A four-dimensional clutter suppression method and system for ultrasonic micro blood flow imaging

Abstract

The application discloses a four-dimensional clutter suppression method and system for ultrasonic micro blood flow imaging, and belongs to the field of ultrasonic imaging. The method is applied to a four-dimensional data set after multi-frame multi-angle plane wave transmission and beam synthesis based on time accumulation, utilizes the angular coherence principle, that is, main lobe components have high coherence in angular space while side lobe components are opposite, and adopts high-order singular value decomposition for clutter suppression. The four-dimensional clutter filtering method based on angles expands the data dimension (increases the feature space), increases the information quantity, is beneficial to blood flow separation, can effectively suppress clutter with fewer frames, can effectively improve the micro blood flow imaging quality in small data sets, requires smaller multi-angle data quantity, has lower calculation cost, and solves the problems in the prior art.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of ultrasonic imaging, and relates to a four-dimensional clutter suppression method and system for ultrasonic micro blood flow imaging. BACKGROUND

[0002] The ultrasonic micro blood vessel imaging technology can realize real-time monitoring of micro blood flow movement and morphological structure changes of micro blood vessels, and has important significance for physiological state evaluation and clinical disease diagnosis and treatment. However, the current micro blood flow detection faces two challenges; one is that the blood flow signal is weak and closer to the noise base; the other is that the tissue movement will destroy the measured blood volume or blood flow level. Therefore, an effective clutter suppression method is the key to micro blood flow detection.

[0003] The traditional frequency domain clutter filter is suitable for filtering out the arterial blood vessel wall movement, but for the low-speed blood flow accompanied by slow tissue movement, it is difficult to effectively distinguish them because the slow time-frequency spectrum of the two overlaps. The filter based on eigenvalue decomposition (SVD) can effectively separate blood flow and clutter by using spatial coherence, and is the most widely used method at present, but its performance is highly dependent on the selection of the threshold value, and the performance is limited in real-time small set (several tens of frames) imaging. The robust principal component analysis (RPCA) and its improved filtering method can enhance the sensitivity of micro blood flow detection, but the required set of samples is large, the computational complexity is high, and the calculation time is long. The high-order SVD method based on pulse data and aperture data can effectively separate blood flow and clutter, and significantly improve the blood flow imaging quality, but the data required by the method is large, and the calculation cost is high, which is not suitable for real-time imaging. At present, there is no method that can effectively suppress clutter in a small data set with low computational complexity, which limits the real-time imaging of ultrasonic micro blood flow. SUMMARY

[0004] The application aims to solve the problem in the prior art that effective clutter suppression cannot be performed in a small data set with low computational complexity, and provides a four-dimensional clutter suppression method and system for ultrasonic micro blood flow imaging.

[0005] To achieve the above-mentioned purpose, the following technical solutions are adopted in the application:

[0006] The application provides a four-dimensional clutter suppression method for ultrasonic micro blood flow imaging, which comprises the following steps:

[0007] Reorganize the four-dimensional IQ data after multi-frame multi-angle beam synthesis to obtain three-dimensional IQ data;

[0008] Decompose the three-dimensional IQ data to obtain a singular vector matrix and mode eigenvalues;

[0009] Select an appropriate cutoff point based on the feature space and pattern eigenvalues ​​of the singular vector matrix;

[0010] The filtered blood flow core tensor is obtained based on the cutoff point. The singular components corresponding to clutter in the core tensor are set to zero, and the filtered blood flow signal is obtained to achieve four-dimensional clutter suppression.

[0011] Preferably, the four-dimensional IQ data obtained by multi-frame, multi-angle beamforming is recombined in Cascorati form to obtain three-dimensional IQ data. ,in, Represents the complex field, spatial sample number , Indicates the number of axial samples (perpendicular to the sensor array). Indicates the number of samples in the lateral direction (along the sensor array). Indicates the number of angles. Indicates the number of time frames.

[0012] Preferably, a higher-order singular value decomposition method is used to decompose the three-dimensional IQ data. Decomposed into ,in, For orthogonal core tensors, For spatial singular vectors, Transmitting singular vectors for angles The orthogonal factor matrix of the time singular vectors. Represents the pattern - n product.

