Denoising methods and apparatus for blood flow images

Abstract

This specification provides a method and apparatus for denoising blood flow images. Based on this method, signal data from multiple channels across multiple rounds targeting a target region can be acquired first; then, multiple initial blood flow images are obtained through image reconstruction; a first data matrix is ​​constructed based on these initial blood flow images; and a second data matrix is ​​obtained by performing a preset filtering process on the first data matrix; a first auxiliary matrix and a second auxiliary matrix are constructed based on the signal data from multiple channels across multiple rounds; a preset correlation operation is performed on the first and second auxiliary matrices to obtain a correlation matrix; and the blood flow image corresponding to the second data matrix is ​​processed using the correlation matrix to obtain the corresponding denoised blood flow image. This fully utilizes the lack of correlation between noise and the correlation between blood flow signals, removing noise from the image while preserving the blood flow signal, resulting in a blood flow image with a high signal-to-noise ratio that clearly displays small vessel signals.

Description

Technical Field

[0001] This specification belongs to the field of image processing technology, and in particular relates to methods and apparatus for denoising blood flow images. Background Technology

[0002] Blood flow images obtained through ultrasound imaging typically have a low signal-to-noise ratio because they are obtained by penetrating deep tissues with non-focused waves. This often makes it difficult to clearly identify small blood vessel signals in the images.

[0003] Existing methods often apply depth-based time gain compensation (TGC) to the relevant signals. However, after compensation, the corresponding background noise is also amplified to the same extent. Since this background noise is spatially variable during depth-related TGC and beamforming, spatially independent noise is mixed into the blood flow signal during subsequent extraction, making it difficult to accurately extract signals from small blood vessels at more distant locations.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This specification provides a method and apparatus for denoising blood flow images, which can make full use of the uncorrelation of noise and the correlation of blood flow signals, effectively removing noise in the image while preserving the blood flow signal, and obtaining a blood flow image with a high signal-to-noise ratio that can clearly display small blood vessel signals.

[0006] This specification provides a method for denoising blood flow images, including:

[0007] Acquire signal data from multiple channels across multiple rounds for a target area; where the signal data from one channel corresponds to one emission angle;

[0008] Based on signal data from multiple channels across multiple rounds, multiple initial blood flow images are obtained through image reconstruction; each initial blood flow image corresponds to one round.

[0009] A first data matrix is ​​constructed based on multiple initial blood flow images; and a second data matrix is ​​obtained by performing a preset filtering process on the first data matrix.

[0010] Based on the signal data from multiple channels in multiple rounds, construct the first auxiliary matrix and the second auxiliary matrix;

[0011] Based on the first auxiliary matrix and the second auxiliary matrix, perform the preset correlation operation to obtain the correlation matrix;

[0012] The blood flow image corresponding to the second data matrix is ​​processed using the correlation matrix to obtain the corresponding denoised blood flow image.

[0013] In one embodiment, a first data matrix is ​​constructed based on multiple initial blood flow images, including:

[0014] Based on the acquisition time, the data corresponding to multiple initial blood flow images are arranged along the time direction to obtain a three-dimensional data matrix, which serves as the first data matrix. The first and second dimensions of the first data matrix are used to represent data information in the planar direction, and the third dimension of the first data matrix is ​​used to represent data information in the time direction.

[0015] In one embodiment, a second data matrix is ​​obtained by performing a preset filtering process on a first data matrix, comprising:

[0016] The first and second dimensions of the first data matrix are converted into one-dimensional data to obtain the third data matrix;

[0017] Determine the eigenvalue matrix and eigenvector matrix for the third data matrix;

[0018] Adjust the eigenvalues ​​in the eigenvalue matrix whose values ​​are greater than a preset first data threshold and / or whose values ​​are less than a preset second data threshold to obtain the adjusted eigenvalue matrix;

[0019] Based on the adjusted eigenvalue matrix and eigenvector matrix, the fourth data matrix is ​​obtained; the fourth data matrix contains image information of blood flow.

[0020] The second data matrix is ​​obtained by performing an inverse transformation on the fourth data matrix.

[0021] In one embodiment, acquiring signal data from multiple channels across multiple rounds for a target area includes: acquiring signal data from multiple channels for the current round of the target area in the following manner:

[0022] The ultrasound imaging system is controlled to start from a specified negative starting angle and change the emission angle of the ultrasound imaging system according to a preset deflection rule until the positive ending angle ends; and the ultrasound imaging system is controlled to emit plane waves to the target area based on different emission angles to acquire signal data of one channel corresponding to the emission angle; wherein the angle value of the specified negative starting angle is the same as the angle value of the positive ending angle.

[0023] In one embodiment, multiple initial blood flow images are obtained through image reconstruction based on signal data from multiple channels across multiple rounds, including:

[0024] The initial blood flow image for the current round is obtained through image reconstruction based on signal data from multiple channels in the current round, using the following method:

[0025] The signal data from each channel are used to calculate the signal intensity of each pixel in the image area based on the transmission and reception time of the ultrasound signal, resulting in multiple frames of ultrasound images; each frame of ultrasound image corresponds to the signal data of one channel in the current round.

[0026] The initial blood flow image for the current round is obtained by superimposing and summing multiple ultrasound images.

[0027] In one embodiment, constructing a first auxiliary matrix and a second auxiliary matrix based on signal data from multiple channels across multiple rounds includes:

[0028] Multiple ultrasound images with negative emission angles are selected from signal data from multiple channels in multiple rounds, and then superimposed and summed to obtain the first type of summation result; the first type of summation result is then arranged along the time direction to obtain the first auxiliary matrix;

[0029] Multiple ultrasound images with positive emission angles are selected from signal data from multiple channels in multiple rounds, and then superimposed and summed to obtain a second type of summation result. The second type of summation result is then arranged along the time direction to obtain a second auxiliary matrix.

