CN111772676B - Ultrafast ultrasonic Doppler spinal cord micro-blood flow imaging system - Google Patents

Abstract

The invention belongs to the technical field of biomedical ultrasonic imaging, and particularly relates to an ultrafast ultrasonic Doppler spinal cord micro-blood flow imaging system. The system comprises two parts, namely hardware and software; the hardware part comprises an ultrasonic wave transmitting and receiving module and an experimental equipment module. The software part comprises ultra-fast ultrasonic imaging, beam synthesis, motion calibration, clutter filtering, doppler imaging, difference solving, correlation analysis and other modules. Firstly, an ultrasonic plane wave transmitting and receiving control module is compiled based on an ultrafast ultrasonic imaging technology and a multi-angle plane wave composite imaging theory, and a hardware device is controlled by computer software to transmit and receive ultrasonic waves; processing the received echo data to finally obtain a Doppler blood flow image; and analyzing parameters such as the speed, the direction and the like of the blood flow. The system also provides a mode for imaging spinal micro-blood flow under the conditions of spinal pressurization and injury, and can perform spinal functional analysis and physiological and pathological analysis.

Description

Ultrafast Ultrasonic Doppler Spinal Cord Microflow Imaging System

technical field

The invention belongs to the technical field of biomedical ultrasonic imaging, in particular to a spinal cord micro-blood flow imaging system.

Background technique

Spinal cord injury is often accompanied by functional nerve injury, and the complete loss of blood flow in the trauma center and the reduction of peripheral blood perfusion can be used to indicate the lesion of spinal cord function injury.

At present, the standard clinical diagnostic methods for spinal injuries include X-ray plain film, computed tomography (CT) and magnetic resonance imaging (MRI). These methods still have the following disadvantages: there is electromagnetic radiation harmful to human health, the cost is high, and the imaging time is long. Compared with these existing methods, ultrasound imaging has the advantages of fast imaging speed, portable equipment, low cost and no ionizing radiation. Ultrafast ultrasonic Doppler blood flow imaging method based on multi-angle plane wave composite imaging, as a new blood flow monitoring technology, has the potential of high-resolution imaging of intraspinal microvessels and related microblood flow. (He Qiong, Luo Jianwen. Research progress of ultra-high-speed ultrasound imaging[J]. Chinese Medical Imaging Technology, 2014,30(8):1251-1255.).

Contents of the invention

The purpose of the present invention is to provide a spinal cord micro blood flow imaging system with high imaging speed, portable equipment, low cost and no ionizing radiation.

The spinal cord micro-flow imaging system provided by the present invention is based on ultra-fast ultrasonic Doppler technology, and the system is divided into two parts: hardware and software. Among them, the hardware part can be divided into two modules: the ultrasonic signal transmitting and receiving module, and the experimental equipment module. The ultrasonic signal transmitting and receiving module includes a waveform generator, an ultrasonic probe, an analog signal amplifier, an analog signal filter, an analog-to-digital converter and a storage unit; the experimental equipment module includes a spinal cord fixation device, a physiological signal detection device, a spinal cord compression device, an electric Stimulation device, oscilloscope and arbitrary waveform generator; the software part includes ultrasonic pulse transmission and reception sequence module, beamforming module, motion calibration module, clutter filtering module, blood flow Doppler imaging module, and difference module. The technologies involved in the system include ultrafast ultrasound imaging technology, ultrasound blood flow Doppler imaging technology, spinal cord functional imaging, etc.

(1) Hardware part

1. Ultrasonic transmitting and receiving module

The ultrasonic transmitting and receiving module includes a waveform generator, an ultrasonic probe, an analog signal amplifier, an analog signal filter, an analog-to-digital converter and a storage unit. The function of this module is to transmit the ultrasonic signal and receive the echo signal.

Preferably, the connection mode and work flow of the hardware module are shown in Figure 2. First, the ultrasonic transmission and reception sequence set by the software controls the waveform generator to generate a transmission waveform, and the ultrasonic probe converts electrical energy into acoustic energy to emit ultrasonic waves; The echo signal is received by the same or different ultrasonic probes, and the echo signal is processed by the analog signal amplifier and analog signal filter, and then the analog-to-digital converter performs analog-to-digital conversion, and sends it to the storage unit to complete the data storage.

