Ultrasound elastographic image generation and processing system and method

Abstract

This application relates to a system and method for generating and processing ultrasound elastography images. The system includes: an ultrasound image acquisition module for acquiring ultrasound images of living tissue at predetermined locations; an ultrasound elastography image generation module, communicatively connected to the ultrasound image acquisition module, for generating ultrasound elastography images based on the USE-GAN image generation model; a display module for displaying the ultrasound elastography images and the ultrasound image; a manual review module for pre-evaluating the elasticity score of the ultrasound elastography images according to an algorithm program, and determining the elasticity score of the ultrasound elastography images generated by the ultrasound elastography image generation module after manual verification; and an output module for outputting the elasticity score of the ultrasound elastography images determined by the manual review module and a manually determined conclusive report related to the elasticity score. The system and method of this application can be used for auxiliary diagnosis of tissues and organs such as the thyroid, liver, breast, and prostate when ultrasound elastography assessment is required.

Description

Technical Field

[0001] This application relates to ultrasound imaging and processing, and in particular, to an ultrasound elastic image generation and processing system and a method for generating and processing ultrasound elastic images using the ultrasound elastic image generation and processing system. Background Technology

[0002] Ultrasonic elastography utilizes sound waves to detect the hardness properties of tissues. Currently, there are two commonly used ultrasonic elastography methods in clinical practice: stress elastography and shear wave elastography. In stress elastography, a specific probe applies pressure to the tissue surface, causing deformation and displacement of the examined tissue. The magnitude of this deformation and displacement is then measured using ultrasound imaging, indirectly reflecting the tissue hardness. In shear wave elastography, an ultrasonic probe emits acoustic radiation pulses that act on the target tissue. The target tissue deforms under the influence of the acoustic radiation pulses, forming shear waves within the tissue, with their direction perpendicular to the probe's pulse direction. These shear waves propagate slowly within the tissue, making them easily received and analyzed by the probe. The shear wave velocity is directly proportional to tissue hardness; therefore, shear wave velocity can be used to assess tissue hardness. The generated elastography image is a color image with different colors. The hardness of the target tissue is determined by the proportion of specific colors, thus obtaining an elasticity score or elasticity hardness value. In the early stages of many diseases, the conventional two-dimensional ultrasound echoes of diseased and healthy tissues are similar and difficult to distinguish. Ultrasonic elastography, by measuring the elastic parameters of diseased tissues, can reveal the differences between the two, providing more information for early disease diagnosis. This technology has been widely used in the ultrasound diagnosis and differential diagnosis of various diseases in clinical practice, such as the assessment of lesions in tissues and organs such as the thyroid, liver, breast, prostate, and musculoskeletal system.

[0003] Ultrasound elastography has high clinical application value, but obtaining elasticity scores of target tissues and organs through ultrasound elastography requires an additional elastography device, and good results can only be achieved after doctors have received professional training and gained relevant experience. Summary of the Invention

[0004] The technical problem to be solved by this application is to connect the ultrasound image acquisition module to a PC and convert ultrasound images into ultrasound elastography images, so as to realize the analysis and judgment of ultrasound images on-site or remotely and to efficiently screen various lesions.

[0005] To address the aforementioned technical problems, according to one aspect of this application, an ultrasound elasticity image generation and processing system is provided, comprising: an ultrasound image acquisition module for acquiring ultrasound images of living tissue at a predetermined location; an ultrasound elasticity image generation module communicatively connected to the ultrasound image acquisition module for generating ultrasound elasticity images from the ultrasound images received by the data storage module based on the USE-GAN image generation model; a display module for displaying the ultrasound elasticity images generated by the ultrasound elasticity image generation module and the ultrasound images output by the ultrasound image acquisition module; a manual review module for performing a pre-evaluation of the elasticity score of the ultrasound elasticity images generated by the ultrasound elasticity image generation module according to an algorithm program, and determining the elasticity score of the ultrasound elasticity images generated by the ultrasound elasticity image generation module under manual verification; and an output module for outputting the elasticity score of the ultrasound elasticity images determined by the manual review module and a conclusive report related to the elasticity score determined manually.

[0006] According to embodiments of this application, the ultrasonic elastography image generation and processing system may further include a data storage module, which is connected to the output module, and stores the conclusive report of the ultrasonic elastography image as a new case in the data storage module for archiving.