[0013] Preferably, the decomposed three-dimensional IQ data An expanded matrix is ​​formed by fixing one dimension and combining the other dimensions. This generates a covariance matrix, which includes the covariance matrix generated from the spatial singular vector matrix. The covariance matrix generated by the time singular vector matrix and the covariance matrix generated by the angular singular vector matrix The expression is as follows:

[0014]

[0015]

[0016]

[0017] in, This is the diagonal eigenvalue matrix of the spatial pattern. This is the diagonal eigenvalue matrix of the angle pattern. This is the diagonal eigenvalue matrix of the time pattern. Transpose of a spatial singular vector is the transpose of the angle transmission singular vector, is the transpose of the time singular vector.

[0018] Preferably, the mode eigenvalues include , and , and are specifically expressed as follows:

[0019]

[0020]

[0021]

[0022] wherein, is a vector of the tensor , are vectors representing space, angle and time, respectively.

[0023] Preferably, the cutoff points are obtained as follows:

[0024] In the spatial domain, two cutoff values , are calculated from the perspectives of the eigenimage correlation and singular value, respectively, to suppress tissue motion and noise; each spatial singular vector is re-formatted into a two-dimensional eigenimage, and then the correlation between adjacent eigenimages is calculated to obtain a Pearson correlation coefficient curve, and the largest mutation point on the curve indicates that the two adjacent eigenimages are affected by tissue clutter and blood flow signal, respectively, and thus the point is a low cutoff point .

[0025] In order to remove noise, a pre-cutoff point is first determined at a relatively smooth part of the Pearson correlation coefficient curve, and then a linear fitting is performed on the corresponding part of the normalized singular value curve, and the point deviating from the linear fitting line is the final high cutoff point .

[0026] In the time domain: two cutoff values , are calculated from the perspectives of the Doppler shift and singular value, respectively, to suppress tissue motion and noise; in order to suppress tissue motion, the time singular vector obtained after HOSVD decomposition is subjected to spectral analysis, and the low cutoff point is the point at which the average Doppler shift exceeds a specified cutoff frequency;

[0027] To remove noise, the principle that higher-order singular values ​​of noise should follow a linear distribution on a numerical scale is first applied. A pre-cutoff point is then determined within the flat portion of the average Doppler shift, indicating the initial attainment of the noise threshold. Next, a linear fit is performed on the corresponding portion of the normalized singular value curve; the point where the curve begins to deviate from the linear fit line is the final high cutoff point. ;

[0028] In the angular domain, based on the principle that the main lobe component has high spatial coherence and is usually distributed in large singular values, the angular cutoff point... A value of 1 or 2 is usually chosen to suppress side lobes.

[0029] Preferably, the filtered blood flow core tensor for:

[0030]

[0031] Filtered blood flow signal , This is the filtered orthogonal core tensor.

[0032] This invention proposes a four-dimensional clutter suppression system for ultrasound micro-blood flow imaging, comprising:

[0033] The data acquisition module is used to reassemble the four-dimensional IQ data after multi-frame multi-angle beamforming to obtain three-dimensional IQ data.

[0034] The pattern feature value acquisition module is used to decompose the three-dimensional IQ data to obtain the singular vector matrix and each pattern feature value.

[0035] The cutoff point acquisition module is used to select a suitable cutoff point based on the feature space and pattern feature values ​​of the singular vector matrix.

[0036] The signal filtering module is used to obtain the filtered blood flow core tensor based on the cutoff point, set the singular components corresponding to clutter in the core tensor to zero, obtain the filtered blood flow signal, and achieve four-dimensional clutter suppression.

[0037] A computer device includes a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of a four-dimensional clutter suppression method for ultrasound micro-blood flow imaging.

[0038] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of a four-dimensional clutter suppression method for ultrasound micro-blood flow imaging.

[0039] Compared with the prior art, the present application has the following beneficial effects:

[0040] The present application provides a four-dimensional clutter suppression method for ultrasonic micro blood flow imaging, which is applied to a four-dimensional data set after multi-frame multi-angle plane wave transmission and beam synthesis based on time accumulation, utilizes the principle of angular coherence, that is, the main lobe component has high coherence in the angle space while the side lobe component is opposite, and adopts high-order singular value decomposition for clutter suppression.