2. Experimental equipment module

The experimental equipment module is used to fix the spinal cord, image the micro blood flow of the spinal cord under different conditions, and monitor the relevant physiological signals.

Preferably, the module includes a spinal cord fixation device, a physiological signal monitoring device, a spinal cord compression device, an electrical stimulation device, an oscilloscope and an arbitrary waveform generator. The spinal cord fixation device is used to fix the spinal cord. The spinal cord compression device can be used to quantitatively pressurize the spinal cord, and the hemodynamic changes under different pressure conditions can be observed through an oscilloscope; Electrical stimulation, explore the relationship between spinal cord nerve function and blood flow changes through spinal cord micro blood flow analysis, so as to conduct related physiological and pathological analysis; and can be supplemented with physiological signal monitoring devices to measure physiological signals such as respiration and heartbeat in real time, and perform respiration and heartbeat correlation. Calibration of motion, and correlation analysis of other physiological signals and blood flow signals.

(2) Software modules

The software modules include an ultrasonic pulse transmitting and receiving sequence module, a beam forming module, a motion calibration module, a clutter filtering module, a blood flow Doppler imaging module, and a difference calculation module.

The software flow of the spinal cord blood flow Doppler imaging system is shown in Figure 1. First, the ultrasonic plane wave transmitting and receiving sequence is programmed by the ultrasonic pulse transmitting and receiving sequence module, and the hardware module is controlled to transmit ultrasonic waves and receive echo signals. The beamforming module performs beamforming on the received echo signals to obtain a B-mode image, and coherently combines the data of the B-mode image obtained by transmitting multiple angle plane waves to obtain a high-quality B-mode image. The position between images is calibrated by a motion calibration module (breathing causes movement of the spinal cord position, so calibration is required). Next, the clutter filtering module performs clutter filtering. It is known that all received signals are y, y=s+t+n, s is the blood flow signal, t is the tissue signal, and n is the noise signal, which needs to be filtered Only by removing the tissue signal and noise signal in the original signal can a clear blood flow signal map be obtained. Next, the blood flow Doppler imaging module performs further processing to obtain a Doppler blood flow map. If the spinal cord is pressurized or electrical stimulation is applied to the observed object, the difference calculation module performs difference calculation on the image data obtained in different states, so as to perform spinal cord function analysis and related physiological and pathological analysis.

(1) Ultrasonic plane wave transmitting and receiving module

Determine the emission and reception sequence of an ultrasonic plane wave, based on ultra-fast ultrasonic imaging technology, to meet the requirements of high frame rate and high-quality imaging.

Specifically, to determine an efficient signal transmission and reception sequence (including the transmission of ultrasonic plane wave pulses and the reception of echo data) for ultrasonic Doppler imaging of spinal cord micro blood flow, a complete The sequence of includes multiple sub-sequences. In each sub-sequence, a set of N inclined plane waves with different angles are emitted, and the B-mode images obtained from the N inclined plane waves can be synthesized into a frame of high-quality B-mode images. The sub-sequence is repeated K times within 1 second, that is, K frames of high-quality composite B-mode images can be obtained per second. Set the sampling time as t seconds, and analyze the echo data obtained within the time t seconds.

In order to be able to observe subtle changes in blood flow between consecutive multi-frame images, and to increase the imaging frame rate as much as possible, it is necessary to calculate the achievable shortest time interval between two transmissions of ultrasonic plane waves, and calculate the theoretical maximum imaging Frame frequency, according to which the plane wave transmitting and receiving sequences are designed. Since the propagation time of ultrasonic waves within a certain distance is restricted by the propagation speed of sound waves in a specific medium, there is a minimum physical limit, and the analysis process is as follows:

Using the linear array probe to scan the area of interest with a compound plane wave, a rectangular imaging area can be observed. In order to obtain clear image information in the imaging area, the time interval between the previous and subsequent plane wave pulse transmissions should be greater than or equal to the time required for ultrasonic waves to go back and forth in the diagonal length of the rectangular area. It is known that the depth of the imaging area is d, the total length of the probe array is L, and the propagation speed of ultrasound in soft tissue is c, then the shortest time interval between two transmissions of ultrasonic plane waves is calculated as follows: The number of a group of inclined plane waves is N, then the theoretical maximum frame rate is: />

(2) Beamforming module: perform beamforming on echo data received for a period of time to obtain B-mode image data of the imaging area. That is, beamforming is performed on each group of echo data of N angles received by the ultrasound probe within a time period of t seconds to obtain continuous multi-frame high-quality composite B-mode images.

