Ultrasound image classification methods, ultrasound imaging equipment, and readable storage media

Abstract

This application relates to the field of ultrasound imaging technology, and discloses a method for classifying ultrasound images, an ultrasound imaging device, and a readable storage medium. The method includes: transmitting an ultrasound signal to a target tissue and acquiring the ultrasound echo signal reflected by the target tissue; forming an ultrasound image corresponding to the target tissue based on the ultrasound echo signal; and automatically classifying the ultrasound image to obtain an image type; wherein the image type includes transverse section images and longitudinal section images. This method can improve classification efficiency, reduce the workload of medical personnel, and thus reduce reliance on the professional expertise of medical personnel.

Description

Technical Field

[0001] This application relates to the field of ultrasound imaging technology, and in particular to methods for classifying ultrasound images, ultrasound imaging equipment, and readable storage media. Background Technology

[0002] In calculating labor monitoring parameters, different cross-sectional images are required for different parameters. The traditional method usually involves doctors manually dividing the acquired images into different cross-sections based on their clinical knowledge and experience after acquiring the intrapartum ultrasound images. Summary of the Invention

[0003] The main technical problem addressed by this application is to provide a method for classifying ultrasound images, an ultrasound imaging device, and a readable storage medium, which can improve classification efficiency, reduce the workload of medical staff, and thus reduce reliance on the professional expertise of medical staff.

[0004] To address the aforementioned issues, this application provides a method for classifying ultrasound images. This method includes: transmitting an ultrasound signal to a target tissue and acquiring ultrasound echo signals reflected from the target tissue; forming an ultrasound image corresponding to the target tissue based on the ultrasound echo signals; and automatically classifying the ultrasound image to obtain its image type. The image type includes cross-sectional images and longitudinal sections.

[0005] The process of transmitting ultrasound signals to the target tissue and collecting ultrasound echo signals reflected by the target tissue includes: transmitting ultrasound signals to the perineal tissue during fetal delivery and collecting ultrasound echo signals reflected by the perineal tissue; the image types include transverse and longitudinal images of the perineal tissue.

[0006] The process of automatically classifying ultrasound images to obtain image types includes: determining the similarity between the ultrasound image and a standard image of each image type; and using the image type of the standard image with a similarity greater than a preset similarity as the image type of the ultrasound image.

[0007] The method further includes displaying ultrasound images, the corresponding image types, and similarity scores on a human-computer interaction interface.

[0008] The process includes, after displaying the ultrasound image, the corresponding image type, and the similarity of the ultrasound image on the human-computer interaction interface, receiving a first selection instruction to save the ultrasound image displayed on the human-computer interaction interface; or receiving a second selection instruction to remove the ultrasound image displayed on the human-computer interaction interface.

[0009] The process of automatically classifying ultrasound images to obtain their image types includes: classifying ultrasound images using an image classification model to obtain the corresponding image type.

[0010] Before automatically classifying the ultrasound image to obtain its image type, the process includes: performing edge detection on the ultrasound image; determining the edge area based on the edge detection results; retaining the ultrasound image if the edge area is greater than or equal to a set area threshold; or removing the ultrasound image if the edge area is less than the set area threshold.

[0011] The process includes automatically classifying ultrasound images to obtain their image types, including automatically classifying the retained ultrasound images to obtain their image types.

[0012] To address the aforementioned problems, this application provides an ultrasound imaging device comprising: an ultrasound probe; a transmitting circuit connected to the ultrasound probe for transmitting ultrasound signals to a target tissue via the ultrasound probe; a receiving circuit connected to the ultrasound probe for acquiring ultrasound echo signals reflected by the target tissue; and a processor connected to the receiving circuit for implementing the method provided by the above technical solution.

[0013] To address the aforementioned problems, one technical solution adopted in this application is to provide a computer-readable storage medium for storing a computer program, which, when executed by a processor, is used to implement the method provided by the above technical solution.

