Ultrasonic Doppler blood flow velocity vector imaging method and system

Abstract

The invention discloses an ultrasonic Doppler blood flow velocity vector imaging method, which comprises the following steps: using a medical ultrasonic probe, transmitting ultrasonic signals to an imaging space region at multiple different angles with different ultrasonic channel delays; the ultrasonic signals propagate and reflect in tissues and blood flows, and the ultrasonic probe receives the reflected ultrasonic signals; processing the received ultrasonic signals, and calculating the Doppler frequency signals of different transmitting and receiving angles of each voxel point in the imaging space region; for each voxel point in the imaging space region, establishing a relationship between the Doppler frequency signal and the blood flow velocity according to the transmitting angle of the ultrasonic signal transmitted by the ultrasonic probe and the receiving angle of the ultrasonic signal received, and establishing a linear matrix equation group; combining the linear matrix equation groups on all voxel points to establish a large-scale sparse matrix system; solving the large-scale sparse matrix system using the least squares method, obtaining the blood flow velocity vectors on all voxel points and generating a blood flow velocity field.

Description

Technical Field

[0001] The invention belongs to the technical field of ultrasonic imaging and relates to an ultrasonic Doppler blood flow velocity vector imaging method. Background Art

[0002] Cardiovascular disease is the number one killer of humans. Blood flow vector imaging has important clinical significance for the diagnosis of cardiovascular diseases, such as determining coronary or carotid artery stenosis, tricuspid / mitral valve stenosis and regurgitation. At present, ultrasonic Doppler technology is often used in clinical practice for blood flow imaging, but the existing Doppler imaging technology can usually only obtain the projection component of blood flow in the direction of the ultrasonic beam. The real blood flow velocity is a vector, and the existing method can only obtain the blood flow velocity projection in the direction of the ultrasonic beam. Blood flow vector imaging can assist doctors in making accurate and comprehensive cardiovascular disease diagnoses. Existing ultrasonic Doppler imaging methods (such as color ultrasound) can only obtain the velocity projection component of blood flow velocity in the direction of the ultrasonic beam. In this method, the ultrasonic transmit beam and the receive beam are usually opposite, and the Doppler frequency obtained along the direction of the ultrasonic beam can be expressed as:

[0003]

[0004] Where f0 is the ultrasonic center frequency, is the actual blood flow velocity, θ is the angle between the ultrasound transmitting beam or receiving beam and the actual blood flow velocity, c is the ultrasound velocity, and f d is the Doppler frequency. According to formula (1), we can calculate That is, the blood flow velocity component along the direction of the ultrasound beam.

[0005] This method has the following disadvantages and shortcomings: it can only obtain the blood flow velocity component along the direction of the ultrasound beam, that is, it cannot obtain the blood flow velocity vector. However, the blood flow situation is complex, especially in the case of complex lesions, the clinical significance of measuring a single blood flow velocity component is limited; the accuracy of Doppler imaging depends on the angle between the ultrasound beam and the blood flow velocity, and the closer the angle is to the vertical angle, the lower the accuracy; in the presence of lesions, the blood flow usually has eddy currents, etc., and the existing imaging methods have low accuracy and limited clinical significance. Summary of the invention

[0006] The technical solution of the present invention is to provide an ultrasonic Doppler blood flow velocity vector imaging method, comprising the following steps:

[0007] S1. Using an ultrasonic probe, setting different transmission delays for different ultrasonic transducer channels, and transmitting ultrasonic signals into an imaging space region at multiple different angles;

[0008] S2, the ultrasonic signal propagates and reflects in the blood flow and tissue, and the ultrasonic probe receives the reflected ultrasonic signal;

[0009] S3, processing the transmitted ultrasound signal and the received ultrasound signal, and calculating the Doppler frequency signal corresponding to different transmission and reception angles of each voxel point in the imaging space area;

[0010] S4. Construct a blood flow velocity vector signal using the Doppler frequency signal and the corresponding transmission angle and receiving angle of the ultrasound signal through the following steps:

[0011] For each voxel point in the imaging space area, the relationship between the Doppler frequency signal and the blood flow velocity is established according to the transmission angle and the receiving angle of the ultrasound signal, and a linear matrix equation group is established by simultaneous equations;

[0012] Combine the sparse matrices at all voxel points to establish a large-scale sparse matrix system;

[0013] The least square method is used to solve the large-scale sparse matrix system to obtain the blood flow velocity vectors of all voxel points and generate the blood flow velocity field;

[0014] S5. Generate a blood flow velocity vector field image on a display device using the blood flow velocity vector field.

