A method, device, medium, and product for motion correction and registration in ultrasound microvascular imaging.

Abstract

This application discloses a motion correction and registration method, device, medium, and product for ultrasound microvascular imaging, relating to the field of ultrasound microvascular imaging. The method includes acquiring ultrasound sequence images that change over time; dividing the ultrasound sequence images using a threshold of the correlation coefficient of data frames; determining reconstructed images of still frames and relatively still frames based on the clustering results of clustering processing according to the correlation coefficients between still frames; performing image registration on the reconstructed images of still frames and relatively still frames respectively to obtain reconstructed images after still frame registration and reconstructed images after relatively still frame registration; and superimposing the reconstructed images after still frame registration, the reconstructed images after relatively still frame registration, and the reconstructed image without respiratory motion to obtain a registered image. This application can improve the performance and quality of ultrasound super-resolution microvascular imaging.

Description

Technical Field

[0001] This application relates to the field of ultrasound microvascular imaging, and in particular to a method, device, medium and product for motion correction and registration in ultrasound microvascular imaging. Background Technology

[0002] Super-resolution imaging (ULM) reconstructs microvascular networks by locating and tracking sparsely distributed microbubble contrast agents, visualizing microvessels ranging from 2μm to 20μm in size, ideally capable of depicting capillary networks. However, the sparse target results in long acquisition times, typically between 1 and 10 minutes. Ideally, during this acquisition time, the imaging target in ULM is completely stationary. However, in actual in vivo experiments, factors such as animal respiration, heartbeat, or external pressure can cause changes in the position and morphology of microvessels in ultrasound images over time, resulting in tissue movement. Since ULM reconstructs images by superimposing the locations of many microbubbles accumulated over time, inter-frame motion significantly affects the visualization of microvascular images. Furthermore, the impact of motion scales in clinical scans is often greater than the super-resolution itself, directly affecting image clarity and accuracy. Therefore, precise motion correction is necessary for the acquired data during long acquisition periods.

[0003] In 2014, Jeffries et al. used in vivo ULM technology and post-processing algorithms to achieve a relatively high-resolution representation of microvessels. However, the human body needs to remain still and the mouse head needs to be fixed. It can be seen that the above method causes some inconvenience to subsequent practical applications.

[0004] In 2017, Foiret et al. achieved maximum axial and lateral resolutions of 2.1 μm and 6.1 μm for rat renal microvessels by excluding frames with excessive respiratory motion and reducing the influence of motion. However, this was based on the assumption that only rigid motion existed in the experiment. In reality, there are some non-rigid motions in human tissues that cannot be ignored, so it is not easy to directly transfer to practical applications.

[0005] In 2019, Piepenbrock et al. proposed applying the optical flow method to non-separated B-mode images. In the process of non-rigid motion estimation, the microbubble region was excluded, and the motion displacement error in the simulation experiment was as low as 6μm. Although the new technology improved the robustness of the results, its steps still involve selecting all combinations of two frames from tens of thousands of frames for synthesis. Therefore, it has not made much improvement on the key point of speed.

[0006] In 2019, Hao et al. proposed a speckle tracking technique based on 2D normalized cross-correlation (NCC) to estimate liver displacement. This technique was used to correct non-rigid motion at the center of microbubbles. Although it effectively improved the accuracy and resolution of mouse liver ULM, the underlying technology still required registration every two frames, so the time cost was not reduced.

[0007] In 2021, Taghavi et al. used B-Mode images for local motion correction and applied the estimated motion field to detect microbubble locations in contrast images of rat kidneys. However, the dynamic effect was not very good, and the accuracy was low in three sets of data due to the movement of the rats.

[0008] It is evident that existing technologies for processing ultrasound microvascular imaging have the following problems:

[0009] (1) Long processing time: Ultrasonic localization microscopy (ULM) usually requires the collection of tens of thousands of frames of ultrasound data to reconstruct the final image. Moreover, most existing registration techniques use arbitrary two frames to be combined for registration, which requires more computation and thus consumes a lot of time.

