Microvessel three-dimensional imaging method and system based on ultrasonic phased array, medium

Abstract

The application discloses a kind of based on microvessel three-dimensional imaging method of ultrasonic phased array, it is related to ultrasonic imaging technology.The application can obtain three-dimensional ultrasonic sequence image according to the ultrasonic image data obtained by multiple spherical wave emission operation, then three-dimensional ultrasonic sequence image is divided into multiple space-time sub-blocks, singular value decomposition is carried out to all space-time sub-blocks, contrast agent signal is extracted, and three-dimensional contrast agent sequence image corresponding to three-dimensional ultrasonic sequence image is obtained according to contrast agent signal;Three-dimensional contrast agent sequence image is carried out binaryzation processing, obtains multiple connected domains, and the microbubble center point corresponding to each connected domain is calculated according to each connected domain, and finally microvessel three-dimensional imaging operation is carried out according to microbubble center point.Each space-time sub-block design can reduce the influence of large blood vessels and physiological movement on data;Three-dimensional contrast agent sequence image is carried out binaryzation processing, can reduce the amount of calculation, improve efficiency, and finally can efficiently and accurately realize the three-dimensional imaging of microvessel.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of ultrasonic imaging, in particular to a microvessel three-dimensional imaging method and system based on an ultrasonic phased array, and a medium. BACKGROUND

[0002] Ultrasound localization microscopy (ULM) is a new type of ultrasonic imaging method, which indirectly depicts the structure of blood vessels and hemodynamics by tracking the trajectory of microbubbles in the blood vessel network. Because the imaging spatial resolution of the technology is high, it has great clinical application value in tumor benign and malignant differentiation and diagnosis and treatment of heart and brain circulation related diseases. ULM relies on rapid acquisition of ultrasonic sequences to reconstruct high-resolution microvessel distribution maps of tissues, but it is challenging to obtain multiple frames of high space-time coherent ultrasonic images in a short time due to the physiological movement of the heart. Correspondingly, ultrasonic super-resolution microvessel imaging technology (SR) uses ultrasonic contrast agent through intravenous injection, uses ultrasonic imaging algorithm to track the flow trajectory of the contrast agent, and superimposes multiple frames of contrast agent trajectories to reconstruct the microvessel morphology and calculate the hemodynamic parameters.

[0003] In 2024, Professor Mengxing Tang's team of Imperial College London used phased array ultrasonic superfast imaging combined with electro-physiological equipment to collect transthoracic echocardiogram sequences to visualize myocardial microcirculation and coronary arteries. However, this microvessel three-dimensional imaging method based on ultrasonic phased array is currently limited to two-dimensional imaging of microvessels, and therefore, it is necessary to propose a new microvessel imaging method based on ultrasonic phased array to realize three-dimensional imaging of microvessels. SUMMARY

[0004] The application provides a microvessel three-dimensional imaging method and system based on an ultrasonic phased array, and a medium, which are used for realizing three-dimensional imaging of microvessels.

[0005] To solve the above technical problems, the application discloses a microvessel three-dimensional imaging method based on an ultrasonic phased array in a first aspect, a two-dimensional ultrasonic phased array includes a plurality of ultrasonic array elements located on the same plane, and the method comprises the following steps.

[0006] In a preset continuous time, a spherical wave emission operation is performed multiple times, each time the spherical wave emission operation comprising: the two-dimensional ultrasonic phased array sequentially emits a spherical wave, after each spherical wave emission, all ultrasonic array elements simultaneously receive echo data, and after the coherent superposition of all echo data, ultrasonic image data is obtained, the ultrasonic image data at least including contrast agent echoes;

[0007] According to the ultrasonic image data obtained in each spherical wave emission operation, a three-dimensional ultrasonic sequence image is obtained.

[0008] segmenting the three-dimensional ultrasound sequence image into a plurality of spatio-temporal sub-blocks, each spatio-temporal sub-block comprising partial time dimension data and partial space dimension data of the three-dimensional ultrasound sequence image, and adjacent two spatio-temporal sub-blocks having partially overlapped time dimension data and space dimension data;

[0009] performing singular value decomposition on all the spatio-temporal sub-blocks, extracting contrast agent signals, and obtaining a three-dimensional contrast agent sequence image corresponding to the three-dimensional ultrasound sequence image according to the contrast agent signals;

[0010] performing binaryzation processing on the three-dimensional contrast agent sequence image to obtain a plurality of connected domains, and calculating a microbubble center point corresponding to each connected domain according to the connected domain;

[0011] performing microvessel three-dimensional imaging operation according to the microbubble center point.

[0012] As an optional implementation, in the first aspect of the present application, the two-dimensional ultrasound phased array comprises a plurality of sub-arrays located in the same plane, each sub-array comprising a plurality of ultrasound array elements located in the same plane, wherein at least one ultrasound array element in each sub-array is determined as a reference array element;

[0013] The two-dimensional ultrasound phased array emits spherical waves corresponding to different virtual point sources each time, and all the virtual point sources have a preset positional relationship so that echo data corresponding to all the virtual point sources can be focused on an imaging origin;

[0014] Moreover, the spherical wave emission operation each time comprises:

[0015] The two-dimensional ultrasound phased array emits spherical waves corresponding to different virtual point sources in sequence, and all the ultrasound array elements simultaneously receive echo data after each spherical wave emission to obtain an echo subset received by each sub-array at each virtual point source;

[0016] All the reference array elements in the two-dimensional ultrasound phased array emit spherical waves corresponding to the imaging origin, and all the reference array elements simultaneously receive echo data to obtain a superposition reference system;

[0017] The echo subsets received by all the sub-arrays at each virtual point source are coherently superimposed based on the superposition reference system to obtain ultrasound image data.

[0018] As an optional implementation, in the first aspect of the present application, the operation of emitting spherical waves by the two-dimensional ultrasound phased array each time comprises:

[0019] Each ultrasound array element on the two-dimensional ultrasound phased array emits an ultrasound wave at a respective preset time point, so that the ultrasound waves emitted by the two-dimensional ultrasound phased array are equivalent to spherical waves emitted from a virtual point source.

[0020] As an optional implementation, in the first aspect of the present application, the pixel point of each frame image in the three-dimensional ultrasound sequence image is , and the time dimension corresponding to the three-dimensional ultrasound sequence image is ;

[0021] Further, the segmentation of the three-dimensional ultrasound sequence image into a plurality of spatio-temporal sub-blocks comprises:

[0022] The segmentation of the three-dimensional ultrasound sequence image into a plurality of spatio-temporal sub-blocks, the spatial dimension data corresponding to any one spatio-temporal sub-block is , and the time dimension data corresponding to any one spatio-temporal sub-block is ; wherein , and the time dimension data and the spatial dimension data corresponding to adjacent two spatio-temporal sub-blocks have partial overlapping data.

[0023] As an optional implementation, in the first aspect of the present application, the three-dimensional imaging operation of the microbubble center point comprises:

[0024] Trajectory tracking is performed on the microbubble center point in each frame three-dimensional contrast agent image to obtain a plurality of three-dimensional microbubble trajectories corresponding to the three-dimensional contrast agent sequence image;

[0025] Superimposition of all the three-dimensional microbubble trajectories corresponding to the three-dimensional contrast agent sequence image is performed to obtain the three-dimensional distribution image of the microvessel.

[0026] As an optional implementation, in the first aspect of the present application, the trajectory tracking of the microbubble center point in each frame three-dimensional contrast agent image comprises:

[0027] A plurality of feature pixel points in each frame three-dimensional contrast agent image are identified, the feature pixel point being a pixel point located at the contour edge of the microvessel;

[0028] A predetermined number of adjacent pixel points around each feature pixel point are obtained, and for each feature pixel point, a feature pattern is fitted according to the feature pixel point and the adjacent pixel points corresponding to the feature pixel point;

[0029] For each feature pattern in each frame three-dimensional contrast agent image, a similar feature pattern corresponding to each feature pattern is determined in the adjacent frame three-dimensional contrast agent image corresponding to the frame three-dimensional contrast agent image;

[0030] According to the similar feature pattern corresponding to each feature pattern in all frames three-dimensional contrast agent images, the trajectory corresponding to each feature point is determined.

