Ultrasonic plane wave imaging method and device
Through preset transmission parameters and multi-angle coherent composite technology, combined with GAN model to generate images, the problem of low resolution and signal-to-noise ratio in ultrasonic plane wave imaging is solved, high-quality and high frame rate ultrasonic imaging is achieved, and application scenarios are expanded .
Patent Information
- Application Number
- CN202411091846.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-11-12
AI Technical Summary
The existing ultrasonic plane wave imaging methods have problems with poor imaging resolution and low signal-to-noise ratio, resulting in limited application scenarios.
Control the ultrasonic transducer to emit ultrasonic plane waves through preset transmission parameters, use delay superposition algorithm and coherent superposition algorithm to generate images, and generate images of additional angles based on the generation adversarial network (GAN) model, perform multi-scale weighted fusion to improve image resolution rate and signal-to-noise ratio.
The resolution, contrast and signal-to-noise ratio of ultrasonic images are improved, the application scenarios are expanded, the actual number of emissions is reduced, and the image quality is improved, so that ultrasonic images have wider application potential at high frame rates.
Smart Images

Figure CN118924332A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of ultrasonic imaging, and in particular to an ultrasonic plane wave imaging method and device. Background Art
[0002] Ultrasound imaging has become one of the main tools for tissue morphology and functional examinations due to its non-invasive, safe and convenient characteristics. Traditional focused wave ultrasound imaging systems have a low frame rate, usually around 20 to 50 frames per second, due to the limitation of the speed of sound waves propagating inside tissues. The development of ultra-high-speed ultrasound imaging technology has provided new possibilities for improving the frame rate of ultrasound images. Although plane wave ultrasound imaging technology can significantly improve the imaging speed, allowing real-time imaging frame rates to reach more than 5,000 frames per second. However, the images obtained by a single plane wave transmission have problems with poor lateral resolution and low mid- and far-field signal-to-noise ratio.
[0003] Although the existing ultrasonic plane wave imaging method can improve the real-time imaging frame rate, its imaging resolution is poor and the signal-to-noise ratio is low, resulting in limited application scenarios. Summary of the invention
[0004] The present invention provides an ultrasonic plane wave imaging method and device to solve the problem that the existing ultrasonic plane wave imaging has poor imaging resolution and low signal-to-noise ratio, which leads to limited application scenarios.
[0005] In order to solve the above technical problems, an embodiment of the present invention provides an ultrasonic plane wave imaging method, comprising:
[0006] Preset transmission parameters according to imaging requirements, and control the ultrasonic transducer to transmit ultrasonic plane waves according to the transmission parameters; the transmission parameters include the angle number and the transmission angle of the transmitted plane wave;
[0007] Acquire the ultrasonic plane wave, perform beam synthesis on the ultrasonic plane wave according to a delay-and-addition algorithm, and acquire the reflection intensity of an imaging point;
[0008] Generate a plurality of single-angle ultrasonic images according to the reflection intensity, the number of angles of the transmitted plane wave and the transmission angle, and superimpose the single-angle ultrasonic images according to a coherent superposition algorithm to obtain a first ultrasonic image;
[0009] Building an image generation model based on a generative adversarial network, and generating a second ultrasound image according to the image generation model and the emission angle;
[0010] The first ultrasonic image and the second ultrasonic image are subjected to multi-scale weighted fusion according to an angle weighted algorithm to obtain a final ultrasonic image.
[0011] The present invention presets the emission parameters of the emitted plane wave according to the imaging requirements, thereby improving the matching of ultrasonic imaging with the actual imaging requirements; the presentation frame rate is improved according to the plane wave imaging, and based on the multi-angle coherent composite technology and the angle weighted variable, the resolution, contrast and signal-to-noise ratio of the image can be effectively improved, and the deficiency of the single-angle plane wave imaging can be compensated; and the GAN model is used to generate images of additional angles, thereby reducing the actual number of transmissions, expanding the number of angles, improving the image quality, and improving the image quality while improving the image frame rate, so that the application scenarios of ultrasonic images are more extensive.
[0012] Furthermore, the transmission parameters are preset according to the imaging requirements, and the ultrasonic transducer is controlled to transmit the ultrasonic plane wave according to the transmission parameters, specifically:
[0013] Determine the imaging target and imaging requirements, and preset the angle number and the transmission frequency of the transmitted plane wave according to the resolution and frame rate of the imaging requirements;
[0014] The deflection angle range for each emission angle is preset according to the size and shape of the imaging target.