[0030] In one embodiment, a preset correlation operation is performed based on the first auxiliary matrix and the second auxiliary matrix to obtain the correlation matrix, including:

[0031] Extract the first vector and the second vector corresponding to the same plane position from the first auxiliary matrix and the second auxiliary matrix respectively, perform relevant calculations, and obtain the relevant value corresponding to the plane position;

[0032] The correlation matrix is ​​obtained by combining the correlation values.

[0033] In one embodiment, the blood flow image corresponding to the second data matrix is ​​processed using a correlation matrix to obtain the corresponding denoised blood flow image, including:

[0034] Multiply the correlation matrix by the blood flow image corresponding to the second data matrix to obtain the corresponding denoised blood flow image.

[0035] This specification also provides a noise reduction device for blood flow images, comprising:

[0036] The acquisition module is used to acquire signal data from multiple channels in multiple rounds for a target area; wherein, the signal data of one channel corresponds to one emission angle;

[0037] The reconstruction module is used to obtain multiple initial blood flow images by image reconstruction based on signal data from multiple channels in multiple rounds; wherein, one initial blood flow image corresponds to one round.

[0038] The filtering module is used to construct a first data matrix based on multiple initial blood flow images; and to obtain a second data matrix based on the first data matrix by performing a preset filtering process.

[0039] A construction module is used to construct a first auxiliary matrix and a second auxiliary matrix based on signal data from multiple channels in multiple rounds;

[0040] The correlation processing module is used to perform preset correlation operations based on the first auxiliary matrix and the second auxiliary matrix to obtain the correlation matrix;

[0041] The denoising module is used to process the blood flow image corresponding to the second data matrix using the correlation matrix to obtain the corresponding denoised blood flow image.

[0042] This specification also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, perform the following steps: acquiring signal data from multiple channels across multiple rounds for a target region; wherein, the signal data from one channel corresponds to one emission angle; obtaining multiple initial blood flow images through image reconstruction based on the signal data from multiple channels across multiple rounds; wherein, one initial blood flow image corresponds to one round; constructing a first data matrix based on the multiple initial blood flow images; and obtaining a second data matrix by performing a preset filtering process on the first data matrix; constructing a first auxiliary matrix and a second auxiliary matrix based on the signal data from multiple channels across multiple rounds; performing a preset correlation operation on the first auxiliary matrix and the second auxiliary matrix to obtain a correlation matrix; and processing the blood flow image corresponding to the second data matrix using the correlation matrix to obtain a corresponding denoised blood flow image.

[0043] Based on the blood flow image denoising method and apparatus provided in this specification, signal data from multiple channels across multiple rounds targeting a target region can be acquired first; then, multiple initial blood flow images can be obtained through image reconstruction; a first data matrix can be constructed based on the multiple initial blood flow images; and a second data matrix can be obtained by performing a preset filtering process on the first data matrix; a first auxiliary matrix and a second auxiliary matrix can be constructed based on the signal data from multiple channels across multiple rounds; a preset correlation operation can be performed on the first auxiliary matrix and the second auxiliary matrix to obtain a correlation matrix; and the blood flow image corresponding to the second data matrix can be processed using the correlation matrix to obtain the corresponding denoised blood flow image. This fully utilizes the lack of correlation between noise and the correlation between blood flow signals, effectively removing noise from the image while preserving the blood flow signal, resulting in a blood flow image with a high signal-to-noise ratio that clearly displays small vessel signals. Attached Figure Description

[0044] To more clearly illustrate the embodiments of this specification, the accompanying drawings used in the embodiments will be briefly introduced below. The drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a schematic flowchart of a blood flow image denoising method provided in one embodiment of this specification;

[0046] Figure 2 This is a schematic diagram of an embodiment of the blood flow image denoising method provided in the embodiments of this specification, applied in a scenario example.

[0047] Figure 3 This is a schematic diagram of an embodiment of the blood flow image denoising method provided in the embodiments of this specification, applied in a scenario example.

[0048] Figure 4 This is a schematic diagram of an embodiment of the blood flow image denoising method provided in the embodiments of this specification, applied in a scenario example.

[0049] Figure 5 This is a schematic diagram of the structural composition of a server provided in one embodiment of this specification;

[0050] Figure 6 This is a schematic diagram of the structural composition of a blood flow image denoising device provided in one embodiment of this specification. Detailed Implementation

[0051] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0052] See Figure 1 and Figure 2 As shown in the embodiments of this specification, a method for denoising blood flow images is provided. In specific implementation, this method may include the following:

[0053] S101: Acquire signal data from multiple channels in multiple rounds targeting the target area; where the signal data of one channel corresponds to one emission angle;

[0054] S102: Based on the signal data from multiple channels in multiple rounds, multiple initial blood flow images are obtained through image reconstruction; wherein, one initial blood flow image corresponds to one round;

[0055] S103: Based on multiple initial blood flow images, a first data matrix is ​​constructed; and based on the first data matrix, a second data matrix is ​​obtained by performing a preset filtering process.

[0056] S104: Construct a first auxiliary matrix and a second auxiliary matrix based on the signal data from multiple channels in multiple rounds;

[0057] S105: Perform a preset correlation operation based on the first auxiliary matrix and the second auxiliary matrix to obtain the correlation matrix;

[0058] S106: Process the blood flow image corresponding to the second data matrix using the correlation matrix to obtain the corresponding denoised blood flow image.

[0059] In some embodiments, when implemented, an ultrasound imaging system based on blood flow imaging technology can be used to acquire signal data from multiple channels in multiple rounds for a target area.

[0060] Specifically, the target area can be a body area with blood vessels in an animal or a body area with blood vessels in a patient.

[0061] The aforementioned multiple rounds include several different rounds, each with a different acquisition time. Each round corresponds to an acquisition cycle, containing signal data from multiple channels within that cycle. Each channel's signal data corresponds to a transmission angle, which can be understood as the ultrasound signal data acquired by the ultrasound imaging system after emitting a plane wave based on that angle.

