Ultrasonic image processing method and device and storage medium

Abstract

The invention relates to the technical field of ultrasonic imaging, and discloses an ultrasonic image processing method and device and a storage medium, and the method comprises the steps: obtaining a collected original ultrasonic image; identifying a cross section image of a blood vessel in the original ultrasonic image; performing amplification processing on the cross section image according to a target super-resolution amplification model obtained by training; and displaying the amplified blood vessel image. According to the method, the cross section image of the blood vessel is recognized and amplified, so that the amplified cross section image of the blood vessel can be displayed to a user, the blood vessel can be displayed to the user more clearly, the user can judge the position of the blood vessel more accurately, accurate positioning of the blood vessel is achieved, the working efficiency of the user can be improved, and the user experience is improved. And misoperation can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of ultrasonic imaging, and particularly relates to a method, device, and storage medium for processing ultrasonic images. Background Art

[0002] PICC (Peripherally Inserted Central Venous Catheters) refers to a catheter inserted through a peripheral vein (such as the basilic vein, cephalic vein, brachial vein, etc.) with the tip of the catheter reaching the superior vena cava, and is widely used in aspects such as tumor chemotherapy, establishment of adult postoperative parenteral nutrition access, and establishment of premature infant nutrition access.

[0003] This process generally relies on an ultrasonic imaging device. For example, medical staff first use the ultrasonic probe of the ultrasonic imaging device to scan the cross-section of the blood vessel to be punctured. After the ultrasonic imaging device processes the scanned data, it displays the processed image to guide the puncture. Then, a guide wire is sent into the blood vessel along the puncture needle. After several operations, the catheter is sent through the intubation sheath; after the initial catheter placement, it is necessary to switch to the ECG (electrocardiogram) mode, observe the state of the P wave in the electrocardiogram signal, and adjust the catheter depth; after the catheter placement is completed, it is necessary to use the ultrasonic probe to scan the cross-section of the neck blood vessel again to observe whether there are problems such as ectopia at the internal jugular vein.

[0004] According to the current ultrasonic image processing method, it is sometimes difficult to determine the position of the blood vessel in the image processed by the ultrasonic imaging device, resulting in subsequent prone to misoperation problems. Summary of the Invention

[0005] In view of this, the present invention provides a method, device, and storage medium for processing ultrasonic images.

[0006] In a first aspect, the present invention provides a method for processing ultrasonic images, including:

[0007] Obtain the acquired original ultrasonic image;

[0008] Identify the cross-sectional image of the blood vessel in the original ultrasonic image;

[0009] Perform magnification processing on the cross-sectional image according to the trained target super-resolution magnification model;

[0010] Display the magnified blood vessel image.

[0011] In some optional embodiments, the performing magnification processing on the cross-sectional image according to the trained target super-resolution magnification model includes:

[0012] Determine the magnification factor that matches the size of the cross-sectional image;

[0013] Input the cross-sectional image into a target super-resolution magnification model to generate a cross-sectional image magnified to the magnification factor; the magnification factor of the target super-resolution magnification model includes at least the magnification factor.

[0014] In some alternative embodiments, the determining the magnification factor that matches the size of the cross-sectional image includes:

[0015] Determine the ratio between the size of the cross-sectional image and the size of the original ultrasound image;

[0016] Determine the matching magnification factor according to the ratio; there is a negative correlation between the magnification factor and the ratio.

[0017] In some alternative embodiments, the determining the ratio between the size of the cross-sectional image and the size of the original ultrasound image includes:

[0018] Determine a first ratio between the width of the cross-sectional image and the width of the original ultrasound image, and a second ratio between the height of the cross-sectional image and the height of the original ultrasound image;

[0019] The determining the matching magnification factor according to the ratio includes:

[0020] In the case where the first ratio is less than a first threshold and the second ratio is less than a second threshold, determine a matching first magnification factor;

[0021] In the case where the first ratio is greater than a third threshold and the second ratio is greater than a fourth threshold, determine a matching second magnification factor;

[0022] Wherein, the first threshold is less than or equal to the third threshold, and the second threshold is less than or equal to the fourth threshold; the first magnification factor is greater than the second magnification factor.

[0023] In some alternative embodiments, the target super-resolution magnification model is obtained by training a super-resolution magnification model to be trained with a pre-determined training sample;

[0024] The training sample includes a sample image as input and a label image as output; the label image is an image of the cross-section of the blood vessel area, and the sample image is an image generated by downsampling the label image.

[0025] In some alternative embodiments, the identifying the cross-sectional image of the blood vessel in the original ultrasound image includes:

[0026] Identify the position boundary values of blood vessels in the original ultrasound image according to a pre-trained segmentation model;

[0027] Identify the cross-sectional image of the blood vessel from the original ultrasound image according to the position boundary values.

[0028] In some alternative embodiments, after identifying the cross-sectional image of the blood vessel in the original ultrasound image, the method further includes: highlighting the cross-sectional image in the original ultrasound image.

[0029] In some alternative embodiments, highlighting the cross-sectional image in the original ultrasound image includes:

[0030] Determine a bounding box that matches the position boundary values;

[0031] Frame the cross-sectional image in the original ultrasound image based on the bounding box.

[0032] In some alternative embodiments, identifying the cross-sectional image of the blood vessel in the original ultrasound image includes:

[0033] Identify the cross-sectional image of the blood vessel in one original ultrasound image collected within a processing cycle; the processing cycle is greater than the sampling cycle of the ultrasound image, and the number of processing cycles is multiple;

[0034] The magnifying the cross-sectional image according to a trained target super-resolution magnification model includes:

[0035] Within the processing cycle, magnify the cross-sectional image in the corresponding original ultrasound image according to the target super-resolution magnification model.

[0036] In some alternative embodiments, before identifying the cross-sectional image of the blood vessel in the original ultrasound image, it further includes:

[0037] Obtain a trigger instruction for identifying and magnifying blood vessels; after obtaining the trigger instruction, then perform the process of identifying the cross-sectional image of the blood vessel in the original ultrasound image.

[0038] In some alternative embodiments, displaying the magnified cross-sectional image includes:

[0039] Display the original ultrasound image and display the magnified cross-sectional image in the original ultrasound image;

[0040] Wherein, the magnified cross-sectional image is located in the edge region of the original ultrasound image; or, the central position of the magnified cross-sectional image corresponds to the central position of the cross-sectional image in the original ultrasound image.