Preferably, after beamforming, coherent composite imaging is performed on each group of image data obtained by emitting multi-angle inclined plane waves, which can effectively improve the signal-to-noise ratio and resolution of the image, and obtain high-quality B-mode images.

(3) Motion calibration module: Breathing will cause the position of the spinal cord to move, and position calibration between images is required.

(4) Clutter filtering module: the received echo data includes three parts, the echo signal of static tissue, the echo signal of blood flow and noise. Therefore, it is necessary to filter the image data after motion calibration to filter out noise and static tissue signal data.

Preferably, in the present invention, the filtering adopts an eigenvalue decomposition method. The method is mainly divided into the following three steps: eigenvalue decomposition, filtering out eigenvectors and eigenvalues corresponding to clutter components, and matrix reconstruction.

(4.1) Decomposition of eigenvalues: firstly, the image data of consecutive multiple frames is constructed into a two-dimensional matrix A _m*n . The eigenvalue decomposition process is as follows:

E(A*A ^T )=λ*U*U ^T ;

Among them, U is the eigenvector matrix of m*m, λ is the diagonal matrix of m*m, and the diagonal elements are the eigenvalues of the matrix.

(4.2) Filter out the eigenvectors and eigenvalues corresponding to the clutter components. It is known that the static tissue clutter components correspond to signal components with larger eigenvalues, and the dynamic blood flow signals correspond to signal components with smaller eigenvalues. Set the first k largest eigenvalues and corresponding eigenvectors to zero to obtain a new eigenvector matrix U _k .

(4.3) Reconstruct the image matrix, as shown in the following formula, Y _m*n is the matrix of the extracted dynamic blood flow components.

Further preferably, k is selected in the following ways: 1. Direct method: directly determine the value of k, and it is recommended that k be 5% to 25% of m (the total number of eigenvalues). 2. Indirect method: Determine the eigenvalue subset corresponding to the clutter component according to the relative distribution between eigenvalues. First, the eigenvalues are arranged in descending order, and the difference between two adjacent eigenvalues λ _i -λ _i+1 drops to a certain threshold σ or the corresponding feature when the ratio λ _i /λ _i+1 drops to a certain threshold δ The serial number i of the value is used as the value of k. The recommended value of σ is 0.15-0.2, and the recommended value of δ is 1.0015-1.002. 3. Doppler frequency shift analysis method: the Doppler frequency shift of blood flow signal is higher, and the Doppler frequency shift of soft tissue signal lower. By calculating the Doppler frequency shift _fi of each eigenvector, the eigenvectors and eigenvalues to be filtered out are determined. The Doppler frequency shift after normalization ranges from 0 to 0.5. It is recommended to take a value between 0.03 and 0.04 as the cutoff frequency of soft tissue signals, and the Doppler frequency shift below the cutoff frequency. The corresponding eigenvalues are set to zero, then the clutter components can be filtered out. The subscript of the eigenvector corresponding to the cutoff frequency is k. The calculation formula of Doppler frequency shift is as follows:

Wherein, PRF is the pulse emission frequency, m is the total length of a feature vector, e _i (j) is the jth element corresponding to the i-th feature vector.

(5) Doppler imaging module: further process the extracted dynamic blood flow signals to obtain Doppler imaging results.

(6) Differentiation module: under different external effects and under normal conditions, changes in the micro-blood flow images of the spinal cord, and perform functional analysis of the spinal cord and related physiological and pathological analysis.