[0062] In some embodiments, the acquisition of signal data from multiple channels in multiple rounds for a target area described above may specifically include the following: acquiring signal data from multiple channels in the current round for the target area in the following manner:

[0063] The ultrasound imaging system is controlled to start from a specified negative starting angle and change the emission angle of the ultrasound imaging system according to a preset deflection rule until the positive ending angle ends; and the ultrasound imaging system is controlled to emit plane waves to the target area based on different emission angles to acquire signal data of one channel corresponding to the emission angle; wherein the angle value of the specified negative starting angle is the same as the angle value of the positive ending angle.

[0064] For details, please refer to Figure 3 As shown, the negative angle can be understood as an angle to the left, and the positive angle can be understood as an angle to the right.

[0065] The specified negative starting angle can specifically refer to the angle at which the ultrasound imaging system begins acquiring signal data in a cycle, for example, -5°; the specified positive ending angle can specifically refer to the angle at which the ultrasound imaging system ends acquiring signal data in a cycle, for example, 5°. The absolute values ​​of the specified negative starting angle and the positive ending angle are the same, which can also be understood as the specified negative starting angle and positive ending angle being symmetrical about the midline. It should be noted that the specified negative starting angle and positive ending angle listed above are only illustrative. In actual implementation, an appropriate negative angle can be set as the negative starting angle, and an appropriate positive angle as the positive starting angle, based on the specific range of the target area and the performance parameters of the ultrasound imaging system. This specification does not impose any limitations on this.

[0066] The aforementioned preset deflection rule can specifically be an equal-interval deflection rule. For example, based on the preset deflection rule, if the interval angle is determined to be 1°, and the first emission angle is a specified negative starting angle of -5°, then the second emission angle is the first emission angle plus the interval angle, i.e., -4°. Similarly, the third, fourth, fifth, sixth, seventh, eighth, ninth, tenth, and eleventh emission angles can be sequentially determined to be -3°, -2°, -1°, 0°, 1°, 2°, 3°, 4°, and 5°, respectively.

[0067] Furthermore, the preset deflection rule can also be a deflection rule that requires symmetrical emission angles. Specifically, for example, depending on the specific situation, the emission angles on the left can be set sequentially as -5° and -3°; correspondingly, based on the preset deflection rule, the emission angles on the right can be symmetrically set sequentially as 3° and 5° according to the emission angles on the left; in addition, the angle corresponding to the centerline, 0°, can also be set as an emission angle.

[0068] Taking the acquisition of signal data from multiple channels in the current round of multiple rounds as an example, in specific implementation, the emission angles can be determined according to the preset deflection rules, combined with the specified negative and positive starting angles, as described above. Then, the ultrasound imaging system is controlled to start emitting a plane wave beam from the specified negative starting angle (i.e., the first emission angle) towards the target area, and the corresponding signal data is acquired as the signal data for one channel corresponding to the first emission angle. After acquiring the signal data for one channel, the ultrasound imaging system is controlled to deflect to the next emission angle, and the above process is repeated to acquire the signal data for one channel corresponding to the next emission angle. This process continues until the ultrasound imaging system is controlled to deflect to the specified positive ending angle (i.e., the last emission angle), emitting a plane wave beam towards the target area, and acquiring the signal data for one channel corresponding to the last emission angle. This completes the acquisition of signal data from multiple channels in the current round.

[0069] Furthermore, it checks whether the number of rounds in the current round is less than the preset number of rounds. If it is determined that the number of rounds in the current round is less than the preset number of rounds, the above process can be repeated to acquire signal data from multiple channels in the next round after the current round. If it is determined that the number of rounds in the current round is equal to the preset number of rounds, the acquisition can be stopped, thus acquiring signal data from multiple channels across multiple rounds.

[0070] Furthermore, when specifically using an ultrasonic imaging system to transmit plane waves, the frame rate can be set to 1500Hz. The ultrasonic transducer used in the aforementioned ultrasonic imaging system can specifically be a linear transducer.

[0071] In some embodiments, after acquiring signal data from multiple channels in multiple rounds using an ultrasound imaging system, the signal data from these multiple channels can be amplified and filtered using the hardware built into the ultrasound imaging system to initially reduce data errors. The amplified and filtered signal data from these multiple channels is then provided to a server (or other computer equipment) for subsequent data processing.

[0072] In some embodiments, the above-mentioned method of obtaining multiple initial blood flow images through image reconstruction based on signal data from multiple channels in multiple rounds can, in specific implementation, include obtaining the initial blood flow image for the current round through image reconstruction based on signal data from multiple channels in the current round in the following manner:

[0073] S1: Calculate the signal intensity of each pixel in the image area based on the transmission and reception times of the ultrasound signal from the signal data of each channel to obtain multiple frames of ultrasound images; where each frame of ultrasound image corresponds to the signal data of one channel in the current round.

[0074] S2: Based on multiple ultrasound images, the initial blood flow image for the current round is obtained by superimposing and summing the images.

[0075] Specifically, for example, based on the signal data from multiple channels in the current round, five ultrasound images corresponding to emission angles of -5°, -3°, 0°, 3°, and 5° can be obtained in the manner described above. The initial blood flow image of the current round can then be obtained by superimposing and summing the five ultrasound images.

[0076] Using the above method, signal data from multiple channels in other rounds can be processed separately, thereby obtaining the initial blood flow images for each round.

[0077] It should be noted that the initial blood flow image obtained using the above method has a low signal-to-noise ratio, and the signals from small blood vessels are inherently weak. These signals are often masked by background noise such as image noise and tissue noise, making it impossible to accurately extract the small blood vessel signals. To obtain a blood flow image with clearer small blood vessel signals, a higher signal-to-noise ratio, and better image quality, further denoising of the initial blood flow image is necessary.