[0015] This technical solution obtains multiple sets of Doppler frequencies by scanning the blood flow at each voxel point position in the imaging space region using multiple ultrasonic beams with different emission angles and reception angles. Subsequently, the obtained multiple sets of Doppler frequencies are post-processed and solved using the least squares method to obtain the blood flow velocity vector imaging in the imaging space region, thereby obtaining multiple components of the blood flow velocity.

[0016] In one embodiment of the technical solution, in the step S1, the ultrasonic probe transmits non-focused plane waves at multiple different angles by controlling the delays of different ultrasonic transducer channels.

[0017] In one embodiment of the technical solution, in the step S1, the ultrasonic probe transmits ultrasonic waves by a dynamic sub-aperture synthesis method, that is, multiple focused ultrasonic waves are used to scan the imaging space area.

[0018] In one embodiment of the technical solution, the ultrasonic probe uses a linear array probe to image the two-dimensional plane. In step S4, the relationship between the Doppler frequency signal and the blood flow velocity is:

[0019]

[0020] Among them, α is the angle between the blood flow velocity direction and the normal vector direction of the ultrasound probe, v is the blood flow velocity amplitude, and the mth emission angle of the voxel point is recorded as θ tm , the nth receiving angle is recorded as θ rn , the Doppler frequency signal using the mth transmission angle and the nth reception angle is recorded as

[0021] For the kth voxel point (k∈{1,2,…,NV}) in the imaging space, The relationship between the blood flow velocity and the linear matrix equations is established, which can be expressed as:

[0022] A k v k =u k (3)

[0023] Among them, the matrix A k ∈R M*N×2 , vector v k ∈R 2 , vector u k ∈R M*N×1 , M is the number of transmitting angles, N is the number of receiving angles; and:

[0024]

[0025]

[0026]

[0027] In one embodiment of the technical solution, the ultrasound probe uses a two-dimensional array probe to perform three-dimensional volume imaging. In step S4, the relationship between the Doppler frequency signal and the blood flow velocity is:

[0028]

[0029] in is the angle between the blood flow velocity direction and the ultrasound probe direction, v is the blood flow velocity amplitude, and the mth emission angle of the voxel point is recorded as Its nth receiving angle is recorded as like Figure 2 The Doppler frequency signal using the mth transmission angle and the nth reception angle is recorded as

[0030] For the kth voxel point (k∈{1,2,…,NV}) in the imaging space, The relationship between the blood flow velocity and the linear matrix equations is established, which can be expressed as:

[0031] A k v k =u k , (8)

[0032] Among them, the matrix A k ∈R M*N×3 , vector v k ∈R 3 Vector u∈R M*N×1, M is the number of transmitting angles, N is the number of receiving angles; and:

[0033]

[0034]

[0035]

[0036] In one embodiment of the technical solution, in step S4, the linear matrix equations of all voxel points in the entire imaging space area are combined to establish a large-scale sparse matrix system, which is recorded as:

[0037] Av=u, (12)

[0038] Where v = [v1 T ,…,v NV T ] T , u=[u1 T ,…,v NV T ] T , A is the block diagonal matrix, NV is the number of voxel points in the imaging space, and there is

[0039]

[0040] In one embodiment of the technical solution, in step S4, the least square method is used to solve formula (12), that is, the least square method formula (14) is solved to obtain the blood flow velocity vector of each voxel point,

[0041] |Av-u|2≤tol, (14)

[0042] The left side of the less than or equal to sign in formula (14) is the binary norm, and the right side is the iterative residual tol. The iterative residual tol is set to 5% to 10%. The solution can obtain the blood flow velocity components of each voxel point in multiple directions. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The above and other objects, features and advantages of the present application will become more apparent through a more detailed description of exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0044] Figure 1 Schematic diagram of the emission angle and the receiving angle when the linear array probe is used in the embodiment.