Features and effects of the system of the present invention

In the present invention, an ultrasonic plane wave transmitting and receiving control module is written based on the theory of ultrafast ultrasonic imaging technology and multi-angle plane wave composite imaging, and the computer software controls hardware equipment to transmit and receive ultrasonic waves; the received echo data is processed, and finally Obtain the Doppler blood flow image; and analyze the parameters such as the speed and direction of the blood flow. The system also provides a mode for imaging the micro blood flow of the spinal cord under the conditions of spinal cord compression, injury and stimulation, and can perform spinal cord functional analysis and physiological and pathological analysis.

The system of the present invention can perform high-frame frequency and high-quality imaging of spinal cord micro-blood flow; and can obtain continuous multi-frame high-resolution images of rapid blood flow to generate dynamic changes in blood flow; compared with traditional micro-blood flow imaging Methods, the system does not need to inject ultrasound contrast agents into blood vessels, and improves the resolution and signal-to-noise ratio of blood flow images through high frame rate imaging and multi-angle plane wave composite imaging. The imaging system can be used for hemodynamic analysis and spinal cord function analysis of the spinal cord.

Description of drawings

Figure 1 is a schematic diagram of the system structure.

Figure 2 is a schematic diagram of the workflow of the ultrasonic transmitting and receiving hardware modules.

Fig. 3 Schematic diagram of the furthest propagation distance of ultrasound in the imaging area.

Fig. 4 is a schematic diagram of an ultrafast ultrasonic plane wave transmitting and receiving sequence in this embodiment.

Fig. 5 is the result after image reconstruction is performed on the echo data obtained by imaging the spinal cord of the rat in this embodiment.

Fig. 6 is a scatter diagram of eigenvalues obtained by performing eigenvalue decomposition on data after beamforming in an embodiment, arranged in descending order.

Fig. 7 is a scatter diagram of the difference (λ _i -λ _i+1 ) between two adjacent eigenvalues after the eigenvalues are arranged in descending order in the embodiment.

Fig. 8 is a scatter diagram of the ratio (λ _i /λ _i+1 ) of two adjacent eigenvalues after the eigenvalues are arranged in descending order in the embodiment.

Fig. 9 is a graph showing the results of Doppler shift analysis for each eigenvector.

Fig. 10 is a dynamic blood flow diagram extracted by using an eigenvalue decomposition algorithm after position calibration of multiple frames of images in the embodiment.

Fig. 11 is a static tissue map separated by eigenvalue decomposition algorithm after position calibration of multiple frames of images in the embodiment.

Fig. 12 is a Doppler blood flow diagram obtained by imaging the spinal cord of a rat in the embodiment.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be described in detail below in conjunction with the embodiments and accompanying drawings. The specific embodiments described here are only used to explain the present invention, not to limit the present invention.

Fig. 1 is a spinal cord micro blood flow ultrasonic imaging system proposed by the present invention, and the system block diagram is also applicable to this embodiment. This embodiment uses the spinal cord micro-flow imaging system proposed by the present invention to image the rat spinal cord, but is not limited to the imaging of the rat spinal cord.

1. Design the ultrasonic transmission and reception sequence

In the present invention, ultrafast ultrasonic plane wave imaging technology is used to perform high frame rate and high quality imaging of spinal cord micro blood flow. Therefore, first of all, it is necessary to determine the shortest time interval required for two adjacent transmissions of inclined plane waves. The longest distance traveled by ultrasonic waves in one imaging is shown in Figure 3. The depth of the imaging area is d=6.308*10 ^-3 m. The total length of the probe array The length L=1.27*10 ^-2 m, the propagation speed of ultrasound in soft tissue c=1.54*10 ³ m/s, the calculated shortest time interval is: Fig. 4 is the ultrasonic plane wave transmission and reception sequence of the present embodiment, the sequence includes a plurality of subsequences, each subsequence transmits a group of oblique plane waves of 27 angles equally spaced and uniformly distributed between -10° and 10°, set The number of subsequences transmitted per second is K=520, the number of inclined plane waves contained in each subsequence N=27, and the time interval between each subsequence is t ₂ , t ₂ =1.2*10 ^-3 (s ).