[0014] To address the aforementioned problems, another technical solution adopted in this application is to provide an ultrasonic device, which includes a processor, a memory coupled to the processor, and an ultrasonic probe; wherein the memory is used to store a computer program, the ultrasonic probe is used to acquire ultrasonic images, and the processor is used to execute the computer program to implement the method provided by the above technical solution.

[0015] To address the aforementioned problems, another technical solution adopted in this application is to provide a computer-readable storage medium for storing a computer program, which, when executed by a processor, implements the method provided by the above technical solution.

[0016] The beneficial effects of this application are as follows: Unlike existing technologies, this application provides a method for classifying ultrasound images. This method includes: transmitting ultrasound signals to a target tissue and acquiring ultrasound echo signals reflected from the target tissue; forming an ultrasound image corresponding to the target tissue based on the ultrasound echo signals; and automatically classifying the ultrasound image to obtain its image type; wherein the image type includes transverse section images and longitudinal section images. By automatically classifying ultrasound images in this way, classification efficiency can be improved, reducing the workload for medical personnel and thus reducing reliance on their expertise. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart of an embodiment of the ultrasound image classification method provided in this application;

[0018] Figure 2 This is a schematic diagram of the structure of an embodiment of the ultrasound imaging device provided in this application;

[0019] Figure 3 This is a schematic flowchart of another embodiment of the ultrasound image classification method provided in this application;

[0020] Figure 4 This is a schematic diagram illustrating an application scenario of the ultrasound image classification method provided in this application.

[0021] Figure 5 This is a schematic flowchart of another embodiment of the ultrasound image classification method provided in this application;

[0022] Figure 6 This is a schematic flowchart of another embodiment of the ultrasound image classification method provided in this application;

[0023] Figure 7 This is a schematic flowchart of another embodiment of the ultrasound image classification method provided in this application;

[0024] Figure 8 This is a schematic diagram illustrating another application scenario of the ultrasound image classification method provided in this application;

[0025] Figure 9 This is a schematic diagram of the structure of an embodiment of the ultrasound imaging device provided in this application;

[0026] Figure 10 This is a schematic diagram of an embodiment of the computer-readable storage medium provided in this application. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It is understood that the specific embodiments described herein are only for explaining this application and not for limiting it. Furthermore, it should be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all structures. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0028] The terms "first," "second," etc., used in this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] See Figure 1 , Figure 1 This is a schematic flowchart of an embodiment of the ultrasound image classification method provided in this application. The method includes:

[0031] Step 11: Emit an ultrasonic signal to the target tissue and collect the ultrasonic echo signal reflected by the target tissue.

[0032] Step 12: Generate an ultrasound image of the target tissue based on the ultrasound echo signal.

[0033] In some embodiments, an ultrasound imaging device may be used to acquire ultrasound images. See also Figure 2 The ultrasound imaging device 100 includes an ultrasound probe 101, a transmitting circuit 102, a receiving circuit 103, a transmit / receive selection switch 104, a processor 105, a display 106, and a memory 107. The transmitting circuit 102 and the receiving circuit 103 can be connected to the ultrasound probe 101 via the transmit / receive selection switch 104. In some embodiments, the transmitting circuit 102, the receiving circuit 103, and the transmit / receive selection switch 104 can be integrated with the ultrasound probe 101.

[0034] During ultrasound imaging, the transmitting circuit 102 sends a delayed-focused transmission pulse with a certain amplitude and polarity to the ultrasound probe 101 via the transmit / receive selection switch 104 to excite the ultrasound probe 101 to emit ultrasonic waves. After a certain delay, the receiving circuit 103 receives the echo of the ultrasonic wave via the transmit / receive selection switch 104, obtains the ultrasonic echo signal, and performs amplification, analog-to-digital conversion, and beamforming on the echo signal. Then, the processed ultrasonic echo signal is sent to the processor 105 for further processing. The processor 105 processes the ultrasonic echo signal to obtain the corresponding ultrasound image.