[0045] Figure 2 It is a spatial schematic diagram of the emission angle and the receiving angle when a two-dimensional array, a curved array or a 1.5D probe is used in the embodiments. DETAILED DESCRIPTION

[0046] The preferred embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0047] This embodiment is an ultrasonic Doppler blood flow velocity vector imaging method, and its application scenario is to image the blood flow in the heart and blood vessels by extracorporeal ultrasound (such as transthoracic ultrasound, etc.) or endoscopic ultrasound (such as transesophageal ultrasound, intracardiac ultrasound, etc.). The probe used can be a surface imaging probe or a volume imaging probe. The surface imaging probe can obtain two vector components of the blood flow velocity in the imaging plane, and the volume imaging probe can obtain three vector components of the blood flow velocity in the imaging body.

[0048] The ultrasound waves of this embodiment can use plane waves or dynamic sub-aperture synthesis methods to transmit ultrasound signals. When using plane waves, there are usually multiple emission angles to transmit non-focused plane wave ultrasound signals. When using dynamic sub-aperture synthesis, multiple focused ultrasound signals are usually used to scan the imaging space area in sequence.

[0049] This embodiment takes intracardiac ultrasound as an example, and uses ultrasound Doppler to perform blood flow velocity vector imaging. The specific steps are as follows:

[0050] 1. Before imaging, ultrasound blood flow signal enhancement can be performed with or without microbubble contrast agents.

[0051] 2. Perform intracardiac ultrasound imaging, insert the intracardiac ultrasound probe into the heart cavity through a peripheral vein minimally invasive intervention, and adjust the position and orientation of the ultrasound window to image the imaging space area (such as the heart cavity).

[0052] In addition, transesophageal ultrasound imaging can also be performed by inserting a transesophageal ultrasound probe into the esophagus and adjusting the position and orientation of the ultrasound window to image an imaging space area, such as the heart.

[0053] 3. Use a multi-subaperture synthesis method or a plane wave method to transmit an ultrasonic transmission signal, and then the ultrasonic probe receives a reflected ultrasonic echo signal; the ultrasonic transmission signal and the received ultrasonic echo signal are processed by a processor to extract signal information, and a blood flow velocity vector signal is constructed using the extracted signal information through the following processing method, wherein the extracted signal information includes but is not limited to a Doppler frequency signal, and a corresponding transmission angle and a receiving angle.

[0054] The specific processing method is that for each voxel point in the imaging space area, assuming that there are M transmission angles of ultrasonic transmission signals and N reception angles of ultrasonic reception signals, M*N groups of Doppler frequency signals can be collected at each voxel point, where the Doppler frequency signal using the mth transmission angle (m is 1 to M) and the nth reception angle (n is 1 to N) is recorded as It is generally obtained using the lag-one autocorrelation phase method, which is common knowledge in the art and will not be introduced here.

[0055] Extracorporeal ultrasound, transesophageal ultrasound and intracardiac ultrasound usually use linear array probes and two-dimensional array probes. The following will introduce the processing of linear array probes and two-dimensional array probes respectively:

[0056] A. Figure 1 As shown in , using a linear array probe with a linear array of transducer units for surface imaging, two blood flow velocity vector components in the imaging plane can be obtained, and the mth emission angle is recorded as θ tm , the nth receiving angle is recorded as θ rn , Doppler frequency signal The relationship with blood flow velocity v is:

[0057]

[0058] Where α is the angle between the blood flow velocity and the normal vector of the ultrasound probe, v is the velocity amplitude, and the kth voxel point (k∈{1,2,…,NV}) in the imaging space region is The relationship between the blood flow velocity and the linear matrix equations is established, which can be expressed as:

[0059] A k v k =u k , (3)

[0060] The matrix A k ∈R M*N×2 , vector v k ∈R 2 , vector u k ∈R M*N×1 . There are as follows:

[0061]

[0062]

[0063]

[0064] B. Figure 2As shown in FIG, using a planar array probe with transducer units in a planar array for volume imaging, three blood flow velocity vector components in the imaging space can be obtained. Expressed in spherical coordinates, and the plane where the transducer unit is located is recorded as the xy plane, then the mth emission angle is recorded as The nth receiving angle is recorded as The three velocity components of each voxel point k (k∈{1,2,…,NV}) in the entire imaging space can be obtained. Among them, The relationship with blood flow velocity is:

[0065]

[0066] in is the angle between the blood flow velocity direction and the ultrasound probe direction, and v is the velocity amplitude. The relationship between the blood flow velocity and the linear matrix equations can also be obtained.