[0015] Furthermore, the obtaining of the ultrasonic plane wave, performing beam synthesis on the ultrasonic plane wave according to a delay-and-addition algorithm, and obtaining the reflection intensity of the imaging point, is specifically as follows:
[0016] Calculate the propagation time of ultrasonic plane waves according to the positions of transmitting array elements, receiving array elements and imaging points;
[0017] The received ultrasonic plane wave is subjected to signal delay processing, and the delayed ultrasonic plane wave is subjected to signal superposition according to a superposition algorithm and a propagation time to obtain the reflection intensity of the imaging point.
[0018] Furthermore, the obtaining of the first ultrasonic image by superimposing the single-angle ultrasonic images according to the coherent superposition algorithm is specifically as follows:
[0019] Calculate the weighted coefficients of each transmission angle, and perform coherent superposition on the single-angle ultrasound images of each transmission angle according to the weighted coefficients to generate a first ultrasound image. Specifically:
[0020]
[0021] Among them, W α is the weighting coefficient of each angle, Wα=cos p (α), α is the emission angle, I α (x,z) is the emission intensity of the imaging point, and p is the control parameter.
[0022] Furthermore, the image generation model is constructed based on the generative adversarial network, and the second ultrasound image is generated according to the image generation model and the emission angle, specifically:
[0023] Building an image generation model based on a generative adversarial network of difference ions, and training the image generation model according to multi-angle ultrasound imaging data;
[0024] And generate a second ultrasound image according to the trained image generation model and the emission angle.
[0025] Furthermore, performing multi-scale weighted fusion on the first ultrasonic image and the second ultrasonic image according to the angle weighted algorithm to obtain the final ultrasonic image is specifically as follows:
[0026] Decomposing the first ultrasonic image and the second ultrasonic image into a plurality of sub-images of different scales respectively;
[0027] Presetting weight coefficients of the sub-images of the first ultrasonic image and the sub-images of the second ultrasonic image, and performing weighted fusion on the first ultrasonic image and the second ultrasonic image at various scales according to the weight coefficients;
[0028] The fused images at each scale are obtained, and all the fused images are reconstructed to generate the final ultrasound image.
[0029] In a second aspect, the present invention provides an ultrasonic plane wave imaging device, comprising: a control module, a beamforming module, a coherent superposition module, a GAN module and a fusion module;
[0030] The control module is used to preset transmission parameters according to imaging requirements, and control the ultrasonic transducer to transmit ultrasonic plane waves according to the transmission parameters; the transmission parameters include the angle number and the transmission angle of the transmitted plane wave;
[0031] The beamforming module is used to obtain the ultrasonic plane wave, perform beamforming on the ultrasonic plane wave according to a delay-and-addition algorithm, and obtain the reflection intensity of the imaging point;
[0032] The coherent superposition module is used to generate a plurality of single-angle ultrasonic images according to the reflection intensity, the number of angles of the transmitted plane wave and the transmission angle, and to superimpose the single-angle ultrasonic images according to the coherent superposition algorithm to obtain a first ultrasonic image;
[0033] The GAN module is used to construct an image generation model based on a generative adversarial network, and generate a second ultrasound image according to the image generation model and the emission angle;
[0034] The fusion module is used to perform multi-scale weighted fusion on the first ultrasonic image and the second ultrasonic image according to an angle weighted algorithm to obtain a final ultrasonic image.
[0035] Furthermore, the coherent superposition module is specifically used for:
[0036] Calculate the weighted coefficients of each transmission angle, and perform coherent superposition on the single-angle ultrasound images of each transmission angle according to the weighted coefficients to generate a first ultrasound image. Specifically:
[0037]
[0038] Among them, W α is the weighting coefficient of each angle, Wα=cos p (α), α is the emission angle, I α (x,z) is the emission intensity of the imaging point, and p is the control parameter.
[0039] Furthermore, the GAN module is specifically used for:
[0040] Building an image generation model based on a generative adversarial network of difference ions, and training the image generation model according to multi-angle ultrasound imaging data;
[0041] And generate a second ultrasound image according to the trained image generation model and the emission angle.