[0007] According to an embodiment of this application, the ultrasound elastic image generation and processing system may further include: performing three-layer downsampling on the ultrasound image through a spatial attention module, performing feature extraction through a six-layer residual module, performing three-layer upsampling, and finally outputting the result through an integrated output channel after the channel attention module.

[0008] According to embodiments of this application, the ultrasonic elastic image generation and processing system may further include: selecting a GAN+L1+Color+VGG19 loss function to make the training process more stable, wherein the GAN loss function is used to generate more realistic images, the L1 loss function focuses on the similarity of each pixel in the image, the Color loss function focuses on the similarity of color distribution in the image, and the VGG19 loss function focuses on the similarity of the overall structure of the image.

[0009] According to embodiments of this application, the communication connection between the data storage module and the ultrasound image acquisition module may include one or both of wired and wireless connections.

[0010] According to embodiments of this application, the ultrasonic elastic image generation module generates ultrasonic elastic images from ultrasonic images received by the data storage module based on the USE-GAN image generation model, which may include simulated stress-based elastic imaging or shear wave elastic imaging.

[0011] According to embodiments of this application, the data storage module, the ultrasound elastography image generation module, the manual review module, and the output module can be integrated into a single PC. In this case, the ultrasound machine and the PC can be remotely connected, or the ultrasound machine and the PC can be located adjacent to each other and connected.

[0012] According to embodiments of this application, the data storage module may have a communication port connected to a database or the cloud. In this case, the data storage module can receive the required data information from the database or the cloud, and the data information stored on the data storage module can be transmitted to the database or the cloud, thereby realizing internet-based information sharing in medical diagnosis.

[0013] According to embodiments of this application, the ultrasound elastography image generation and processing system may further include an optimization module, which continuously optimizes the USE-GAN image generation model to improve accuracy and model stability.

[0014] According to embodiments of this application, the ultrasound image acquisition module may include a conventional B-ultrasound machine.

[0015] According to another aspect of this application, a method for generating and processing ultrasound elasticity images using an ultrasound elasticity image generation and processing system is provided. The ultrasound elasticity image generation and processing system includes an ultrasound image acquisition module, an ultrasound elasticity image generation module, a display module, a manual review module, and an output module. The method includes the following steps: S1, acquiring an ultrasound image of a living tissue at a predetermined location using the ultrasound image acquisition module; S2, generating an ultrasound elasticity image from the ultrasound image using the image generation module based on the USE-GAN image generation model; S3, displaying the ultrasound elasticity image generated by the ultrasound elasticity image generation module and the ultrasound image output by the ultrasound image acquisition module using the display module; S4, performing a pre-evaluation of the elasticity score of the ultrasound elasticity image generated by the ultrasound elasticity image generation module according to an algorithm program using the manual review module, and determining the elasticity score of the ultrasound elasticity image generated by the ultrasound elasticity image generation module under manual verification; and S5, outputting the elasticity score of the ultrasound elasticity image determined by the manual review module and a conclusive report related to the elasticity score determined by the manual review module using the output module.

[0016] According to embodiments of this application, the ultrasonic elastic image generation and processing system may further include a data storage module. In the ultrasonic elastic image generation and processing method, the data storage module is connected to the output module, and the conclusive report of the ultrasonic elastic image is stored as a new case in the data storage module for archiving.

[0017] According to embodiments of this application, the ultrasonic elastic image generation and processing method may further include: performing three-layer downsampling on the ultrasonic image through a spatial attention module, performing feature extraction through a six-layer residual module, performing three-layer upsampling, and finally outputting the result through an integrated output channel after the channel attention module.

[0018] According to embodiments of this application, the ultrasonic elastic image generation and processing method may further include: selecting a GAN+L1+Color+VGG19 loss function to make the training process more stable, wherein the GAN loss function is used to generate more realistic images, the L1 loss function focuses on the similarity of each pixel in the image, the Color loss function focuses on the similarity of color distribution in the image, and the VGG19 loss function focuses on the similarity of the overall structure of the image.

[0019] According to embodiments of this application, the communication connection between the data storage module and the ultrasound image acquisition module may include one or both of wired and wireless connections.

[0020] According to embodiments of this application, the ultrasonic elastic image generation module generates ultrasonic elastic images from ultrasonic images received by the data storage module based on the USE-GAN image generation model, which may include simulated stress-based elastic imaging or shear wave elastic imaging.