[0035] The display 106 is connected to the processor 105. For example, the processor 105 can be connected to the display 106 via an external input / output port. The display 106 can detect user input information, which may include, for example, control commands for ultrasonic wave transmission and reception timing, operation input commands for initiating still image capture, dynamic video capture, and / or dynamic image storage, or other command types. The display 106 may include one or more of the following: keyboard, mouse, scroll wheel, trackball, mobile input device (such as a mobile device with a touch screen, a mobile phone, etc.), multi-function knob, buttons, etc. Therefore, the corresponding external input / output port can be a wireless communication module, a wired communication module, or a combination of both. The external input / output port can also be implemented based on USB, bus protocols such as CAN, and / or wired network protocols.

[0036] The display 106 also includes a screen that can display ultrasound images acquired by the processor 105. Furthermore, while displaying ultrasound images, the screen can also provide a graphical user interface for human-computer interaction. One or more controlled objects can be set on the graphical interface, allowing the user to input operation commands through the display 106 to control these controlled objects and perform corresponding control operations. For example, icons can be displayed on the graphical interface, and the user can operate these icons using a human-computer interaction device to perform specific functions, such as the function of storing dynamic images while simultaneously capturing still images / movie clips. In practical applications, the screen can be a touchscreen display. Furthermore, the display in this embodiment may include one screen or multiple screens.

[0037] In other embodiments of this application, the processor 105 is also configured to receive an instruction to store the ultrasound image, and in response to the instruction to store a dynamic image, a static image, or a short video of the ultrasound image, thereby facilitating a user (e.g., a doctor) to browse and review it for diagnosis.

[0038] The ultrasound imaging device 100 can be of the amplitude modulation type, the spot scanning type, or the grayscale modulation type.

[0039] Step 13: Automatically classify the ultrasound images to obtain the image types; the image types include transverse images and longitudinal images.

[0040] In some embodiments, an image classification model can be used to classify ultrasound images to obtain the corresponding image types. These image types include cross-sectional images and longitudinal section images.

[0041] In this embodiment, an ultrasound signal is emitted to the target tissue, and the ultrasound echo signal reflected by the target tissue is collected; an ultrasound image corresponding to the target tissue is formed based on the ultrasound echo signal; the ultrasound image is automatically classified to obtain the image type of the ultrasound image; wherein the image type includes cross-sectional images and longitudinal images, the automatic classification of ultrasound images can improve the efficiency of ultrasound image classification, reduce the operation of medical staff, and thus reduce the reliance on the professionalism of medical staff.

[0042] See Figure 3 , Figure 3 This is a schematic flowchart of another embodiment of the ultrasound image classification method provided in this application. The method includes:

[0043] Step 31: During the delivery process, an ultrasound signal is emitted into the perineal tissue, and the ultrasound echo signal reflected by the perineal tissue is collected.

[0044] In some embodiments, a three-dimensional volume probe or a two-dimensional convex array probe covered with a disposable sterile isolation sleeve may be placed at the perineum to perform step 31.

[0045] Step 32: Generate an ultrasound image of the target tissue based on the ultrasound echo signal.

[0046] Step 33: Determine the similarity between the ultrasound images and the standard images for each image type.

[0047] In this process, a standard image for each image type can be determined in advance. Then, features are extracted from the standard images to obtain the corresponding standard features.

[0048] Then, feature extraction is performed on the ultrasound image to obtain target features. The similarity between the target features and standard features is compared to determine the similarity between the ultrasound image and the standard image for each image type.

[0049] Step 34: Use the image type of the standard image with a similarity greater than the preset similarity as the image type of the ultrasound image; wherein, the image type includes transverse and longitudinal images of the perineal tissue.

[0050] In some embodiments, since multiple image types exist, multiple similarity scores will be obtained. The image type with a similarity score greater than a preset similarity score can be used as the image type of the ultrasound image. For example, the similarity between ultrasound image A and standard image a is A1, and the similarity between ultrasound image A and standard image b is B1. Where A1 is greater than the preset similarity score, the image type of standard image a is used as the image type of ultrasound image A.

[0051] If the similarity scores are all less than or equal to the preset similarity scores, it means that the ultrasound image does not meet the requirements and can be deleted.