[0010] (2) Direct ultrasound data registration is difficult: ultrasound images are based on tissue scattering echo signals. In some cases, tissue structure information is not obvious. Direct motion registration of ultrasound data may result in large errors.

[0011] (3) Low data utilization: Based on the large amount of data mentioned above, some technologies may discard some data, such as some microbubble individuals, or ignore some tissue movement patterns (only considering rigid registration and not non-rigid registration, etc.) in order to reduce the running time. This will greatly reduce the data utilization.

[0012] (4) The scope of applicable scenarios is relatively narrow: Most of the previous technologies were only suitable for objects that were stationary or as stationary as possible, such as making objects hold their breath or fixing a certain part, which greatly limited the versatility and flexibility of the technology.

[0013] Therefore, considering the above problems, there is an urgent need to provide a new microvascular image registration method that can make full use of the large and effective amount of data while ensuring the accuracy and stability of the algorithm and shortening the data processing time. Summary of the Invention

[0014] The purpose of this application is to provide a motion correction and registration method, device, medium, and product for ultrasound microvascular imaging, which can improve the performance and quality of ultrasound super-resolution microvascular imaging.

[0015] To achieve the above objectives, this application provides the following solution:

[0016] In a first aspect, this application provides a motion correction and registration method for ultrasound microvascular imaging, the motion correction and registration method for ultrasound microvascular imaging comprising:

[0017] Acquire ultrasound sequence images that change over time; the ultrasound sequence images include: sequence images of microbubble scattering flow and sequence images of tissue scattering caused by respiratory motion;

[0018] The ultrasound sequence images are divided using a threshold of the data frame correlation coefficient to obtain reconstructed images with and without respiratory motion; the threshold of the data frame correlation coefficient is used to determine whether respiratory motion exists in the ultrasound sequence images.

[0019] The reconstructed images of still frames and relative still frames are determined by clustering based on the correlation coefficients between still frames. The still frames represent the resting state between two breaths, while the relative still frames represent the state between inhalation and exhalation.

[0020] Image registration is performed on the reconstructed images of still frames and relatively still frames respectively to obtain the reconstructed images after still frame registration and the reconstructed images after relatively still frame registration.

[0021] The reconstructed image after registration of still frames, the reconstructed image after registration of relative still frames, and the reconstructed image without respiratory motion are superimposed to obtain the registered image.

[0022] Optionally, the step of dividing the ultrasound sequence images using a threshold of the data frame correlation coefficient to obtain reconstructed images with and without respiratory motion specifically includes:

[0023] Ultrasound sequence images are segmented using a threshold based on the correlation coefficient of data frames;

[0024] The segmented ultrasound sequence images were subjected to SVD filtering, localization, tracking and reconstruction to obtain reconstructed images with and without respiratory motion.

[0025] Optionally, the image registration method is a registration method that employs a two-stage rigid and affine transformation followed by symmetric normalization deformation.

[0026] Optionally, the motion correction and registration method in ultrasound microvascular imaging further includes:

[0027] A model of ultrasound microvascular network was simulated using the MATLAB ultrasound toolbox and COMSOL Multiphysics software, and then evaluated using the ultrasound microvascular network model.

[0028] Optionally, the ultrasound microvascular network model is jointly simulated using the MATLAB ultrasound toolbox and COMSOL Multiphysics software, specifically including:

[0029] Obtain the geometric model of the microvessels;

[0030] The geometric model was imported into COMSOL Multiphysics for finite element simulation to obtain the blood flow velocity field in the microvessels and the displacement field caused by the respiratory motion of the tissue.

[0031] Based on the blood flow velocity field within microvessels and the displacement field caused by respiratory motion leading to tissue movement, the MUST toolkit is used to simulate the ULM imaging process, resulting in a co-simulated ultrasound microvascular network model. This co-simulated ultrasound microvascular network model is used to obtain time-varying sequence images of microbubble scatterer flow and sequence images of tissue scatterers moving due to respiratory motion.