[0042] Furthermore, the fusion module is specifically used for:
[0043] Decomposing the first ultrasonic image and the second ultrasonic image into a plurality of sub-images of different scales respectively;
[0044] Presetting weight coefficients of the sub-images of the first ultrasonic image and the sub-images of the second ultrasonic image, and performing weighted fusion on the first ultrasonic image and the second ultrasonic image at various scales according to the weight coefficients;
[0045] The fused images at each scale are obtained, and all the fused images are reconstructed to generate the final ultrasound image. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] Figure 1 A schematic diagram of a flow chart of an ultrasonic plane wave imaging method provided by an embodiment of the present invention;
[0047] Figure 2 A schematic structural diagram of an ultrasonic plane wave imaging device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0049] Embodiment 1
[0050] Please refer to Figure 1 , Figure 1 A schematic flow chart of an ultrasonic plane wave imaging method provided by an embodiment of the present invention includes steps 101 to 105, which are specifically as follows:
[0051] Step 101: preset transmission parameters according to imaging requirements, and control the ultrasonic transducer to transmit ultrasonic plane waves according to the transmission parameters; the transmission parameters include the angle number and the transmission angle of the transmitted plane wave;
[0052] In this embodiment, the transmission parameters are preset according to the imaging requirements, and the ultrasonic transducer is controlled to transmit the ultrasonic plane wave according to the transmission parameters, specifically:
[0053] Determine the imaging target and imaging requirements, and preset the angle number and the transmission frequency of the transmitted plane wave according to the resolution and frame rate of the imaging requirements;
[0054] The deflection angle range for each emission angle is preset according to the size and shape of the imaging target.
[0055] In this embodiment, the transmission parameters and transmission timing are preset according to the imaging requirements, and the transmission parameters include the number of plane wave angles (N), the transmission angle α, and the deflection angle and transmission frequency of each transmission angle. And according to each transmission angle and transmission timing, each transmission array element is controlled to simultaneously transmit ultrasonic pulses to form a plane wave.
[0056] In this embodiment, the transmission parameters and the transmission timing can be set manually or automatically according to imaging requirements.
[0057] In this embodiment, parameters are set according to imaging requirements, including: determining imaging targets and imaging requirements, wherein imaging targets include heart, blood vessels, tumors, etc.; imaging requirements include high resolution, high frame rate, deep tissue imaging, etc.
[0058] In this embodiment, the emission angle is determined according to the imaging requirements and imaging targets. When high-resolution imaging is required, more emission angles can be selected to improve image details. When high-frame rate imaging is required, fewer emission angles can be selected to reduce the amount of calculation and data processing time.
[0059] In this embodiment, the deflection angle of each emission angle is determined according to the imaging target, and an appropriate deflection angle range is selected according to the size and shape of the target area. For example, for a larger imaging area, a larger deflection angle range can be selected. The deflection angle range can be adjusted according to the weighting coefficient formula Wα=cosp(α) for a larger center angle weight. A smaller deflection angle range helps to improve the image quality of the center area.
[0060] In this embodiment, the transmission frequency is determined according to the imaging requirement and the imaging target. When the imaging requirement is high-resolution imaging, a higher transmission frequency is required to capture more details. When the imaging target is deep tissue imaging, the transmission frequency needs to be appropriately reduced to increase the penetration depth of ultrasound.
[0061] In this embodiment, the emission timing of each emission angle needs to be set accordingly, and the emission timing includes uniform timing and adaptive timing. In most cases, the timing of the emission angle is set to a uniform timing, that is, the emission time intervals of each angle are the same. This method is simple and easy to implement. In some cases, adaptive timing can be used according to specific imaging requirements and characteristics of the target area. For example, in order to reduce motion artifacts, the emission frequency can be increased and the time interval can be reduced when the target area moves quickly.
[0062] In this embodiment, uniform timing setting helps reduce motion artifacts and improve image quality. In particular, in applications that require high frame rates, such as cardiac imaging, uniform timing can capture details of fast motion. Adaptive timing can adjust the transmission time interval according to actual conditions, reduce signal overlap and interference, and optimize image quality.
[0063] In this embodiment, by reasonably setting the timing of the emission angles, a balance can be found between the amount of calculation and the image quality. For example, in deep tissue imaging, the emission interval time can be appropriately increased to reduce the complexity of data processing.
[0064] In this embodiment, the imaging requirements and target area are imported in advance, and by analyzing the size, shape, position and imaging purpose of the target area, such as high-resolution imaging, deep tissue imaging, etc., the computer can automatically optimize the emission angle and frequency to achieve the best imaging effect.
[0065] In this embodiment, after the transmission parameters are set by the computer, the transmission parameters can also be adjusted manually.
[0066] In this embodiment, after setting the transmission parameters and transmission timing, the ultrasonic transducer array is controlled to transmit ultrasonic plane waves according to a predetermined angle and timing. During the control, the transmission timing of all array elements is ensured to be synchronized to ensure that the propagation of ultrasonic waves in the medium forms a plane wave.
