Ultrasound image processing method and apparatus for adnexa uteri

Abstract

This application relates to a method, apparatus, device, storage medium, and computer program product for processing ultrasound images of the uterus and adnexa. The method acquires the corresponding uterine and adnexal contours based on ultrasound cross-sectional images of the uterus and adnexa, and then determines the type of ultrasound cross-sectional image of the uterus and adnexa. The first type of image includes longitudinal and transverse cross-sectional images of the uterus, while the second type includes cross-sectional images of the ovaries and longitudinal cross-sections of the cervix. The subsequent measurement process is determined based on the type of ultrasound cross-sectional image of the uterus and adnexa. For the first type of image, measurements are performed based on the extracted skeleton lines and the contour of the uterus and adnexa; for the second type of image, measurements are performed based on the contour of the uterus and adnexa. By identifying different types of images and performing subsequent measurements, the corresponding measurement data of the uterus and adnexa are obtained more accurately.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, device, storage medium, and computer program product for processing ultrasound images of the uterus and adnexa. Background Technology

[0002] Ultrasound examination is the most widely used tool by gynecologists for disease diagnosis, preoperative evaluation, and postoperative follow-up. Measuring the size of the uterus and ovaries is the most basic examination component. In daily clinical practice, significant variations in measurements are common among different doctors, causing considerable confusion for both gynecologists and patients, and hindering mutual verification and communication among colleagues. The reasons for this are: differences in experience among doctors lead to variations in standardized images; individual differences exist in the identification of measurement points; and the heavy workload may further increase the likelihood of doctors neglecting the preservation of standard sections and related measurements.

[0003] Therefore, there is an urgent need for a method that can accurately measure data of the uterine adnexa. Summary of the Invention

[0004] Therefore, it is necessary to provide an accurate method, apparatus, device, computer-readable storage medium, and computer program product for processing ultrasound images of the uterus and adnexa, addressing the aforementioned technical problems.

[0005] In a first aspect, this application provides a method for processing ultrasound images of the uterine adnexa. The method includes:

[0006] Obtain ultrasound cross-sectional images of the uterus and adnexa;

[0007] The ultrasound cross-sectional image of the uterus and adnexa is segmented to obtain the outline of the uterus and adnexa;

[0008] If the ultrasound cross-sectional image of the uterine adnexa is a first type image, then the skeleton line of the uterine adnexa is extracted, and measurements are performed based on the extracted skeleton line and the uterine adnexa contour.

[0009] If the ultrasound cross-sectional image of the uterus and adnexa is a type II image, then measurements are performed based on the outline of the uterus and adnexa.

[0010] The first type of image includes a longitudinal section image of the uterus and a transverse section image of the uterus, while the second type of image includes a section image of the ovary and a longitudinal section image of the cervix.

[0011] In one embodiment, the ultrasound cross-sectional image of the uterine adnexa includes a longitudinal cross-sectional image of the uterus; the step of performing feature segmentation on the ultrasound cross-sectional image of the uterine adnexa to obtain the contour of the uterine adnexa includes: performing feature segmentation on the longitudinal cross-sectional image of the uterus to obtain the contour of the uterine body and the contour of the endometrium; extracting the skeleton line of the contour of the uterine adnexa and measuring based on the extracted skeleton line and the contour of the uterine adnexa includes: extracting the skeleton line of the endometrial contour in the contour of the uterine adnexa to obtain the endometrial skeleton line; and measuring based on the endometrial skeleton line, the contour of the uterine body, and the contour of the endometrium to obtain measurement data of the cervix.

[0012] In one embodiment, the measurement of the cervix based on the endometrial skeleton line, the uterine body contour, and the endometrial contour includes: determining the largest pixel in the image corresponding to the endometrial skeleton line; extending the largest pixel along a direction perpendicular to the endometrial skeleton line to obtain the intersection of the extended line and the endometrial contour; calculating the Euclidean distance between the intersection of the extended line and the endometrial contour to obtain the endometrial thickness; performing fitting processing based on the endometrial skeleton line to obtain the endometrial centerline; obtaining the intersection of the uterine body contour and the endometrial contour to obtain the first intersection point; determining the intersection of the endometrial centerline and the uterine body contour to obtain the second intersection point; and calculating the Euclidean distance between the first intersection point and the second intersection point to obtain the major axis length of the uterine body.

[0013] In one embodiment, the method further includes: determining the long diameter of the uterine body based on the first intersection point and the second intersection point; drawing a perpendicular line to the long diameter of the uterine body based on the long diameter of the uterine body, determining the longest perpendicular line formed by the perpendicular line to the long diameter of the uterine body and the outline of the uterine body, and obtaining the anteroposterior diameter of the uterine body; and calculating the length of the anteroposterior diameter of the uterine body.