[0078] In some embodiments, the first data matrix is ​​constructed based on multiple initial blood flow images. In specific implementations, it may include the following:

[0079] Based on the acquisition time, the data corresponding to multiple initial blood flow images are arranged along the time direction to obtain a three-dimensional data matrix, which serves as the first data matrix. The first and second dimensions of the first data matrix represent data information in the planar direction, while the third dimension represents data information in the time direction. Each data value in the first data matrix represents the data information of a point in the target region along the planar direction at the corresponding acquisition time.

[0080] In some embodiments, see Figure 4 As shown, the second data matrix obtained by performing a preset filtering process on the first data matrix can, in specific implementation, include the following:

[0081] S1: Convert the first and second dimension data in the first data matrix into one dimension data to obtain the third data matrix;

[0082] S2: Determine the eigenvalue matrix and eigenvector matrix for the third data matrix;

[0083] S3: Adjust the eigenvalues ​​in the eigenvalue matrix whose values ​​are greater than a preset first data threshold and / or whose values ​​are less than a preset second data threshold to obtain the adjusted eigenvalue matrix;

[0084] S4: Based on the adjusted eigenvalue matrix and eigenvector matrix, the fourth data matrix is ​​obtained; the fourth data matrix contains image information of blood flow.

[0085] S5: Perform an inverse transformation on the fourth data matrix to obtain the second data matrix.

[0086] Wherein, the aforementioned preset first data threshold is greater than the preset second data domain. The blood flow image corresponding to the aforementioned second data matrix can be understood as the blood flow image obtained after performing preset wave processing on the initial blood flow image.

[0087] In specific adjustments, the eigenvalues ​​in the eigenvalue matrix that are greater than a preset first data threshold and / or less than a preset second data threshold can be adjusted to 0 to obtain the adjusted eigenvalue matrix. Alternatively, a preset first number of eigenvalues ​​in the eigenvalue matrix can be adjusted to 0, and / or a preset second number of eigenvalues ​​in the eigenvalue matrix that are lower in the eigenvalue matrix can be adjusted to 0.

[0088] In specific implementation, the aforementioned eigenvalue matrix is ​​adjusted by setting eigenvalues ​​greater than a preset first data threshold and / or eigenvalues ​​less than a preset second data threshold. This adjustment can include the following: evaluating the noise level of the initial blood flow image; determining the eigenvalue adjustment range based on the noise level of the initial image; and adjusting the values ​​of the preset first data threshold and / or the preset second data threshold based on the eigenvalue adjustment range. Thus, the specific number of eigenvalues ​​adjusted in the eigenvalue matrix can be controlled by adjusting the values ​​of the preset first data threshold and / or the preset second data threshold according to the eigenvalue adjustment range.

[0089] Specifically, for example, when the initial image has a high level of noise, the eigenvalue adjustment range can be set relatively large. This will adjust a relatively large number of eigenvalues ​​in the eigenvalue matrix to 0, thus focusing more on reducing the data information involving noise in the image. Conversely, when the initial image has a low level of noise, the eigenvalue adjustment range can be set relatively small. This will adjust a relatively small number of eigenvalues ​​in the eigenvalue matrix to 0, thus focusing more on preserving the data information in the image.

[0090] In specific adjustments, historical experience can be incorporated. Furthermore, adjustments can be made and tested simultaneously; based on test feedback, previous adjustments can be further refined. By following this method multiple times, a better eigenvalue matrix for the broom can be obtained.

[0091] The second data matrix obtained based on the adjusted eigenvalue matrix and eigenvector matrix in the above manner can effectively suppress noisy image information compared to the previous first data, and can better preserve the required blood flow image information in a targeted manner.

[0092] In some embodiments, the construction of the first auxiliary matrix and the second auxiliary matrix based on signal data from multiple channels in multiple rounds may include the following:

[0093] S1: Select multiple frames of ultrasound images with negative emission angles from the signal data of multiple channels in multiple rounds, and sum them to obtain the first type of summation result; then arrange the first type of summation result along the time direction to obtain the first auxiliary matrix;

[0094] S2: Select multiple ultrasound images with positive emission angles from the signal data of multiple channels in multiple rounds, and sum them to obtain the second type of summation result; then arrange the second type of summation result along the time direction to obtain the second auxiliary matrix.

[0095] Specifically, for example, multiple ultrasound images corresponding to emission angles of -5° and -3° can be superimposed and summed to obtain a first type of summation result. This first type of summation result is then arranged along the time direction to obtain a three-dimensional data matrix, which serves as the first auxiliary matrix. Similarly, multiple ultrasound images corresponding to emission angles of 5° and 3° can be superimposed and summed to obtain a second type of summation result. This second type of summation result is then arranged along the time direction to obtain another three-dimensional data matrix, which serves as the second auxiliary matrix. The third dimension of both the first and second auxiliary matrices is used to represent data information along the time direction.

[0096] In some embodiments, the construction of the first auxiliary matrix and the second auxiliary matrix based on signal data from multiple channels in multiple rounds may further include: selecting multiple ultrasound images with a corresponding negative first emission angle from the signal data from multiple channels in multiple rounds; arranging the multiple ultrasound images with a corresponding negative first emission angle along the time direction to obtain the first auxiliary matrix; selecting multiple sub-images with a corresponding positive second emission angle from the signal data from multiple channels in multiple rounds; and arranging the multiple ultrasound images with a corresponding positive second emission angle along the time direction to obtain the second auxiliary matrix; wherein the angle values ​​of the first angle and the second angle are the same.

[0097] Specifically, for example, multiple rounds of ultrasound images corresponding to the emission angle -5° (negative first angle) can be directly acquired; and the ultrasound images can be arranged along the time direction to obtain a first auxiliary matrix; similarly, multiple rounds of ultrasound images corresponding to the emission angle 5° (positive second angle) can be directly acquired; and the ultrasound images can be arranged along the time direction to obtain a second auxiliary matrix.