[0031] The second aspect of the present application discloses an ultrasonic imaging system, which comprises:

[0032] A two-dimensional ultrasonic phased array comprising a plurality of ultrasonic array elements located in the same plane, the two-dimensional ultrasonic phased array being used for:

[0033] Within a preset continuous time, a spherical wave transmission operation is performed multiple times, each time the spherical wave transmission operation comprising: the two-dimensional ultrasonic phased array sequentially emits spherical waves, after each spherical wave transmission, all ultrasonic array elements simultaneously receive echo data, and after the coherent superposition of all echo data, ultrasonic image data is obtained, the ultrasonic image data at least comprising contrast agent echoes;

[0034] A host computer, which is in communication connection with the two-dimensional ultrasonic phased array, and is used for:

[0035] According to the ultrasonic image data obtained by each spherical wave transmission operation, a three-dimensional ultrasonic sequence image is obtained;

[0036] The three-dimensional ultrasonic sequence image is divided into a plurality of spatio-temporal sub-blocks, each spatio-temporal sub-block comprising partial time dimension data and partial space dimension data of the three-dimensional ultrasonic sequence image, and there is partial overlapping data in the time dimension data and the space dimension data corresponding to adjacent two spatio-temporal sub-blocks;

[0037] Singular value decomposition is performed on all the spatio-temporal sub-blocks to extract contrast agent signals, and according to the contrast agent signals, a three-dimensional contrast agent sequence image corresponding to the three-dimensional ultrasonic sequence image is obtained;

[0038] The three-dimensional contrast agent sequence image is subjected to binaryzation processing to obtain a plurality of connected domains, and according to each connected domain, a microbubble center point corresponding to the connected domain is calculated;

[0039] According to the microbubble center point, a microvessel three-dimensional imaging operation is performed.

[0040] The third aspect of the present application discloses a microvessel three-dimensional imaging device based on an ultrasonic phased array, a two-dimensional ultrasonic phased array comprising a plurality of ultrasonic array elements located in the same plane, the device comprising:

[0041] A transmission module, configured to, within a preset continuous time, control the execution of a spherical wave transmission operation multiple times;

[0042] Each time the spherical wave transmission operation comprises: controlling the two-dimensional ultrasonic phased array to sequentially emit spherical waves, after each spherical wave transmission, all ultrasonic array elements simultaneously receive echo data, and after the coherent superposition of all echo data, ultrasonic image data is obtained, the ultrasonic image data at least comprising contrast agent echoes;

[0043] The signal receiving and acquisition module is used to obtain three-dimensional ultrasound sequence images based on the ultrasound image data obtained from each spherical wave transmission operation;

[0044] The data segmentation module is used to segment the three-dimensional ultrasound sequence image into multiple spatiotemporal sub-blocks. Each spatiotemporal sub-block includes part of the time dimension data and part of the spatial dimension data of the three-dimensional ultrasound sequence image, and there is partial overlap between the time dimension data and spatial dimension data of two adjacent spatiotemporal sub-blocks.

[0045] The singular value decomposition module is used to perform singular value decomposition on all the spatiotemporal sub-blocks, extract the contrast agent signal, and obtain the three-dimensional contrast agent sequence image corresponding to the three-dimensional ultrasound sequence image based on the contrast agent signal.

[0046] The center point localization module is used to perform binarization processing on the three-dimensional contrast agent sequence image to obtain multiple connected components, and calculate the microbubble center point corresponding to each connected component.

[0047] An imaging module is used to perform three-dimensional imaging of microvessels based on the center point of the microbubbles.

[0048] As an optional implementation, in a third aspect of the invention, the two-dimensional ultrasonic phased array includes a plurality of subarrays located on the same plane, each subarray including a plurality of ultrasonic array elements located on the same plane, wherein at least one ultrasonic array element on each subarray is determined as a reference array element;

[0049] Each spherical wave emitted by the two-dimensional ultrasonic phased array corresponds to a different virtual point source, and all virtual point sources have a preset positional relationship so that the echo data corresponding to all virtual point sources can be focused on the imaging origin.

[0050] And, each of the spherical wave emission operations includes:

[0051] The two-dimensional ultrasonic phased array is controlled to emit spherical waves corresponding to different virtual point sources in sequence. After each spherical wave is emitted, all ultrasonic array elements simultaneously receive echo data to obtain the echo subset received by each subarray at each virtual point source.

[0052] All reference elements on the two-dimensional ultrasonic phased array emit spherical waves corresponding to the imaging origin, and all reference elements simultaneously receive echo data to obtain a superimposed reference system.

[0053] Ultrasonic image data is obtained by coherently superimposing the echo subsets received by all subarrays at each virtual point source based on the superposition reference frame.

[0054] As an optional implementation, in a third aspect of the invention, the operation of the two-dimensional ultrasonic phased array emitting spherical waves each time includes:

[0055] Each ultrasonic element on the two-dimensional ultrasonic phased array emits ultrasonic waves at its own preset time point, so that the ultrasonic waves emitted by the two-dimensional ultrasonic phased array are equivalent to spherical waves emitted from a virtual point source.

[0056] As an optional implementation, in a third aspect of the invention, the number of pixels in each frame of the three-dimensional ultrasound sequence image is... The time dimension corresponding to the three-dimensional ultrasound sequence image is ;

[0057] Furthermore, the specific operation method by which the data segmentation module segments the three-dimensional ultrasound sequence image into multiple spatiotemporal sub-blocks includes:

[0058] The three-dimensional ultrasound sequence image is segmented into multiple spatiotemporal sub-blocks, and the spatial dimension data corresponding to any one spatiotemporal sub-block is: The time dimension data corresponding to any spatiotemporal sub-block is ;in Furthermore, there is some overlap between the time dimension data and the spatial dimension data of two adjacent spatiotemporal sub-blocks.

[0059] As an optional implementation, in a third aspect of the present invention, the imaging module performs three-dimensional microvascular imaging based on the microbubble center point, including:

[0060] Trajectory tracking is performed on the center point of microbubbles in each frame of the 3D contrast agent image to obtain multiple 3D microbubble trajectories corresponding to the 3D contrast agent sequence images;

[0061] By superimposing all the three-dimensional microbubble trajectories corresponding to the three-dimensional contrast agent sequence image, the three-dimensional distribution image of the microvessels is obtained.

[0062] As an optional implementation, in a third aspect of the present invention, the specific operation method of the imaging module for tracking the trajectory of the microbubble center point in each frame of the three-dimensional contrast agent image includes:

[0063] Identify multiple feature pixels in each frame of the 3D contrast agent image, wherein the feature pixels are pixels located at the edge of the microvascular contour;

[0064] Obtain a preset number of neighboring pixels around each feature pixel. For each feature pixel, fit a feature image based on the feature pixel and its corresponding neighboring pixels.

[0065] For each feature graphic in each frame of the 3D contrast agent image, the similar feature graphic corresponding to each feature graphic is determined in the adjacent frames of the 3D contrast agent image corresponding to that frame.

[0066] Based on the similar feature patterns corresponding to each feature pattern in all frames of 3D contrast agent images, the trajectory corresponding to each feature point is determined.

[0067] A fourth aspect of this invention discloses a three-dimensional microvascular imaging system based on an ultrasound phased array, the system comprising:

[0068] Memory containing executable program code;

[0069] A processor coupled to the memory;

[0070] The processor calls the executable program code stored in the memory to execute the microvascular three-dimensional imaging method based on ultrasound phased array disclosed in the first aspect of the present invention.

[0071] The fifth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute the microvascular three-dimensional imaging method based on ultrasound phased array disclosed in the first aspect of the present invention.