[0067] Step 102: Acquire the ultrasonic plane wave, perform beam synthesis on the ultrasonic plane wave according to a delay-and-addition algorithm, and acquire the reflection intensity of the imaging point;
[0068] In this embodiment, the ultrasonic signal emitted by the transmitting array element is received and preprocessed, and the preprocessing includes signal conversion, filtering and analog-to-digital conversion operations. Specifically, each receiving array element receives the ultrasonic signal scattered back by the tissue and converts it into an electrical signal. The received electrical signal is amplified and filtered to remove noise and useless frequency components. The filtered electrical signal is converted into a data signal for beamforming operation.
[0069] In this embodiment, the obtaining of the ultrasonic plane wave, performing beam synthesis on the ultrasonic plane wave according to a delay-and-addition algorithm, and obtaining the reflection intensity of the imaging point are specifically as follows:
[0070] Calculate the propagation time of ultrasonic plane waves according to the positions of transmitting array elements, receiving array elements and imaging points;
[0071] The received ultrasonic plane wave is subjected to signal delay processing, and the delayed ultrasonic plane wave is subjected to signal superposition according to a superposition algorithm and a propagation time to obtain the reflection intensity of the imaging point.
[0072] In this embodiment, the propagation time τ of the ultrasonic wave from the transmitting element to the receiving element is calculated according to the positions of the transmitting and receiving elements and the position of the imaging point. Specifically:
[0073]
[0074] Among them, x0 and x1 are the X-axis coordinates of the transmitting array element and the receiving array element respectively, x and z are the coordinates of the imaging point, and c is the propagation speed of ultrasound in the medium.
[0075] In this embodiment, the received ultrasonic signal is delayed so that the signals of different propagation paths are aligned at the same time point, and the delayed signals are superimposed according to the superposition algorithm to generate the reflection intensity of the imaging point and generate single-angle ultrasonic images at each emission angle. Specifically:
[0076]
[0077] Where M is the number of receiving array elements, RF i (x,τ i ) is the signal strength of the i-th receiving element after delay, and I(x,z) is the single-angle ultrasound image.
[0078] Step 103: generating a plurality of single-angle ultrasonic images according to the reflection intensity, the number of angles of the transmitted plane wave and the transmission angle, and superimposing the single-angle ultrasonic images according to a coherent superposition algorithm to obtain a first ultrasonic image;
[0079] In this embodiment, a single-angle ultrasonic image I is obtained at each transmission angle α. α (x,z), get N images I α (x,z).
[0080] In this embodiment, the single-angle ultrasound images are superimposed according to the coherent superposition algorithm to obtain the first ultrasound image, specifically:
[0081] Calculate the weighted coefficients of each transmission angle, and perform coherent superposition on the single-angle ultrasound images of each transmission angle according to the weighted coefficients to generate a first ultrasound image. Specifically:
[0082]
[0083] Among them, W α is the weighting coefficient of each angle, Wα=cos p (α), α is the emission angle, I α (x,z) is the emission intensity of the imaging point, and p is the control parameter.
[0084] In this embodiment, N images are coherently superimposed and the weighted variable W of the emission angle is calculated. α , and obtain a first ultrasonic image.
[0085] In this embodiment, a weighting coefficient is set according to each emission angle, and the relationship between the weighting coefficient and the deflection angle is specifically:
[0086] Wα=cosp(α) (4)
[0087] Where: Wα is the weighting coefficient corresponding to the deflection angle α, and α is the deflection angle of the plane wave emission. p is a preset control parameter used to adjust the distribution of the weighting coefficient; when p=1, the weighting coefficient Wα is cos(α), indicating linear attenuation. When p>1, the weighting coefficient decays faster, which means a greater contribution to the image at the center angle and a smaller contribution to the image at a large deflection angle. When p<1, the weighting coefficient decays slower, which means a more balanced contribution to the images at all angles.
[0088] Step 104: constructing an image generation model based on a generative adversarial network, and generating a second ultrasound image according to the image generation model and the emission angle;
[0089] In this embodiment, the image generation model is constructed based on the generative adversarial network, and the second ultrasound image is generated according to the image generation model and the emission angle, specifically:
[0090] Building an image generation model based on a generative adversarial network of difference ions, and training the image generation model according to multi-angle ultrasound imaging data;
[0091] And generate a second ultrasound image according to the trained image generation model and the emission angle.
[0092] In this embodiment, an image generation model is constructed based on a generative adversarial network (GAN) model of diffus ion, and the image generation model is trained according to the multi-angle imaging data actually collected, so that it can generate ultrasound images at different angles. A second ultrasound image is generated according to the image generation model.
[0093] Step 105: Perform multi-scale weighted fusion on the first ultrasonic image and the second ultrasonic image according to an angle weighted algorithm to obtain a final ultrasonic image.