[0052] In some embodiments, the ultrasound image, the corresponding image type, and the similarity can be displayed on the human-computer interaction interface.

[0053] In other embodiments, a corresponding confirmation option can be set in the human-computer interaction interface so that when medical staff see the image type and similarity of the ultrasound image, they can manually judge whether the requirements are met. If the requirements are met, they can select the confirmation option to save the ultrasound image.

[0054] Specifically, in combination Figure 4 Explanation:

[0055] exist Figure 4 The system displays ultrasound images, their corresponding image types, and similarity scores on the human-computer interaction interface. It also provides corresponding controls for selection, such as a save control and a no-save control.

[0056] When a medical professional selects the save control, a first selection instruction is generated. In response to the first selection instruction, the ultrasound image displayed on the human-computer interaction interface is saved.

[0057] When the medical staff selects not to save the control, a second selection instruction is generated. In response to the second selection instruction, the ultrasound image displayed on the human-computer interaction interface is removed.

[0058] In this embodiment, automatic classification of ultrasound images can improve the efficiency of ultrasound image classification, reduce the operation of medical staff, and thus reduce the reliance on the professionalism of medical staff. Furthermore, the use of manual processing can improve the accuracy of ultrasound image classification.

[0059] See Figure 5 , Figure 5 This is a schematic flowchart of another embodiment of the ultrasound image classification method provided in this application. The method includes:

[0060] Step 51: Emit an ultrasonic signal to the target tissue and collect the ultrasonic echo signal reflected by the target tissue.

[0061] Step 52: Generate an ultrasound image of the target tissue based on the ultrasound echo signal.

[0062] Step 53: Perform edge detection on the ultrasound image.

[0063] First, straight lines are detected in the ultrasound images.

[0064] In ultrasound images, the region containing an object is roughly fan-shaped, defined by two straight lines and an arc. Therefore, to determine the edges of an ultrasound image, it is necessary to identify the corresponding straight lines and arcs. Thus, straight line detection is required in ultrasound images.

[0065] For example, the ultrasound image is first subjected to edge detection, then binarized, then mapped to Hough space, local maxima are taken, thresholds are set, interfering lines are filtered out, and then lines are drawn and corner points are calibrated based on these maxima.

[0066] Specifically, the intersection points of multiple curves corresponding to multiple points on the ultrasound image are identified. If the number of curves intersecting at the same intersection point exceeds a set threshold, a straight line is determined based on the parameters of the intersection point.

[0067] A straight line can be represented by the number of curves intersecting at a single point in a plane. The more curves intersecting at a single point, the more points constitute the straight line. A threshold is set to define how many curves intersect at a single point, thus indicating the detection of a straight line. Therefore, the Hough line transform is used to track the intersections of curves corresponding to each point in an ultrasound image. When the number of curves intersecting at a single point exceeds the threshold, the parameters represented by that intersection point constitute a straight line.

[0068] Furthermore, circular arc detection is performed on the ultrasound images.

[0069] For example, select a target arc. After edge detection, a series of coordinate points are obtained. These coordinate points can be fitted into an arc, but the arc is inaccurate at this time.

[0070] Then, based on the distance from the point on the target arc to the center and the error between the distance and the radius of the target arc, the center position and radius of the target arc are confirmed.

[0071] Specifically, determine the difference between the square of the distance from a point on the target arc to the center of the arc and the square of the radius of the target arc.

[0072] Since the arc is determined, the center can be determined accordingly, and the square of the distance from a point on the arc to the center can be calculated. Once the center is determined, the radius of the target arc can be determined, and the square of the radius can then be calculated.

[0073] Therefore, the difference between the square of the distance from a point on the target arc to the center and the square of the radius of the target arc can be determined.

[0074] For example, if M represents the square of the distance from a point on the target arc to the center of the arc, and N represents the square of the radius of the target arc, then the difference in squares P = MN.

[0075] Then determine the sum of squared errors of the squared differences corresponding to multiple points on the target arc.