[0072] Compared with existing technologies, the microvascular three-dimensional imaging method based on ultrasound phased array of the present invention can obtain a three-dimensional ultrasound sequence image from ultrasound image data obtained by multiple spherical wave emission operations. Then, the three-dimensional ultrasound sequence image is segmented into multiple spatiotemporal sub-blocks, and singular value decomposition is performed on all spatiotemporal sub-blocks to extract contrast agent signals. Based on the contrast agent signals, a three-dimensional contrast agent sequence image corresponding to the three-dimensional ultrasound sequence image is obtained. The three-dimensional contrast agent sequence image is binarized to obtain multiple connected components. The microbubble center point corresponding to each connected component is calculated, and finally, microvascular three-dimensional imaging is performed based on the microbubble center points. The scheme of superimposing multiple three-dimensional spherical waves improves the detection breadth without sacrificing detection accuracy. The design of each spatiotemporal sub-block reduces the influence of large blood vessels and physiological movements on the data. The overlapping data of adjacent spatiotemporal sub-blocks reflects the continuity of blood flow, preserving the continuity and intrinsic relationship of the ultrasound sequence image. The binarization of the three-dimensional contrast agent sequence image reduces computation and improves efficiency. Ultimately, the embodiments of the present invention can efficiently and accurately achieve three-dimensional imaging of microvessels. Attached Figure Description

[0073] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0074] Figure 1 This is a schematic flowchart of a microvascular three-dimensional imaging method based on ultrasound phased array disclosed in an embodiment of the present invention;

[0075] Figure 2 This is a schematic diagram illustrating different detection ranges corresponding to different point sources, using two dimensions as an example, as disclosed in the embodiments of the present invention.

[0076] Figure 3 This is a schematic diagram of a two-dimensional phased array disclosed in an embodiment of the present invention;

[0077] Figure 4 This is a schematic diagram of the detection range of a spherical wave emitted by a single virtual point source as disclosed in an embodiment of the present invention;

[0078] Figure 5 This is a spatial relationship diagram between a single point source and a two-dimensional ultrasonic phased array disclosed in an embodiment of the present invention;

[0079] Figure 6 This is a schematic diagram of the use of a two-dimensional miniature ultrasound phased array to detect myocardium, as disclosed in an embodiment of the present invention;

[0080] Figure 7 This is a schematic diagram of the structure of a microvascular three-dimensional imaging system based on an ultrasound phased array disclosed in an embodiment of the present invention. Detailed Implementation

[0081] To enable those skilled in the art to better understand the present invention, 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.

[0082] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention 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, apparatus, product, or end 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 ends.

[0083] 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 the invention. 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.

[0084] Example 1

[0085] Please see Figure 1 , Figure 1 This is a schematic flowchart of a microvascular three-dimensional imaging method based on ultrasound phased array disclosed in an embodiment of the present invention. Wherein, Figure 1 The described three-dimensional microvascular imaging method based on ultrasound phased array can be integrated into a certain three-dimensional imaging device, which can in turn be integrated into a cloud server or a local server. Figure 1 The described microvascular three-dimensional imaging method based on ultrasound phased array can also be integrated into a host computer that communicates with a two-dimensional ultrasound phased array. For example... Figure 1 As shown, this three-dimensional microvascular imaging method based on ultrasound phased array can include the following operations:

[0086] Step 101: Perform the spherical wave emission operation multiple times within a preset continuous time period.

[0087] In this embodiment of the invention, each spherical wave emission operation may include: a two-dimensional ultrasonic phased array sequentially emitting spherical waves; after each spherical wave emission, all ultrasonic array elements simultaneously receive echo data; and after coherently superimposing all echo data, ultrasonic image data is obtained. The ultrasonic image data may include at least contrast agent echoes.

[0088] In this embodiment of the invention, a two-dimensional ultrasonic phased array may include multiple ultrasonic elements located on the same plane. When emitting ultrasonic waves, if all ultrasonic elements in the two-dimensional ultrasonic phased array emit ultrasonic waves simultaneously, the ultrasonic waves emitted by the entire ultrasonic phased array are three-dimensional plane waves, with their wavefronts spreading in a planar manner. If each ultrasonic element in the two-dimensional ultrasonic phased array emits ultrasonic waves at different preset times, different types of ultrasonic waves can be emitted, such as three-dimensional plane waves propagating obliquely at a certain angle, or three-dimensional spherical waves with a point source. In this embodiment of the invention, each ultrasonic element in the two-dimensional ultrasonic phased array is selected and controlled to emit ultrasonic waves at different preset times, thereby emitting three-dimensional spherical waves with a point source. Spherical waves have the advantages of a wide detection range and the ability to be focused.

[0089] While three-dimensional spherical wave detection increases the detection range, it also reduces accuracy. To address this issue, this application proposes a scheme that coherently superimposes multiple three-dimensional spherical waves, thus improving detection range without sacrificing accuracy. For example, each spherical wave corresponds to a specific centerline direction. A two-dimensional ultrasonic phased array can sequentially emit spherical waves along different centerline directions, thereby covering a larger detection space while maintaining a certain detection density, enabling refined detection. Figure 2 As shown, taking a two-dimensional spherical wave emitted by a one-dimensional ultrasonic phased array as an example, the spherical wave corresponding to the first point source and the spherical wave corresponding to the second point source each detect different directions. The combination of the two corresponds to a wider detection direction. The same conclusion is reached when the above principle is applied to a three-dimensional spherical wave emitted by a two-dimensional ultrasonic phased array.

[0090] Step 102: Obtain three-dimensional ultrasound sequence images based on the ultrasound image data obtained from each spherical wave emission operation.

[0091] By performing spherical wave emission operations multiple times within a preset continuous time period, several consecutive three-dimensional ultrasound images of the contrast agent flowing with the blood can be obtained. These consecutive three-dimensional ultrasound images constitute a three-dimensional ultrasound sequence image. This sequence image not only contains the ultrasound echo of the contrast agent, but also information about the contrast agent flowing with the blood, thus providing more reliable data support for subsequent processing.

[0092] Step 103: Divide the three-dimensional ultrasound sequence image into multiple spatiotemporal sub-blocks.

[0093] In this optional embodiment, each spatiotemporal sub-block may include partial temporal dimension data and partial spatial dimension data of a three-dimensional ultrasound sequence image, and there is partial overlap between the temporal dimension data and spatial dimension data of two adjacent spatiotemporal sub-blocks.

[0094] Because tissue and contrast agent signals have different spatiotemporal coherence, tissue corresponds to a high spatiotemporal coherence signal in the image sequence, while contrast agent signals correspond to a low spatiotemporal coherence signal. Therefore, singular value decomposition can be used to separate them.

[0095] In ultrasound imaging, tissues and flowing contrast agents (such as microbubbles) exhibit different spatiotemporal properties in image sequences:

[0096] Organization: Due to the relatively small movement of the structure, it changes little between consecutive image frames, exhibiting high spatiotemporal coherence (strong correlation in time and space).

[0097] Contrast agent: As blood flows rapidly, its position and intensity change dramatically over time, exhibiting low spatiotemporal coherence (weak correlation in time and space).

[0098] This difference provides a basis for using mathematical tools (such as singular value decomposition) to separate the two types of signals.

[0099] In this embodiment of the invention, if singular value decomposition is directly performed on the ultrasound sequence image, it is easily affected by large blood vessels and physiological movements, which may result in the loss of some microvascular information. To solve this problem, this embodiment of the invention divides the three-dimensional ultrasound sequence image into multiple spatiotemporal sub-blocks. Each spatiotemporal sub-block may include part of the temporal dimension data and part of the spatial dimension data of the three-dimensional ultrasound sequence image, and there is partial overlap in the temporal and spatial dimension data of two adjacent spatiotemporal sub-blocks. After segmentation, the influence of large blood vessels and physiological movements on the data can be reduced. The overlap in data between adjacent spatiotemporal sub-blocks can reflect the continuity of blood flow, thus preserving the continuity and intrinsic relationship of the ultrasound sequence image.

[0100] Step 104: Perform singular value decomposition on all spatiotemporal sub-blocks, extract contrast agent signals, and obtain the three-dimensional contrast agent sequence image corresponding to the three-dimensional ultrasound sequence image based on the contrast agent signals.