[0098] In some embodiments, the above-mentioned correlation operation based on the first auxiliary matrix and the second auxiliary matrix is ​​performed to obtain the correlation matrix. In specific implementation, it may include the following: extracting the first vector and the second vector corresponding to the same plane position from the first auxiliary matrix and the second auxiliary matrix respectively, performing correlation calculation to obtain the correlation value corresponding to the plane position; and combining the correlation values ​​to obtain the correlation matrix.

[0099] In some embodiments, the above-described processing of the blood flow image corresponding to the second data matrix using the correlation matrix to obtain the corresponding denoised blood flow image may specifically include: multiplying the correlation matrix with the blood flow image corresponding to the second data matrix to obtain the corresponding denoised blood flow image. This effectively utilizes the lack of correlation between noise and the correlation between blood flow signals, preserving the blood flow signal while effectively removing noise, resulting in a blood flow image with a high signal-to-noise ratio and good image quality.

[0100] In some embodiments, after obtaining the corresponding denoised blood flow image, the method may further include the following: performing image comparison between the denoised blood flow image and a preset healthy blood flow image template to obtain the corresponding comparison result; and detecting whether there is an anomaly in the target area based on the comparison result.

[0101] As can be seen from the above, the blood flow image denoising method provided in the embodiments of this specification can first acquire signal data from multiple channels in multiple rounds targeting the target region; then, through image reconstruction, obtain multiple initial blood flow images; construct a first data matrix based on the multiple initial blood flow images; and obtain a second data matrix by performing a preset filtering process on the first data matrix; construct a first auxiliary matrix and a second auxiliary matrix based on the signal data from multiple channels in multiple rounds; perform a preset correlation operation on the first auxiliary matrix and the second auxiliary matrix to obtain a correlation matrix; and process the blood flow image corresponding to the second data matrix using the correlation matrix to obtain the corresponding denoised blood flow image. This fully utilizes the uncorrelation of noise and the correlation of blood flow signals, effectively removing noise from the image while preserving the blood flow signal, resulting in a blood flow image with a high signal-to-noise ratio that clearly displays small vessel signals. Furthermore, it can reduce denoising processing costs and improve overall processing efficiency.

[0102] In a specific scenario example, see Figure 2 As shown, the blood flow image denoising method provided in this manual can be applied to specifically acquire and denoise blood flow images.

[0103] S1: An ultrasound imaging system is used to transmit and receive ultrasound signals. The ultrasound waves are transmitted using a multi-angle plane wave transmission method, as shown in the following diagram. Figure 3 As shown, it can be emitted first along a specified negative direction angle (e.g., a specified negative starting angle), and the deflection angle of subsequent emissions increases by an equal amount. The number of emissions is generally set to an odd number. The ultrasonic transducer used is a linear transducer. The frame rate for a single emission must reach more than 1500 Hz. The ultrasonic imaging hardware system amplifies and filters the acquired signals and then transmits them to the computer terminal for reconstruction processing in the software.

[0104] S2: Image Reconstruction. The data acquired by the ultrasound imaging system is raw channel data (e.g., signal data from multiple channels in multiple rounds), which cannot be directly used for imaging. Due to the time delay between each channel during ultrasound signal transmission, the reconstruction process involves calculating the signal intensity of each pixel in the imaging area based on the transmission and reception times of the ultrasound signal from the raw channel data acquired in each frame. Finally, each reconstructed frame within a complete acquisition set is summed (e.g., if a sub-acquisition set contains -5°, -3°, 0, 3°, and 5°, the data from these transmissions can be reconstructed, calculated, and then summed) to obtain an image from one acquisition (e.g., the initial blood flow image corresponding to one round).

[0105] S3: After reconstructing the time-series acquired dataset, a three-dimensional data matrix is ​​obtained. The third dimension represents the time direction. This data matrix is ​​then subjected to a pre-defined SVD filter, specifically: the first two dimensions of the three-dimensional matrix are transformed into one dimension, resulting in a two-dimensional matrix (e.g., the third data matrix), where the second dimension represents information in the time direction. The eigenvalues ​​and their corresponding eigenvectors of this matrix are then calculated, and the information corresponding to the preceding and following eigenvalues ​​is removed. After multiple adjustments, eigenvalues ​​and their corresponding eigenvectors retaining only blood flow information are obtained. Then, through inverse transformation and other processing, image data containing blood flow information is obtained (e.g., the second data matrix).

[0106] S4: Take the single-frame data reconstructed from S2 corresponding to each emission angle, sum the data for negative or positive emission angles (e.g., sum the reconstructed data for -5° and -3°, and sum the reconstructed data for 5° and 3°), and then arrange the reconstructed data for negative or positive emission along the time direction to form two three-dimensional matrices (e.g., a first auxiliary matrix and a second auxiliary matrix). The third dimension is also along the time direction. In the first two-dimensional plane of the two three-dimensional matrices, each pixel represents a signal along the time direction. Then, perform correlation calculation on the two signals corresponding to the same pixel position (at the same time) to obtain a correlation value; after calculating for all pixels, a two-dimensional correlation matrix can be obtained.

[0107] S5: Multiplying the matrix obtained after the preset SVD filtering with the correlation matrix yields the desired denoised image (e.g., a denoised blood flow image). This utilizes the lack of correlation between noise and the correlation between blood flow signals to remove noise while preserving the blood flow signal.

[0108] Based on the above method, blood flow signal noise can be effectively filtered out. Specifically, a correlation matrix can be obtained by performing correlation calculations on signals at different emission angles, and then the correlation matrix can be used for filtering. Furthermore, the filtering method employed is well-suited for both plane waves and scattered waves.

[0109] The above scenario examples also verify that the blood flow image denoising method provided in the embodiments of this specification can indeed effectively remove noise in the image while preserving the blood flow signal by making full use of the uncorrelation of noise and the correlation of blood flow signal, resulting in a blood flow image with a high signal-to-noise ratio that can clearly show the small blood vessel signal.