[0041] In a second aspect, the present invention provides an electronic device, including: a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the ultrasound image processing method according to the first aspect or any corresponding embodiment thereof.

[0042] In a third aspect, the present invention provides an ultrasound imaging device, including: an ultrasound probe and a host; the ultrasound probe is configured to acquire an original ultrasound image; the host is configured to execute the ultrasound image processing method according to the first aspect or any corresponding embodiment thereof.

[0043] In a fourth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to cause a computer to execute the ultrasound image processing method according to the first aspect or any corresponding embodiment thereof.

[0044] By identifying the cross-sectional image of a blood vessel and magnifying the cross-sectional image, the present invention can display the magnified cross-sectional image of the blood vessel to the user, enabling the blood vessel to be more clearly displayed to the user, allowing the user to more accurately judge the position of the blood vessel, realizing precise positioning of the blood vessel, not only improving the work efficiency of the user, but also reducing misoperations. By magnifying the cross-sectional image, the influence of image noise on blood vessel positioning can be reduced, and it is also convenient for the user to observe whether the catheter is in place; moreover, only the cross-sectional image can be magnified, which can reduce the amount of data for the magnification process and ensure the processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the related art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the related art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0046] Figure 1 is a flowchart of the ultrasound image processing method according to an embodiment of the present invention;

[0047] Figure 2 is a flowchart of another ultrasound image processing method according to an embodiment of the present invention;

[0048] Figure 3It is a schematic flowchart of another method for processing an ultrasonic image according to an embodiment of the present invention;

[0049] Figure 4 It is a schematic diagram of identifying and magnifying a cross-sectional image of a blood vessel according to an embodiment of the present invention;

[0050] Figure 5 It is a schematic structural diagram of an ultrasonic imaging device according to an embodiment of the present invention;

[0051] Figure 6 It is a block diagram of the structure of a processing device for ultrasonic images according to an embodiment of the present invention;

[0052] Figure 7 It is a schematic hardware structure diagram of an electronic device according to an embodiment of the present invention. Detailed implementation manners

[0053] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0054] According to an embodiment of the present invention, an embodiment of a method for processing an ultrasonic image is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0055] In this embodiment, a method for processing an ultrasonic image is provided, which can be used in an ultrasonic forming device, and the ultrasonic imaging device can display an ultrasonic image. Figure 1 It is a flowchart of a method for processing an ultrasonic image according to an embodiment of the present invention. As Figure 1 shown, this process includes the following steps S101 to step S104.

[0056] Step S101, obtain the collected original ultrasonic image.

[0057] Among them, during the operation of the ultrasonic imaging device, it can generate a corresponding ultrasonic image based on ultrasonic signals; generally, the ultrasonic imaging device can directly display the collected ultrasonic image, while in this embodiment, the collected ultrasonic image is further processed. For the convenience of description, this ultrasonic image is referred to as the original ultrasonic image.

[0058] In this embodiment, the original ultrasound image is an ultrasound image containing blood vessels; and the ultrasound imaging device collects a cross-section of the blood vessel, that is, the cross-section of the blood vessel is shown in the original ultrasound image. For example, during the PICC catheterization process, the corresponding blood vessel can be scanned by the ultrasound imaging device to generate an original ultrasound image containing the cross-section of the blood vessel.

[0059] For example, the ultrasound probe of the ultrasound imaging device can be controlled to emit ultrasonic waves to the object to be measured and receive the ultrasonic waves returned from the object to be measured to obtain an echo signal; the echo signal is subjected to front-end processing, which may specifically include time gain compensation, analog-to-digital conversion, beam synthesis, demodulation, filtering, envelope extraction, logarithmic compression, etc., so as to form an original image signal. Then, post-processing such as gain control, dynamic range compression, coordinate transformation, etc. is performed to form the final ultrasound image, that is, the original ultrasound image. Among them, the object to be measured can be a biological tissue containing blood vessels, such as the area where PICC catheterization is required.

[0060] Step S102: Identify the cross-sectional image of the blood vessel in the original ultrasound image.

[0061] In this embodiment, when performing blood vessel recognition on the original ultrasound image, the blood vessel area in the original ultrasound image can be recognized, so that the image of the area where the blood vessel is located in the original ultrasound image can be determined. Among them, the cross-section of the blood vessel is shown in the original ultrasound image, and the image of the area where the blood vessel is located is called the cross-sectional image. It can be understood that the cross-sectional image is an image including the cross-section of the blood vessel.

[0062] Step S103: Perform magnification processing on the cross-sectional image according to the trained target super-resolution magnification model.

[0063] In this embodiment, a magnification model capable of realizing image magnification is pre-trained; among them, the magnification model is realized based on the super-resolution model. Specifically, a preset super-resolution magnification model can be trained to generate a super-resolution magnification model capable of realizing image magnification. For the convenience of description, the trained super-resolution magnification model is called the target super-resolution magnification model, that is, the target super-resolution magnification model has the function of image magnification.

[0064] After the cross-sectional image of the blood vessel is recognized, the cross-sectional image is input into the target super-resolution magnification model, and the magnification processing of the cross-sectional image can be realized; based on the output result of the target super-resolution magnification model, the magnified cross-sectional image can be obtained. For example, the cross-sectional image can be magnified to an image of a specific size.

[0065] Step S104: Display the magnified blood vessel image.

[0066] In this embodiment, the ultrasonic imaging device has a display function and can display the magnified blood vessel image, that is, it can display the magnified blood vessel area to the user, facilitating the user to observe the blood vessels more clearly. Among them, it can only display the magnified blood vessel image to the user, or it can also display the original ultrasonic image and the magnified blood vessel image to the user at the same time.

[0067] Since there is generally a lot of noise in ultrasonic images, these noises will interfere with the user's positioning of blood vessels in the ultrasonic images. Although the current denoising processing methods can reduce the influence of noise to a certain extent, the denoising effect is limited, and the ultrasonic image can only be displayed after denoising, which also affects the real-time display of the ultrasonic image. In this embodiment, by magnifying the cross-sectional image of the blood vessel, the blood vessel characteristics can be highlighted to a certain extent, thereby reducing the interference of image noise and being more beneficial for positioning blood vessels. In addition, after identifying the cross-sectional image of the area where the blood vessel is located, only this cross-sectional image can be magnified, that is, it is not necessary to magnify the complete original ultrasonic image, which can reduce the amount of data for magnification processing and ensure that the displayed magnified cross-sectional image has a certain real-time performance. And since the cross-sectional image is a part of the original ultrasonic image, the original ultrasonic image can still be displayed simultaneously when the magnified cross-sectional image is displayed, so that the ultrasonic image can be displayed in real time.