[0041] Further, in the selection of the characteristic space cutoff value, if the cutoff value for filtering out tissue motion is too small, the tissue interference cannot be filtered out, if the cutoff value for filtering out tissue motion is too large, the blood signal will be lost, if the cutoff value for filtering out noise is too small, the blood signal will be lost, and if the cutoff value for filtering out noise is too large, the noise interference cannot be filtered out.

[0042] The present application provides a four-dimensional clutter suppression system for ultrasonic micro blood flow imaging, which divides the system into a data acquisition module, a mode characteristic value acquisition module, a cutoff point acquisition module and a signal filtering module, acquires the filtered blood flow signal, and realizes four-dimensional clutter suppression. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments, and it should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0044] Figure 1 is a flow chart of the four-dimensional clutter suppression method for ultrasonic micro blood flow imaging provided by the present application.

[0045] Figure 2 is a result comparison chart of different clutter suppression methods provided by the present application.

[0046] Figure 3 is a four-dimensional clutter suppression system diagram for ultrasonic micro blood flow imaging provided by the present application. DETAILED DESCRIPTION

[0047] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings. Obviously, the described embodiments are only some but not all of the embodiments of the present application. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0048] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative work fall within the scope of protection of the present application.

[0049] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0050] In the description of the embodiments of the present application, it should be noted that, if the orientation or position relationship indicated by the terms "upper", "lower", "horizontal", "inner" and the like is based on the orientation or position relationship shown in the drawings, or is the orientation or position relationship when the product of the present application is usually placed, and is only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the indicated device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, the terms "first", "second" and the like are only used for differentiation and cannot be understood as indicating or implying relative importance.

[0051] In addition, if the term "horizontal" appears, it does not mean that the component must be absolutely horizontal, but can be slightly inclined. For example, "horizontal" only means that its direction is relatively more horizontal than "vertical", and does not mean that the structure must be completely horizontal, but can be slightly inclined.

[0052] In the description of the embodiments of the present application, it should also be noted that, unless otherwise explicitly specified and limited, if the terms "arrange", "mount", "connect", "connect" appear, they should be understood in a broad sense, for example, can be fixedly connected, can be detachably connected, or integrally connected; can be mechanically connected, can be electrically connected; can be directly connected, can be indirectly connected through an intermediate medium, or can be the communication inside two elements. For those of ordinary skill in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0053] The present application will be described in further detail below with reference to the accompanying drawings:

[0054] In view of the defects of the prior art or the improvement needs, the purpose of the present application is to provide a four-dimensional clutter suppression method for ultrasonic micro blood flow imaging, as shown in the accompanying drawings, comprising the following steps: Figure 1

[0055] S1, reorganize the four-dimensional IQ data after multi-frame multi-angle beam synthesis to obtain three-dimensional IQ data;

[0056] The four-dimensional IQ data after multi-frame multi-angle beam synthesis is reorganized in Casorati form to obtain three-dimensional IQ data , wherein represents the complex domain, and the number of spatial samples , represents the number of axial samples (perpendicular to the sensor array), represents the number of transverse samples (along the sensor array), represents the number of angles, represents the number of time frames.

[0057] S2, decompose the three-dimensional IQ data to obtain a singular vector matrix and each mode eigenvalue;

[0058] The three-dimensional IQ data is decomposed by using a high-order singular value decomposition method, and the three-dimensional IQ data is decomposed into , wherein is an orthogonal core tensor, is a spatial singular vector, is an angle transmission singular vector, is a time singular vector orthogonal factor matrix, represents a mode-n product.

[0059] The decomposed three-dimensional IQ data is formed into an unfolded matrix by fixing one dimension and combining other dimensions, and a covariance matrix is generated, which includes a covariance matrix generated by the spatial singular vector matrix, a covariance matrix generated by the time singular vector matrix, and a covariance matrix generated by the angle singular vector matrix, and the expression is as follows:

[0060]

[0061]

[0062]

[0063] wherein is a diagonal eigenvalue matrix of the spatial mode, diagonal eigenvalue matrix of angle mode, diagonal eigenvalue matrix of time mode, transpose of spatial singular vector, transpose of angle transmission singular vector, transpose of time singular vector.