[0101] Singular Value Decomposition (SVD) is a linear algebraic method that decomposes a matrix into three submatrices: (left singular vector): representing spatial features (such as pixel position relationships); (singular value matrix): a diagonal matrix where singular values ​​are arranged in descending order, representing the principal component energy of the data; and (right singular vector): representing temporal features (such as inter-frame variations). Its core idea is to extract the primary patterns (corresponding to large singular values) and secondary patterns (corresponding to small singular values) from the data using SVD, thereby separating signals with different levels of coherence.

[0102] In this embodiment of the invention, a high-coherence signal (tissue) exhibits slow changes or periodic patterns in a time series, with energy concentrated in a few principal components (large singular values). A low-coherence signal (contrast agent) manifests as rapid random motion, with energy dispersed among multiple small singular values. The components corresponding to the singular vectors exhibit high-frequency, unstructured characteristics.

[0103] Therefore, this embodiment of the invention can perform singular value decomposition on all spatiotemporal sub-blocks to extract the three-dimensional contrast agent signal, and then obtain the three-dimensional contrast agent sequence image corresponding to the three-dimensional ultrasound sequence image based on the three-dimensional contrast agent signal. The three-dimensional ultrasound sequence image is composed of multiple three-dimensional ultrasound images in consecutive frames, which includes the echo situation of the contrast agent in the microvessels within a preset continuous time. In steps 103-104, the echo situation of the contrast agent in the microvessels within the preset continuous time is extracted from the three-dimensional ultrasound sequence image through a series of methods to obtain the three-dimensional contrast agent sequence image.

[0104] Step 105: Binarize the three-dimensional contrast agent sequence image to obtain multiple connected components, and calculate the microbubble center point corresponding to each connected component.

[0105] In this embodiment of the invention, the three-dimensional contrast agent sequence image is binarized, with pixel values ​​being only 0 or 1, thereby separately displaying the microbubble signal in the contrast agent. Because the echo signal of the contrast agent microbubble is much stronger than the echo of blood vessels and other tissues, the echo of the contrast agent microbubble appears as bright spots, which are ideal spatial scattering points in the three-dimensional ultrasound image. The diameter of the microbubble is 1-3 micrometers, which can be regarded as ideal scattering points in the imaging space, and its ultrasound signal can be considered as the system point spread function, generally considered to be a three-dimensional Gaussian model.

[0106] Localization of the contrast agent center point is the core of the imaging algorithm. Current methods based on Gaussian point spread function models and radial symmetry require numerous image frames, resulting in long acquisition and post-processing times. In this embodiment, however, the binarization process can begin by setting a threshold. Pixels with values ​​greater than this threshold are uniformly set to 1, and those with values ​​less than the threshold are uniformly set to 0. This allows for the extraction of multiple connected components from the 3D contrast agent sequence image. Each connected component corresponds to a 3D microbubble echo. Based on these connected components, the microbubble center point is determined; for example, the centroid of the connected component can be calculated using a centroid algorithm. This embodiment of the invention reduces computational load and improves efficiency.

[0107] Step 106: Perform three-dimensional imaging of microvessels based on the center point of the microbubble.

[0108] In this embodiment of the invention, the distribution of microbubbles can indirectly represent the distribution of microvessels; a three-dimensional microbubble distribution represents a three-dimensional microvessel distribution. For example, all microbubble center points can be fitted and superimposed to obtain the three-dimensional distribution of microvessels. Based on this, the velocity of the microbubble center points can be calculated to obtain the blood flow velocity.

[0109] As can be seen, the microvascular three-dimensional imaging method based on ultrasound phased array in this embodiment of the invention can obtain a three-dimensional ultrasound sequence image from ultrasound image data obtained by multiple spherical wave emission operations. Then, the three-dimensional ultrasound sequence image is divided into multiple spatiotemporal sub-blocks, and singular value decomposition is performed on all spatiotemporal sub-blocks to extract contrast agent signals. Based on the contrast agent signals, a three-dimensional contrast agent sequence image corresponding to the three-dimensional ultrasound sequence image is obtained. The three-dimensional contrast agent sequence image is binarized to obtain multiple connected components. The microbubble center point corresponding to each connected component is calculated, and finally, a microvascular three-dimensional imaging operation is performed based on the microbubble center points. The imaging scheme of coherently superimposing multiple three-dimensional spherical waves improves the detection breadth without sacrificing detection accuracy. The design of each spatiotemporal sub-block reduces the influence of large blood vessels and physiological movements on microvascular detection. The overlapping data between adjacent spatiotemporal sub-blocks effectively reduces the interference of random noise signals on vascular imaging. The binarization of the three-dimensional contrast agent sequence image reduces computational load and improves efficiency. Ultimately, this embodiment of the invention can efficiently and accurately achieve three-dimensional imaging of microvessels.

[0110] In this embodiment of the invention, obtaining ultrasound image data by coherently superimposing all echo data will result in a certain amount of computation. For example, when coherently superimposing echo data corresponding to the left front, right front, upper front, and lower front directions of the four beam center lines, the ultrasound image data at the overlapping points need to be identified first, and then the data is combined based on the coherence relationship.

[0111] To address the technical problem of high computational complexity in the coherent superposition process, in one optional embodiment, the two-dimensional ultrasonic phased array may include multiple subarrays located on the same plane, and each subarray may include multiple ultrasonic array elements located on the same plane, wherein at least one ultrasonic array element on each subarray is determined as a reference array element.

[0112] In this optional embodiment, examples are given as follows: Figure 3 As shown, for some special application scenarios, such as the special miniature two-dimensional phased array for intracardiac ultrasound imaging, a subarray scheme can be adopted. Specifically, for an 8×8 two-dimensional ultrasound phased array, it is divided into 4 subarrays a, b, c, and d. Each subarray includes 4×4 ultrasound array elements. Each subarray has at least one reference array element. For example, the reference array elements can be four centrally symmetric ultrasound array elements A, B, C, and D.

[0113] In this optional embodiment, the spherical wave emitted by the two-dimensional ultrasonic phased array corresponds to a different virtual point source each time. For a single virtual point source, the detection range of its emitted spherical wave is as follows: Figure 4 As shown, all virtual point sources have a preset positional relationship so that the echo data corresponding to all virtual point sources can be focused on the imaging origin. Optionally, the two-dimensional ultrasound phased array emits four spherical waves with centerline directions corresponding to the left front, right front, upper front, and lower front, respectively. These four spherical wave pairs can be focused on the same imaging origin, that is, the centerlines of the four beam directions corresponding to the left front, right front, upper front, and lower front can intersect at a point.

[0114] Furthermore, each spherical wave emission operation may include:

[0115] A two-dimensional ultrasonic phased array sequentially emits spherical waves corresponding to different virtual point sources. After each spherical wave is emitted, all ultrasonic array elements simultaneously receive echo data, thus obtaining the echo subset received by each subarray at each virtual point source.

[0116] In a two-dimensional ultrasound phased array, all reference elements emit spherical waves corresponding to the imaging origin, and all reference elements simultaneously receive echo data, resulting in a superimposed reference system. During each spherical wave emission operation, not only is a full-scale spherical wave emission operation performed, but also a detection operation based solely on the reference elements. For the detection operation of the reference elements, the received echo data is used only as a reference and not for actual microvascular imaging. Specifically, all reference elements in the two-dimensional ultrasound phased array emit spherical waves corresponding to the imaging origin, and all reference elements simultaneously receive echo data. A simple combination of all these echo data yields a superimposed reference system. This superimposed reference system is actually a wide-range but insufficiently precise echo data set. By filling in the more precise echo data from each subarray, a wide-range and high-precision echo data set can be obtained.

[0117] The ultrasound image data is obtained by coherently stacking the subsets of echoes received by all subarrays at each virtual point source, based on a stacking reference frame. Since this coherent stacking operation is performed with the reference of the stacking reference frame, the computational cost of coherent stacking is greatly reduced.