[0014] In one embodiment, the ultrasound cross-sectional image of the uterine adnexa includes a transverse cross-sectional image of the uterus; the step of performing feature segmentation on the ultrasound cross-sectional image of the uterine adnexa to obtain the contour of the uterine adnexa includes: performing feature segmentation on the transverse cross-sectional image of the uterus to obtain the contour of the uterine body and the contour of the endometrium; extracting the skeleton line of the contour of the uterine adnexa and measuring based on the extracted skeleton line and the contour of the uterine adnexa includes: extracting the skeleton line of the endometrial contour in the contour of the uterine adnexa to obtain the endometrial skeleton line; generating a horizontal line parallel to the endometrial skeleton line based on the endometrial skeleton line; determining the longest straight line formed by the horizontal line and the contour of the uterine body to obtain the transverse diameter of the uterine body; and calculating the length of the transverse diameter of the uterine body.

[0015] In one embodiment, the ultrasound cross-sectional image of the uterine adnexa includes an ovarian cross-sectional image; the step of performing feature segmentation on the ultrasound cross-sectional image of the uterine adnexa to obtain the contour of the uterine adnexa includes: performing feature segmentation on the ovarian cross-sectional image to obtain the ovarian contour and the contour of the largest follicle; the measurement based on the contour of the uterine adnexa includes: performing distance calculation based on the ovarian contour to determine the longest distance between the ovarian contours, obtaining the ovarian long axis and the ovarian long axis length; performing distance calculation based on the contour of the largest follicle to determine the longest distance between the largest follicle contours, obtaining the largest follicle long axis and the largest follicle long axis length.

[0016] In one embodiment, the method further includes: drawing a perpendicular line from the ovarian long axis to the ovarian long axis, determining the longest perpendicular line from the ovarian long axis formed by the perpendicular line from the ovarian long axis and the ovarian contour, obtaining the ovarian short axis, and calculating the length of the ovarian short axis; drawing a perpendicular line from the long axis of the largest follicle to the follicle long axis, determining the longest perpendicular line from the follicle long axis formed by the perpendicular line from the follicle long axis and the follicle contour, obtaining the largest follicle short axis, and calculating the length of the largest follicle short axis.

[0017] In one embodiment, the ultrasound cross-sectional image of the uterine adnexa includes a longitudinal cross-sectional image of the cervix; the step of performing feature segmentation on the ultrasound cross-sectional image of the uterine adnexa to obtain the contour of the uterine adnexa includes: performing feature segmentation on the longitudinal cross-sectional image of the cervix to obtain the cervical contour and the cervical endometrial contour; the measurement based on the contour of the uterine adnexa includes: determining the internal cervical os and the external cervical os based on the cervical endometrial contour; determining the cervical major diameter and the length of the cervical major diameter based on the internal cervical os and the external cervical os; drawing a perpendicular line to the cervical major diameter based on the cervical major diameter, determining the longest perpendicular line to the cervical major diameter formed by the perpendicular line to the cervical contour, obtaining the anteroposterior diameter of the cervix, and calculating the length of the anteroposterior diameter of the cervix.

[0018] In one embodiment, the process of extracting the skeleton line of the uterine adnexa and measuring based on the extracted skeleton line and the uterine adnexa contour includes: obtaining a set of uterine adnexa contour points; generating a uterine adnexa contour mask based on the set of uterine adnexa contour points; and performing distance transformation and inversion operations on the uterine adnexa contour mask to obtain the uterine adnexa skeleton line.

[0019] Secondly, this application also provides a uterine adnexa ultrasound image processing device. The device includes:

[0020] The acquisition module is used to acquire ultrasound cross-sectional images of the uterine adnexa;

[0021] The segmentation module is used to perform feature segmentation on the ultrasound cross-sectional image of the uterine adnexa to obtain the outline of the uterine adnexa.

[0022] The first measurement module is used to extract the skeleton line of the uterine adnexa if the type of the uterine adnexa is the first type, and to perform measurement based on the extracted skeleton line and the uterine adnexa contour to obtain the measurement data of the first type of uterine adnexa.

[0023] The second measurement module is used to measure the outline of the uterine adnexa if the type of the uterine adnexa is the second type, and to obtain the measurement data of the second type of uterine adnexa.

[0024] The first type of uterine adnexa includes a longitudinal section of the uterus and a transverse section of the uterus, while the second type of uterine adnexa includes a section of the ovary and a longitudinal section of the cervix.