[0068] For example, during the PICC catheterization process, after imaging the cross-section of the blood vessel based on the ultrasonic imaging device, it is necessary to determine the position of the blood vessel and observe whether the catheter is in the blood vessel. However, there are many interfering noises in the ultrasonic image, which easily affect the user's judgment. In this embodiment, after collecting the ultrasonic image, the cross-sectional image of the blood vessel in the ultrasonic image can be automatically identified, and this cross-sectional image can be magnified and the magnified cross-sectional image can be displayed to the user. By observing the magnified cross-sectional image, the user can more accurately determine the characteristics of the blood vessel area and facilitate observing whether the catheter is correctly located in the blood vessel, assisting the user to judge whether the catheter is in place and enabling a more accurate and efficient catheterization process to be completed.

[0069] The method for processing ultrasonic images provided in this embodiment can display the magnified cross-sectional image of the blood vessel to the user by identifying the cross-sectional image of the blood vessel and magnifying this cross-sectional image, which can more clearly display the blood vessel to the user, enabling the user to more accurately judge the position of the blood vessel, realizing accurate blood vessel positioning, not only improving the work efficiency of the user, but also reducing misoperations. By magnifying the cross-sectional image, the influence of image noise on blood vessel positioning can be reduced, and it is also convenient for the user to observe whether the catheter is in place. And only the cross-sectional image can be magnified, which can reduce the amount of data for magnification processing and ensure the processing efficiency.

[0070] In this embodiment, a method for processing ultrasonic images is provided, which can be used in ultrasonic forming devices. Figure 2is a flowchart of a method for processing ultrasonic images according to an embodiment of the present invention, as Figure 2 shown, this process includes the following steps.

[0071] Step S201, obtain the collected original ultrasonic image.

[0072] Among them, for details, please refer to Figure 1 step S101 of the embodiment shown, which will not be elaborated here.

[0073] In some alternative embodiments, the user can independently select whether to magnify and display blood vessels. When it is necessary to magnify and display blood vessels, the user can input a corresponding instruction to activate the function of automatically identifying and magnifying blood vessels. Specifically, before step S201 "identifying the cross-sectional image of blood vessels in the original ultrasonic image" above, the method further includes the following step A1.

[0074] Step A1, obtain a trigger instruction for identifying and magnifying blood vessels. After obtaining the trigger instruction, then execute the process of step S201 "identifying the cross-sectional image of blood vessels in the original ultrasonic image" above.

[0075] Specifically, when the user operates the ultrasonic imaging device, the ultrasonic imaging device can collect ultrasonic images in real time and display them; if it is necessary to magnify and display blood vessels currently, for example, it is necessary to magnify and display blood vessels during the PICC catheterization process, the user can actively input a trigger instruction for identifying and magnifying blood vessels to activate the function of identifying and magnifying blood vessels. Among them, the user can input the trigger instruction by means of buttons (physical buttons or virtual buttons), voice, gestures, etc. This embodiment does not limit the way for the user to input the trigger instruction.

[0076] After the ultrasonic imaging device obtains the trigger instruction, it can respond and activate the function of identifying and magnifying blood vessels, that is, it can identify the collected original ultrasonic image to identify the cross-sectional image of blood vessels therein, and then magnify the cross-sectional image and display the magnified cross-sectional image to the user.

[0077] Step S202, identify the cross-sectional image of blood vessels in the original ultrasonic image.

[0078] Among them, for details, please refer to Figure 1 step S102 of the embodiment shown, which will not be elaborated here.

[0079] Step S203, perform magnification processing on the cross-sectional image according to the trained target super-resolution magnification model.

[0080] Among them, based on the actual situation of the cross-sectional image, an appropriate magnification factor can be determined, and based on this, the cross-sectional image is magnified. Specifically, the above step S203 includes the following steps S2031 to step S2032.

[0081] Step S2031, determine the magnification factor that matches the size of the cross-sectional image.

[0082] In this embodiment, the identified cross-sectional image has a certain size, which can represent the size of the cross-sectional image, and specifically can include width, height, area, etc. Based on the size of the cross-sectional image, the corresponding magnification factor is determined. In order to achieve a better magnification effect, the smaller the size of the cross-sectional image, the larger the corresponding magnification factor, that is, there is a negative correlation between the size of the cross-sectional image and the magnification factor.

[0083] Step S2032, input the cross-sectional image into the target super-resolution magnification model to generate a cross-sectional image magnified to the magnification factor; the magnification factor of the target super-resolution magnification model includes at least the magnification factor.

[0084] In this embodiment, the magnification factor of the trained target super-resolution magnification model includes at least the magnification factor, that is, the target super-resolution magnification model supports the magnification factor and can magnify the input image according to the magnification factor. Input the cross-sectional image containing the blood vessel area into the target super-resolution magnification model, and the generated magnified cross-sectional image is an image magnified by the magnification factor on the basis of the original cross-sectional image.

[0085] In some optional embodiments, the above step S2031 "determine the magnification factor that matches the size of the cross-sectional image" specifically includes the following steps B1 to step B2.

[0086] Step B1, determine the ratio between the size of the cross-sectional image and the size of the original ultrasound image.

[0087] In this embodiment, the size of the original ultrasound image is generally fixed. For example, the size of the original ultrasound image is consistent with the size of the display area used to display the ultrasound image in the display screen of the ultrasound imaging device, and this display area is generally fixed. The larger the size of the cross-sectional image, the larger its proportion in the original ultrasound image, that is, the larger the ratio between the size of the cross-sectional image and the size of the original ultrasound image.

[0088] Step B2, determine the matching magnification factor according to the ratio; there is a negative correlation between the magnification factor and the ratio.

[0089] In this embodiment, if the above ratio is larger, the determined magnification is smaller. Specifically, if the above ratio is large, it indicates that the size of the cross-sectional image is also large. At this time, magnifying the cross-sectional image with a smaller magnification can better display the blood vessel to the user, and the ultrasonic imaging device can also display the magnified cross-sectional image, avoiding exceeding the display range of the ultrasonic imaging device. If the above ratio is small, it indicates that the size of the cross-sectional image is also small. At this time, a larger magnification needs to be set to magnify the cross-sectional image so that the user can see a larger cross-sectional image of the blood vessel, ensuring the magnification effect.