[0076] Therefore, the corresponding squared difference can be obtained for each point on the target arc, and these squared differences can be added together to calculate the sum of squared errors.

[0077] To determine the center position and radius of the target arc, the sum of squared errors is minimized.

[0078] When the sum of squared errors is minimized, the center position and radius of the target arc can be determined. This target arc is the best-fit arc.

[0079] Then, determine an arc based on the center position and the radius of the arc.

[0080] Once the center position and radius of the arc are determined, an optimal arc can be identified.

[0081] There are two arcs in the ultrasound image, and the above method can be used to obtain the two optimal arcs.

[0082] Then, the edge detection results of the ultrasound image are determined based on the results of line detection and arc detection. At this point, the detected lines and arcs can be used as the edge detection results.

[0083] Step 54: Determine the edge area based on the edge detection results.

[0084] Finally, the edge area is determined based on the edge detection results.

[0085] Specifically, based on the above-mentioned line detection and arc detection, the edge region formed by the first arc, the first line, the second arc, and the second line can be determined.

[0086] An ultrasound image contains two arcs, which can be used to obtain the first and second arcs using the method described above. Similarly, an ultrasound image contains two straight lines, which can be used to obtain the first and second straight lines using the method described above.

[0087] Then determine the area of ​​the first sector between the ultrasonic probe and the first arc; and determine the area of ​​the second sector between the ultrasonic probe and the second arc.

[0088] The ultrasound probe can be used as the center of the first and second circular arcs. Therefore, the area of ​​the first sector formed between the first circular arc and the ultrasound probe can be obtained based on the first circular arc. The area of ​​the second sector formed between the second circular arc and the ultrasound probe can be obtained based on the second circular arc.

[0089] Then, based on the difference between the areas of the first and second sectors, the edge area of ​​the edge region is determined. That is, the area formed by the first arc, the first straight line, the second arc, and the second straight line is the difference between the areas of the first and second sectors.

[0090] Step 55: If the edge area is greater than or equal to a set area threshold, the ultrasound image is retained.

[0091] If the edge area is less than a set area threshold, the ultrasound image is removed.

[0092] Step 56: Automatically classify the retained ultrasound images to obtain the image types of the ultrasound images; among which, the image types include transverse section images and longitudinal section images.

[0093] Step 56: Any of the above embodiments has the same or similar technical solutions, which will not be repeated here.

[0094] In this embodiment, edge detection is performed on ultrasound images to remove those that do not meet the requirements in advance, thereby reducing the amount of ultrasound image data to be automatically classified later, further improving classification efficiency, reducing the workload of medical staff, and thus reducing the reliance on the professionalism of medical staff.

[0095] See Figure 6 , Figure 6 This is a schematic flowchart of another embodiment of the ultrasound image classification method provided in this application. This embodiment illustrates the training of the image classification model. The method includes:

[0096] Step 61: Input the training dataset into the image classification model to train the image classification model.

[0097] The ultrasound images in the training dataset are labeled with corresponding tags.

[0098] Step 62: Input the test dataset into the image classification model to test the image classification model.

[0099] In one application scenario, combined Figures 7-8 Explanation:

[0100] Figure 7 This is a schematic flowchart of another embodiment of the ultrasound image classification method provided in this application. The method includes:

[0101] Step 71: Acquire transperineal ultrasound data during childbirth and generate ultrasound images based on the ultrasound data.

[0102] A three-dimensional volume probe or a two-dimensional convex array probe covered with a disposable sterile isolation sleeve is placed at the perineum to acquire a dataset of intrapartum ultrasound images. Here, the intrapartum ultrasound image data acquired by the three-dimensional volume probe contains at least one volumetric three-dimensional image, while the ultrasound image data acquired by the two-dimensional convex array probe is a segment of intrapartum ultrasound video.

[0103] Step 72: Screen the ultrasound images.

[0104] From the acquired intrapartum ultrasound image data, an algorithm was used to filter out a dataset containing images of the fetal head being cleared.

[0105] Step 73: Perform data augmentation on the selected ultrasound images.