[0094] In this embodiment, the multi-scale weighted fusion of the first ultrasonic image and the second ultrasonic image is performed according to the angle weighted algorithm to obtain the final ultrasonic image, specifically:
[0095] Decomposing the first ultrasonic image and the second ultrasonic image into a plurality of sub-images of different scales respectively;
[0096] Presetting weight coefficients of the sub-images of the first ultrasonic image and the sub-images of the second ultrasonic image, and performing weighted fusion on the first ultrasonic image and the second ultrasonic image at various scales according to the weight coefficients;
[0097] The fused images at each scale are obtained, and all the fused images are reconstructed to generate the final ultrasound image.
[0098] In this embodiment, the first ultrasonic image and the second ultrasonic image are decomposed into a plurality of sub-images of different scales according to a multi-scale decomposition method; the multi-scale decomposition method includes a Laplace pyramid algorithm and a wavelet transform algorithm.
[0099] In this embodiment, weighting coefficients are defined for the first ultrasonic image and the second ultrasonic image at each scale. Assume that the first ultrasonic image is Ireal and the second ultrasonic image is Igan, and the weighting coefficients at the i-th scale are Wreali and Wgani, respectively, and Wreali+Wgani=1.
[0100] In this embodiment, the weighting coefficients can be optimized and set at different scales according to the quality and signal-to-noise ratio of the image. For example, for high-frequency detail levels, the weight of the first ultrasound image can be larger to retain more details; for low-frequency levels, the weight of the second ultrasound image can be larger to reduce noise.
[0101] In this embodiment, at each scale, the first ultrasonic image and the second ultrasonic image are weighted fused respectively. Assuming that the first ultrasonic image at the i-th scale is Ireali, the second ultrasonic image is Igani, and the fused image is Ifinali, then:
[0102] Ifinali(x,z)=Wreali·Ireali(x,z)+Wgani·Igani (x,z) (5)
[0103] Here, x and z are the coordinates of the image.
[0104] In this embodiment, fused images at various scales are acquired, and all fused images are reconstructed to generate a final ultrasound image.
[0105] In this embodiment, the image is decomposed into different scales (or frequency bands), weighted fusion is performed at each scale, and then the images at each scale are reconstructed to obtain a final high-quality image. This can better preserve the details and layer information of the image, and is suitable for tasks such as image quality improvement and denoising.
[0106] In this embodiment, after the final ultrasound image is acquired, the image is further subjected to denoising and contrast enhancement operations so as to apply the ultrasound image to actual imaging requirements.
[0107] In this embodiment, the transmission parameters of the transmitted plane wave are preset according to the imaging requirements, so as to improve the matching of ultrasonic imaging with the actual imaging requirements; the presentation frame rate is improved according to the plane wave imaging, and based on the multi-angle coherent composite technology and the angle weighted variable, the resolution, contrast and signal-to-noise ratio of the image can be effectively improved, so as to make up for the deficiency of the single-angle plane wave imaging, and use the GAN model to generate images of additional angles, thereby reducing the actual number of transmissions, expanding the number of angles, improving the image quality, improving the image quality while improving the image frame rate, and making the application scenarios of ultrasonic images more extensive.
[0108] Please refer to Figure 2 , Figure 2 A schematic structural diagram of an ultrasonic plane wave imaging device provided by an embodiment of the present invention includes: a control module 201, a beamforming module 202, a coherent superposition module 203, a GAN module 204 and a fusion module 205;
[0109] The control module 201 is used to preset transmission parameters according to imaging requirements, and control the ultrasonic transducer to transmit ultrasonic plane waves according to the transmission parameters; the transmission parameters include the angle number and the transmission angle of the transmitted plane wave;
[0110] The beamforming module 202 is used to obtain the ultrasonic plane wave, perform beamforming on the ultrasonic plane wave according to a delay-and-addition algorithm, and obtain the reflection intensity of the imaging point;
[0111] The coherent superposition module 203 is used to generate a plurality of single-angle ultrasound images according to the reflection intensity, the number of angles of the transmitted plane wave and the transmission angle, and to superimpose the single-angle ultrasound images according to the coherent superposition algorithm to obtain a first ultrasound image;
[0112] The GAN module 204 is used to construct an image generation model based on a generative adversarial network, and generate a second ultrasound image according to the image generation model and the emission angle;
[0113] The fusion module 205 is used to perform multi-scale weighted fusion on the first ultrasonic image and the second ultrasonic image according to an angle weighted algorithm to obtain a final ultrasonic image.
[0114] In this embodiment, the control module is used to:
[0115] Determine the imaging target and imaging requirements, and preset the number of angles, the transmission angle, and the transmission frequency of the transmitted plane wave according to the resolution and frame rate of the imaging requirements;
[0116] The deflection angle range for each emission angle is preset according to the size and shape of the imaging target.