[0118] The above solutions are illustrated with examples as follows:

[0119] like Figure 3 As shown, all elements on a two-dimensional ultrasonic phased array sequentially emit spherical waves corresponding to different virtual point sources. After each spherical wave emission, all ultrasonic elements simultaneously receive echo data, obtaining the subset of echoes received by each subarray at each virtual point source:

[0120] Set a = {left front echo data; right front echo data; top front echo data; bottom front echo data}

[0121] Set b = {left front echo data; right front echo data; top front echo data; bottom front echo data}

[0122] Set c = {left front echo data; right front echo data; top front echo data; bottom front echo data}

[0123] Set d = {left front echo data; right front echo data; top front echo data; bottom front echo data}

[0124] All reference elements A, B, C, and D on the two-dimensional ultrasonic phased array emit spherical waves corresponding to the imaging origin, and all reference elements simultaneously receive echo data to obtain a superimposed reference frame P.

[0125] Ultrasound image data are obtained by coherently superimposing sets a, b, c, and p based on reference frame P.

[0126] As can be seen, this optional embodiment, by designing subarrays and reference elements on the subarrays, can use the superposition reference frame obtained by the reference elements to assist the coherent superposition operation of echo data during the coherent superposition process, thereby reducing the computational load of coherent superposition and improving the efficiency of coherent superposition.

[0127] In yet another alternative embodiment, the operation of the two-dimensional ultrasonic phased array emitting spherical waves each time may include:

[0128] Each ultrasonic element on a two-dimensional ultrasonic phased array emits ultrasonic waves at its own preset time point, so that the ultrasonic waves emitted by the ultrasonic phased array are equivalent to spherical waves emitted from a virtual point source.

[0129] In this optional embodiment, spherical waves can be emitted by a time-delayed transmission method. It is known that sound waves emitted from the same point will experience a delay when they reach a certain plane. Utilizing this principle, spherical waves can be simulated by setting different arrays to emit ultrasonic waves with different time delays.

[0130] Optionally, several virtual point sources (spherical wave sources) can be placed behind the two-dimensional ultrasound array. Each virtual point source can be used as a sound source. A schematic diagram of a single virtual point source is shown below. Figure 5 As shown, the emission of spherical waves can be controlled by setting the element delay. The delay of the i-th element can be expressed by the following formula.

[0131]

[0132]

[0133] Wherein, the coordinates of the virtual point source are The coordinates of the i-th transducer element are d represents the distance from the surface of the virtual point source transducer. Let be the distance from the current array element to the virtual point source, and c be the speed of sound in the specific medium. This is the delay for the i-th array element.

[0134] The aforementioned delayed emission method enables efficient and simple emission of spherical waves from a two-dimensional ultrasonic phased array.

[0135] In yet another optional embodiment, the number of pixels in each frame of the three-dimensional ultrasound sequence image is 100. That is, corresponding to the spatial dimension, the time dimension corresponding to the ultrasound sequence image is ;

[0136] Furthermore, segmenting a three-dimensional ultrasound sequence image into multiple spatiotemporal sub-blocks may include:

[0137] The three-dimensional ultrasound sequence image is segmented into multiple spatiotemporal sub-blocks, and the spatial dimension data corresponding to any one spatiotemporal sub-block is: The time dimension data corresponding to any spatiotemporal sub-block is ;in Furthermore, there is some overlap between the time dimension data and the spatial dimension data of two adjacent spatiotemporal sub-blocks.

[0138] In this embodiment of the invention, the contrast agent is fluid, and within the time dimension of ultrasound detection, the flowing contrast agent remains at almost the same position in the ultrasound image. Therefore, this application employs a block-based operation, where the contrast agent signal corresponding to the same part of the microvessels is processed only in a small block, and there is an overlap between the time dimension data and the spatial dimension data between two consecutive spatiotemporal sub-blocks, thereby preserving the inherent connection of information.

[0139] In yet another optional embodiment, performing three-dimensional microvascular imaging based on the microbubble center point may include:

[0140] Trajectory tracking is performed on the center point of microbubbles in each frame of the 3D contrast agent image to obtain multiple 3D microbubble trajectories corresponding to the 3D contrast agent sequence images;

[0141] By superimposing all the three-dimensional microbubble trajectories corresponding to the three-dimensional contrast agent sequence images, a three-dimensional distribution image of microvessels is obtained.

[0142] In this optional embodiment, the Kuhn-Munkres allocation algorithm can be used to pair the closest contrast signals between two adjacent frames. This algorithm uses the distance between each microbubble in two adjacent frames as a weight, and pairs microbubbles in two adjacent frames based on the idea of ​​bipartite graphs to minimize the total weight. Finally, by setting a threshold, matches with excessive distance or zero distance are removed to achieve the tracking of contrast signals in three-dimensional space.

[0143] In this optional embodiment, further optionally, tracing the trajectory of the microbubble center point in each frame of the three-dimensional contrast agent image may include:

[0144] Multiple feature pixels in each frame of the 3D contrast agent image are identified. Feature pixels are those located at the edge of the microvessel contour. Because there are fewer pixels at the contour edge and they are more recognizable, selecting pixels at the contour edge can select feature points more efficiently.

[0145] A preset number of neighboring pixels are acquired around each feature pixel. For each feature pixel, a feature shape is fitted based on the feature pixel and its corresponding neighboring pixels. This optional embodiment does not search for the nearest pixel in the next frame as a matching point for a single pixel, but rather combines multiple pixels around the feature pixel to achieve matching of multiple pixels. Because the interval between two adjacent frames is short, the combination formed by multiple pixels has not been destroyed or deformed during the flow of the contrast agent. Therefore, matching the shape formed by multiple pixels (which can be a three-dimensional shape or a two-dimensional shape obtained by cross-section from a certain angle) is more accurate than matching a single pixel.

[0146] For each feature pattern in each frame of the 3D contrast agent image, similar feature patterns corresponding to each feature pattern are identified in the adjacent frames of the 3D contrast agent image corresponding to that frame. Generally, in two frames within a short time interval, the feature patterns change little with the flow of the contrast agent and can be easily identified.

[0147] Based on the similar feature patterns corresponding to each feature pattern in all frames of 3D contrast agent images, the trajectory corresponding to each feature point is determined.

[0148] As can be seen, this optional embodiment can achieve more accurate microbubble center point trajectory tracking, thereby achieving more accurate microvascular three-dimensional imaging.

[0149] Example 2

[0150] Based on Embodiment 1, this invention specifically discloses a method for realizing four-dimensional anatomical structure and rapid three-dimensional functional imaging of myocardial microcirculation in the cardiac chamber using an interventional ultrasound catheter. This invention aims to provide effective clinical diagnosis and postoperative monitoring solutions for the diagnosis of structural heart disease and cardiac electrophysiological surgery, as well as myocardial microcirculation-related diseases, such as non-coronary artery occlusive coronary artery disease, heart transplantation, and post-myocardial infarction reflow assessment.

[0151] Currently, commonly used clinical imaging methods for the circulatory system include MRA, DSA, and Doppler ultrasound imaging. However, visualizing microcirculation remains a significant challenge due to the constraints between the physical diffraction limit and clinically permissible radiation doses. Ultrasound localization microscopy (ULM) fills this technological gap, but existing ULM requires reconstructing the microbubble center point trajectory from multiple ultrasound image sequences and superimposing all trajectories to obtain a microcirculation distribution map. This requires a long post-processing time, and since the heart is constantly beating, acquiring tens of thousands of high spatiotemporal coherence images requires a long acquisition time. Furthermore, the heart is surrounded by the sternum at considerable depth, necessitating a trade-off between imaging depth and resolution. Generally, lower frequencies are used, making it more difficult to capture flowing contrast agent signals.

[0152] Subsequently, two-dimensional ultrasound images show the cross-sectional structure of tissues and cannot provide a holistic view of blood flow throughout the tissue. Due to their high planar dependence, if a suitable tissue cross-section cannot be obtained, there may be missed diagnoses in clinical applications. In addition, because organisms exhibit physiological motion, if the direction of motion deviates from the imaging plane, conventional motion correction algorithms are difficult to correct, thus introducing additional artifacts into the super-resolution image.