[0032] Optionally, the ultrasound microvascular network model is jointly simulated using the MATLAB ultrasound toolbox and COMSOL Multiphysics software, and then includes:

[0033] The ultrasound microvascular network model was optimized.

[0034] Secondly, this application also provides a motion correction and registration device for ultrasound microvascular imaging, the motion correction and registration device for ultrasound microvascular imaging comprising:

[0035] An ultrasound sequence image acquisition module is used to acquire ultrasound sequence images that change over time; the ultrasound sequence images include: sequence images of microbubble scattering flow and sequence images of tissue scattering caused by respiratory motion.

[0036] An ultrasound sequence image segmentation module is used to segment ultrasound sequence images using a threshold of the data frame correlation coefficient to obtain reconstructed images with respiratory motion and reconstructed images without respiratory motion; the threshold of the data frame correlation coefficient is used to determine whether respiratory motion exists in the ultrasound sequence image.

[0037] The module for determining the reconstructed images of still frames and relative still frames is used to determine the reconstructed images of still frames and relative still frames based on the clustering results of clustering processing according to the correlation coefficient between still frames; the still frame is the resting state between two breaths; the relative still frame is the state between inhalation and exhalation.

[0038] The image registration module is used to register the reconstructed image of the still frame and the reconstructed image of the relative still frame respectively, so as to obtain the reconstructed image after still frame registration and the reconstructed image after relative still frame registration.

[0039] The image determination module after registration is used to overlay the reconstructed image after registration of still frames, the reconstructed image after registration of relative still frames, and the reconstructed image without respiratory motion to obtain the registered image.

[0040] Thirdly, this application also provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the motion correction and registration method in ultrasound microvascular imaging.

[0041] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the motion correction and registration method in ultrasound microvascular imaging.

[0042] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the motion correction and registration method in ultrasound microvascular imaging.

[0043] According to the specific embodiments provided in this application, this application has the following technical effects:

[0044] This application provides a motion correction and registration method, device, medium, and product for ultrasound microvascular imaging. It determines the reconstructed images of still frames and relatively still frames by segmenting ultrasound sequence images using a threshold of data frame correlation coefficients and by clustering the results based on the correlation coefficients between still frames. This provides a frame rate segmentation optimization registration scheme, which effectively reduces interference from respiratory motion and obtains better ultrasound super-resolution microvascular images while significantly improving processing speed. Furthermore, this application does not restrict whether the object is stationary, thus improving its versatility. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram of a motion correction and registration method in ultrasound microvascular imaging according to an embodiment of this application;

[0047] Figure 2 This is a schematic diagram of the image registration process in one embodiment of this application;

[0048] Figure 3 This is a schematic diagram of frame segmentation optimization registration results for different registration methods in one embodiment of this application;

[0049] Figure 4 This is a schematic diagram showing the results of different registration methods under a registration network in one embodiment of this application;

[0050] Figure 5 This is a schematic diagram showing the results of different registration methods under a registration network in another embodiment of this application;

[0051] Figure 6 This is a schematic diagram showing the comparison results of rat liver microvascular images before and after motion correction. Detailed Implementation

[0052] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0053] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0054] In one exemplary embodiment, such as Figure 1 As shown, a motion correction and registration method for ultrasound microvascular imaging is provided. This method is executed by a computer device, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is described using a server as an example, and includes the following steps S101 to S105. Wherein:

[0055] S101, acquire ultrasound sequence images that change over time; the ultrasound sequence images include: sequence images of microbubble scattering flow and sequence images of tissue scattering caused by respiratory motion;

[0056] S102, the ultrasound sequence images are divided using the data frame correlation coefficient threshold Corr_th2 to obtain reconstructed images with and without respiratory motion. The data frame correlation coefficient threshold Corr_th2 is used to determine whether respiratory motion exists in the ultrasound sequence images; that is, data frames with correlation coefficients less than the threshold Corr_th2 exhibit respiratory motion. Respiratory motion can be divided into two phases: exhalation and inhalation, and the displacement directions generated in these two phases are inconsistent, thus further subdividing the data frames in the motion phase. Therefore, dividing the ultrasound sequence images is essentially dividing the data frames within the ultrasound sequence images.