[0117] In this embodiment, the beamforming module is used to:
[0118] Calculate the propagation time of ultrasonic plane waves according to the positions of transmitting array elements, receiving array elements and imaging points;
[0119] The received ultrasonic plane wave is subjected to signal delay processing, and the delayed ultrasonic plane wave is subjected to signal superposition according to a superposition algorithm and a propagation time to obtain the reflection intensity of the imaging point.
[0120] In this embodiment, the coherent superposition module is specifically used for:
[0121] Calculate the weighted coefficients of each transmission angle, and perform coherent superposition on the single-angle ultrasound images of each transmission angle according to the weighted coefficients to generate a first ultrasound image. Specifically:
[0122]
[0123] Among them, W αis the weighting coefficient of each angle, Wα=cos p (α), α is the emission angle, I α (x,z) is the emission intensity of the imaging point, and p is the control parameter.
[0124] In this embodiment, the GAN module is specifically used for:
[0125] Building an image generation model based on a generative adversarial network of difference ions, and training the image generation model according to multi-angle ultrasound imaging data;
[0126] And generate a second ultrasound image according to the trained image generation model and the emission angle.
[0127] In this embodiment, the fusion module is specifically used for:
[0128] Decomposing the first ultrasonic image and the second ultrasonic image into a plurality of sub-images of different scales respectively;
[0129] Presetting weight coefficients of the sub-images of the first ultrasonic image and the sub-images of the second ultrasonic image, and performing weighted fusion on the first ultrasonic image and the second ultrasonic image at various scales according to the weight coefficients;
[0130] The fused images at each scale are obtained, and all the fused images are reconstructed to generate the final ultrasound image.
[0131] The embodiment of the present invention also provides an ultrasonic plane wave imaging system, which includes an ultrasonic transducer array, a transmission control module, a reception processing module, a data processing and image reconstruction module, and a display module.
[0132] In this embodiment, the ultrasonic transducer array includes a transmitting array element and a receiving array element.
[0133] In this embodiment, the transmission control module is used to control the ultrasonic transducer array to transmit ultrasonic plane waves according to a predetermined angle and timing. Specifically, it includes initialization parameters, control signal generation and synchronization control. The initialization parameters are the number of angles (N) for transmitting plane waves, the deflection angle (N) of each angle and the transmission frequency set according to imaging requirements. The control signal generation is to generate a control signal for controlling the ultrasonic transducer array to transmit plane waves. For each transmission angle (α), each array element is controlled to simultaneously transmit ultrasonic pulses according to the set timing. The synchronization control is to ensure that the transmission timing of all array elements is synchronized to ensure that the propagation of ultrasonic waves in the medium forms a plane wave.
[0134] In this embodiment, the receiving and processing module is used to receive the ultrasonic signal reflected by the tissue and perform preliminary processing, including signal reception, amplification and filtering, and analog-to-digital conversion operations. The signal reception is that each array element receives the ultrasonic signal scattered by the tissue and converts it into an electrical signal. The amplification and filtering is to amplify and filter the received electrical signal to remove noise and useless frequency components. The analog-to-digital conversion is to convert the analog signal into a digital signal to facilitate subsequent data processing.
[0135] In this embodiment, the data processing and image reconstruction module is used to process and reconstruct the received ultrasound signal to generate a high-quality ultrasound image. Specifically, it includes beamforming, multi-angle coherent compounding, and GAN model expansion angle. The beamforming is to process the received ultrasound signal according to a predetermined algorithm to obtain the reflection intensity of each point in the imaging area. In order to further improve the image quality, the multi-angle coherent compounding coherently superimposes the plane wave imaging results of multiple different angles to coherently superimpose N single-angle ultrasound images, and introduces the angle weighting variable W. α The GAN model uses a generative adversarial network (GAN) model based on diffusion to generate images at more angles, decomposes the image into different scales (or frequency bands), performs weighted fusion at each scale, and then reconstructs the images at each scale to obtain the final high-quality image.
[0136] In this embodiment, the display module is used to display the reconstructed ultrasound image so that the ultrasound image can be applied to actual imaging needs. It includes image processing and image display. The image processing is to perform necessary image processing on the reconstructed image, such as denoising, contrast enhancement, etc. The image display is to display the processed image on the screen and provide necessary image operation functions, such as zooming in, zooming out, moving, etc.