[0064] mode eigenvalue includes , and , which are specifically expressed as follows:

[0065]

[0066]

[0067]

[0068] wherein, is a vector of tensor , are vectors representing space, angle and time respectively.

[0069] S3, selecting appropriate cut-off points according to the eigenvalue space of singular vector matrix and mode eigenvalue;

[0070] The cut-off points are obtained as follows:

[0071] In the spatial domain, two cut-off values , are calculated from the perspectives of feature image correlation and singular value respectively to suppress tissue motion and noise; each spatial singular vector is re-formatted into a two-dimensional feature image, and then the correlation between adjacent feature images is calculated to obtain a Pearson correlation coefficient curve, and the largest mutation point on the curve indicates that the adjacent two feature images are affected by tissue clutter and blood flow signal respectively, so the point is the low cut-off point ;

[0072] In order to remove noise, first determine the pre-cut-off point when the Pearson correlation coefficient curve starts to reach smoothness in the relatively smooth part of the curve, and then linearly fit the corresponding part of the normalized singular value curve, and the point where the curve starts to deviate from the linear fitting line is the final high cut-off point ;

[0073] In the time domain: from the perspectives of Doppler shift and singular value, two cut-off values , are calculated to suppress tissue motion and noise respectively; in order to suppress tissue motion, the time singular vector obtained after HOSVD decomposition is subjected to spectral analysis, and the low cut-off point is calculated according to the average Doppler shift the points where the average Doppler frequency exceeds a specified cutoff frequency;

[0074] In order to remove noise, first of all, according to the principle that the high-order singular value of noise should follow the linear distribution under the number scale, the pre-cutoff point where the noise threshold is first reached is determined in the flat part of the average Doppler shift, and then linear fitting is performed on the corresponding part of the normalized singular value curve, and the point where the linear fitting line deviates is the final high cutoff point ;

[0075] In the angle domain, according to the principle that the main lobe component has high spatial coherence and is usually distributed in a large singular value, the angle cutoff point is usually selected as 1 or 2 to suppress the sidelobe.

[0076] S4, obtaining the filtered blood flow core tensor according to the cutoff point, setting the singular component corresponding to the clutter in the core tensor to zero, obtaining the filtered blood flow signal, and realizing four-dimensional clutter suppression.

[0077] The filtered blood flow core tensor is:

[0078]

[0079] The filtered blood flow signal , is the filtered orthogonal core tensor.

[0080] The specific steps are as follows:

[0081] Step 1, load a plurality of four-dimensional IQ data after multi-angle beam synthesis , wherein represents the number of axial samples (perpendicular to the sensor array), represents the number of transverse samples (along the sensor array), represents the number of angles, represents the number of time frames.

[0082] Step 2, reorganize the spatial samples of the data set into ( ) in the Casorati form. The reorganized data becomes three dimensions of space, time and angle.

[0083] Step 3, perform high-order SVD (HOSVD) decomposition, i.e. Tucker decomposition, on the reshaped signal.

[0084] The specific steps of HOSVD decomposition are: first, the original multi-angle transmission data is formed into an unfolded matrix by fixing one dimension and combining the other dimensions. Covariance matrix is generated and eigen decomposition is performed to obtain singular vector matrix:

[0085]

[0086]

[0087]

[0088] where, , , are the diagonal eigenvalue matrices of the three modes respectively.

[0089] Three-dimensional multi-angle transmission data is decomposed into: where, is the orthogonal core tensor, , , are orthogonal factor matrices, representing spatial singular vectors, angle transmission singular vectors, and time singular vectors respectively, denotes the mode-n product, and matrix is the product of matrix and the tensor expanded by mode-n, defined as

[0090] The set of mode-n singular values can also be represented by the Frobenius norm of the core tensor :

[0091]

[0092] where the eigenvalues of each mode are arranged in descending order.

[0093] Step 4: Select appropriate cutoff points in each eigen space to suppress tissue clutter signals and reduce noise levels.