[0090] Optionally, the above step B1, "determine the ratio between the size of the cross-sectional image and the size of the original ultrasonic image" may include the following step B11.

[0091] Step B11, determine the first ratio between the width of the cross-sectional image and the width of the original ultrasonic image, and the second ratio between the height of the cross-sectional image and the height of the original ultrasonic image.

[0092] In this embodiment, the dimensions required for determining the magnification include the width and the length, and corresponding ratios can be determined based on the width and the length. Specifically, if the width and height of the cross-sectional image are w and h respectively, and the width and height of the original ultrasonic image are W and H respectively, the ratio R1 = w / W between the width w of the cross-sectional image and the width W of the original ultrasonic image can be determined, and this ratio R1 is called the first ratio. Similarly, the ratio R2 = h / H between the height h of the cross-sectional image and the height H of the original ultrasonic image can be determined, and this ratio R2 is called the second ratio. It can be understood that w < W and h < H.

[0093] And the above step B2, "determine the matching magnification according to the ratio" may include the following step B21 and step B22.

[0094] Step B21, in the case where the first ratio is less than the first threshold and the second ratio is less than the second threshold, determine the matching first magnification.

[0095] Step B22, in the case where the first ratio is greater than the third threshold and the second ratio is greater than the fourth threshold, determine the matching second magnification. Among them, the first threshold is less than or equal to the third threshold, the second threshold is less than or equal to the fourth threshold; the first magnification is greater than the second magnification.

[0096] In this embodiment, thresholds for the first ratio, namely a first threshold th1 and a third threshold th3, are preset, and the first threshold th1 ≤ the third threshold th3; and thresholds for the second ratio, namely a second threshold th2 and a fourth threshold th4, are preset, and the second threshold th2 ≤ the fourth threshold th4. Among them, the first threshold th1 and the second threshold th2 may be the same, and the third threshold th3 and the fourth threshold th4 may be the same.

[0097] If the first ratio R1 is less than the first threshold th1 and the second ratio R2 is less than the second threshold th2, it indicates that both the width and height of the cross-sectional image are small. At this time, a larger magnification factor, namely a first magnification factor, can be determined. If the first ratio R1 is greater than the third threshold th3 and the second ratio R2 is greater than the fourth threshold th4, it indicates that both the width and height of the cross-sectional image are large. At this time, a smaller magnification factor, namely a second magnification factor, can be determined, and the first magnification factor is greater than the second magnification factor. When the two ratios meet other conditions, the first magnification factor can be set, or the second magnification factor can be set, or a third magnification factor between the first magnification factor and the second magnification factor can be set. This embodiment does not limit this.

[0098] Among them, both the first magnification factor and the second magnification factor are magnification factors of a target super-resolution magnification model. For example, a target super-resolution magnification model that can magnify both the first magnification factor and the second magnification factor can be pre-trained, that is, the target super-resolution magnification model supports multiple magnification factors; or, a target super-resolution magnification model that can magnify the first magnification factor and a target super-resolution magnification model that can magnify the second magnification factor can be pre-trained, that is, the corresponding magnification factors are achieved based on multiple target super-resolution magnification models.

[0099] For example, the target super-resolution magnification model A supports 2-fold super-resolution magnification, and the target super-resolution magnification model B supports 4-fold super-resolution magnification; the preset first threshold th1, second threshold th2, third threshold th3, and fourth threshold th4 are all 1 / 16. If the first ratio R1 between the width of the cross-sectional image and the width of the original ultrasonic image and the second ratio R2 between the height of the cross-sectional image and the height of the original ultrasonic image are both less than 1 / 16, then 4-fold can be used as the currently required magnification factor, that is, the first magnification factor is 4-fold. At this time, the target super-resolution magnification model B can be called to magnify the cross-sectional image by 4-fold based on the target super-resolution magnification model B. If the first ratio R1 and the second ratio R2 are both greater than 1 / 16, then 2-fold can be used as the currently required magnification factor, that is, the second magnification factor is 2-fold. At this time, the target super-resolution magnification model A can be called to magnify the cross-sectional image by 2-fold based on the target super-resolution magnification model A.

[0100] In this embodiment, the super-resolution magnification model supports one or more magnification ratios. Based on the size of the cross-sectional image, an appropriate magnification factor can be determined, and this magnification factor can be supported by the super-resolution magnification model. Thus, based on this super-resolution magnification model, the magnification process can be simply carried out, with not only high processing efficiency, but also the required functions of the super-resolution magnification model being single and the structure being simple, making it easy to train. The appropriate magnification factor is determined based on the ratio between the size of the cross-sectional image and the size of the original ultrasound image. Thus, when extracting the cross-sectional image from the original ultrasound images of different sizes, reasonable magnification processing can be achieved, and it can be applicable to different ultrasound imaging devices, with strong applicability.

[0101] Optionally, the target super-resolution magnification model is obtained by training the super-resolution magnification model to be trained with a pre-determined training sample. Specifically, an appropriate training sample can be pre-determined, and based on this training sample, the super-resolution magnification model is trained to obtain the required target super-resolution magnification model.

[0102] Among them, the training sample includes a sample image as the input and a label image as the output. To be able to train the super-resolution magnification model, a low-resolution image needs to be used as the input and a high-resolution image as the output to achieve model training. Although in practice, a relatively large number of images of the cross-section of the blood vessel area can be obtained, if the images of the cross-section of the blood vessel area are used as the input of the model, a higher-resolution image corresponding to the images of the cross-section of the blood vessel area is also required, which is relatively difficult to achieve.

[0103] In this embodiment, the image of the cross-section of the blood vessel area is used as the output of the model, and the image generated by downsampling it is used as the input of the model, and training is carried out based on this, that is, the label image is the image of the cross-section of the blood vessel area, and the sample image is the image generated by downsampling the label image.

[0104] Specifically, multiple ultrasound images containing the blood vessel area can be pre-obtained, and the blood vessel areas in these ultrasound images can be intercepted to obtain the images of the cross-section of the blood vessel area. These images of the cross-section of the blood vessel area can be used as the output images of the model, that is, the label images. And by downsampling the label image, a lower-resolution image can be generated, and the generated image can be used as the input of the model, that is, the sample image.