[0021] According to embodiments of this application, the data storage module, the ultrasound elastography image generation module, the manual review module, and the output module can be integrated into a single PC. In this case, the ultrasound machine and the PC can be remotely connected, or the ultrasound machine and the PC can be located adjacent to each other and connected.

[0022] According to embodiments of this application, the data storage module may have a communication port that connects to a database or the cloud.

[0023] According to embodiments of this application, the ultrasound elastography image generation and processing system may further include an optimization module, which continuously optimizes the USE-GAN image generation model to improve accuracy and model stability.

[0024] According to embodiments of this application, the ultrasonic elastic image generation and processing method may further include an image preprocessing step before the step of generating the ultrasonic elastic image, wherein the image preprocessing step may crop and extract features from the real-time image.

[0025] According to embodiments of this application, the ultrasound image acquisition module may include a conventional B-ultrasound machine.

[0026] Compared with the prior art, the technical solution provided by the embodiments of this application can achieve at least the following beneficial effects:

[0027] Without changing the existing ultrasound machine and PC, information sharing can be achieved by connecting the ultrasound machine and PC via the Internet, enabling remote operation regardless of whether the ultrasound machine and PC are adjacent or far apart.

[0028] The ultrasound image generated by the ultrasound machine is used by software stored on the PC's storage module to generate an ultrasound elastography image using the USE-GAN image generation model. This ultrasound elastography image is a virtual image that presents the elasticity score of the detected active tissue. The elasticity score represents the stiffness of the detected active tissue. Therefore, when the elasticity score exceeds a certain threshold, doctors can combine this score with other parameters to make a more accurate diagnosis of the patient's lesions.

[0029] Because ultrasound imaging is a relatively inexpensive imaging technology and can be applied to the examination of various soft tissues, the automatic generation of ultrasound elastography images from the ultrasound data allows for the direct determination of elastography scores, which greatly assists doctors in assessing lesions. This enables timely and rapid screening of various lesions, thereby reducing screening costs.

[0030] In the case of telemedicine, the PC can clearly receive ultrasound images transmitted from a remote ultrasound machine, thus enabling telemedicine as if the patient were in the same location. Attached Figure Description

[0031] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings of the embodiments will be briefly described below. Obviously, the drawings described below only relate to some embodiments of this application, and are not intended to limit this application.

[0032] Figure 1 This is a block diagram of an ultrasonic elastic image generation and processing system according to an embodiment of this application.

[0033] Figure 2 This is an optimized module diagram of the ultrasonic elastic image generation and processing system according to an embodiment of this application.

[0034] Figure 3 This is a step block diagram of the ultrasonic elasticity image generation and processing method according to an embodiment of this application.

[0035] Figure 4 This is a model sampling architecture diagram of the ultrasonic elastic image generation and processing method according to an embodiment of this application. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the described embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0037] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this application pertains. The terms “first,” “second,” and similar terms used in the specification and claims of this patent application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms “an” or “a,” and similar terms, do not indicate a limitation of quantity, but rather indicate the presence of at least one.

[0038] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0039] Figure 1 This is a block diagram of an ultrasonic elastic image generation and processing system according to an embodiment of this application.

[0040] See Figure 1 This application provides an ultrasound elasticity image generation and processing system, including an ultrasound image acquisition module, an ultrasound elasticity image generation module, a display module, a manual review module, and an output module. These are described in detail below.

[0041] The ultrasound image acquisition module acquires ultrasound images of living tissue at a predetermined location using ultrasound. According to embodiments of this application, the ultrasound image acquisition module may be, for example, a conventional B-mode ultrasound machine; therefore, in the following description, the ultrasound image acquisition module may be directly referred to as a conventional B-mode ultrasound machine or a B-mode ultrasound machine. However, embodiments of this application are not limited to this, and the ultrasound image acquisition module may also be any other device that acquires ultrasound images of living tissue at a predetermined location using ultrasound, such as A-mode ultrasound, M-mode ultrasound, D-mode ultrasound, color Doppler ultrasound, etc.