[0025] Thirdly, this application also provides a uterine adnexa ultrasound image processing device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0026] Obtain ultrasound cross-sectional images of the uterus and adnexa;

[0027] The ultrasound cross-sectional image of the uterus and adnexa is segmented to obtain the outline of the uterus and adnexa;

[0028] If the ultrasound cross-sectional image of the uterine adnexa is a first type image, then the skeleton line of the uterine adnexa is extracted, and measurements are performed based on the extracted skeleton line and the uterine adnexa contour.

[0029] If the ultrasound cross-sectional image of the uterus and adnexa is a type II image, then measurements are performed based on the outline of the uterus and adnexa.

[0030] The first type of image includes a longitudinal section image of the uterus and a transverse section image of the uterus, while the second type of image includes a section image of the ovary and a longitudinal section image of the cervix.

[0031] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0032] Obtain ultrasound cross-sectional images of the uterus and adnexa;

[0033] The ultrasound cross-sectional image of the uterus and adnexa is segmented to obtain the outline of the uterus and adnexa;

[0034] If the ultrasound cross-sectional image of the uterine adnexa is a first type image, then the skeleton line of the uterine adnexa is extracted, and measurements are performed based on the extracted skeleton line and the uterine adnexa contour.

[0035] If the ultrasound cross-sectional image of the uterus and adnexa is a type II image, then measurements are performed based on the outline of the uterus and adnexa.

[0036] The first type of image includes a longitudinal section image of the uterus and a transverse section image of the uterus, while the second type of image includes a section image of the ovary and a longitudinal section image of the cervix.

[0037] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0038] Obtain ultrasound cross-sectional images of the uterus and adnexa;

[0039] The ultrasound cross-sectional image of the uterus and adnexa is segmented to obtain the outline of the uterus and adnexa;

[0040] If the ultrasound cross-sectional image of the uterine adnexa is a first type image, then the skeleton line of the uterine adnexa is extracted, and measurements are performed based on the extracted skeleton line and the uterine adnexa contour.

[0041] If the ultrasound cross-sectional image of the uterus and adnexa is a type II image, then measurements are performed based on the outline of the uterus and adnexa.

[0042] The first type of image includes a longitudinal section image of the uterus and a transverse section image of the uterus, while the second type of image includes a section image of the ovary and a longitudinal section image of the cervix.

[0043] The aforementioned method, apparatus, device, storage medium, and computer program product for processing ultrasound images of the uterus and adnexa acquire ultrasound cross-sectional images of the uterus and adnexa; perform feature segmentation on the ultrasound cross-sectional images of the uterus and adnexa to obtain the contours of the uterus and adnexa; if the ultrasound cross-sectional image of the uterus and adnexa is a first type image, then extract the skeleton lines of the contours of the uterus and adnexa, and perform measurements based on the extracted skeleton lines and the contours of the uterus and adnexa; if the ultrasound cross-sectional image of the uterus and adnexa is a second type image, then perform measurements based on the contours of the uterus and adnexa; wherein, the first type image includes longitudinal and transverse uterine cross-sectional images, and the second type image includes ovarian and longitudinal cervical cross-sectional images. The entire scheme is based on ultrasound cross-sectional images of the uterus and adnexa to obtain the corresponding contours of the uterus and adnexa, then determines the type of ultrasound cross-sectional image of the uterus and adnexa, and determines the subsequent measurement process based on the type of ultrasound cross-sectional image of the uterus and adnexa. For the first type image, measurements are performed based on the extracted skeleton lines and the contours of the uterus and adnexa; for the second type image, measurements are performed based on the contours of the uterus and adnexa. By recognizing different types of images and performing subsequent measurements, the corresponding measurement data of the uterus and adnexa are obtained more accurately. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of this application or the conventional technology, the drawings used in the description of the embodiments or the conventional technology will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is an application environment diagram of the ultrasound image processing method for the uterine adnexa in one embodiment;

[0046] Figure 2 This is a flowchart illustrating a method for processing ultrasound images of the uterine adnexa in one embodiment;

[0047] Figure 3 This is a schematic diagram of a longitudinal section image of the uterus in one embodiment;

[0048] Figure 4 This is a schematic diagram of the outline of the uterine body and the outline of the endometrium in one embodiment;

[0049] Figure 5 This is a schematic diagram of the endometrial contour in one embodiment;

[0050] Figure 6 This is a schematic diagram of the long diameter of the uterine cervix in one embodiment;

[0051] Figure 7 This is a schematic diagram of the anteroposterior diameter of the uterus and cervix in one embodiment.