[0057] S102 specifically includes:

[0058] Ultrasound sequence images are segmented using a threshold based on the correlation coefficient of data frames;

[0059] The segmented ultrasound sequence images were subjected to SVD filtering, localization, tracking and reconstruction to obtain reconstructed images with and without respiratory motion.

[0060] S103, based on the clustering results of the correlation coefficient between still frames, determine the reconstructed image of the still frame and the reconstructed image of the relative still frame; the still frame is the resting state between two breaths; the relative still frame is the state between inhalation and exhalation; that is, based on the slope analysis of the direction of respiratory movement, determine the slope of the inter-frame correlation curve with respiratory movement stage, and obtain the data frame number of the two states of exhalation and inhalation respectively.

[0061] S104, perform image registration on the reconstructed image of the still frame and the reconstructed image of the relative still frame, respectively, to obtain the reconstructed image after still frame registration and the reconstructed image after relative still frame registration; wherein, the image registration process is as follows: Figure 2 As shown; the image registration method is a registration method that first performs rigid and affine two-stage transformations and then performs symmetric normalization deformation.

[0062] S105, the reconstructed image after registration of the still frame, the reconstructed image after registration of the relative still frame, and the reconstructed image without respiratory motion are superimposed to obtain the registered image.

[0063] like Figure 3 The frame-segmentation optimization registration results for different registration methods are shown. The frame-segmentation optimization registration scheme can essentially eliminate artifacts caused by respiratory motion, resulting in a clearer display of microvascular structures. For image similarity assessment, the SSIM and MSE indices were used to evaluate the registration results in the images. The evaluation indices were calculated three times and then averaged. The statistical results are shown in Table 1.

[0064] Table 1. Performance comparison of microvascular model registration using number segmentation optimization under different registration methods.

[0065]

[0066]

[0067] As shown in Table 1, the SSIM and MSE indices reveal that for the complex microvascular network model 2, the registration method employing a two-stage rigid and affine transformation followed by SyN deformation improves the SSIM value by approximately 0.14 compared to the unregistered SSIM value, and reduces the MSE to approximately 0.436. Experimental results demonstrate that using a frame segmentation-optimized registration scheme for ultrasound image registration can more effectively improve the performance of super-resolution microvascular imaging and enhance image quality. The entire process takes only 5-6 hours, approximately 60% shorter than other algorithms, significantly reducing time costs.

[0068] To verify and evaluate the effectiveness of the motion correction and registration method in ultrasound microvascular imaging provided in this application, this application also includes:

[0069] A model of ultrasound microvascular network was simulated using the MATLAB ultrasound toolbox and COMSOL Multiphysics software, and then evaluated using the ultrasound microvascular network model.

[0070] In another exemplary embodiment, the ultrasound microvascular network model is jointly simulated using the MATLAB ultrasound toolbox and COMSOL Multiphysics software, specifically including:

[0071] S1, Obtain the geometric model of the microvessels;

[0072] S2. Import the geometric model into COMSOL Multiphysics for finite element simulation. Use theoretical knowledge such as fluid mechanics and solid mechanics to obtain the blood flow velocity field in microvessels and the displacement field caused by respiratory motion.

[0073] S3. Based on the blood flow velocity field within the microvessels and the displacement field caused by respiratory motion leading to tissue movement, the MUST toolkit is used to simulate the ULM imaging process, resulting in a co-simulated ultrasound microvascular network model. The co-simulated ultrasound microvascular network model is used to obtain time-varying sequence images of microbubble scatterer flow and sequence images of tissue scatterers moving due to respiratory motion.