[0042] The ultrasound elastography image generation module generates ultrasound elastography images from ultrasound images received by the data storage module based on the USE-GAN image generation model. According to embodiments of this application, the model used for ultrasound elastography image generation is a USE-GAN image generation model based on a U-Net network. The USE-GAN model extracts features from the ultrasound images to generate the corresponding ultrasound elastography images. Ultrasound elastography can be used to study tumors and metastatic diseases that cannot be detected by traditional ultrasound, and can be applied to breast, thyroid, and prostate areas. Tissue elasticity depends on its molecular and microstructure. Clinicians qualitatively evaluate and diagnose breast masses through palpation, based on the close correlation between tissue stiffness or elasticity and the histopathology of the lesion. Similar to the qualitative evaluation and diagnosis by clinicians through palpation, elastography technology provides images of tissue stiffness, that is, information about the tissue characteristics of the lesion. Based on the different elasticity scores of different tissues, the degree of deformation after being subjected to external pressure varies. The changes in the amplitude of echo signal movement before and after pressure are converted into real-time color images. Tissues with low elasticity scores and large displacement changes after pressure are displayed in red; tissues with high elasticity scores and small displacement changes after pressure are displayed in blue; and tissues with moderate elasticity scores are displayed in green. The image color reflects the tissue's stiffness. Elastography technology broadens ultrasound images, compensating for the limitations of conventional ultrasound, and can more vividly display and locate lesions.

[0043] The display module can display the elastic image generated by the elastic image generation module and the ultrasound image output by the ultrasound image acquisition module. The display module can be a black and white or color display.

[0044] The manual review module performs a preliminary evaluation of the elasticity score of the ultrasound elasticity images generated by the ultrasound elasticity image generation module based on the algorithm program, and determines the elasticity score of the ultrasound elasticity images generated by the ultrasound elasticity image generation module under manual intervention and verification. The manual review module, also known as the error correction module, is the process of correcting and verifying the ultrasound elasticity images generated by the ultrasound elasticity image generation module, thereby ensuring that the elasticity score of the ultrasound elasticity images formed from ultrasound images is accurate.

[0045] The output module outputs the elasticity score of the ultrasound elastography image determined by the manual review module, along with a manually determined conclusive report related to the elasticity score. It also stores the conclusive report of the ultrasound elastography image as a new case study in the data storage module for archiving. After completing the conclusive report related to the elasticity score, the ultrasound image, ultrasound elastography image, and elasticity score are stored as the basic data information of the new case study in the data storage module for archiving. This process continuously accumulates experience, enabling in-depth machine learning and training the ultrasound elastography image generation module to more accurately generate ultrasound elastography images and their elasticity scores for ultrasound images in future work.

[0046] According to embodiments of this application, the ultrasound elastography image generation and processing system may further include a data storage module. The data storage module is connected to the output module and stores conclusive reports of the ultrasound elastography images as new cases for archiving.

[0047] According to embodiments of this application, the data storage module can be a volatile or non-volatile memory, and therefore may include a hard disk, a portion of a hard disk partition, flash memory, cache memory, etc. Preferably, the data storage module includes at least a cache portion for receiving and processing ultrasound images and a hard disk portion for storing data, and is preferably part of PC computer hardware resources. According to embodiments of this application, the communication connection between the data storage module and the ultrasound image acquisition module may include one or both of wired and wireless connections. However, embodiments of this application are not limited to this, and the communication connection between the data storage module and the ultrasound image acquisition module may also be a non-real-time signal transmission connection. For example, transferring ultrasound images to the data storage module via a removable hard disk, USB flash drive, email, etc., or transmitting ultrasound images in real time or indirectly via communication tools such as WeChat, are also within the scope of protection of this application.

[0048] Figure 2 This is an optimized module diagram of the ultrasonic elastic image generation and processing system according to an embodiment of this application.

[0049] like Figure 2 As shown, according to an embodiment of this application, the ultrasonic elastic image generation module generates ultrasonic elastic images from the ultrasonic images received by the data storage module based on the USE-GAN image generation model, which may include simulated stress-based elastic imaging or shear wave elastic imaging.

[0050] Currently, there are two commonly used ultrasound elastography methods in clinical practice: stress elastography and shear wave elastography. In stress elastography, a specific probe applies pressure to the tissue surface, causing deformation and displacement. The magnitude of this deformation and displacement is then measured using ultrasound imaging, indirectly reflecting tissue stiffness. In shear wave elastography, an ultrasound probe emits acoustic radiation pulses that act on the target tissue. The target tissue deforms under the influence of these pulses, creating shear waves perpendicular to the probe's pulse direction. These shear waves propagate slowly within the tissue, making them easily received and analyzed by the probe. Since shear wave velocity is directly proportional to tissue stiffness, it can be used to assess tissue stiffness.