2. Transmit plane wave and receive echo signal

Gently puncture the spinal cord with a fine needle to cause local injury. Fix the ultrasound probe directly above the spinal cord, set the acquisition time to 1s, collect and store the echo data. The schematic diagram of the workflow of the ultrasonic signal transmitting and receiving module is shown in Figure 2. The ultrasonic signal is first transmitted by the waveform generator to generate the transmission waveform, and the probe converts the electrical energy into acoustic energy to transmit ultrasonic waves; the echo signal is received by the same Or different ultrasonic probes, and the echo signal is processed by the analog signal amplifier and filter, followed by analog-to-digital conversion and data storage.

3. Beamforming

The beamforming algorithm is used to reconstruct the image of the echo data, and the 27 B-mode images obtained from each group are coherently superimposed to synthesize a high-quality B-mode image. The sub-sequence is repeated 520 times within 1 second, that is, 520 frames of high-quality composite B-mode images can be obtained per second. In this embodiment, the sampling duration is set to 1s, and 520 frames of composite B-mode images can be obtained after data processing. FIG. 5 is an imaging result of a frame of composite B-mode images obtained by beam synthesis in this embodiment.

4. Motion Calibration

Other factors such as heartbeat and electrical stimulation can cause the position of the spinal cord to move, so position calibration between multiple images is performed.

5. Clutter filtering

(1) Construct the data of 520 consecutive frames of composite B-mode images after motion calibration into a two-dimensional matrix A _m*n , and the eigenvalue decomposition process is as follows: E(A* ^AT )=λ*U* ^UT , where , U is the eigenvector matrix of m*m, S is the diagonal matrix of m*m, and the diagonal elements are eigenvalues. In this embodiment, the 520 eigenvalues obtained through eigenvalue decomposition are arranged in descending order as shown in FIG. 6 .

(2) Since the clutter components correspond to signal components with large eigenvalues, the first k largest eigenvalues are set to zero, and then the matrix is reconstructed to achieve clutter filtering. There are several ways to choose k:

I. Direct value selection method: k is directly selected according to the percentage p of the total number of eigenvalues, and the recommended value of p is 5% to 20%, that is, 26 to 130. In this embodiment, the total number of feature values m is 520, and the suggested value of k is 26-104.

II. Indirect value selection method: After the eigenvalues are arranged in descending order, the difference between two adjacent eigenvalues λ _i -λ _i+1 drops to a certain threshold σ or the ratio λ _i /λ _i+1 drops to a certain The serial number of the corresponding eigenvalue at the threshold δ is used as the value of k. The recommended value of σ is 0.15~0.2. Figure 7 is a scatter diagram of the difference between two adjacent eigenvalues λ _i -λ _i+1 after the eigenvalues are arranged in descending order in the embodiment. The recommended value of k in this embodiment is obtained from the figure The value is 30 to 102; δ is recommended to take a value of 1.0015 to 1.002. Figure 8 is a scatter diagram of the ratio λ _i /λ _i+1 of two adjacent eigenvalues after the eigenvalues are arranged in descending order in the embodiment, and k can be obtained from the figure The value ranges from 30 to 102.

III. Spectrum analysis method: Doppler frequency shift analysis is performed for each eigenvector, and the result is shown in FIG. 9 . As shown in the figure, the amplitude of the signal component with a relatively low Doppler frequency shift is relatively large, corresponding to the soft tissue area; in order not to filter out the blood flow signal component mixed in the clutter signal, it is recommended to take k at a Doppler frequency shift of The serial number of the feature vector corresponding to 0.03~0.04, k can take the value of 42~102.

The calculation formula of Doppler frequency shift is as follows: where PRF is the pulse emission frequency, N is the sample length, e _i (j) is the jth element corresponding to the i-th eigenvector.

(3) Reconstruct the image matrix, as shown in the following formula: Y _m*n is the matrix of the extracted dynamic blood flow components. Figure 10 is a dynamic blood flow map extracted by using the eigenvalue decomposition algorithm, and Figure 11 is a static tissue map separated by the eigenvalue decomposition algorithm.

6. Doppler flow chart

The dynamic blood flow diagram is further processed to obtain a Doppler blood flow diagram, as shown in FIG. 12 . The microvascular structure of the rat spinal cord can be seen in the picture, and the local micro blood flow loss caused by the mechanical injury of fine needle insertion at -2.5mm in the horizontal direction and -1.5mm to -2mm in the vertical direction can be observed.