[0105] For example, based on a large amount of clinical data containing cross-sections of blood vessels, the blood vessel regions can be intercepted therefrom to generate a label image. The label image is downsampled N times in both the horizontal and vertical directions to generate a downsampled image, i.e., a sample image. Using a super-resolution model, such as the SRCNN (Super-Resolution Convolutional Neural Network) model, etc., training can be performed based on these sample images and label images to obtain a trained target super-resolution magnification model, which can magnify an image N times.

[0106] In this embodiment, the acquired original image is used as the output of the model, and the image generated by downsampling is used as the input of the model. A large number of training samples can be determined relatively simply, facilitating the training of the super-resolution magnification model.

[0107] Step S204, display the magnified blood vessel image.

[0108] Among them, for details, please refer to Figure 1 Step S104 of the illustrated embodiment, which will not be elaborated here.

[0109] The method for processing ultrasonic images provided in this embodiment can assist users in determining the blood vessel position, improving work efficiency, and reducing misoperations; using the super-resolution magnification model to achieve a specific magnification ratio, the implementation method is simple, the processing efficiency is high, and it is easy to train the super-resolution magnification model. Determining an appropriate magnification multiple based on the ratio between the size of the cross-sectional image and the size of the original ultrasonic image can be applicable to different ultrasonic imaging devices, with strong applicability.

[0110] In this embodiment, a method for processing ultrasonic images is provided, which can be used in ultrasonic forming equipment. Figure 3 It is a flowchart of the method for processing ultrasonic images according to an embodiment of the present invention. As Figure 3 shown, this process includes the following steps.

[0111] Step S301, acquire the acquired original ultrasonic image.

[0112] Among them, for details, please refer to Figure 1 Step S101 of the illustrated embodiment, which will not be elaborated here.

[0113] Step S302, identify the cross-sectional image of the blood vessel in the original ultrasonic image.

[0114] Among them, a segmentation model capable of identifying blood vessels can be pre-trained, and based on this segmentation model, the cross-sectional image of the blood vessel is identified. Specifically, the above step S302 "identify the cross-sectional image of the blood vessel in the original ultrasonic image" may include steps S3021 to S3022.

[0115] Step S3021: Identify the position boundary values of blood vessels in the original ultrasound image according to a pre-trained segmentation model.

[0116] In this embodiment, ultrasound images containing blood vessel cross-sections can be collected in advance, and the blood vessel regions therein can be marked to make labels. Then, based on the ultrasound images with labels, a segmentation model capable of identifying blood vessel regions can be obtained through training. Alternatively, a dataset marked with blood vessels can also be used for training to obtain the required segmentation model. Among them, the segmentation model can be an object detection model or a semantic segmentation model, such as the Unet model, etc. This embodiment does not make any limitations in this regard.

[0117] Input the collected original ultrasound image into the segmentation model, and the location of blood vessels in the original ultrasound image can be identified. Specifically, the position boundary values of the blood vessels can be determined, and these position boundary values are the boundary values of the blood vessel region in the original ultrasound image.

[0118] For example, the position boundary values can include four position boundary values: top, bottom, left, and right. The top and bottom boundary values can represent the position of the blood vessel region in the original ultrasound image in the longitudinal (height direction), and the left and right boundary values can represent the position of the blood vessel region in the original ultrasound image in the transverse (width direction). For example, if the blood vessel region is rectangular, the vertex coordinates of the upper left corner and the lower right corner of the rectangle can be used as the corresponding position boundary values.

[0119] Step S3022: Identify the cross-sectional image of the blood vessel from the original ultrasound image according to the position boundary values.

[0120] In this embodiment, the position boundary values can mark the position of the blood vessel region in the original ultrasound image. Therefore, based on these position boundary values, the image of the blood vessel region, that is, the cross-sectional image of the blood vessel, can be identified from the original ultrasound image. Among them, the cross-sectional image can be a rectangular image or a circular image. This embodiment does not limit the shape of the cross-sectional image, as long as it contains the blood vessel region.

[0121] Optionally, the cross-sectional image can be highlighted in the original ultrasound image to facilitate the user to visually observe the position of the identified cross-sectional image. Specifically, after the above step S302 "Identify the cross-sectional image of the blood vessel in the original ultrasound image", the method can further include the following step C1.

[0122] Step C1: Highlight the cross-sectional image in the original ultrasound image.

[0123] Among them, to facilitate the user to locate the cross-sectional image of the blood vessel, the cross-sectional image can be highlighted in the original ultrasound image. For example, a highlighting mark can be set in the original ultrasound image, and the highlighting mark corresponds to the position boundary value of the blood vessel, so that the highlighting mark can be set at the position where the cross-sectional image is located. By adding the highlighting mark, the cross-sectional image in the original ultrasound image can be highlighted.

[0124] For example, the highlighting mark can be the outer contour line of a rectangle, a circle or other shapes, or a color block with a certain transparency; the highlighting mark can be fused with the original ultrasound image, or can be a visual floating layer such as a mask set on the upper layer of the original ultrasound image.

[0125] Optionally, step C1 of "highlighting the cross-sectional image in the original ultrasound image" can specifically include the following steps C11 to C12.

[0126] Step C11, determine a bounding box that matches the position boundary value.

[0127] Step C12, frame the cross-sectional image in the original ultrasound image based on the bounding box.

[0128] In this embodiment, based on the position boundary value of the blood vessel, a bounding box that can enclose the blood vessel area can be determined; and based on this bounding box, the blood vessel area in the original ultrasound image can be framed, that is, the cross-sectional image can be framed. After automatically identifying the cross-sectional image, the position of the cross-sectional image in the original ultrasound image can be marked based on the bounding box, so that the cross-sectional image can be highlighted; based on this bounding box, the user can initially determine the blood vessel area, which is also convenient for the user to initially judge whether the identified blood vessel area is correct.

[0129] Among them, the bounding box can be a rectangular box; for example, when the segmentation model identifies the blood vessel area, it can use a bounding box to describe the position of the blood vessel area, and this bounding box can be used as the bounding box. Or, the bounding box can also be a circular box or an elliptical box. Based on the position boundary value, the center and length of the blood vessel area can be determined, so that a circular or elliptical bounding box that can enclose the blood vessel area can be generated. The shape of the bounding box in this embodiment is not limited.

[0130] Step S303, perform magnification processing on the cross-sectional image according to the trained target super-resolution magnification model.