[0110] This specification also provides a server, including a processor and a memory for storing processor-executable instructions. Specifically, the processor can perform the following steps according to the instructions: acquiring signal data from multiple channels across multiple rounds for a target area; wherein, the signal data from one channel corresponds to one emission angle; obtaining multiple initial blood flow images through image reconstruction based on the signal data from multiple channels across multiple rounds; wherein, one initial blood flow image corresponds to one round; constructing a first data matrix based on the multiple initial blood flow images; obtaining a second data matrix by performing a preset filtering process on the first data matrix; constructing a first auxiliary matrix and a second auxiliary matrix based on the signal data from multiple channels across multiple rounds; performing a preset correlation operation on the first auxiliary matrix and the second auxiliary matrix to obtain a correlation matrix; and processing the blood flow image corresponding to the second data matrix using the correlation matrix to obtain a corresponding denoised blood flow image.

[0111] To execute the above instructions more accurately, please refer to... Figure 5 As shown in the embodiments of this specification, another specific server is also provided, wherein the server includes a network communication port 501, a processor 502 and a memory 503, and the above structures are connected by internal cables so that the various structures can perform specific data interaction.

[0112] Specifically, the network communication port 501 can be used to acquire signal data from multiple channels in multiple rounds targeting the target area; wherein, the signal data of one channel corresponds to one transmission angle.

[0113] The processor 502 is specifically configured to obtain multiple initial blood flow images through image reconstruction based on signal data from multiple channels across multiple rounds; wherein each initial blood flow image corresponds to one round; construct a first data matrix based on the multiple initial blood flow images; and obtain a second data matrix by performing a preset filtering process on the first data matrix; construct a first auxiliary matrix and a second auxiliary matrix based on signal data from multiple channels across multiple rounds; perform a preset correlation operation on the first auxiliary matrix and the second auxiliary matrix to obtain a correlation matrix; and process the blood flow image corresponding to the second data matrix using the correlation matrix to obtain the corresponding denoised blood flow image.

[0114] The memory 503 can be used to store the corresponding instruction program.

[0115] In this embodiment, the network communication port 501 can be a virtual port bound to different communication protocols, thereby enabling the sending or receiving of different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0116] In this embodiment, the processor 502 can be implemented in any suitable manner. For example, the processor can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc. This specification is not limiting.

[0117] In this embodiment, the memory 503 may include multiple layers. In a digital system, anything that can store binary data can be a memory. In an integrated circuit, a circuit with storage function but no physical form is also called a memory, such as RAM, FIFO, etc. In a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0118] This specification also provides a computer-readable storage medium for the above-described blood flow image denoising method. The computer-readable storage medium stores computer program instructions that, when executed, implement the following: acquiring signal data from multiple channels across multiple rounds for a target region; wherein each channel's signal data corresponds to a transmission angle; obtaining multiple initial blood flow images through image reconstruction based on the signal data from multiple channels across multiple rounds; wherein each initial blood flow image corresponds to one round; constructing a first data matrix based on the multiple initial blood flow images; obtaining a second data matrix by performing a preset filtering process on the first data matrix; constructing a first auxiliary matrix and a second auxiliary matrix based on the signal data from multiple channels across multiple rounds; performing a preset correlation operation on the first auxiliary matrix and the second auxiliary matrix to obtain a correlation matrix; and processing the blood flow image corresponding to the second data matrix using the correlation matrix to obtain the corresponding denoised blood flow image.

[0119] In this embodiment, the storage medium includes, but is not limited to, Random Access Memory (RAM), Read-Only Memory (ROM), cache, hard disk drive (HDD), or memory card. The memory can be used to store computer program instructions. The network communication unit can be an interface configured according to standards specified in the communication protocol for network connection communication.

[0120] In this embodiment, the specific functions and effects implemented by the program instructions stored in the computer-readable storage medium can be explained in comparison with other embodiments, and will not be repeated here.

[0121] See Figure 6 As shown, at the software level, this specification also provides a denoising device for blood flow images, which may specifically include the following structural modules:

[0122] The acquisition module 601 is specifically used to acquire signal data from multiple channels in multiple rounds for a target area; wherein, the signal data of one channel corresponds to one emission angle;

[0123] The reconstruction module 602 is specifically used to obtain multiple initial blood flow images by image reconstruction based on signal data from multiple channels in multiple rounds; wherein, one initial blood flow image corresponds to one round.

[0124] The filtering module 603 is specifically used to construct a first data matrix based on multiple initial blood flow images; and to obtain a second data matrix based on the first data matrix by performing a preset filtering process.

[0125] The construction module 604 can be used to construct a first auxiliary matrix and a second auxiliary matrix based on signal data from multiple channels in multiple rounds.

[0126] The relevant processing module 605 can be used to perform preset relevant operations based on the first auxiliary matrix and the second auxiliary matrix to obtain the relevant matrix;

[0127] The denoising module 606 can be used to process the blood flow image corresponding to the second data matrix using the correlation matrix to obtain the corresponding denoised blood flow image.

[0128] In some embodiments, when the filtering module 603 is specifically implemented, a first data matrix can be constructed based on multiple initial blood flow images in the following manner: according to the acquisition time, the data corresponding to multiple initial blood flow images are arranged along the time direction to obtain a three-dimensional data matrix, which serves as the first data matrix; wherein, the first and second dimensions of the first data matrix are used to represent data information in the planar direction, and the third dimension of the first data matrix is ​​used to represent data information in the time direction.

[0129] In some embodiments, when the filtering module 603 is specifically implemented, it can obtain a second data matrix based on the first data matrix by performing a preset filtering process as follows: converting the first-dimensional data and the second-dimensional data in the first data matrix into one-dimensional data to obtain a third data matrix; determining the eigenvalue matrix and eigenvector matrix for the third data matrix; adjusting the eigenvalues ​​in the eigenvalue matrix whose values ​​are greater than a preset first data threshold and / or whose values ​​are less than a preset second data threshold to obtain an adjusted eigenvalue matrix; obtaining a fourth data matrix based on the adjusted eigenvalue matrix and eigenvector matrix; wherein the fourth data matrix contains blood flow image information; and performing an inverse transformation on the fourth data matrix to obtain the second data matrix.