[0074] S2 is followed by:

[0075] The ultrasound microvascular network model was optimized to more closely resemble reality, specifically including microbubble displacement due to respiration and continuous microbubble inflow. Finally, the simulated ultrasound images underwent post-processing to reconstruct super-resolution ultrasound microvascular images, as shown in the final image. Figure 4 As shown.

[0076] The ultrasound sequence images of the previously simulated ULM imaging process were reconstructed. However, the quality of the reconstructed ULM images degraded due to respiratory motion. Therefore, motion registration of the ULM imaging data was performed to improve the quality of the moving images.

[0077] Common registration algorithms include affine registration, symmetric normalization (SyN) registration, and elastic registration. These algorithms can process various types of medical images, including MRI, CT, and ultrasound.

[0078] Affine registration adjusts the shape and position of an image through linear transformations such as translation, rotation, scaling, and shearing. It is efficient in handling simple image registration problems. In contrast, rigid registration is limited to translation and rotation operations and does not include shearing and scaling. Affine registration algorithms aim to find suitable affine transformation parameters so that points in two images correspond spatially before and after registration. However, affine registration methods cannot capture nonlinear relationships between images and do not include factors related to structural deformation, thus having certain limitations.

[0079] Symmetric Normalization (SyN) registration is one of the most commonly used registration algorithms. It is a nonlinear registration method based on deformation fields, capable of capturing complex anatomical variations. Image registration is achieved by minimizing similarity metrics between images (such as mutual information and normalized cross-correlation) and the regularization term of the deformation field, making the two images as similar as possible after registration. The SyN deformation algorithm has many advantages, such as considering the transformations between two images, better handling interchangeability and symmetry between images, capturing complex nonlinear transformations between images, being suitable for handling large image differences and deformations, and being adaptable to specific situations. It exhibits good adaptability and flexibility in different types of image registration, considering both global transformations and local deformations between images, and offering higher registration accuracy and precision compared to affine transformations.

[0080] The above registration methods each have their own advantages and disadvantages. Combining these registration methods can sometimes achieve better registration results by taking advantage of each other's strengths and compensating for each other's weaknesses. The following is a detailed analysis of the combinations.

[0081] According to such Figure 4 The ULM image with tissue motion, reconstructed from the original image, is then image-registered with a ULM image reconstructed without tissue motion. The registration results include: the mapping matrix estimated by linear transformation, the mapping relationship estimated by nonlinear transformation, the result of registering the reference image to the moving image, and the result of registering the moving image to the reference image. Next, the mapping relationship is applied to the filtered microbubble IQ data with tissue motion displacement to obtain motion-corrected microbubble IQ data. Finally, microbubble localization, tracking, and image reconstruction are performed on the corrected microbubble image to obtain the motion-corrected ULM image.

[0082] Registration methods with different combinations were used to obtain, for example... Figure 4 and Figure 5 The results are shown. Among them, Figure 4 and Figure 5 The reference images used in the study are all reconstructed images of the same batch of microbubble scatterers under tissue motion conditions, and the deformation algorithm used is the SyN deformation registration method.

[0083] The following results based on in vivo imaging data further illustrate the effectiveness of this application. First, cross-sectional images of rat livers were acquired, microbubbles were injected, and post-processing and super-resolution image reconstruction were performed. Then, the registration scheme was optimized using frame segmentation. Rigid and affine transformations were performed first, followed by SyN deformation registration. Respiratory state data frames were separated, motion correction was performed, and the images were superimposed.