[0051] However, according to embodiments of this application, the elastic image generation module generates an elastic image from the ultrasound image received by the data storage module based on the USE-GAN image generation model. There are mainly two methods for generating elastic images: pressure-based and shear wave-based. The elastic image generated according to embodiments of this application is a virtual elastic image generated on the ultrasound image based on the USE-GAN image generation model. This virtual elastic image can simulate an elastic imaging map generated using elastic imaging technology and has a high degree of consistency with the elastic image generated using elastic imaging technology.

[0052] Although not shown in the accompanying drawings, according to embodiments of this application, the data storage module, ultrasound elastography image generation module, manual review module, and output module can be integrated into a single PC. In this case, the ultrasound image acquisition module (conventional B-ultrasound machine) and the PC can be remotely connected, or the B-ultrasound machine and the PC can be adjacent to each other and connected. When the B-ultrasound machine and the PC are adjacent to each other and connected, the B-ultrasound examination time will be greatly shortened, and various nodules can be efficiently screened for benign or malignant tumors during the almost simultaneous generation of ultrasound elastography images. When the B-ultrasound machine and the PC are remotely connected, telemedicine can be conveniently realized, helping remote areas share medical resources from developed areas. Therefore, the ultrasound elastography image generation and processing system according to embodiments of this application will make a significant contribution to the sharing of medical resources in telemedicine.

[0053] like Figure 2 As shown in the embodiments of this application, the data storage module may have a communication port connected to a database or the cloud. In this case, the data storage module can receive the required data information from the database or the cloud, and the data information stored on the data storage module can be transmitted to the database or the cloud, thereby realizing internet and information sharing in medical diagnosis.

[0054] like Figure 2 As shown in the embodiments of this application, the ultrasound elastography image generation and processing system may further include an optimization module. This optimization module continuously optimizes the USE-GAN image generation model to improve accuracy and model stability. The optimization module stores learning software, update software, and upgrade software. The optimization module has or supports deep learning capabilities. The optimization module is interconnected with the ultrasound elastography image generation module and continuously optimizes the USE-GAN image generation model on the ultrasound elastography image generation module.

[0055] The following describes, with reference to the accompanying drawings, a method for generating and processing ultrasonic elastic images using an ultrasonic elastic image generation and processing system according to an embodiment of this application.

[0056] Figure 3 This is a step block diagram of the ultrasonic elasticity image generation and processing method according to an embodiment of this application.

[0057] See Figure 3 According to embodiments of this application, a method for generating and processing ultrasonic elastic images using an ultrasonic elastic image generation and processing system is provided.

[0058] As referenced above Figure 1 and 2 As described, the ultrasound elasticity image generation and processing system according to this application includes an ultrasound image acquisition module, an ultrasound elasticity image generation module, a display module, a manual review module, and an output module. Preferably, the ultrasound elasticity image generation and processing system according to this application may further include a data storage module, a database, an optimization module, a printing module, and may also be connected to the cloud. To avoid redundancy, in the following description, references to previously mentioned... Figure 1 and 2 Features that are the same or similar to those described in the technical solutions will not be described again.

[0059] like Figure 3 As shown, the ultrasonic elasticity image generation and processing method according to this application includes the following steps: S1, acquiring an ultrasonic image of a living tissue at a predetermined location through an ultrasonic image acquisition module; S2, generating an ultrasonic elasticity image from the ultrasonic image using an image generation module based on the USE-GAN image generation model; S3, displaying the ultrasonic elasticity image generated by the ultrasonic elasticity image generation module and the ultrasonic image output by the ultrasonic image acquisition module through a display module; S4, performing an elasticity score pre-evaluation on the ultrasonic elasticity image generated by the ultrasonic elasticity image generation module according to an algorithm program through a manual review module, and determining the elasticity score of the ultrasonic elasticity image generated by the ultrasonic elasticity image generation module under manual verification; and S5, outputting the elasticity score of the ultrasonic elasticity image determined by the manual review module and a conclusive report related to the elasticity score determined by the manual review module through an output module.

[0060] According to embodiments of this application, the ultrasonic elastic image generation and processing system may further include a data storage module. In the ultrasonic elastic image generation and processing method, the data storage module is connected to the output module, and the conclusive report of the ultrasonic elastic image is stored as a new case in the data storage module for archiving.