[0137] In this embodiment, in order to ensure the accuracy and reliability of the system, system calibration and optimization are required. Specifically, the system is calibrated before use to ensure the position and timing of the transmitting and receiving array elements are accurate. According to different imaging requirements, the transmission angle, transmission frequency and other parameters are optimized to achieve the best imaging effect.
[0138] In this embodiment, plane wave technology and parallel processing methods are used to improve the imaging frame rate, and through multi-angle coherent composite technology and angle weighted variables, the resolution, contrast and signal-to-noise ratio of the image can be effectively improved to make up for the shortcomings of single-angle plane wave imaging. In addition, by generating images of additional angles based on the diffus ion-based GAN model, the actual number of transmissions is reduced, while the number of angles is expanded and the image quality is improved. High-quality real-time imaging is achieved at a high frame rate to meet clinical imaging requirements for dynamic tissues and fast-moving targets. Improve the application scenarios of ultrasound imaging. For example, it is applied to cardiac imaging, blood flow imaging and other ultrasonic functional imaging fields.
[0139] In one embodiment, in the field of cardiac imaging, due to the rapid movement of the heart, the frame rate and image quality requirements of the imaging system are very high. Using the ultrasonic plane wave imaging method, imaging can be achieved at a speed of 5000 frames per second to capture the details of the rapid movement of the heart. Through multi-angle coherent composite technology and angle weighted variables, high-resolution cardiac structure images are obtained to clearly display the details of the heart. At the same time, the signal-to-noise ratio of the image is significantly improved, which enhances the contrast between the cardiac tissue and the surrounding tissue, and helps doctors make accurate diagnoses. The GAN model based on diffus ion is used to generate images at additional angles, which reduces the actual number of transmissions while ensuring image quality.
[0140] In one embodiment, in the field of blood flow imaging, real-time performance and image quality are equally important. Using the ultrasonic plane wave imaging method, the dynamic changes of blood flow can be displayed in real time, and the blood flow velocity and direction can be visualized. And through multi-angle coherence composite technology and angle weighted variables, the contrast between the blood flow signal and the surrounding tissue is enhanced, and the blood flow path is clearly displayed. At the same time, the high signal-to-noise ratio and high resolution make blood flow imaging more accurate and reduce the possibility of imaging errors. In addition, the use of a GAN model based on diffusion to generate images with extended angles reduces the actual number of transmissions and improves imaging efficiency.
[0141] In this embodiment, the transmission parameters of the transmitted plane wave are preset according to the imaging requirements, so as to improve the matching of ultrasonic imaging with the actual imaging requirements; the presentation frame rate is improved according to the plane wave imaging, and based on the multi-angle coherent composite technology and the angle weighted variable, the resolution, contrast and signal-to-noise ratio of the image can be effectively improved, so as to make up for the deficiency of the single-angle plane wave imaging, and use the GAN model to generate images of additional angles, thereby reducing the actual number of transmissions, expanding the number of angles, improving the image quality, improving the image quality while improving the image frame rate, and making the application scenarios of ultrasonic images more extensive.
[0142] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.
Claims
1. An ultrasonic plane wave imaging method, It is characterized in that include: Preset transmission parameters according to imaging requirements, and control the ultrasonic transducer to transmit ultrasonic plane waves according to the transmission parameters; the transmission parameters include the angle number and the transmission angle of the transmitted plane wave; Acquire the ultrasonic plane wave, perform beam synthesis on the ultrasonic plane wave according to a delay-and-addition algorithm, and acquire the reflection intensity of the imaging point; Generate a plurality of single-angle ultrasonic images according to the reflection intensity, the number of angles of the transmitted plane wave and the transmission angle, and superimpose the single-angle ultrasonic images according to a coherent superposition algorithm to obtain a first ultrasonic image; Building an image generation model based on a generative adversarial network, and generating a second ultrasound image according to the image generation model and the emission angle; The first ultrasonic image and the second ultrasonic image are subjected to multi-scale weighted fusion according to an angle weighted algorithm to obtain a final ultrasonic image.
2. An ultrasonic plane wave imaging method as claimed in claim 1, It is characterized in that The transmitting parameters are preset according to the imaging requirements, and the ultrasonic transducer is controlled to transmit the ultrasonic plane wave according to the transmitting parameters, specifically: Determine the imaging target and imaging requirements, and preset the number of angles, the emission angle, and the emission frequency of the emitted plane wave according to the resolution and frame rate of the imaging requirements; The deflection angle range for each emission angle is preset according to the size and shape of the imaging target.