[0084] Figure 6 The image displays four sets of before-and-after microvascular motion correction comparisons, divided into two rows. The first row shows the reconstructed image without motion correction, and the second row shows the reconstructed image obtained using a frame segmentation optimized registration scheme, employing a two-stage rigid and affine transformation followed by SyN deformation registration. Figure 6 As indicated by the red and white arrows, the frame-segmentation optimized registration scheme reduces the impact of motion artifacts, revealing clearer microvascular branches. Similarly, as shown in the white box, the frame-segmentation optimized registration scheme more accurately assesses the thickness of microvessels. Therefore, the following conclusions can be drawn from the above results: the frame-segmentation optimized registration scheme can significantly reduce motion artifacts and obtain clearly branched microvascular images. This demonstrates that this application can improve the performance of ultrasound super-resolution microvascular imaging.

[0085] The results in Table 2 show that the frame segmentation optimization registration scheme can effectively reduce the interference caused by respiratory motion, reduce the average thickness of microvessels by about 2.2 times, and improve the number of terminal points and the average porosity of blood vessels, effectively improving the quality of ultrasound super-resolution microvascular images. At the same time, the entire process takes only 6-7 hours, which is nearly 60% shorter than the time taken by other similar techniques.

[0086] Table 2 Comparison of Quantitative Parameters of Microvascular Characteristics

[0087]

[0088] As can be seen, this application can reduce the interference of respiratory movements and obtain better ultrasound high-resolution microvascular images while significantly improving the operating speed. In other words, this application can shorten data processing time and improve imaging speed.

[0089] Based on the same inventive concept, this application also provides a motion correction and registration device for ultrasound microvascular imaging to implement the motion correction and registration method described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the motion correction and registration device for ultrasound microvascular imaging provided below can be found in the limitations of the motion correction and registration method for ultrasound microvascular imaging described above, and will not be repeated here.

[0090] In one exemplary embodiment, a motion correction and registration device for ultrasound microvascular imaging is provided, comprising:

[0091] An ultrasound sequence image acquisition module is used to acquire ultrasound sequence images that change over time; the ultrasound sequence images include: sequence images of microbubble scattering flow and sequence images of tissue scattering caused by respiratory motion.

[0092] An ultrasound sequence image segmentation module is used to segment ultrasound sequence images using a threshold of the data frame correlation coefficient to obtain reconstructed images with respiratory motion and reconstructed images without respiratory motion; the threshold of the data frame correlation coefficient is used to determine whether respiratory motion exists in the ultrasound sequence image.

[0093] The module for determining the reconstructed images of still frames and relative still frames is used to determine the reconstructed images of still frames and relative still frames based on the clustering results of clustering processing according to the correlation coefficient between still frames; the still frame is the resting state between two breaths; the relative still frame is the state between inhalation and exhalation.

[0094] The image registration module is used to register the reconstructed image of the still frame and the reconstructed image of the relative still frame respectively, so as to obtain the reconstructed image after still frame registration and the reconstructed image after relative still frame registration.

[0095] The image determination module after registration is used to overlay the reconstructed image after registration of still frames, the reconstructed image after registration of relative still frames, and the reconstructed image without respiratory motion to obtain the registered image.

[0096] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal. The computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a motion correction and registration method in ultrasound microvascular imaging.

[0097] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0098] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0099] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0100] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0101] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0102] In this application, all actions to acquire signals, information, or data are carried out in compliance with the relevant data protection laws and policies of the country where the location is situated, and with the authorization granted by the owner of the relevant device.

[0103] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, 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 this application.

Claims

1. A motion correction and registration method for ultrasound microvascular imaging, characterized in that, The motion correction and registration method in ultrasound microvascular imaging includes: Acquire ultrasound sequence images that change over time; the ultrasound sequence images include: sequence images of microbubble scattering flow and sequence images of tissue scattering caused by respiratory motion; The ultrasound sequence images are divided using a threshold of the data frame correlation coefficient to obtain reconstructed images with and without respiratory motion; the threshold of the data frame correlation coefficient is used to determine whether respiratory motion exists in the ultrasound sequence images. The reconstructed images of still frames and relative still frames are determined by clustering based on the correlation coefficients between still frames. The still frames represent the resting state between two breaths, while the relative still frames represent the state between inhalation and exhalation. Image registration is performed on the reconstructed images of still frames and relatively still frames respectively to obtain the reconstructed images after still frame registration and the reconstructed images after relatively still frame registration. The reconstructed image after registration of still frames, the reconstructed image after registration of relative still frames, and the reconstructed image without respiratory motion are superimposed to obtain the registered image.