[0052] Figure 8 This is a schematic diagram of a cross-sectional image of the uterus in one embodiment;

[0053] Figure 9 This is a schematic diagram of the outline of the uterine body and the outline of the endometrium in another embodiment;

[0054] Figure 10 This is a schematic diagram of the transverse diameter of the uterine body in one embodiment;

[0055] Figure 11 This is a schematic diagram of a cross-sectional image of the ovary in one embodiment;

[0056] Figure 12 This is a schematic diagram of the ovarian outline and the outline of the largest follicle in one embodiment;

[0057] Figure 13 This is a schematic diagram of the long axis of the ovary and the long axis of the largest follicle in one embodiment;

[0058] Figure 14 This is a schematic diagram of the short diameter of the ovary and the short diameter of the largest follicle in one embodiment;

[0059] Figure 15 This is a schematic diagram of the cervical contour and the cervical endometrium contour in one embodiment.

[0060] Figure 16 This is a schematic diagram of the cervical long diameter in one embodiment;

[0061] Figure 17 This is a schematic diagram of the anterior and posterior diameter of the cervix in one embodiment;

[0062] Figure 18 This is a flowchart illustrating a method for processing ultrasound images of the uterine adnexa in another embodiment;

[0063] Figure 19 This is a structural block diagram of a uterine adnexa ultrasound image processing device in one embodiment;

[0064] Figure 20 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0066] The ultrasound image processing method for the uterine adnexa provided in this application embodiment can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed on a cloud or other network server. Terminal 102 acquires ultrasound cross-sectional images of the uterus and adnexa, and performs feature segmentation on the ultrasound cross-sectional images of the uterus and adnexa based on the instance segmentation model trained by server 104 to obtain the contour of the uterus and adnexa. If the ultrasound cross-sectional image of the uterus and adnexa is a first type image, the skeleton line of the uterus and adnexa contour is extracted, and measurements are performed based on the extracted skeleton line and the uterus and adnexa contour. If the ultrasound cross-sectional image of the uterus and adnexa is a second type image, measurements are performed based on the uterus and adnexa contour. The first type image includes longitudinal and transverse cross-sectional images of the uterus, and the second type image includes cross-sectional images of the ovaries and longitudinal cross-sectional images of the cervix. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0067] In one embodiment, such as Figure 2 As shown, a method for processing ultrasound images of the uterus and adnexa is provided, which can be applied to... Figure 1 Taking terminal 102 as an example, the explanation includes the following steps:

[0068] Step 202: Obtain ultrasound cross-sectional images of the uterine adnexa.

[0069] Among them, transvaginal ultrasound images of the uterus and adnexa refer to transvaginal ultrasound images of the uterus and adnexa. These images provide gynecologists with useful information during gynecological ultrasound examinations and diagnoses, such as the length and transverse diameter of the uterus, the thickness of the endometrium, and the type of uterine position (e.g., folded, anteroposterior). The adnexa include the uterus, ovaries, and cervix.

[0070] Specifically, the terminal sends a command to the medical ultrasound scanning device to acquire ultrasound cross-sectional images of the uterus and adnexa. When the medical ultrasound scanning device receives the command, it performs an ultrasound beam scan on the female user's vagina to obtain an ultrasound cross-sectional image of the uterus and adnexa.

[0071] Step 204: Perform feature segmentation on the ultrasound cross-sectional image of the uterus and adnexa to obtain the outline of the uterus and adnexa.

[0072] Specifically, the terminal performs feature segmentation on the ultrasound cross-sectional images of the uterus and adnexa based on the trained instance segmentation model to obtain the contours of the uterus and adnexa. The instance segmentation model is trained on a Mask-RCNN (Mask-Region Convolutional Neural Networks) network and a dataset of ultrasound cross-sectional images of the uterus and adnexa. The dataset of ultrasound cross-sectional images of the uterus and adnexa includes cross-sectional images of the uterus and adnexa as well as polygonal regions of different uterine structures.

[0073] Step 206: If the ultrasound cross-sectional image of the uterus and adnexa is a first type image, then the skeleton line of the uterus and adnexa is extracted, and the extracted skeleton line and the uterus and adnexa contour are measured.

[0074] The first type of image includes longitudinal and transverse uterine sections, while the second type of image includes ovarian and cervical sections.