[0131] Among them, for details, please refer to Figure 1 step S102 of the embodiment shown, or Figure 2 step S202 of the embodiment shown, which will not be elaborated here.

[0132] In some alternative embodiments, the ultrasonic imaging device can collect ultrasonic images in real time and display the collected ultrasonic images to the user, that is, the ultrasonic imaging device can collect multiple ultrasonic images. During the catheterization process, the position of the ultrasonic probe of the ultrasonic imaging device is generally relatively fixed, and the frequency of changing the position of the ultrasonic probe is low. Therefore, only one of the ultrasonic images can be magnified. Specifically, one of the ultrasonic images can be used as the original ultrasonic image, and the original ultrasonic image can be recognized and magnified to be able to display the magnified cross-sectional image of the blood vessel to the user.

[0133] For example, each time the user inputs a trigger instruction, one ultrasonic image can be recognized and magnified, and the corresponding magnified cross-sectional image of the blood vessel can be displayed; the user can control the update frequency of the magnified cross-sectional image by controlling the frequency of inputting the trigger instruction.

[0134] Alternatively, after the user activates the function of recognizing and magnifying the blood vessel, the ultrasonic imaging device can automatically recognize and magnify the current ultrasonic image at regular intervals, so as to automatically update the displayed magnified cross-sectional image. Specifically, the above step S302 "recognize the cross-sectional image of the blood vessel in the original ultrasonic image" may include the following steps D1; and the above step S303 "magnify the cross-sectional image according to the trained target super-resolution magnification model" may include the following steps D2.

[0135] Step D1: Recognize the cross-sectional image of the blood vessel in one original ultrasonic image collected during the processing period; the processing period is longer than the sampling period of the ultrasonic image, and the number of processing periods is multiple.

[0136] Step D2: During the processing period, magnify the cross-sectional image in the corresponding original ultrasonic image according to the target super-resolution magnification model.

[0137] Among them, the ultrasonic imaging device collects and generates ultrasonic images at a certain period, and this period is called the sampling period; for example, if the sampling period is 50 ms, the ultrasonic imaging device can generate one ultrasonic image every 50 ms and display the ultrasonic image in real time. Due to the limited processing performance of the ultrasonic imaging device, it is difficult to recognize and magnify each ultrasonic image. Therefore, in this embodiment, some of the ultrasonic images are recognized and magnified.

[0138] Specifically, a period larger than the sampling period is set, that is, the processing period; generally, this processing period is not less than 2 times the sampling period.

[0139] During this processing cycle, the ultrasonic imaging device can acquire multiple ultrasonic images, and use one ultrasonic image within this processing cycle as the original ultrasonic image. Then, it can identify the cross-sectional image of the blood vessel in the original ultrasonic image and perform magnification processing on it, so as to display the magnified cross-sectional image within this processing cycle. In the next processing cycle, it is still possible to identify and magnify one of the original ultrasonic images according to the above steps D1 and D2, so as to update the displayed magnified cross-sectional image and ensure that the magnified cross-sectional image has a certain degree of real-time performance.

[0140] For example, the sampling period of the ultrasonic image is 50 ms, and the processing cycle can be set to 200 ms. Within a processing cycle of 200 ms, four ultrasonic images can be acquired; one of them is used as the original ultrasonic image, and the magnified cross-sectional image can be generated and displayed to the user. That is, every 200 ms, the magnified cross-sectional image can be updated. Without consuming too much processing performance, it is possible to achieve real-time display of the magnified cross-sectional image.

[0141] Step S304: Display the magnified blood vessel image.

[0142] Among them, for details, please refer to Figure 1 Step S104 of the illustrated embodiment, which will not be elaborated here.

[0143] In some optional embodiments, the above step of "displaying the magnified cross-sectional image" may include: displaying the original ultrasonic image and displaying the magnified cross-sectional image in the original ultrasonic image. Among them, the magnified cross-sectional image is located in the edge area of the original ultrasonic image; or, the center position of the magnified cross-sectional image corresponds to the center position of the cross-sectional image in the original ultrasonic image.

[0144] In this embodiment, the original ultrasonic image and the magnified cross-sectional image can be displayed simultaneously; specifically, the magnified cross-sectional image can be embedded into the original ultrasonic image for image fusion, so that when the original ultrasonic image is displayed, the magnified cross-sectional image located in the original ultrasonic image can be displayed.

[0145] Among them, the magnified cross-sectional image can be located in the edge area of the original ultrasonic image, and the edge area can specifically be the upper left area, the upper right area, etc.; for example, the display screen of the ultrasonic imaging device displays the original ultrasonic image, and the magnified cross-sectional image is displayed in the upper right of the display screen. Generally, the blood vessel is located at the center position of the ultrasonic image. Displaying the magnified cross-sectional image in the edge area can display the cross-sectional images before and after magnification simultaneously.

[0146] For example, if only one ultrasound image is recognized and magnified, after generating the magnified cross-sectional image, the magnified cross-sectional image can be displayed in the edge area. Moreover, the ultrasound image can also be displayed in real time. That is, the ultrasound imaging device can display the fixed magnified cross-sectional image and the ultrasound image it acquires in real time. The user can observe the blood vessel area before magnification in real time and also view the magnified blood vessel area of a certain frame.

[0147] Alternatively, the magnified cross-sectional image can also be embedded into the original ultrasound image in a way that covers the original blood vessel area. Specifically, the center position of the magnified cross-sectional image corresponds to the center position of the cross-sectional image in the original ultrasound image. This display method is more suitable for the situation where the magnified cross-sectional image can be updated in real time.

[0148] For example, if a corresponding magnified cross-sectional image is generated every processing cycle, the magnified cross-sectional image can be embedded into the original ultrasound image and aligned with the center of the original cross-sectional image to obtain the finally displayed image and display it to the user.

[0149] Figure 4 The schematic diagram of the process of recognizing and magnifying an original ultrasound image is shown. As Figure 4 shown, for the original ultrasound image 400, it can be recognized and processed based on the segmentation model to segment out the cross-sectional image 401 containing blood vessels. Among them, when the cross-sectional image 401 is recognized, the position of the cross-sectional image 401 in the original ultrasound image 400 can be marked in the original ultrasound image in the form of a bounding box. Figure 4 The white box in the upper left image is a kind of bounding box.