[0130] In some embodiments, when the acquisition module 601 is specifically implemented, it can acquire signal data of multiple channels for the current round of the target area in the following manner: control the ultrasound imaging system to start from a specified negative starting angle, change the emission angle of the ultrasound imaging system according to a preset deflection rule, until the positive ending angle ends; and control the ultrasound imaging system to emit plane waves to the target area based on different emission angles, so as to acquire signal data of one channel corresponding to the emission angle; wherein, the angle value of the specified negative starting angle is the same as the angle value of the positive ending angle.

[0131] In some embodiments, when the reconstruction module 602 is specifically implemented, multiple initial blood flow images can be obtained by image reconstruction based on signal data from multiple channels in multiple rounds in the following manner: the initial blood flow image of the current round is obtained by image reconstruction based on signal data from multiple channels in the current round in the following manner: the signal intensity of each pixel in the image area is calculated based on the transmission and reception time of the ultrasound signal from the signal data of each channel to obtain multiple frames of ultrasound images; wherein, each frame of ultrasound image corresponds to the signal data of one channel in the current round; the initial blood flow image of the current round is obtained by superimposing and summing the multiple frames of ultrasound images.

[0132] In some embodiments, when the above-mentioned construction module 604 is specifically implemented, a first auxiliary matrix and a second auxiliary matrix can be constructed based on the signal data of multiple channels in multiple rounds in the following manner: multiple frames of ultrasound images with corresponding negative emission angles are selected from the signal data of multiple channels in multiple rounds, and summed to obtain a first type of summation result; the first type of summation result is arranged along the time direction to obtain a first auxiliary matrix; multiple frames of ultrasound images with corresponding positive emission angles are selected from the signal data of multiple channels in multiple rounds, and summed to obtain a second type of summation result; the second type of summation result is arranged along the time direction to obtain a second auxiliary matrix.

[0133] In some embodiments, when the above-mentioned correlation processing module 605 is specifically implemented, it can perform a preset correlation operation based on the first auxiliary matrix and the second auxiliary matrix in the following manner to obtain the correlation matrix: extract the first vector and the second vector corresponding to the same plane position from the first auxiliary matrix and the second auxiliary matrix respectively, perform correlation calculation to obtain the correlation value corresponding to the plane position; combine the correlation values ​​to obtain the correlation matrix.

[0134] In some embodiments, when the denoising processing module 606 is specifically implemented, the blood flow image corresponding to the second data matrix can be processed using the correlation matrix in the following manner to obtain the corresponding denoised blood flow image: multiply the correlation matrix with the blood flow image corresponding to the second data matrix to obtain the corresponding denoised blood flow image.

[0135] It should be noted that the units, devices, or modules described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. For ease of description, the above devices are described by dividing them into various modules according to their functions. Of course, in implementing this specification, the functions of each module can be implemented in one or more software and / or hardware, or the module that implements the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection between the devices or units shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0136] As can be seen from the above, the blood flow image denoising device provided in the embodiments of this specification can make full use of the uncorrelation of noise and the correlation of blood flow signals, effectively remove noise in the image while retaining the blood flow signal, and obtain a blood flow image with a high signal-to-noise ratio that can clearly display small blood vessel signals.

[0137] While this specification provides the steps of operation for the methods described in the embodiments or flowcharts, more or fewer steps may be included based on conventional or non-inventive means. The order of steps listed in the embodiments is merely one possible order of execution among many steps and does not represent the only possible order. In actual device or client product execution, the methods shown in the embodiments or drawings may be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment, or even a distributed data processing environment). The terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitations, the presence of other identical or equivalent elements in a process, method, product, or apparatus that includes said elements is not excluded. The terms "first," "second," etc., are used to denote names and do not indicate any particular order.

[0138] Those skilled in the art will also know that, besides implementing the controller using purely computer-readable program code, the same functions can be achieved by logically programming the method steps, making the controller function as logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers (PLCs), and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the devices within it used to implement various functions can also be considered structures within that hardware component. Alternatively, the devices used to implement various functions can be considered as both software modules implementing the method and structures within a hardware component.

[0139] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer-readable storage media, including storage devices.

[0140] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this specification can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions of this specification can essentially be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, mobile terminal, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments of this specification.

[0141] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. This specification can be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0142] Although this specification has been described by way of examples, those skilled in the art will recognize that many variations and modifications are possible without departing from the spirit of this specification, and it is intended that the appended claims cover such variations and modifications without departing from the spirit of this specification.

Claims

1. A method for denoising blood flow images, characterized in that, include: Acquire signal data from multiple channels across multiple rounds for a target area; where the signal data from one channel corresponds to one emission angle; Based on signal data from multiple channels across multiple rounds, multiple initial blood flow images are obtained through image reconstruction; each initial blood flow image corresponds to one round. A first data matrix is ​​constructed based on multiple initial blood flow images; and a second data matrix is ​​obtained by performing a preset filtering process on the first data matrix. Based on the signal data from multiple channels in multiple rounds, construct the first auxiliary matrix and the second auxiliary matrix; Based on the first auxiliary matrix and the second auxiliary matrix, perform the preset correlation operation to obtain the correlation matrix; The blood flow image corresponding to the second data matrix is ​​processed using the correlation matrix to obtain the corresponding denoised blood flow image; The acquisition of signal data from multiple channels across multiple rounds targeting the target area includes: acquiring signal data from multiple channels in the current round targeting the target area in the following manner: controlling the ultrasound imaging system to start from a specified negative starting angle, changing the emission angle of the ultrasound imaging system according to a preset deflection rule until the positive ending angle ends; and controlling the ultrasound imaging system to emit plane waves to the target area based on different emission angles to acquire signal data from one channel corresponding to that emission angle; wherein the specified negative starting angle is the same as the positive ending angle; and the preset deflection rule requires symmetrical emission angles. The step of constructing a first auxiliary matrix and a second auxiliary matrix based on signal data from multiple channels in multiple rounds includes: selecting multiple ultrasound images with a corresponding negative first emission angle from the signal data from multiple channels in multiple rounds; arranging the multiple ultrasound images with a corresponding negative first emission angle along the time direction to obtain a first auxiliary matrix; selecting multiple sub-images with a corresponding positive second emission angle from the signal data from multiple channels in multiple rounds; arranging the multiple ultrasound images with a corresponding positive second emission angle along the time direction to obtain a second auxiliary matrix; wherein the angle values ​​of the first angle and the second angle are the same.

