region is the k-th sub-region in the first image, and the second sub-region is the k-th sub-region in the second image, where k is a positive integer; Step 5: Based on the displacement information, perform deformation processing on the second image; Step 6: Reduce the size of the window sub-region in Step 2, and repeat Step 2-Step 5 until the iteration accuracy requirement is met to obtain the target velocity vector distribution image.

7. An ultrasonic device, characterized in that, It includes an ultrasound probe and an ultrasound host, and also includes a velocity imaging device as described in claim 6.

8. An ultrasonic device according to claim 7, characterized in that, The ultrasonic probe is selected according to the array element structure, including but not limited to ultrasonic linear array / ring array or convex array probes, ultrasonic surface array probes, and rotating ultrasonic linear array / ring array or convex array probes. The ultrasonic probe is selected according to the material, including but not limited to PZT ceramics, composite materials, single crystal materials, CMUT or PMUT.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, the processor performs the steps of the method as described in any one of claims 1-5.

Images