[0150] For the recognized cross-sectional image 401, input it into the target super-resolution magnification model, and the cross-sectional image 401 can be magnified to generate the magnified cross-sectional image 402. Then the magnified cross-sectional image 402 can be embedded into the original ultrasound image 400 for image fusion, so that the original ultrasound image 400 embedded with the magnified cross-sectional image 402 can be displayed to the user. As Figure 4 shown in the lower left image, the magnified cross-sectional image 402 and the original cross-sectional image 401 can be centered.

[0151] The processing method of the ultrasound image provided in this embodiment can assist the user to determine the blood vessel position, improve work efficiency, and reduce misoperations; according to the segmentation model, the blood vessel area can be recognized more accurately, so as to recognize the corresponding cross-sectional image; every processing cycle, recognition and magnification can be performed once, and real-time display of the magnified cross-sectional image can be achieved without consuming too much processing performance.

[0152] In this embodiment, an ultrasonic imaging device is provided. As Figure 5 shown, the ultrasonic imaging device includes an ultrasonic probe 501 and a host 502. Among them, the ultrasonic probe is configured to collect an original ultrasonic image; the host 502 is configured to execute the ultrasonic image processing method provided in the above embodiment. For example, the host 502 is configured to execute: obtaining the original ultrasonic image collected by the ultrasonic probe 501; identifying the cross-sectional image of blood vessels in the original ultrasonic image; performing magnification processing on the cross-sectional image according to the trained target super-resolution magnification model; and displaying the magnified blood vessel image.

[0153] Optionally, as shown in Figure 5 the figure, the ultrasonic probe 501 may include a human-computer interaction device 5011; the human-computer interaction device 5011 is configured to receive a trigger instruction input by a user for identifying and magnifying blood vessels.

[0154] Among them, when the user needs to start the function of identifying and magnifying blood vessels, a corresponding trigger instruction can be input based on the human-computer interaction device 5011. For example, the human-computer interaction device 5011 can be a button (physical button, virtual button, etc.), a voice collection device, a gesture recognition device, etc. The user can input a trigger instruction by operating the button, issuing a voice command, a gesture command, etc. to start the function of identifying and magnifying blood vessels.

[0155] In this embodiment, a processing device for ultrasonic images is further provided. This device is used to implement the above embodiment and preferred implementation manners, and those that have been described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0156] This embodiment provides a processing device for ultrasonic images. As Figure 6 shown, it includes:

[0157] An acquisition module 601, configured to acquire the collected original ultrasonic image;

[0158] An identification module 602, configured to identify the cross-sectional image of blood vessels in the original ultrasonic image;

[0159] An amplification module 603, configured to perform magnification processing on the cross-sectional image according to the trained target super-resolution magnification model;

[0160] A display module 604, configured to display the magnified blood vessel image.

[0161] In some alternative embodiments, the magnification module 603 magnifies the cross-sectional image according to a trained target super-resolution magnification model, including:

[0162] Determine a magnification factor that matches the size of the cross-sectional image;

[0163] Input the cross-sectional image into the target super-resolution magnification model to generate a cross-sectional image magnified to the magnification factor; the magnification factor of the target super-resolution magnification model at least includes the magnification factor.

[0164] In some alternative embodiments, the magnification module 603 determines a magnification factor that matches the size of the cross-sectional image, including:

[0165] Determine the ratio between the size of the cross-sectional image and the size of the original ultrasound image;

[0166] Determine a matching magnification factor according to the ratio; there is a negative correlation between the magnification factor and the ratio.

[0167] In some alternative embodiments, the magnification module 603 determines the ratio between the size of the cross-sectional image and the size of the original ultrasound image, including:

[0168] Determine a first ratio between the width of the cross-sectional image and the width of the original ultrasound image, and a second ratio between the height of the cross-sectional image and the height of the original ultrasound image;

[0169] The magnification module 603 determines a matching magnification factor according to the ratio, including:

[0170] In the case where the first ratio is less than a first threshold and the second ratio is less than a second threshold, determine a matching first magnification factor;

[0171] In the case where the first ratio is greater than a third threshold and the second ratio is greater than a fourth threshold, determine a matching second magnification factor;

[0172] Wherein, the first threshold is less than or equal to the third threshold, and the second threshold is less than or equal to the fourth threshold; the first magnification factor is greater than the second magnification factor.

[0173] In some alternative embodiments, the target super-resolution magnification model is obtained by training a super-resolution magnification model to be trained with a pre-determined training sample;

[0174] The training samples include a sample image as the input and a label image as the output; the label image is an image of the cross-section of the blood vessel area, and the sample image is an image generated by downsampling the label image.

[0175] In some alternative embodiments, the recognition module 602 recognizes the cross-sectional image of the blood vessel in the original ultrasound image, including:

[0176] Recognize the position boundary values of the blood vessel in the original ultrasound image according to a pre-trained segmentation model;

[0177] Recognize the cross-sectional image of the blood vessel from the original ultrasound image according to the position boundary values.

[0178] In some alternative embodiments, the recognition module 602 is further configured to highlight the cross-sectional image in the original ultrasound image after recognizing the cross-sectional image of the blood vessel in the original ultrasound image.

[0179] In some alternative embodiments, the recognition module 602 highlights the cross-sectional image in the original ultrasound image, including: determining a bounding box that matches the position boundary values; framing the cross-sectional image in the original ultrasound image based on the bounding box.

[0180] In some alternative embodiments, the recognition module 602 recognizes the cross-sectional image of the blood vessel in the original ultrasound image, including:

[0181] Recognize the cross-sectional image of the blood vessel in one original ultrasound image collected during a processing period; the processing period is greater than the sampling period of the ultrasound image, and the number of processing periods is multiple;

[0182] The magnification module 603 magnifies the cross-sectional image according to a trained target super-resolution magnification model, including:

[0183] During the processing period, magnify the cross-sectional image in the corresponding original ultrasound image according to the target super-resolution magnification model.

[0184] In some alternative embodiments, a trigger module is further included, configured to obtain a trigger instruction for recognizing and magnifying the blood vessel before the recognition module 602 recognizes the cross-sectional image of the blood vessel in the original ultrasound image; after the trigger module obtains the trigger instruction, the recognition module 602 then performs the process of recognizing the cross-sectional image of the blood vessel in the original ultrasound image.