[0153] Based on the shortcomings of existing technologies, this invention innovatively proposes a three-dimensional imaging method for myocardial microcirculation based on a two-dimensional micro-ultrasound phased array. First, a two-dimensional micro-ultrasound phased array combined with subarray beamforming technology is used to compress the number of channels and echo data, thereby achieving 4D ultrasound imaging. Then, multiple frames of three-dimensional myocardial ultrasound image sequences containing microbubbles are acquired. Next, the innovative three-dimensional weighted average microbubble localization method proposed in this invention is used, which can effectively shorten post-processing time. Furthermore, compared to other microbubble center point localization methods, the three-dimensional weighted localization algorithm can reconstruct results similar to other methods using fewer frames, thus greatly reducing data acquisition time. Then, the located microbubble center points are tracked inter-frame to obtain several microbubble motion trajectories. Finally, all microbubble trajectories are superimposed to obtain a three-dimensional myocardial microcirculation image.

[0154] The detailed disclosure of the technical solution of the embodiments of the present invention is as follows:

[0155] (1) Data collection

[0156] In this embodiment of the invention, the two-dimensional ultrasonic phased array divides all array elements into several (N) subarrays. The beamforming delay is decomposed into coarse delay and fine delay. Within the subarray, the fine delay is used for precise calculation, and several echo data are combined into a set of data. Then, the coarse delay is used for beamforming of the entire imaging area. This reduces the number of probe cables and the amount of beamforming data by N times, greatly improving the imaging frame rate.

[0157] While focused-line scanning can improve the imaging frame rate, further improvements are needed due to the rapid movement of microbubbles within the vascular network. This invention utilizes spatial spherical wave coherent composite imaging, transmitting and receiving multiple sets of spherical wave and echo data, and coherently superimposing them to achieve a trade-off between imaging frame rate and signal-to-noise ratio. Specifically, several virtual point sources (spherical wave sources) can be positioned behind a two-dimensional ultrasound array. Each virtual point source can act as a sound source, and the spherical wave transmission can be controlled by setting the array element delay. Figure 5 As shown, the delay of the i-th array element can be expressed by the following formula.

[0158]

[0159]

[0160] Wherein, the coordinates of the virtual point source are The coordinates of the i-th transducer element are d represents the distance from the surface of the virtual point source transducer. Let be the distance from the current array element to the virtual point source, and c be the speed of sound in the specific medium. This is the delay for the i-th array element.

[0161] Due to the large amount of echo data in three-dimensional ultrasound imaging, and the limitations of the data transmission and storage capacity of the ultrasound imaging system, it is optional to collect 200 frames of coherent composite echo data per group, for a total of 50 groups.

[0162] A schematic diagram of the two-dimensional miniature ultrasound phased array used to detect myocardial tissue in an embodiment of the present invention is shown below. Figure 6 As shown.

[0163] (2) Trivial Value Decomposition

[0164] Since the spatiotemporal coherence of tissue and flow microbubble signals differs, the wall filtering algorithm can be used to remove tissue signals. Traditional singular value decomposition algorithms calculate global singular values, and the spatiotemporal coherence corresponding to larger blood vessels is extremely low, which can easily overwhelm microvascular signals.

[0165] Therefore, this embodiment of the invention utilizes a three-dimensional block singular value decomposition wall filtering algorithm to effectively extract microvascular signals. The dimension of the image sequence is... The entire time-series image is divided into several segments of size [missing information]. The sub-blocks, in which Furthermore, there is overlap between the sub-blocks. A three-dimensional matrix is ​​reconstructed for each sub-block, with the first dimension being spatial and the other being temporal. Singular value decomposition is then performed on this two-dimensional matrix using the following formula:

[0166]

[0167] in It is the two-dimensional decomposition matrix of the nth sub-block, where U and V are respectively A matrix composed of singular value vectors in space and time. It is a singular value matrix. Represents the conjugate transpose of a matrix. High-order thresholds for singular value truncation. and low-order threshold Based on the second derivative of the singular value curve, the point with the largest change in curvature is considered the critical point between blood scattering and tissue scattering. After performing singular value decomposition on each sub-block, the microbubble signal of each pixel in the image can be calculated. :

[0168]

[0169] Where N is the number of target pixels. The total number of overlapping sub-blocks, It is the microbubble signal of the nth sub-block. These are the singular values ​​corresponding to the truncation threshold. Embodiments of this invention can effectively reduce the influence of extremely low spatiotemporal coherence signals (such as physiological motion and large-diameter blood vessels) on microvascular signals, thereby improving microvascular imaging and detection rates.

[0170] (3) Microbubble center point positioning

[0171] Microbubble center point localization is the core of imaging algorithms. Currently, the most commonly used methods are the Gaussian model based on point spread function and the localization method based on radial symmetry. These methods require a long time to acquire multiple image sequences and a long post-processing time, which greatly limits their clinical application.

[0172] This invention proposes a novel three-dimensional microbubble center localization method: a three-dimensional weighted average center point localization method. Since the diameter of a microbubble is 1-3 micrometers, it can be considered an ideal scattering point in the imaging space, and its ultrasonic signal can be considered as the system's point spread function, generally considered a three-dimensional Gaussian model. First, a threshold is selected to binarize the microbubble signal image. Because multiple microbubble signals may overlap into a connected region, only one centroid can be detected, potentially leading to a large localization error. Therefore, morphological erosion and dilation are first performed on the microbubbles to separate irregularly shaped connected regions, thus allowing the detection of multiple microbubble signal centroids. After binarization, erosion, and dilation, the region area is calculated. Then, a weighted average is calculated between the original image pixel values ​​and the region area to obtain the weighted centroid of the microbubble center point. This yields the reduced set of points within the microbubble contour. Its center of mass The following methods can be used to calculate:

[0173]

[0174] Where N is the number of points. for The normalized pixel value.

[0175] (4) Microbubble center point tracking and particle velocity measurement

[0176] After extracting the center coordinates of the microbubble signal in each frame, the Kuhn-Munkres allocation algorithm can be used to pair the nearest contrast signals between adjacent frames. This algorithm uses the distance between each microbubble in adjacent frames as a weight, and based on the idea of ​​a bipartite graph, pairs microbubbles in adjacent frames to minimize the total weight. Finally, a threshold is set to remove matches with excessively large distances or zero distances, thereby achieving the tracking of contrast signals in three-dimensional space. Furthermore, the velocity of the i-th microbubble can be measured sequentially as follows:

[0177]

[0178] in It is the reciprocal of the frame rate. This refers to the velocity of the microbubbles in the first frame. Using the method described in this embodiment of the invention, the blood flow velocity within biological tissue can be obtained. This method has the advantages of fast operation, ability to monitor hemodynamics, and simple algorithm that is easy to implement.

[0179] Example 3

[0180] This invention discloses an ultrasound imaging system, which may include:

[0181] A two-dimensional ultrasonic phased array can include multiple ultrasonic array elements located on the same plane. Two-dimensional ultrasonic phased arrays are used for:

[0182] Within a preset continuous time period, spherical wave emission operations are performed multiple times. Each spherical wave emission operation may include: a two-dimensional ultrasonic phased array sequentially emitting spherical waves; after each spherical wave emission, all ultrasonic array elements simultaneously receive echo data; and after coherently superimposing all echo data, ultrasonic image data is obtained. The ultrasonic image data may include at least contrast agent echoes.

[0183] The host computer, which communicates with the two-dimensional ultrasonic phased array, is used for:

[0184] Based on the ultrasound image data obtained from each spherical wave emission operation, a three-dimensional ultrasound sequence image is obtained;

[0185] The three-dimensional ultrasound sequence image is segmented into multiple spatiotemporal sub-blocks. Each spatiotemporal sub-block can include part of the time dimension data and part of the spatial dimension data of the three-dimensional ultrasound sequence image, and there is partial overlap between the time dimension data and spatial dimension data of two adjacent spatiotemporal sub-blocks.

[0186] Singular value decomposition is performed on all spatiotemporal sub-blocks to extract contrast agent signals. Based on the contrast agent signals, three-dimensional contrast agent sequence images corresponding to three-dimensional ultrasound sequence images are obtained.