[0106] Data augmentation is performed on the selected images, mainly including, but not limited to, the following steps: using methods such as cropping, rotation, flipping, color jittering, translation, and scaling for data augmentation. This increases the diversity of ultrasound images and prevents overfitting.

[0107] Step 74: Preprocess the ultrasound images.

[0108] This section includes: image smoothing and size normalization.

[0109] (1) Image smoothing.

[0110] During the probe scanning process, due to the influence of the external environment, most of the acquired images contain various noises. The first step is to smooth the images to reduce noise at the edges, eliminate isolated points, and fill isolated holes, thus paving the way for subsequent operations.

[0111] Commonly used smoothing methods include mean filtering, median filtering, and Gaussian filtering. Mean filtering results in significant edge blurring, and median filtering also produces some edge blurring. While Gaussian filtering also results in some edge blurring, it effectively removes Gaussian noise compared to mean and median filtering, thus smoothing the image. Therefore, this invention selects Gaussian filtering as the image smoothing operator.

[0112] (2) Size normalization processing.

[0113] After image processing such as smoothing, images typically require size normalization, which involves scaling the image to the same size according to the requirements of subsequent image processing algorithms. This method addresses this image problem; it is simple to operate, highly feasible, and preserves the original shape and information of features while minimizing image resolution distortion.

[0114] For example, nearest neighbor interpolation, bilinear interpolation, and bicubic interpolation can be used. Nearest neighbor interpolation has the advantage of low computational cost and fast processing speed, but its disadvantage is that the interpolated image exhibits a checkerboard effect, resulting in discontinuous pixels and jagged edges at pixel transitions. Bilinear interpolation produces high-quality images without pixel discontinuities and optimizes image edge performance. Although its computation is more complex, its speed is acceptable in practical applications. Bicubic interpolation, on the other hand, has a higher computational cost and slower processing speed, but it can achieve a magnified effect of high-resolution images compared to bilinear interpolation.

[0115] This application may use bilinear interpolation.

[0116] The above methods can improve the quality of the acquired images and remove interference from other tissues and noise.

[0117] Step 75: Train the image classification model.

[0118] (1) The preprocessed ultrasound image data is divided into training set and test set in an 8:2 ratio.

[0119] (2) Input the training dataset into the image classification model. By training the image classification model, the unknown parameters are updated cyclically during the training process, and the training is repeated continuously to obtain the optimal model parameters.

[0120] (3) Input the test set into the trained model for testing and obtain the results of the image classification model.

[0121] Step 76: Classify the ultrasound images using the trained image classification model.

[0122] The acquired intrapartum ultrasound images are input into the classification model trained in step 75 to obtain classification images of different sections.

[0123] The classification model automatically categorizes different sections, replacing the traditional method where doctors manually classify sections, saving doctors' time and effort and improving midwifery efficiency.

[0124] Step 77: Display the classified ultrasound images.

[0125] Display the images classified in step 76, along with their similarity probabilities, such as... Figure 8As shown.

[0126] See Figure 9 , Figure 9 This is a schematic diagram of an embodiment of the ultrasound imaging device provided in this application. The ultrasound imaging device 100 includes: an ultrasound probe 101, a transmitting circuit 102, a receiving circuit 103, and a processor 105.

[0127] The transmitting circuit 102 is connected to the ultrasonic probe 101 and is used to transmit ultrasonic signals to the target tissue through the ultrasonic probe 101.

[0128] The receiving circuit 103 is connected to the ultrasonic probe 101 and is used to acquire the ultrasonic echo signal reflected by the target tissue.

[0129] Processor 105 is connected to receiver circuit 103 to implement the following method:

[0130] An ultrasonic signal is emitted to the target tissue, and the ultrasonic echo signal reflected by the target tissue is acquired; an ultrasonic image corresponding to the target tissue is formed based on the ultrasonic echo signal; the ultrasonic image is automatically classified to obtain the image type of the ultrasonic image; wherein, the image type includes cross-sectional image and longitudinal section image.