[0094] In the spatial domain, two cutoff values , are calculated to suppress tissue motion and noise respectively from the perspectives of eigen image correlation and singular values. To suppress tissue motion, each spatial singular vector is reformatted into a two-dimensional eigen image, and then the correlation between adjacent eigen images is calculated to obtain the Pearson correlation coefficient curve. The largest mutation point on the curve indicates that the adjacent two eigen images are affected by tissue clutter and blood flow signals respectively, so this point is the low cutoff point To remove noise, the pre-threshold of noise is first determined in the flat part of the curve of Pearson correlation coefficient, and then the linear fitting is performed in the corresponding part of the curve of normalized singular value, and the point where the curve begins to deviate from the linear fitting line is the final high threshold ;

[0095] In time domain, two thresholds are calculated from the perspectives of Doppler shift and singular value , respectively to suppress tissue motion and noise. To suppress tissue motion, the time singular vector obtained after HOSVD decomposition is subjected to spectral analysis, and the low threshold is the point where the average Doppler shift exceeds the specified threshold frequency. To remove noise, the pre-threshold of noise is first determined in the flat part of the average Doppler shift, and then the linear fitting is performed in the corresponding part of the curve of normalized singular value, and the point where the curve begins to deviate from the linear fitting line is the final high threshold ;

[0096] In angle domain, according to the principle that the main lobe component has high spatial coherence and is usually distributed in a large singular value, the angle threshold is usually selected as 1 or 2 to suppress the sidelobe.

[0097] Step 5, after selecting the appropriate threshold in each feature space, the singular component corresponding to the clutter in the core tensor is set to zero to achieve the filtering effect. The filtered blood flow core tensor is:

[0098]

[0099] The filtered blood flow signal is:

[0100]

[0101] Preferably, the four-dimensional IQ data after beam synthesis has data dimensions of scanning width (along the sensor array), scanning depth (perpendicular to the sensor array), angle, and time, wherein the time sample amount is 10 frames, preferably 100 frames, the calculation speed is fast, the image quality is high, and it is beneficial to real-time imaging.

[0102] The four-dimensional clutter filtering method adopted by the present application is influenced by the following factors: the number of angles and the angle range, the number of frames, and the selection of the feature space cutoff value. Preferably, the number of angles is less than 31, and the angle range is less than 10°. Within the preferred number of angles and range, the more the number of angles, the better the filtering effect, and the more the calculation amount; conversely, the fewer the number of angles, the worse the filtering effect, and the less the calculation amount. Preferably, the number of frames is greater than 15. Within the preferred number of frames, the more the number of frames, the better the filtering effect, and the more the calculation amount; conversely, the fewer the number of frames, the worse the filtering effect, and the less the calculation amount. In the selection of the feature space cutoff value, if the cutoff value for filtering out tissue motion is too small, the tissue disturbance cannot be filtered out, and if the cutoff value for filtering out tissue motion is too large, the blood signal will be lost, if the cutoff value for filtering out noise is too small, the blood signal will be lost, and if the cutoff value for filtering out noise is too large, the noise disturbance cannot be filtered out.

[0103] As shown in Figure 2 , it is a comparison chart of the results of different clutter suppression methods. The phantom is a 0.6mm wall-free phantom, which is configured by 10% gelatin solution and 5% agar. The blood flow velocity is 1mm / s, and an external mechanical vibrator is added to make the tissue motion velocity 5mm / s. Figure 2 , a is the superfast power Doppler image filtered by the traditional SVD, and it can be seen that the image is obviously affected by clutter, Figure 2 , b is the superfast power Doppler image filtered by the method of the present application, and the image effect is obviously better than a in Figure 2 .

[0104] Overall, compared with the prior art, the above technical solutions conceived by the present application adopt an angle-based four-dimensional clutter filtering method, expand the feature space, can effectively suppress clutter with fewer frames, significantly improve the quality of micro blood flow imaging, and the angle-based four-dimensional data is small, and the calculation cost is low. Reasonable selection of each cutoff point of the four-dimensional clutter suppression method for ultrasonic micro blood flow imaging provided by the present application can effectively exert the performance of the filter. In the spatial and temporal domain, two cutoff points are selected to suppress tissue motion and noise, respectively, and in the angle domain, one cutoff point is selected to suppress incoherent artifacts and noise. In addition, the ranks of mode-n are not necessarily the same, and the core tensor is not diagonal, so that decomposition by HOSVD can increase flexibility in defining the ranks of the clutter and blood subspaces.