3. An ultrasonic plane wave imaging method as claimed in claim 1, It is characterized in that The obtaining of the ultrasonic plane wave, performing beam synthesis on the ultrasonic plane wave according to a delay-and-addition algorithm, and obtaining the reflection intensity of the imaging point, is specifically as follows: Calculate the propagation time of ultrasonic plane waves according to the positions of transmitting array elements, receiving array elements and imaging points; The received ultrasonic plane wave is subjected to signal delay processing, and the delayed ultrasonic plane wave is subjected to signal superposition according to a superposition algorithm and a propagation time to obtain the reflection intensity of the imaging point.
4. An ultrasonic plane wave imaging method as claimed in claim 3, It is characterized in that The step of superimposing the single-angle ultrasound images according to the coherent superposition algorithm to obtain the first ultrasound image is specifically as follows: Calculate the weighted coefficients of each transmission angle, and perform coherent superposition on the single-angle ultrasound images of each transmission angle according to the weighted coefficients to generate a first ultrasound image. Specifically: Among them, W α is the weighting coefficient of each angle, Wα=cos p (α), α is the emission angle, I α (x,z) is the emission intensity of the imaging point, and p is the control parameter.
5. An ultrasonic plane wave imaging method as claimed in claim 4, It is characterized in that The image generation model is constructed based on the generative adversarial network, and the second ultrasound image is generated according to the image generation model and the emission angle, specifically: Building an image generation model based on a diffusion-based generative adversarial network, and training the image generation model according to multi-angle ultrasound imaging data; And generate a second ultrasound image according to the trained image generation model and the emission angle.
6. An ultrasonic plane wave imaging method as claimed in claim 4, It is characterized in that The performing multi-scale weighted fusion on the first ultrasonic image and the second ultrasonic image according to the angle weighted algorithm to obtain the final ultrasonic image is specifically as follows: Decomposing the first ultrasonic image and the second ultrasonic image into a plurality of sub-images of different scales respectively; Presetting weight coefficients of the sub-images of the first ultrasonic image and the sub-images of the second ultrasonic image, and performing weighted fusion on the first ultrasonic image and the second ultrasonic image at various scales according to the weight coefficients; The fused images at each scale are obtained, and all the fused images are reconstructed to generate the final ultrasound image.
7. An ultrasonic plane wave imaging device, It is characterized in that include: Control module, beamforming module, coherent superposition module, GAN module and fusion module; The control module is used to preset transmission parameters according to imaging requirements, and control the ultrasonic transducer to transmit ultrasonic plane waves according to the transmission parameters; the transmission parameters include the angle number and the transmission angle of the transmitted plane wave; The beamforming module is used to obtain the ultrasonic plane wave, perform beamforming on the ultrasonic plane wave according to a delay-and-addition algorithm, and obtain the reflection intensity of the imaging point; The coherent superposition module is used to generate a plurality of single-angle ultrasonic images according to the reflection intensity, the number of angles of the transmitted plane wave and the transmission angle, and to superimpose the single-angle ultrasonic images according to the coherent superposition algorithm to obtain a first ultrasonic image; The GAN module is used to construct an image generation model based on a generative adversarial network, and generate a second ultrasound image according to the image generation model and the emission angle; The fusion module is used to perform multi-scale weighted fusion on the first ultrasonic image and the second ultrasonic image according to an angle weighted algorithm to obtain a final ultrasonic image.
8. An ultrasonic plane wave imaging device as claimed in claim 7, It is characterized in that The coherent superposition module is specifically used for: Calculate the weighted coefficients of each transmission angle, and perform coherent superposition on the single-angle ultrasound images of each transmission angle according to the weighted coefficients to generate a first ultrasound image. Specifically: Among them, W α is the weighting coefficient of each angle, Wα=cos p (α), α is the emission angle, I α (x,z) is the emission intensity of the imaging point, and p is the control parameter.
9. An ultrasonic plane wave imaging device as claimed in claim 8, It is characterized in that The GAN module is specifically used for: Building an image generation model based on a diffusion-based generative adversarial network, and training the image generation model according to multi-angle ultrasound imaging data; And generate a second ultrasound image according to the trained image generation model and the emission angle.
10. An ultrasonic plane wave imaging device as claimed in claim 9, It is characterized in that The fusion module is specifically used for: Decomposing the first ultrasonic image and the second ultrasonic image into a plurality of sub-images of different scales respectively; Presetting weight coefficients of the sub-images of the first ultrasonic image and the sub-images of the second ultrasonic image, and performing weighted fusion on the first ultrasonic image and the second ultrasonic image at various scales according to the weight coefficients; The fused images at each scale are obtained, and all the fused images are reconstructed to generate the final ultrasound image.
Citation Information
Cited By
Portable ultrasonic imaging method based on multi-angle plane waves and neural network
CN121330101A