[0187] The three-dimensional contrast agent sequence image is binarized to obtain multiple connected components. The microbubble center point corresponding to each connected component is calculated.

[0188] Three-dimensional imaging of microvessels is performed based on the center point of the microbubble.

[0189] The host computer can also be used to implement other steps disclosed in Embodiments 1 and 2 of the present invention to achieve the final three-dimensional imaging of microvessels.

[0190] Example 4

[0191] This invention discloses a microvascular three-dimensional imaging device based on an ultrasound phased array. This device can communicate with a two-dimensional ultrasound phased array, which may include multiple ultrasound array elements located on the same plane. The device may include:

[0192] The transmitting module is used to control and execute spherical wave transmission operations multiple times within a preset continuous time period;

[0193] Each spherical wave transmission operation may include: controlling a two-dimensional ultrasonic phased array to sequentially emit spherical waves; after each spherical wave transmission, all ultrasonic array elements simultaneously receive echo data; and after coherently superimposing all echo data, ultrasonic image data is obtained. The ultrasonic image data may include at least contrast agent echoes.

[0194] The signal receiving and acquisition module is used to obtain three-dimensional ultrasound sequence images based on the ultrasound image data obtained from each spherical wave transmission operation;

[0195] The data segmentation module is used to segment three-dimensional ultrasound sequence images into multiple spatiotemporal sub-blocks. Each spatiotemporal sub-block can include part of the time dimension data and part of the spatial dimension data of the three-dimensional ultrasound sequence image, and there is partial overlap between the time dimension data and spatial dimension data of two adjacent spatiotemporal sub-blocks.

[0196] The singular value decomposition module is used to perform singular value decomposition on all spatiotemporal sub-blocks, extract contrast agent signals, and obtain three-dimensional contrast agent sequence images corresponding to three-dimensional ultrasound sequence images based on the contrast agent signals.

[0197] The center point localization module is used to binarize the three-dimensional contrast agent sequence image to obtain multiple connected components, and calculate the microbubble center point corresponding to each connected component.

[0198] The imaging module is used for three-dimensional imaging of microvessels based on the center point of the microbubble.

[0199] In an optional embodiment, the two-dimensional ultrasonic phased array may include multiple subarrays located on the same plane, each subarray may include multiple ultrasonic array elements located on the same plane, wherein at least one ultrasonic array element on each subarray is determined as a reference array element;

[0200] Each spherical wave emitted by the two-dimensional ultrasonic phased array corresponds to a different virtual point source. All virtual point sources have a preset positional relationship so that the echo data corresponding to all virtual point sources can be focused on the imaging origin.

[0201] Furthermore, each spherical wave emission operation may include:

[0202] The two-dimensional ultrasonic phased array is controlled to emit spherical waves corresponding to different virtual point sources in sequence. After each spherical wave is emitted, all ultrasonic array elements simultaneously receive echo data, and the subset of echoes received by each subarray at each virtual point source is obtained.

[0203] All reference elements on a two-dimensional ultrasonic phased array emit spherical waves corresponding to the imaging origin, and all reference elements simultaneously receive echo data to obtain a superimposed reference system.

[0204] Ultrasound image data are obtained by coherently superimposing the subsets of echoes received by all subarrays at each virtual point source based on a superposition reference frame.

[0205] In another alternative embodiment, the operation of the two-dimensional ultrasonic phased array emitting spherical waves each time may include:

[0206] Each ultrasonic element on a two-dimensional ultrasonic phased array emits ultrasonic waves at its own preset time point, so that the ultrasonic waves emitted by the ultrasonic phased array are equivalent to spherical waves emitted from a virtual point source.

[0207] In yet another optional embodiment, the number of pixels in each frame of the three-dimensional ultrasound sequence image is 100. The time dimension corresponding to the ultrasound sequence images is ;

[0208] Furthermore, the specific operation method by which the data segmentation module segments the three-dimensional ultrasound sequence image into multiple spatiotemporal sub-blocks may include:

[0209] The three-dimensional ultrasound sequence image is segmented into multiple spatiotemporal sub-blocks, and the spatial dimension data corresponding to any one spatiotemporal sub-block is: The time dimension data corresponding to any spatiotemporal sub-block is ;in Furthermore, there is some overlap between the time dimension data and the spatial dimension data of two adjacent spatiotemporal sub-blocks.

[0210] In yet another optional embodiment, the imaging module performs three-dimensional imaging of microvessels based on the microbubble center point, which may include:

[0211] Trajectory tracking is performed on the center point of microbubbles in each frame of the 3D contrast agent image to obtain multiple 3D microbubble trajectories corresponding to the 3D contrast agent sequence images;

[0212] By superimposing all the three-dimensional microbubble trajectories corresponding to the three-dimensional contrast agent sequence images, a three-dimensional distribution image of microvessels is obtained.

[0213] In another optional embodiment, the specific operation method of the imaging module to track the trajectory of the microbubble center point in each frame of the 3D contrast agent image may include:

[0214] Identify multiple feature pixels in each frame of the 3D contrast agent image. Feature pixels are pixels located at the edge of the microvascular contour.

[0215] Obtain a preset number of neighboring pixels around each feature pixel. For each feature pixel, fit the feature image based on the feature pixel and its corresponding neighboring pixels.

[0216] For each feature graphic in each frame of the 3D contrast agent image, the similar feature graphic corresponding to each feature graphic is determined in the adjacent frames of the 3D contrast agent image corresponding to that frame.

[0217] Based on the similar feature patterns corresponding to each feature pattern in all frames of 3D contrast agent images, the trajectory corresponding to each feature point is determined.

[0218] Example 5

[0219] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a microvascular three-dimensional imaging system based on an ultrasound phased array, as disclosed in an embodiment of the present invention. Figure 7 As shown, the microvascular three-dimensional imaging system based on ultrasound phased array may include:

[0220] Memory 201 storing executable program code;

[0221] Processor 202 coupled to memory 201;

[0222] The processor 202 calls the executable program code stored in the memory 201 to execute the steps in the three-dimensional microvascular imaging method based on ultrasound phased array described in Embodiment 1 or Embodiment 2 of the present invention.

[0223] Example 6

[0224] This invention discloses a computer storage medium storing computer instructions. When these computer instructions are invoked, they are used to execute the steps in the three-dimensional microvascular imaging method based on ultrasound phased array described in Embodiment 1 or Embodiment 2 of this invention.

[0225] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, 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 according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0226] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0227] Finally, it should be noted that the microvascular three-dimensional imaging method and system based on ultrasound phased array disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for three-dimensional imaging of microvessels based on ultrasound phased array, characterized in that, A two-dimensional ultrasonic phased array includes multiple subarrays located on the same plane, each subarray including multiple ultrasonic elements located on the same plane, wherein at least one ultrasonic element on each subarray is designated as a reference element; the spherical wave emitted by the two-dimensional ultrasonic phased array each time corresponds to a different virtual point source, and all virtual point sources have a preset positional relationship so that the echo data corresponding to all virtual point sources can be focused on the imaging origin, the method including: Within a preset continuous time period, spherical wave emission operations are performed multiple times. Each spherical wave emission operation includes: a two-dimensional ultrasonic phased array sequentially emitting spherical waves corresponding to different virtual point sources; after each spherical wave emission, all ultrasonic array elements simultaneously receive echo data to obtain a subset of echoes received by each subarray at each virtual point source; all reference array elements on the two-dimensional ultrasonic phased array emit spherical waves corresponding to the imaging origin, and all reference array elements simultaneously receive echo data to obtain a superimposed reference frame; and the echo subsets received by all subarrays at each virtual point source are coherently superimposed based on the superimposed reference frame to obtain ultrasonic image data, wherein the ultrasonic image data includes at least contrast agent echoes. Based on the ultrasound image data obtained from each spherical wave emission operation, a three-dimensional ultrasound sequence image is obtained; The three-dimensional ultrasound sequence image is divided into multiple spatiotemporal sub-blocks. Each spatiotemporal sub-block includes part of the time dimension data and part of the spatial dimension data of the three-dimensional ultrasound sequence image. Furthermore, there is partial overlap between the time dimension data and spatial dimension data of two adjacent spatiotemporal sub-blocks. Singular value decomposition is performed on all the spatiotemporal sub-blocks to extract the contrast agent signal, and the three-dimensional contrast agent sequence image corresponding to the three-dimensional ultrasound sequence image is obtained based on the contrast agent signal. The three-dimensional contrast agent sequence image is binarized to obtain multiple connected components, and the microbubble center point corresponding to each connected component is calculated based on each connected component. Three-dimensional imaging of microvessels is performed based on the center point of the microbubble.