Function and effect of this embodiment:

In this embodiment, the spinal cord micro-flow ultrasonic imaging system proposed by the present invention is applied, and based on the ultra-fast ultrasonic Doppler blood flow imaging technology, continuous multiple frames of high-quality spinal cord blood flow images can be obtained, and each frame of image can be observed The micro blood flow of the spinal cord can be detected, and the dynamic changes of the blood flow can be seen over a period of time. After further processing, the Doppler image of the blood flow can be obtained. In this embodiment, it can also be observed that after the spinal cord is subjected to mechanical trauma, the loss of local micro-blood flow in the injured area can be observed, so as to intuitively reflect the injury of the spinal cord. Blood flow imaging and related functional analysis.

The above embodiments are preferred examples of the present invention, and are not intended to limit the protection scope of the present invention.

Claims (4)

1. An ultrafast ultrasonic Doppler spinal cord micro-blood flow imaging system is characterized by comprising two parts, namely hardware and software; the hardware part comprises an ultrasonic signal transmitting and receiving module and an experimental equipment module; the ultrasonic signal transmitting and receiving module comprises a waveform generator, an ultrasonic probe, an analog signal amplifier, an analog signal filter, an analog-to-digital converter and a storage unit; the experimental equipment module comprises a spinal cord fixing device, a physiological signal detection device, a spinal cord pressurizing device, an electric stimulation device, an oscillograph and an arbitrary waveform generator; the software part comprises an ultrasonic pulse transmitting and receiving sequence module, a beam synthesis module, a motion calibration module, a clutter filtering and taking module, a blood flow Doppler imaging module and a difference solving module;

the workflow of the hardware part is as follows: firstly, an ultrasonic plane wave transmitting and receiving sequence set by software controls a waveform generator to generate a transmission waveform, and an ultrasonic probe converts electric energy into acoustic energy to transmit ultrasonic waves; the receiving of the echo signals is completed by the same or different ultrasonic probes, the echo signals are processed by an analog signal amplifier and an analog signal filter, then the echo signals are subjected to analog-to-digital conversion by an analog-to-digital converter, and the echo signals are sent to a storage unit to complete data storage;

in the experimental equipment module, the spinal cord fixing device is used for fixing spinal cords; quantitatively pressurizing the spinal cord with a spinal cord pressurizing device, and observing the change of hemodynamics under different pressure conditions through an oscillograph; or applying electric stimulation to the observed object by using an electric stimulation device, and exploring the relation between the spinal cord nerve function and the blood flow change through spinal cord micro-blood flow analysis so as to perform relevant physiological and pathological analysis; the physiological signal detection device is used for measuring respiration, heartbeat signals and other physiological signals in real time, and carrying out motion calibration related to respiration and heartbeat and correlation analysis of other physiological signals and blood flow signals;

the workflow of the software part is as follows: firstly, an ultrasonic plane wave transmitting and receiving sequence is compiled by an ultrasonic pulse transmitting and receiving sequence module, and a control hardware part transmits ultrasonic waves and receives echo signals; the beam synthesis module performs beam synthesis on the received echo signals to obtain a B-mode image, and performs coherent compositing on data of the B-mode image obtained by transmitting a plurality of angle ultrasonic plane waves to obtain a high-quality composite B-mode image; calibrating the position between the images by a motion calibration module; then, clutter filtering is carried out by a clutter filtering module, and then, further processing is carried out by a blood flow Doppler imaging module to obtain a Doppler blood flow graph; applying pressure to spinal cord or applying electric stimulation to an observation object, and carrying out differencing calculation on the obtained image data under different states by a differencing module so as to carry out spinal cord function analysis and related physiological and pathological analysis;

in the ultrasonic signal transmitting and receiving module, an ultrasonic plane wave transmitting and receiving sequence is determined, so that the requirements of high frame frequency and high quality imaging are met; a complete sequence comprises a plurality of subsequences, in each subsequence, a group of N oblique ultrasonic plane waves with different angles are transmitted, and a frame of high-quality composite B-mode image can be synthesized by B-mode images obtained by the N oblique ultrasonic plane waves; repeating the sub-sequence for K times within 1 second to obtain a K-frame high-quality composite B-mode image every second; setting the sampling time length as t seconds, and analyzing echo data obtained in the time t seconds;