[0105] The four-dimensional clutter suppression system for ultrasonic micro blood flow imaging proposed by the present application, as shown in Figure 3 , comprises a data acquisition module, a mode eigenvalue acquisition module, a cutoff point acquisition module, and a signal filtering module.

[0106] The data acquisition module is used for reorganizing four-dimensional IQ data after multi-frame multi-angle beam synthesis to obtain three-dimensional IQ data.

[0107] The mode characteristic value acquisition module is used for decomposing the three-dimensional IQ data to obtain a singular vector matrix and mode characteristic values.

[0108] The cutoff point acquisition module is used for selecting a suitable cutoff point according to a characteristic space of the singular vector matrix and the mode characteristic values.

[0109] The signal filtering module is used for obtaining a filtered blood flow core tensor according to the cutoff point, setting zero singular components corresponding to clutter in the core tensor, obtaining a filtered blood flow signal, and realizing four-dimensional clutter suppression.

[0110] An embodiment of the present application provides a terminal device, the terminal device of the embodiment comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor. The processor implements the steps in each of the method embodiments when executing the computer program. Alternatively, the processor implements the functions of each module / unit in each of the system embodiments when executing the computer program.

[0111] The computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present application.

[0112] The terminal device can be a desktop computer, a notebook computer, a palm computer, a cloud server and other computing devices. The terminal device can include, but is not limited to, a processor and a memory.

[0113] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0114] The memory can be used to store the computer program and / or modules, and the processor realizes various functions of the terminal device by running or executing the computer program and / or modules stored in the memory, and calling data stored in the memory.

[0115] The modules / units integrated in the terminal device, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can implement the steps of each method embodiment when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The computer readable medium can include any entity or system, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code. It should be noted that the computer readable medium can include or exclude contents according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electric carrier signals and telecommunication signals.

[0116] The application provides a four-dimensional clutter suppression method for ultrasonic micro blood flow imaging, and a programming environment of the method is MATLAB. The method is applied to a four-dimensional data set after multi-frame multi-angle plane wave transmission and beam synthesis based on time accumulation. The method uses an angular coherence principle, that is, main lobe components have high coherence in an angular space and side lobe components are opposite, to perform clutter suppression. The four-dimensional clutter filtering method based on angles expands data dimensions (increases a feature space), increases information quantity, is beneficial to blood flow separation, can effectively suppress clutter with fewer frames, can effectively improve micro blood flow imaging quality in small data sets, needs small multi-angle data quantity, has low calculation cost, and solves problems in the prior art.