[0061] According to embodiments of this application, the ultrasound image can be a conventional B-mode ultrasound image. However, embodiments of this application are not limited to this, and the ultrasound image acquisition module can also be any other device that acquires ultrasound images of living tissue at a predetermined location via ultrasound, such as A-mode ultrasound, M-mode ultrasound, D-mode ultrasound, color Doppler ultrasound, etc. Therefore, the ultrasound image can also be an A-mode ultrasound, M-mode ultrasound, D-mode ultrasound, color Doppler ultrasound, etc.

[0062] According to embodiments of this application, the display module can display the ultrasound elastography image generated by the ultrasound elastography image generation module and the ultrasound image output by the ultrasound image acquisition module, thereby providing doctors with an intuitive ultrasound elastography image.

[0063] According to embodiments of this application, the communication connection between the data storage module and the ultrasound image acquisition module may include one or both of wired and wireless connections. However, embodiments of this application are not limited to this; the communication connection between the data storage module and the ultrasound image acquisition module may also be a non-real-time signal transmission connection. For example, ultrasound images can be transferred to the data storage module via a removable hard drive, USB flash drive, email, etc., or ultrasound images can be transmitted in real time or indirectly via communication tools such as WeChat. Therefore, the connection mentioned herein also includes transferring ultrasound images through necessary transmission means, which is also within the scope of protection of this application.

[0064] According to embodiments of this application, the ultrasonic elastic image generation module generates ultrasonic elastic images from ultrasonic images received by the data storage module based on the USE-GAN image generation model, which may include simulated stress-based elastic imaging or shear wave elastic imaging.

[0065] As mentioned earlier, the ultrasound elastography image generation module generates ultrasound elastography images from the ultrasound images received by the data storage module based on the USE-GAN image generation model. According to embodiments of this application, the model used for ultrasound elastography image generation is a USE-GAN image generation model based on a U-Net network. The USE-GAN model extracts features from the ultrasound images to generate corresponding ultrasound elastography images; therefore, ultrasound elastography images are also called ultrasound elastography images or ultrasound elastography imaging. Ultrasound elastography can be used to study tumors and metastatic diseases that cannot be detected by traditional ultrasound, and can be applied to breast, thyroid, and prostate areas. Tissue elasticity depends on its molecular and microstructure. Clinicians qualitatively evaluate and diagnose breast masses through palpation, based on the close correlation between tissue stiffness or elasticity and the histopathology of the lesion. Similar to the qualitative evaluation and diagnosis by clinicians through palpation, elastography technology provides images of tissue stiffness, that is, information about the tissue characteristics of the lesion. Based on the different elasticity scores of different tissues, the degree of deformation after being subjected to external force varies. The changes in the amplitude of echo signal movement before and after compression are converted into real-time color images. Tissues with low elasticity scores and large displacement changes after compression are displayed in red, tissues with high elasticity scores and small displacement changes after compression are displayed in blue, and tissues with medium elasticity scores are displayed in green. The image color reflects the tissue's stiffness. Elastography technology broadens ultrasound images, compensating for the shortcomings of conventional ultrasound, and can more vividly display and locate lesions. Therefore, the ultrasound elastic image generation and processing method according to the embodiments of this application uses this mechanism to generate ultrasound elastic images from ultrasound images received by the data storage module based on the USE-GAN image generation model, thereby simulating the ultrasound elastic image formed by tissue deformation after being subjected to external force.

[0066] According to embodiments of this application, the data storage module, the ultrasound elastography image generation module, the manual review module, and the output module can be integrated into a single PC. In this case, the ultrasound machine and the PC can be connected remotely, or the ultrasound machine and the PC can be positioned adjacent to each other and connected via communication. The ultrasound elastography image generation and processing method according to embodiments of this application can process ultrasound images to form ultrasound elastography images using both adjacent placement of the ultrasound machine and the PC and remote communication.

[0067] According to embodiments of this application, the data storage module may have a communication port connected to a database or the cloud. As mentioned above, the data storage module generates ultrasound elastography images from ultrasound images received by the data storage module based on the USE-GAN image generation model. Simultaneously, the data storage module possesses machine deep learning capabilities. When the data storage module has a communication port connected to a database or the cloud, the USE-GAN image generation model is continuously updated and upgraded; therefore, the ultrasound elastography image generation and processing method according to embodiments of this application is correspondingly more efficient and accurate.