2. The motion correction and registration method in ultrasound microvascular imaging according to claim 1, characterized in that, The process of dividing ultrasound sequence images using a threshold based on the correlation coefficient of data frames to obtain reconstructed images with and without respiratory motion specifically includes: Ultrasound sequence images are segmented using a threshold based on the correlation coefficient of data frames; The segmented ultrasound sequence images were subjected to SVD filtering, localization, tracking and reconstruction to obtain reconstructed images with and without respiratory motion.

3. The motion correction and registration method in ultrasound microvascular imaging according to claim 1, characterized in that, The image registration method is a registration method that uses two-stage rigid and affine transformations followed by symmetric normalization deformation.

4. The motion correction and registration method in ultrasound microvascular imaging according to claim 1, characterized in that, The motion correction and registration method in ultrasound microvascular imaging further includes: A model of ultrasound microvascular network was simulated using the MATLAB ultrasound toolbox and COMSOL Multiphysics software, and then evaluated using the ultrasound microvascular network model.

5. The motion correction and registration method in ultrasound microvascular imaging according to claim 4, characterized in that, The ultrasound microvascular network model was jointly simulated using the MATLAB ultrasound toolbox and COMSOL Multiphysics software, specifically including: Obtain the geometric model of the microvessels; The geometric model was imported into COMSOL Multiphysics for finite element simulation to obtain the blood flow velocity field in the microvessels and the displacement field caused by the respiratory motion of the tissue. Based on the blood flow velocity field within microvessels and the displacement field caused by respiratory motion leading to tissue movement, the MUST toolkit is used to simulate the ULM imaging process, resulting in a co-simulated ultrasound microvascular network model. This co-simulated ultrasound microvascular network model is used to obtain time-varying sequence images of microbubble scatterer flow and sequence images of tissue scatterers moving due to respiratory motion.

6. The motion correction and registration method in ultrasound microvascular imaging according to claim 5, characterized in that, The ultrasound microvascular network model was jointly simulated using the MATLAB ultrasound toolbox and COMSOL Multiphysics software, and the following steps were also included: The ultrasound microvascular network model was optimized.

7. A motion correction and registration device for ultrasound microvascular imaging, characterized in that, The motion correction and registration device in ultrasound microvascular imaging includes: An ultrasound sequence image acquisition module is used to acquire ultrasound sequence images that change over time; the ultrasound sequence images include: sequence images of microbubble scattering flow and sequence images of tissue scattering caused by respiratory motion. An ultrasound sequence image segmentation module is used to segment ultrasound sequence images using a threshold of the data frame correlation coefficient to obtain reconstructed images with respiratory motion and reconstructed images without respiratory motion; the threshold of the data frame correlation coefficient is used to determine whether respiratory motion exists in the ultrasound sequence image. The module for determining the reconstructed images of still frames and relative still frames is used to determine the reconstructed images of still frames and relative still frames based on the clustering results of clustering processing according to the correlation coefficient between still frames; the still frame is the resting state between two breaths; the relative still frame is the state between inhalation and exhalation. The image registration module is used to register the reconstructed image of the still frame and the reconstructed image of the relative still frame respectively, so as to obtain the reconstructed image after still frame registration and the reconstructed image after relative still frame registration. The image determination module after registration is used to overlay the reconstructed image after registration of still frames, the reconstructed image after registration of relative still frames, and the reconstructed image without respiratory motion to obtain the registered image.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the motion correction and registration method in ultrasound microvascular imaging according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the motion correction and registration method in ultrasound microvascular imaging as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the motion correction and registration method in ultrasound microvascular imaging as described in any one of claims 1-6.

Images