[0117] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application. The present application can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A four-dimensional clutter suppression method for ultrasound micro-blood flow imaging, characterized in that, Includes the following steps: The four-dimensional IQ data obtained by multi-frame multi-angle beamforming is recombined to obtain three-dimensional IQ data. The three-dimensional IQ data is decomposed to obtain the singular vector matrix and the feature values ​​of each mode; Select an appropriate cutoff point based on the feature space and pattern eigenvalues ​​of the singular vector matrix; The filtered blood flow core tensor is obtained based on the cutoff point. The singular components corresponding to clutter in the core tensor are set to zero to obtain the filtered blood flow signal, thus achieving four-dimensional clutter suppression. The three-dimensional IQ data is decomposed using a higher-order singular value decomposition method. Decomposed into ,in, For orthogonal core tensors, For spatial singular vectors, Transmitting singular vectors for angles The orthogonal factor matrix of the time singular vectors. Represents the pattern - n product; Represents the complex field, spatial sample number , Indicates the number of samples along the axial direction, perpendicular to the sensor array; This indicates the number of samples in the horizontal direction, along the sensor array; Indicates the number of angles. Indicates the number of time frames; The cutoff point is obtained as follows: In the spatial domain, two cutoff values ​​are calculated from the perspectives of feature image correlation and singular values. , Suppressing tissue movement and noise respectively; Each spatial singular vector The image is reformatted into a two-dimensional feature image, and then the correlation between adjacent feature images is calculated to obtain the Pearson correlation coefficient curve. The largest abrupt change on the curve indicates that the two adjacent feature images are affected by tissue clutter and blood flow signals, respectively; therefore, this point is the low cutoff point. ; To remove noise, a preliminary cutoff point is first determined in the relatively smooth portion of the Pearson correlation coefficient curve to indicate the start of smoothing. Then, a linear fit is performed on the corresponding portion of the normalized singular value curve, and the point where the curve begins to deviate from the linear fit line is the final high cutoff point. ; In the time domain: Calculate two cutoff values ​​from the perspectives of Doppler frequency shift and singular values. , Tissue motion and noise are suppressed separately; to suppress tissue motion, the temporal singular vector obtained after HOSVD decomposition is analyzed. Perform spectral analysis and calculate the average Doppler shift and low cutoff point. The point where the average Doppler frequency exceeds the specified cutoff frequency; To remove noise, the principle that higher-order singular values ​​of noise should follow a linear distribution on a numerical scale is first applied. A pre-cutoff point is then determined within the flat portion of the average Doppler shift, indicating the initial attainment of the noise threshold. Next, a linear fit is performed on the corresponding portion of the normalized singular value curve; the point where the curve begins to deviate from the linear fit line is the final high cutoff point. ; In the angular domain, based on the principle that the main lobe component has high spatial coherence and is usually distributed in large singular values, the angular cutoff point... A value of 1 or 2 is usually chosen to suppress side lobes.

2. The four-dimensional clutter suppression method for ultrasound micro-blood flow imaging according to claim 1, characterized in that, The four-dimensional IQ data obtained by multi-frame, multi-angle beamforming is reconstructed in Cascorati form to obtain three-dimensional IQ data. .

3. The four-dimensional clutter suppression method for ultrasound micro-blood flow imaging according to claim 1, characterized in that, Decomposed 3D IQ data An expanded matrix is ​​formed by fixing one dimension and combining the other dimensions. This generates a covariance matrix, which includes the covariance matrix generated from the spatial singular vector matrix. The covariance matrix generated by the time singular vector matrix and the covariance matrix generated by the angular singular vector matrix The expression is as follows: in, This is the diagonal eigenvalue matrix of the spatial pattern. This is the diagonal eigenvalue matrix of the angle pattern. This is the diagonal eigenvalue matrix of the time pattern. Transpose of a spatial singular vector For the transpose of the singular vector of angle transmission, It is the transpose of the time singular vector.

4. The four-dimensional clutter suppression method for ultrasound micro-blood flow imaging according to claim 3, characterized in that, Pattern feature values ​​include , and Specifically, it is expressed as follows: in, For tensor A vector, These are vectors representing space, angle, and time, respectively.

5. The four-dimensional clutter suppression method for ultrasound micro-blood flow imaging according to claim 1, characterized in that, Filtered blood flow core tensor for: Filtered blood flow signal , This is the filtered orthogonal core tensor.

6. A four-dimensional clutter suppression system for ultrasound micro-blood flow imaging, characterized in that, The four-dimensional clutter suppression method for ultrasound micro-blood flow imaging according to any one of claims 1 to 5 includes: The data acquisition module is used to reassemble the four-dimensional IQ data after multi-frame multi-angle beamforming to obtain three-dimensional IQ data. The pattern feature value acquisition module is used to decompose the three-dimensional IQ data to obtain the singular vector matrix and each pattern feature value. The cutoff point acquisition module is used to select a suitable cutoff point based on the feature space and pattern feature values ​​of the singular vector matrix. The signal filtering module is used to obtain the filtered blood flow core tensor based on the cutoff point, set the singular components corresponding to clutter in the core tensor to zero, obtain the filtered blood flow signal, and achieve four-dimensional clutter suppression.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes a computer program, it implements the steps of the four-dimensional clutter suppression method for ultrasound micro-blood flow imaging as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the four-dimensional clutter suppression method for ultrasound micro-blood flow imaging as described in any one of claims 1 to 5.

Images