[0185] In some alternative embodiments, the display module 604 displays the magnified cross-sectional image, including:

[0186] Display the original ultrasound image and display the magnified cross-sectional image in the original ultrasound image;

[0187] Wherein, the magnified cross-sectional image is located in the edge area of the original ultrasound image; alternatively, the center position of the magnified cross-sectional image corresponds to the center position of the cross-sectional image in the original ultrasound image.

[0188] The further function descriptions of the above-mentioned various modules and units are the same as those in the corresponding above embodiments, and will not be repeated here.

[0189] The processing device of the ultrasound image in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0190] The embodiment of the present invention also provides an electronic device having the above-mentioned Figure 6 processing device of the ultrasound image shown.

[0191] Please refer to Figure 7 , Figure 7 which is a schematic structural diagram of an electronic device provided by an optional embodiment of the present invention. As shown in Figure 7 , the electronic device includes: one or more processors 10, a memory 20, and an interface for connecting each component, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common main board or installed in other ways as needed. The processor can process instructions executed within the electronic device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 7 In

[0192] processor 10 is taken as an example.

[0193] Among them, the memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the methods shown in the above embodiments.

[0194] The memory 20 may include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the electronic device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 may optionally include a memory remotely provided with respect to the processor 10, and these remote memories may be connected to the electronic device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0195] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk, or a solid-state drive; the memory 20 may further include a combination of the above types of memories.

[0196] The electronic device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 may be connected through a bus or other means, Figure 7 Taking connection through a bus as an example.

[0197] The input device 30 can receive input digital or character information, and generate key signal inputs related to the user settings and function controls of the electronic device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (such as an LED), and a tactile feedback device (such as a vibration motor), etc. The above display device includes but is not limited to a liquid crystal display, a light-emitting diode, a display, and a plasma display. In some alternative embodiments, the display device may be a touch screen.

[0198] Embodiments of the present invention also provide a computer-readable storage medium. The method according to the embodiments of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0199] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for processing an ultrasonic image, characterized in that, The method includes: Obtaining the acquired original ultrasound image; Identifying the cross-sectional image of blood vessels in the original ultrasound image; Performing magnification processing on the cross-sectional image according to the trained target super-resolution magnification model; Displaying the magnified blood vessel image.

2. The method according to claim 1, wherein The performing magnification processing on the cross-sectional image according to the trained target super-resolution magnification model includes: Determining the magnification factor that matches the size of the cross-sectional image; Inputting the cross-sectional image into the target super-resolution magnification model to generate a cross-sectional image magnified to the magnification factor; the magnification factor of the target super-resolution magnification model at least includes the magnification factor.

3. The method according to claim 2, wherein The determining the magnification factor that matches the size of the cross-sectional image includes: Determining the ratio between the size of the cross-sectional image and the size of the original ultrasound image; Determining the matching magnification factor according to the ratio; there is a negative correlation between the magnification factor and the ratio.

4. The method according to claim 3, wherein The determining the ratio between the size of the cross-sectional image and the size of the original ultrasound image includes: Determining a first ratio between the width of the cross-sectional image and the width of the original ultrasound image, and a second ratio between the height of the cross-sectional image and the height of the original ultrasound image; The determining the matching magnification factor according to the ratio includes: When the first ratio is less than a first threshold and the second ratio is less than a second threshold, determining a matching first magnification factor; When the first ratio is greater than a third threshold and the second ratio is greater than a fourth threshold, determining a matching second magnification factor; Wherein, the first threshold is less than or equal to the third threshold, and the second threshold is less than or equal to the fourth threshold; the first magnification factor is greater than the second magnification factor.

5. The method according to claim 1, characterized in that, The target super-resolution magnification model is obtained by training a super-resolution magnification model to be trained with a predetermined training sample; The training sample includes a sample image as input and a label image as output; the label image is an image of the cross-section of the blood vessel area, and the sample image is an image generated by downsampling the label image.

6. The method according to claim 1, wherein, The identifying the cross-sectional image of blood vessels in the original ultrasound image includes: According to a pre-trained segmentation model, identifying the position boundary values of blood vessels in the original ultrasound image; Identifying the cross-sectional image of blood vessels from the original ultrasound image according to the position boundary values.

7. The method according to claim 6, characterized in that, After the identifying the cross-sectional image of blood vessels in the original ultrasound image, it further includes: Highlighting the cross-sectional image in the original ultrasound image.

8. The method according to claim 7, wherein The highlighting the cross-sectional image in the original ultrasound image includes: Determining a bounding box that matches the position boundary values; Framing the cross-sectional image in the original ultrasound image based on the bounding box.

9. The method according to claim 1, characterized in that, The identifying the cross-sectional image of blood vessels in the original ultrasound image includes: Identifying the cross-sectional image of blood vessels in one original ultrasound image acquired within a processing period; the processing period is greater than the sampling period of the ultrasound image, and the number of the processing periods is multiple; Performing magnification processing on the cross-sectional image according to the trained target super-resolution magnification model includes: During the processing period, performing magnification processing on the cross-sectional image in the corresponding original ultrasound image according to the target super-resolution magnification model.

10. The method according to claim 1, characterized in that, Before identifying the cross-sectional image of a blood vessel in the original ultrasound image, it further includes: Obtaining a trigger instruction for identifying and magnifying a blood vessel; after obtaining the trigger instruction, then performing the process of identifying the cross-sectional image of a blood vessel in the original ultrasound image.

11. The method according to claim 1, characterized in that, Displaying the magnified cross-sectional image includes: Displaying the original ultrasound image and displaying the magnified cross-sectional image in the original ultrasound image; wherein, the magnified cross-sectional image is located in the edge area of the original ultrasound image; or, the center position of the magnified cross-sectional image corresponds to the center position of the cross-sectional image in the original ultrasound image.

12. An electronic device, characterized in that, It includes: A memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the method for processing an ultrasound image according to any one of claims 1 to 11.

13. An ultrasonic imaging device, characterized in that, It includes: An ultrasound probe and a host; The ultrasound probe is configured to collect an original ultrasound image; The host is configured to execute the method for processing an ultrasound image according to any one of claims 1 to 11.

14. The ultrasonic imaging device according to claim 13, wherein The ultrasound probe includes a human-computer interaction device; The human-computer interaction device is configured to receive a trigger instruction input by a user for identifying and magnifying a blood vessel.

15. A computer-readable storage medium, characterized in that, Computer instructions are stored on the computer-readable storage medium, and the computer instructions are used to cause a computer to execute the method for processing an ultrasound image according to any one of claims 1 to 11.

Images