[0068] To achieve deep learning of the USE-GAN image generation model, according to embodiments of this application, the ultrasound elastography image generation and processing system may further include an optimization module. This optimization module continuously optimizes the USE-GAN image generation model to improve accuracy and model stability. Simultaneously, the optimization module itself also possesses good deep learning capabilities and is convenient for updates, upgrades, and modifications.

[0069] According to embodiments of this application, the ultrasonic elastic image generation and processing method may further include an image preprocessing step prior to the ultrasonic elastic image generation step. The image preprocessing step can crop and extract features from the real-time image. The image preprocessing step is a necessary step in image processing, and its main function is to correct image deviations and facilitate model analysis, for example, by enlarging or reducing the image size.

[0070] Figure 4 This is a model sampling architecture diagram of the ultrasonic elastic image generation and processing method according to an embodiment of this application.

[0071] See Figure 4 According to embodiments of this application, the ultrasonic elastic image generation and processing method may further include: performing three-layer downsampling on the ultrasonic image through a spatial attention module, performing feature extraction through a six-layer residual module, performing three-layer upsampling, and finally outputting the result through an integrated output channel after the channel attention module.

[0072] like Figure 4 As shown, the ultrasonic elastography image generation and processing system according to an embodiment of this application further includes a spatial attention module, a downsampling module, a residual module, a channel attention module, and an upsampling module. For distinction, these modules are... Figure 4 Different colors are used to represent the images. In the ultrasonic elastography image generation and processing method according to the embodiments of this application, three-layer downsampling, six-layer residual feature extraction, and three-layer upsampling are performed. Through the above sampling operations, a... Figure 4 The rightmost color ultrasound elastography image is shown. This color ultrasound elastography image can provide clues for doctors or operators, and allow for further analysis and judgment of key areas, thereby making a diagnosis.

[0073] According to embodiments of this application, the ultrasonic elastic image generation and processing method may further include: selecting a GAN+L1+Color+VGG19 loss function to make the training process more stable, wherein the GAN loss function is used to generate more realistic images, the L1 loss function focuses on the similarity of each pixel in the image, the Color loss function focuses on the similarity of color distribution in the image, and the VGG19 loss function focuses on the similarity of the overall structure of the image. This method involves further training and optimization of the USE-GAN image generation model.

[0074] Compared with the prior art, the technical solution of the ultrasonic elastography image generation and processing system and method provided in the embodiments of this application can achieve at least the following beneficial effects:

[0075] Without changing the existing ultrasound machine and PC, information sharing can be achieved by connecting the ultrasound machine and PC via the Internet, enabling remote operation regardless of whether the ultrasound machine and PC are adjacent or far apart.

[0076] The ultrasound image generated by the ultrasound machine is used by software stored on the PC's storage module to generate an ultrasound elastography image using the USE-GAN image generation model. This ultrasound elastography image is a virtual image that presents the elasticity score of the detected active tissue. The elasticity score represents the degree of elastic deformation of the detected active tissue, i.e., its stiffness. Therefore, when the elasticity score exceeds a certain threshold, doctors can combine this score with other parameters to make a more accurate diagnosis of the patient's lesions.

[0077] Because ultrasound imaging is a relatively inexpensive imaging technology and can be applied to the examination of various soft tissues, the automatic generation of ultrasound elastography images from ultrasound data allows for the direct determination of elastography scores, greatly aiding doctors in diagnosing lesions. This enables timely and rapid screening of various lesions, thereby reducing screening costs.

[0078] In the case of telemedicine, the PC can clearly receive ultrasound images transmitted from a remote ultrasound machine, thus enabling telemedicine as if the patient were in the same location.

[0079] The above description is merely an exemplary embodiment of this application and is not intended to limit the scope of protection of this application. The scope of protection of this application is determined by the appended claims.