firstly, calculating the shortest time interval which can be achieved between two ultrasonic plane waves, calculating the theoretical highest imaging frame frequency, designing ultrasonic plane wave transmitting and receiving sequences according to the theoretical highest imaging frame frequency, knowing that the depth of an imaging area is d, the total length of a probe array is L, and the propagation speed of ultrasonic waves in soft tissues is c, and calculating the shortest time interval between the two ultrasonic plane waves:the number of the group of inclined ultrasonic plane waves is N, and the highest frame frequency is: />

2. The ultra-fast ultrasonic doppler spinal cord micro-blood flow imaging system according to claim 1, wherein in the beam synthesis module, beam synthesis is performed on echo data received for a period of time to obtain data of a B-mode image of an imaging region; carrying out wave beam synthesis on each group of echo data of N angles received by an ultrasonic probe within t seconds to obtain a continuous multi-frame high-quality composite B-mode image;

after beam synthesis, coherent composite imaging is carried out on image data obtained by each group of emitted multi-angle inclined ultrasonic plane waves, so that the signal-to-noise ratio and resolution of the image are effectively improved, and a high-quality composite B-mode image is obtained.

3. The ultra-fast ultrasound doppler spinal cord micro-blood imaging system according to claim 2, wherein the clutter filtering module: because the received echo data comprises three parts, the echo signal of static tissue, the echo signal of blood flow and noise, the image data after motion calibration needs to be filtered, and the noise and the data of the static tissue signal are filtered;

the filtering adopts a characteristic value decomposition method, which comprises the following three steps of characteristic value decomposition, characteristic vector and characteristic value corresponding to clutter component filtering and matrix reconstruction;

(4.1) eigenvalue decomposition: first, three-dimensional image data of successive frames is constructed as a two-dimensional matrix, denoted A _m*n The eigenvalue decomposition process is as follows:

E(A*A ^T )＝λ*U*U ^T ；

wherein U is a characteristic vector matrix of m, lambda is a diagonal matrix of m, and diagonal elements are matrix characteristic values;

(4.2) filtering out the characteristic vector and the characteristic value corresponding to the clutter component, and knowing the signal component with larger characteristic value corresponding to the static tissue clutter component and the signal component with smaller characteristic value corresponding to the dynamic blood flow signal; zero-setting the first k largest eigenvalues and corresponding eigenvectors to obtain a new eigenvector matrix U _k ；

(4.3) reconstructing an image matrix, the calculation formula being:

Y _m*n m represents the total number of eigenvalues as a matrix of extracted dynamic blood flow components.

4. The ultra-fast ultrasonic doppler spinal cord micro-blood flow imaging system according to claim 3, wherein k is selected from the following methods in the clutter filtering module:

(1) The direct method comprises the following steps: directly determining a k value, wherein the k is 5% -25% of the total m of the characteristic values;

(2) And (3) an indirect method: determining a characteristic value subset corresponding to the clutter components according to the relative distribution among the characteristic values; firstly, the characteristic values are arranged in descending order, and the difference lambda between two adjacent characteristic values _i -λ _i+1 Falling to a certain threshold sigma or ratio lambda _i /λ _i+1 Taking the serial number i of the corresponding characteristic value as the value of k when the characteristic value is lowered to a certain threshold delta; the value of sigma is 0.15-0.2, and the value of delta is 1.0015-1.002;

(3) Doppler frequency shift analysis: by calculating Doppler shifts f of the respective eigenvectors _i Determining a feature vector and a feature value to be filtered; taking a certain value of 0.03-0.04 as the cut-off frequency of the soft tissue signal, and setting the eigenvector of the Doppler frequency shift below the cut-off frequency and the eigenvalue corresponding to the eigenvector to zero to remove clutter components; the subscript of the feature vector corresponding to the cut-off frequency is k; the calculation formula of the Doppler shift is as follows:

wherein PRF is pulse transmission frequency, e _i (j) Is the corresponding j-th element in the i-th feature vector.