2. The microvascular three-dimensional imaging method based on ultrasound phased array according to claim 1, characterized in that, The operation of the two-dimensional ultrasonic phased array emitting spherical waves each time includes: Each ultrasonic element on the two-dimensional ultrasonic phased array emits ultrasonic waves at its own preset time point, so that the ultrasonic waves emitted by the two-dimensional ultrasonic phased array are equivalent to spherical waves emitted from a virtual point source.

3. The microvascular three-dimensional imaging method based on ultrasound phased array according to claim 1, characterized in that, The number of pixels in each frame of the three-dimensional ultrasound sequence image is The time dimension corresponding to the three-dimensional ultrasound sequence image is ; And, the step of segmenting the three-dimensional ultrasound sequence image into multiple spatiotemporal sub-blocks includes: The three-dimensional ultrasound sequence image is segmented into multiple spatiotemporal sub-blocks, and the spatial dimension data corresponding to any one spatiotemporal sub-block is: The time dimension data corresponding to any spatiotemporal sub-block is ;in Furthermore, there is some overlap between the time dimension data and the spatial dimension data of two adjacent spatiotemporal sub-blocks.

4. The microvascular three-dimensional imaging method based on ultrasound phased array according to any one of claims 1 and 3, characterized in that, The step of performing three-dimensional microvascular imaging based on the microbubble center point includes: Trajectory tracking is performed on the center point of microbubbles in each frame of the 3D contrast agent image to obtain multiple 3D microbubble trajectories corresponding to the 3D contrast agent sequence images; By superimposing all the three-dimensional microbubble trajectories corresponding to the three-dimensional contrast agent sequence image, a three-dimensional distribution image of microvessels is obtained.

5. The microvascular three-dimensional imaging method based on ultrasound phased array according to any one of claims 4, characterized in that, Tracing the trajectory of the microbubble center points in each frame of the 3D contrast agent image, including: Identify multiple feature pixels in each frame of the 3D contrast agent image, wherein the feature pixels are pixels located at the edge of the microvascular contour; Obtain a preset number of neighboring pixels around each feature pixel. For each feature pixel, fit a feature image based on the feature pixel and its corresponding neighboring pixels. For each feature graphic in each frame of the 3D contrast agent image, the similar feature graphic corresponding to each feature graphic is determined in the adjacent frames of the 3D contrast agent image corresponding to that frame. Based on the similar feature patterns corresponding to each feature pattern in all frames of 3D contrast agent images, the trajectory corresponding to each feature point is determined.

6. A microvascular three-dimensional imaging device based on ultrasound phased array, characterized in that, A two-dimensional ultrasonic phased array includes multiple subarrays located on the same plane, each subarray including multiple ultrasonic elements located on the same plane, wherein at least one ultrasonic element in each subarray is designated as a reference element; the spherical wave emitted by the two-dimensional ultrasonic phased array each time corresponds to a different virtual point source, and all virtual point sources have a preset positional relationship so that the echo data corresponding to all virtual point sources can be focused on the imaging origin; the device includes: The transmitting module is used to control and execute spherical wave transmission operations multiple times within a preset continuous time period; Each spherical wave emission operation includes: a two-dimensional ultrasonic phased array sequentially emitting spherical waves corresponding to different virtual point sources; after each spherical wave emission, all ultrasonic array elements simultaneously receive echo data to obtain a subset of echoes received by each subarray at each virtual point source; all reference array elements on the two-dimensional ultrasonic phased array emit spherical waves corresponding to the imaging origin, and all reference array elements simultaneously receive echo data to obtain a superimposed reference frame; and the echo subsets received by all subarrays at each virtual point source are coherently superimposed based on the superimposed reference frame to obtain ultrasonic image data, wherein the ultrasonic image data includes at least contrast agent echoes. The signal receiving and acquisition module is used to obtain three-dimensional ultrasound sequence images based on the ultrasound image data obtained from each spherical wave transmission operation; The data segmentation module is used to segment the three-dimensional ultrasound sequence image into multiple spatiotemporal sub-blocks. Each spatiotemporal sub-block includes part of the time dimension data and part of the spatial dimension data of the three-dimensional ultrasound sequence image, and there is partial overlap between the time dimension data and spatial dimension data of two adjacent spatiotemporal sub-blocks. The singular value decomposition module is used to perform singular value decomposition on all the spatiotemporal sub-blocks, extract the contrast agent signal, and obtain the three-dimensional contrast agent sequence image corresponding to the three-dimensional ultrasound sequence image based on the contrast agent signal. The center point localization module is used to perform binarization processing on the three-dimensional contrast agent sequence image to obtain multiple connected components, and calculate the microbubble center point corresponding to each connected component. An imaging module is used to perform three-dimensional imaging of microvessels based on the center point of the microbubbles.

7. An ultrasound imaging system, characterized in that, The ultrasound imaging system includes: A two-dimensional ultrasonic phased array includes multiple subarrays located on the same plane. Each subarray includes multiple ultrasonic elements located on the same plane. At least one ultrasonic element on each subarray is designated as a reference element. The spherical wave emitted by the two-dimensional ultrasonic phased array corresponds to a different virtual point source each time. All virtual point sources have a preset positional relationship so that the echo data corresponding to all virtual point sources can be focused on the imaging origin. The two-dimensional ultrasonic phased array is used for: Within a preset continuous time period, spherical wave emission operations are performed multiple times. Each spherical wave emission operation includes: a two-dimensional ultrasonic phased array sequentially emitting spherical waves corresponding to different virtual point sources; after each spherical wave emission, all ultrasonic array elements simultaneously receive echo data to obtain a subset of echoes received by each subarray at each virtual point source; all reference array elements on the two-dimensional ultrasonic phased array emit spherical waves corresponding to the imaging origin, and all reference array elements simultaneously receive echo data to obtain a superimposed reference frame; and the echo subsets received by all subarrays at each virtual point source are coherently superimposed based on the superimposed reference frame to obtain ultrasonic image data, wherein the ultrasonic image data includes at least contrast agent echoes. The host computer, which is communicatively connected to the two-dimensional ultrasonic phased array, is used for: Based on the ultrasound image data obtained from each spherical wave emission operation, a three-dimensional ultrasound sequence image is obtained; The three-dimensional ultrasound sequence image is divided into multiple spatiotemporal sub-blocks. Each spatiotemporal sub-block includes part of the time dimension data and part of the spatial dimension data of the three-dimensional ultrasound sequence image. Furthermore, there is partial overlap between the time dimension data and spatial dimension data of two adjacent spatiotemporal sub-blocks. Singular value decomposition is performed on all the spatiotemporal sub-blocks to extract the contrast agent signal, and the three-dimensional contrast agent sequence image corresponding to the three-dimensional ultrasound sequence image is obtained based on the contrast agent signal. The three-dimensional contrast agent sequence image is binarized to obtain multiple connected components, and the microbubble center point corresponding to each connected component is calculated based on each connected component. Three-dimensional imaging of microvessels is performed based on the center point of the microbubble.

8. A microvascular three-dimensional imaging system based on ultrasound phased array, characterized in that, The system includes: a memory storing executable program code; a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the three-dimensional microvascular imaging method based on ultrasound phased array as described in any one of claims 1-5.

9. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the microvascular three-dimensional imaging method based on ultrasound phased array as described in any one of claims 1-5.

Images