Claims

1. An ultrasonic elastography image generation and processing system, comprising: The ultrasound image acquisition module acquires ultrasound images of living tissue at a predetermined location using ultrasound. An ultrasound elasticity image generation module is communicatively connected to an ultrasound image acquisition module, and generates an ultrasound elasticity image from the ultrasound image based on the USE-GAN image generation model. The GAN+L1+Color+VGG19 loss function is loaded onto the ultrasonic elastic image generation module. The GAN+L1+Color+VGG19 loss function is selected on the USE-GAN image generation model to make the training process more stable. Among them, the GAN loss function is used to generate more realistic images, the L1 loss function focuses on the similarity of each pixel in the image, the Color loss function focuses on the similarity of color distribution in the image, and the VGG19 loss function focuses on the similarity of the overall structure of the image. The display module is a color display that displays the ultrasonic elastic image generated by the ultrasonic elastic image generation module and the ultrasonic image output by the ultrasonic image acquisition module. The manual review module performs a preliminary evaluation of the elasticity score of the ultrasonic elastic image generated by the ultrasonic elastic image generation module based on the algorithm program, and determines the elasticity score of the ultrasonic elastic image generated by the ultrasonic elastic image generation module under manual verification. The output module outputs the elasticity score of the ultrasound elastography image determined by the manual review module, as well as a conclusive report related to the elasticity score determined manually. A data storage module is connected to the output module and stores the conclusive report of the ultrasound elastography image as a new case in the data storage module for archiving. Spatial attention module, downsampling module, residual module, channel attention module, and upsampling module; and The optimization module continuously optimizes the USE-GAN image generation model to improve accuracy and model stability. The optimization module stores learning software, update software, and upgrade software. The optimization module has deep learning capabilities. The optimization module is interconnected with the ultrasound elasticity image generation module to continuously optimize the USE-GAN image generation model on the ultrasound elasticity image generation module. A method for processing ultrasound elastography images includes the following steps: S1. Acquire ultrasound images of living tissue at a predetermined location using the ultrasound image acquisition module; S2. An ultrasonic elastic image is generated from the ultrasonic image using an image generation module based on the USE-GAN image generation model; S3. Display the ultrasonic elastic image generated by the ultrasonic elastic image generation module and the ultrasonic image output by the ultrasonic image acquisition module through the display module; S4. The ultrasonic elastic image generated by the ultrasonic elastic image generation module is pre-evaluated by the manual review module according to the algorithm program, and the elastic score of the ultrasonic elastic image generated by the ultrasonic elastic image generation module is determined under the condition of manual verification. S5. Output the elasticity score of the ultrasound elasticity image determined by the manual review module and the conclusive report related to the elasticity score determined by the manual review module through the output module. S6. The ultrasound image is subjected to three-layer downsampling, six-layer residual feature extraction, and three-layer upsampling via a spatial attention module. Finally, the result is output through the integrated output channel after the channel attention module. S7. Select the GAN+L1+Color+VGG19 loss function to make the training process more stable. The GAN loss function is used to generate more realistic images, the L1 loss function focuses on the similarity of each pixel in the image, the Color loss function focuses on the similarity of the color distribution in the image, and the VGG19 loss function focuses on the similarity of the overall structure of the image. The USE-GAN image generation model includes performing three-layer downsampling on the ultrasound image through a spatial attention module, feature extraction through a six-layer residual module, three-layer upsampling, and finally outputting the result through an integrated output channel after the channel attention module.

2. The ultrasonic elastography image generation and processing system as described in claim 1, wherein, The communication connection between the data storage module and the ultrasound image acquisition module can be a wired connection or a wireless connection.

3. The ultrasonic elastography image generation and processing system as described in claim 1, wherein, The ultrasonic elastic image generation module generates ultrasonic elastic images, including simulated stress-based elastic imaging, based on the USE-GAN image generation model and the ultrasonic images received by the data storage module.

4. The ultrasonic elastography image generation and processing system as described in claim 1, wherein, The ultrasonic elastic image generation module generates ultrasonic elastic images, including simulated shear wave elastic imaging, based on the USE-GAN image generation model and the ultrasonic images received by the data storage module.

5. The ultrasonic elastography image generation and processing system as described in claim 1, wherein, The data storage module, the ultrasonic elastography image generation module, the manual review module, and the output module are integrated into a single PC.

6. The ultrasonic elastography image generation and processing system as described in claim 5, wherein, The ultrasound image acquisition module is connected to the PC via a remote communication connection.

7. The ultrasonic elastography image generation and processing system as described in claim 1, wherein, The data storage module has a communication port that connects to a database or the cloud.

8. The ultrasonic elastography image generation and processing system as described in claim 1, wherein, The ultrasound image acquisition module includes a conventional B-ultrasound machine.

9. The ultrasonic elastic image generation and processing system as described in claim 1 further includes an image preprocessing step before the ultrasonic elastic image generation step, wherein the image preprocessing step performs cropping and feature extraction on the real-time image.

Images