[0067] A k v k =u k , (8)

[0068] The matrix A k ∈R M*N×3 , vector v k ∈R 3 , vector u∈R M*N×1 , are as follows:

[0069]

[0070]

[0071]

[0072] 4. The blood flow velocity component in the third step is the blood flow velocity vector calculation at a single voxel point. In order to measure the blood flow velocity vectors of all voxel points in the entire imaging space, it is necessary to combine the linear matrix equations at all voxel points to establish a large-scale sparse matrix system, which is recorded as:

[0073] Av=u (12)

[0074] Where v = [v1 T ,…,v NV T ] T , u=[u1 T ,…,u NV T ] T , A is the block diagonal matrix, NV is the number of voxel points in the imaging space, and:

[0075]

[0076] 5. Use the least squares method to solve the large-scale sparse matrix system. Since A is an asymmetric large-scale sparse ill-conditioned matrix (usually an overdetermined linear system), the method based on elimination is slow and often unstable. In order to improve the speed and robustness of solving formula (12), this embodiment proposes to use the sparse matrix GMRES (Generalized minimal residual method) method to solve the least squares formula (14):

[0077] |Av-u|2≤tol (14)

[0078] The left side of the less than or equal sign in formula (14) is the second norm, and the right side is the iterative residual. The iterative residual is generally set to 5% to 10% to prevent overfitting in the solution of the pathological system. The solution result v is the blood flow velocity component at each voxel point of the vectorization. The singular value decomposition SVD method can also be used to solve formula (14). However, the calculation efficiency of the SVD method is much lower than that of GMRES, and this patent will not introduce it in detail.

[0079] 5. Generate a blood flow velocity vector field from the calculated blood flow velocity vector signal of the voxels in the imaging area and display it on a display device.

[0080] As an alternative to this embodiment, the planar transducer unit array can be replaced with other forms of transducer arrays, such as curved arrays, annular arrays, 1.5D probes, etc. The various parameters of the array are not limited, and arrays with different frequencies, numbers of array elements, array element spacing, array element distribution schemes, etc. can be selected according to various application environments.

[0081] This embodiment uses the GMRES method to calculate multiple velocity components. As an alternative to this embodiment, other methods based on least squares elimination may also be used to solve the system, such as singular value decomposition and the like.

[0082] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. An ultrasonic Doppler blood flow velocity vector imaging method, characterized in that: The following steps are involved: S1. Ultrasonic probe uses a two-dimensional array probe to perform three-dimensional volume imaging, sets different transmission delays for different ultrasound channels, and transmits ultrasound signals to the imaging space area at multiple different angles; S2, the ultrasonic signal propagates and reflects in the tissue and blood flow, and the ultrasonic probe receives the reflected ultrasonic signal; S3, processing the received ultrasound signal, and calculating the Doppler frequency signal corresponding to different transmission and reception angles of each voxel point in the imaging space area; S4. Construct a blood flow velocity vector signal through the following steps: For each voxel point in the imaging space area, the relationship between the Doppler frequency signal and the blood flow velocity is established according to the transmission angle of the ultrasound probe transmitting the ultrasound signal and the receiving angle of the ultrasound signal received, which is expressed as: in is the angle between the blood flow velocity direction and the ultrasound probe direction, v is the blood flow velocity amplitude, and the mth emission angle of the voxel point is recorded as Its nth receiving angle is recorded as The Doppler frequency signal using the mth transmission angle and the nth reception angle is recorded as f0 is the ultrasonic center frequency, c is the ultrasonic velocity; For the kth voxel point (k∈{1,2,…,NV}) in the imaging space, The relationship between the blood flow velocity and the linear matrix equations is established, which can be expressed as: A k v k =u k , (8) Among them, the matrix A k ∈R M*N×3 , vector v k ∈R 3 , vector u∈R M*N×1 , M is the number of transmitting angles, N is the number of receiving angles; the following formula is obtained: The linear matrix equations at all voxel points are combined to establish a large-scale sparse matrix system, which is recorded as: Av=u, (12) Where v = [v1 T ,…,v NV T ] T , u=[u1 T ,…,u Nj T ] T , A is the block diagonal matrix, NV is the number of voxel points in the imaging space; The least squares method is used to solve the large-scale sparse matrix system to obtain the blood flow velocity vectors at all voxel points and generate the blood flow velocity field, specifically: The least square method is used to solve formula (12), that is, the least square method formula (14) is used to obtain the blood flow velocity vector of each voxel point, |Av-u|2≤tol, (14) The left side of the less than or equal to sign in formula (14) is the second norm, and the right side is the iteration residual tol. The iteration residual tol is set to 5% to 10% to prevent overfitting when solving formula (14); S5. Generate a blood flow velocity vector field image on a display device using the blood flow velocity vector field.