2. The method according to claim 1, characterized in that, Based on multiple initial blood flow images, a first data matrix is ​​constructed, including: Based on the acquisition time, the data corresponding to multiple initial blood flow images are arranged along the time direction to obtain a three-dimensional data matrix, which serves as the first data matrix. The first and second dimensions of the first data matrix are used to represent data information in the planar direction, and the third dimension of the first data matrix is ​​used to represent data information in the time direction.

3. The method according to claim 2, characterized in that, Based on the first data matrix, a second data matrix is ​​obtained by performing a preset filtering process, including: The first and second dimensions of the first data matrix are converted into one-dimensional data to obtain the third data matrix; Determine the eigenvalue matrix and eigenvector matrix for the third data matrix; Adjust the eigenvalues ​​in the eigenvalue matrix whose values ​​are greater than a preset first data threshold and / or whose values ​​are less than a preset second data threshold to obtain the adjusted eigenvalue matrix; Based on the adjusted eigenvalue matrix and eigenvector matrix, the fourth data matrix is ​​obtained; the fourth data matrix contains image information of blood flow. The second data matrix is ​​obtained by performing an inverse transformation on the fourth data matrix.

4. The method according to claim 1, characterized in that, Based on signal data from multiple channels across multiple rounds, multiple initial blood flow images are obtained through image reconstruction, including: The initial blood flow image for the current round is obtained through image reconstruction based on signal data from multiple channels in the current round, using the following method: The signal data from each channel are used to calculate the signal intensity of each pixel in the image area based on the transmission and reception time of the ultrasound signal, resulting in multiple frames of ultrasound images; each frame of ultrasound image corresponds to the signal data of one channel in the current round. The initial blood flow image for the current round is obtained by superimposing and summing multiple ultrasound images.

5. The method according to claim 4, characterized in that, Based on signal data from multiple channels across multiple rounds, a first auxiliary matrix and a second auxiliary matrix are constructed, including: Multiple ultrasound images with negative emission angles are selected from signal data from multiple channels in multiple rounds, and then superimposed and summed to obtain the first type of summation result; the first type of summation result is then arranged along the time direction to obtain the first auxiliary matrix; Multiple ultrasound images with positive emission angles are selected from signal data from multiple channels in multiple rounds, and then superimposed and summed to obtain a second type of summation result. The second type of summation result is then arranged along the time direction to obtain a second auxiliary matrix.

6. The method according to claim 1, characterized in that, Based on the first auxiliary matrix and the second auxiliary matrix, a preset correlation operation is performed to obtain the correlation matrix, including: Extract the first vector and the second vector corresponding to the same plane position from the first auxiliary matrix and the second auxiliary matrix respectively, perform relevant calculations, and obtain the relevant value corresponding to the plane position; The correlation matrix is ​​obtained by combining the correlation values.

7. The method according to claim 1, characterized in that, The blood flow image corresponding to the second data matrix is ​​processed using the correlation matrix to obtain the corresponding denoised blood flow image, including: Multiply the correlation matrix by the blood flow image corresponding to the second data matrix to obtain the corresponding denoised blood flow image.

8. A noise reduction device for blood flow images, characterized in that, include: The acquisition module is used to acquire signal data from multiple channels in multiple rounds for a target area; wherein, the signal data of one channel corresponds to one emission angle; The reconstruction module is used to obtain multiple initial blood flow images by image reconstruction based on signal data from multiple channels in multiple rounds; wherein, one initial blood flow image corresponds to one round. The filtering module is used to construct a first data matrix based on multiple initial blood flow images; and to obtain a second data matrix based on the first data matrix by performing a preset filtering process. A construction module is used to construct a first auxiliary matrix and a second auxiliary matrix based on signal data from multiple channels in multiple rounds; The correlation processing module is used to perform preset correlation operations based on the first auxiliary matrix and the second auxiliary matrix to obtain the correlation matrix; The denoising module is used to process the blood flow image corresponding to the second data matrix using the correlation matrix to obtain the corresponding denoised blood flow image; Specifically, the acquisition module acquires signal data from multiple channels of the target area in the current cycle in the following manner: It controls the ultrasound imaging system to start from a specified negative starting angle, and according to a preset deflection rule, changes the emission angle of the ultrasound imaging system until it ends at a positive ending angle; and it controls the ultrasound imaging system to emit plane waves to the target area based on different emission angles, so as to acquire signal data from one channel corresponding to that emission angle; wherein the specified negative starting angle is the same as the positive ending angle; and the preset deflection rule requires symmetrical emission angles. The construction module is specifically used for: selecting multiple ultrasound images with a negative first emission angle from signal data from multiple channels in multiple rounds; arranging the multiple ultrasound images with a negative first emission angle along the time direction to obtain a first auxiliary matrix; selecting multiple sub-images with a positive second emission angle from signal data from multiple channels in multiple rounds; arranging the multiple ultrasound images with a positive second emission angle along the time direction to obtain a second auxiliary matrix; wherein the angle values ​​of the first angle and the second angle are the same.

9. A computer-readable storage medium, characterized in that, It stores computer instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.

Images