[0131] It is understood that the processor 105 is also used to execute computer programs to implement the methods of any of the above embodiments, as detailed in any of the above embodiments, which will not be repeated here.

[0132] See Figure 10 , Figure 10 This is a schematic diagram of an embodiment of the computer-readable storage medium provided in this application. The computer-readable storage medium 110 is used to store a computer program 111, which, when executed by a processor, implements the following methods:

[0133] An ultrasonic signal is emitted to the target tissue, and the ultrasonic echo signal reflected by the target tissue is acquired; an ultrasonic image corresponding to the target tissue is formed based on the ultrasonic echo signal; the ultrasonic image is automatically classified to obtain the image type of the ultrasonic image; wherein, the image type includes cross-sectional image and longitudinal section image.

[0134] It is understood that when the computer program 111 is executed by the processor, it is also used to implement the method of any of the above embodiments. For details, please refer to any of the above embodiments, which will not be repeated here.

[0135] In the several embodiments provided in this application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0136] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0137] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0138] If the integrated units in the other embodiments described above are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0139] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method of classifying an ultrasound image, characterized by, The method includes: An ultrasonic signal is emitted toward the target tissue, and the ultrasonic echo signal reflected by the target tissue is collected. An ultrasound image corresponding to the target tissue is generated based on the ultrasound echo signal; The ultrasound images are automatically classified to obtain the image type of the ultrasound images; wherein, the image type includes cross-sectional images and longitudinal images; The step of emitting ultrasonic signals to the target tissue and acquiring the ultrasonic echo signals reflected by the target tissue includes: During fetal delivery, ultrasound signals are emitted into the perineal tissue, and the ultrasound echo signals reflected by the perineal tissue are collected. The image types include transverse and longitudinal cross-sectional images of the perineal tissue; Before automatically classifying the ultrasound image to obtain its image type, the process includes: Edge detection is performed on the ultrasound image; The edge area is determined based on the edge detection results. The edge area is the area of ​​the edge region formed between the arc and the ultrasonic probe. The edge area is determined based on the sector area between the ultrasonic probe and the arc. If the edge area is greater than or equal to a set area threshold, the ultrasound image is retained; or, In response to the edge area being less than the set area threshold, ultrasound images that do not meet the requirements are removed before the automatic classification is performed, so as to reduce the amount of ultrasound image data to be automatically classified in the future.

2. The method of claim 1, wherein, The automatic classification of the ultrasound images to obtain the image type of the ultrasound images includes: Determine the similarity between the ultrasound image and a standard image of each image type; The image type of the ultrasound image is the standard image whose similarity is greater than the preset similarity.

3. The method of claim 2, wherein, The method further includes: The ultrasound image, the corresponding image type, and the similarity are displayed on the human-computer interaction interface.

4. The method of claim 3, wherein, After displaying the ultrasound image, the corresponding image type, and the similarity on the human-computer interaction interface, the process includes: Upon receiving the first selection instruction, the ultrasound image displayed on the human-computer interaction interface is saved; Alternatively, a second selection instruction may be received to remove the ultrasound image displayed on the human-computer interaction interface.

5. The method of claim 1, wherein, The automatic classification of the ultrasound images to obtain the image type of the ultrasound images includes: The ultrasound images are classified using an image classification model to obtain the corresponding image type.

6. The method of claim 5, wherein, The automatic classification of the ultrasound images to obtain the image type of the ultrasound images includes: The retained ultrasound images are automatically classified to obtain the image type of the ultrasound images.

7. An ultrasound imaging device, characterized by The ultrasound imaging device includes: Ultrasonic probe; A transmitting circuit, connected to the ultrasound probe, is used to transmit ultrasound signals to the target tissue through the ultrasound probe; A receiving circuit, connected to the ultrasound probe, is used to acquire the ultrasound echo signal reflected by the target tissue; A processor, connected to the receiving circuit, is configured to implement the method as described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium is configured to store a computer program, which when executed by a processor, is configured to implement the method according to any one of claims 1-6.

Images