2. The method for ultrasonic Doppler blood flow velocity vector imaging according to claim 1, characterized in that: In the step S1, by controlling the transmission delay of different ultrasound transducer channels, the ultrasound probe transmits non-focused plane waves at multiple different angles to scan the imaging space.

3. The method for ultrasonic Doppler blood flow velocity vector imaging according to claim 1, characterized in that: In the step S1, the ultrasonic probe transmits ultrasonic waves by a dynamic sub-aperture synthesis method, that is, multiple focused ultrasonic waves are used to sequentially scan the imaging space area.

4. An ultrasonic Doppler blood flow velocity vector imaging system, characterized in that: An ultrasonic probe is included, and different transmission delays are set for different ultrasonic channels to transmit ultrasonic signals to an imaging space region at multiple different angles; the ultrasonic signal propagates and reflects in tissues and blood flow, and the ultrasonic probe receives the reflected ultrasonic signal; The processor processes the transmitted ultrasound signal and receives the reflected ultrasound signal, calculates the Doppler frequency signal of each voxel point at different transmission and reception angles in the imaging space area, and constructs the blood flow velocity vector signal through the following steps: For each voxel point in the imaging space area, the relationship between the Doppler frequency signal and the blood flow velocity is established according to the transmission angle of the ultrasound probe transmitting the ultrasound signal and the receiving angle of the ultrasound signal received, and a linear matrix is ​​established, which is expressed as: in is the angle between the blood flow velocity direction and the ultrasound probe direction, v is the blood flow velocity amplitude, and the mth emission angle of the voxel point is recorded as Its nth receiving angle is recorded as The Doppler frequency signal using the mth transmission angle and the nth reception angle is recorded as f0 is the ultrasonic center frequency, c is the ultrasonic velocity; For the kth voxel point (k∈{1,2,…,NV}) in the imaging space, The relationship between the blood flow velocity and the linear matrix equations is established, which can be expressed as: A k v k =u k , (8) Among them, the matrix A k ∈R M*N×3 , vector v k ∈R 3 , vector u∈R M*N×1 , M is the number of transmitting angles, N is the number of receiving angles; the following formula is obtained: The linear matrices at all voxel points are combined to establish a large-scale sparse matrix system, which is recorded as: Av=u, (12) Where v = [v1 T ,…,v NV T ] T , u=[u1 T ,…,u NV T ] T , A is the block diagonal matrix, NV is the number of voxel points in the imaging space; The least squares method is used to solve the large-scale sparse matrix system to obtain the blood flow velocity vectors at all voxel points and generate the blood flow velocity field, specifically: The least square method is used to solve formula (12), that is, the least square method formula (14) is used to obtain the blood flow velocity vector of each voxel point, |Av-u|2≤tol, (14) The left side of the less than or equal to sign in formula (14) is the second norm, and the right side is the iteration residual tol. The iteration residual tol is set to 5% to 10% to prevent overfitting when solving formula (14); and The display device is configured to display the blood flow velocity vector field.

Images