[0075] Specifically, the terminal determines the type of the ultrasound image of the uterus and adnexa. If the ultrasound image of the uterus and adnexa is a longitudinal section or a transverse section of the uterus, then the ultrasound image of the uterus and adnexa is a first-type image. Next, the skeleton line of the uterus and adnexa contour is extracted. Based on the skeleton line and the uterus and adnexa contour, the measurement location of the uterus and adnexa is determined. The length of the measurement location is calculated to obtain the measurement value of the uterus and adnexa.

[0076] Step 208: If the ultrasound cross-sectional image of the uterus and adnexa is a type II image, then measurements are taken based on the contour of the uterus and adnexa.

[0077] Specifically, the terminal determines the type of ultrasound image of the uterus and adnexa. If the ultrasound image of the uterus and adnexa is an ovarian section or a longitudinal section of the cervix, then the ultrasound image of the uterus and adnexa is a type II image. Then, based on the contour of the uterus and adnexa, the measurement location of the uterus and adnexa is determined, and the length of the measurement location is calculated to obtain the measurement value of the uterus and adnexa.

[0078] In the aforementioned method for processing ultrasound images of the uterus and adnexa, an ultrasound cross-sectional image of the uterus and adnexa is acquired; feature segmentation is performed on the ultrasound cross-sectional image of the uterus and adnexa to obtain the contour of the uterus and adnexa; if the ultrasound cross-sectional image of the uterus and adnexa is a first type image, the skeleton line of the contour of the uterus and adnexa is extracted, and measurements are performed based on the extracted skeleton line and the contour of the uterus and adnexa; if the ultrasound cross-sectional image of the uterus and adnexa is a second type image, measurements are performed based on the contour of the uterus and adnexa; wherein, the first type image includes longitudinal and transverse uterine cross-sectional images, and the second type image includes ovarian and longitudinal cervical cross-sectional images. The entire scheme is based on the ultrasound cross-sectional image of the uterus and adnexa to obtain the corresponding contour of the uterus and adnexa, and then determines the type of ultrasound cross-sectional image of the uterus and adnexa, and determines the subsequent measurement process according to the type of ultrasound cross-sectional image of the uterus and adnexa. For the first type image, measurements are performed based on the extracted skeleton line and the contour of the uterus and adnexa; for the second type image, measurements are performed based on the contour of the uterus and adnexa. By recognizing different types of images and performing subsequent measurements, the corresponding measurement data of the uterus and adnexa are obtained more accurately.

[0079] In an optional embodiment, feature segmentation of the ultrasound cross-sectional image of the uterus and adnexa to obtain the contour of the uterus and adnexa includes: feature segmentation of the longitudinal cross-sectional image of the uterus to obtain the contour of the uterine body and the contour of the endometrium; extraction of the skeleton line of the contour of the uterus and adnexa, and measurement based on the extracted skeleton line and the contour of the uterus and adnexa includes: extraction of the skeleton line of the endometrial contour in the contour of the uterus and adnexa to obtain the endometrial skeleton line; and measurement based on the endometrial skeleton line, the contour of the uterine body, and the contour of the endometrium to obtain measurement data of the uterus and cervix.

[0080] Among them, the ultrasound images of the uterus and adnexa include longitudinal cross-sectional images of the uterus, such as... Figure 3 The image shown is an example of a longitudinal section of the uterus.

[0081] Specifically, the terminal performs feature segmentation on the longitudinal section image of the uterus based on an instance segmentation model to obtain the contours of the uterine body and the endometrium, such as... Figure 4 The image shows the segmented outline of the uterine body and the internal endometrial outline. Next, the terminal uses a skeleton line extraction algorithm to extract the skeleton line from the endometrial outline within the uterine adnexa outline, obtaining the endometrial skeleton line. Finally, based on the endometrial skeleton line, the uterine body outline, and the endometrial outline, the measurement location of the uterine body is determined, yielding the measurement data of the uterine body.

[0082] In an optional embodiment, measurements of the cervix are obtained based on the endometrial skeleton line, the uterine body contour, and the endometrial contour, including: determining the largest pixel in the image corresponding to the endometrial skeleton line; extending the largest pixel along a direction perpendicular to the endometrial skeleton line to obtain the intersection of the extended line and the endometrial contour; calculating the Euclidean distance between the intersection of the extended line and the endometrial contour to obtain the endometrial thickness; performing fitting processing based on the endometrial skeleton line to obtain the endometrial centerline; obtaining the intersection of the uterine body contour and the endometrial contour to obtain the first intersection point; determining the intersection of the endometrial centerline and the uterine body contour to obtain the second intersection point; and calculating the Euclidean distance between the first and second intersection points to obtain the major axis length of the uterine body.

[0083] Specifically, the largest pixel in the image corresponding to the fitted line of the endometrial skeleton line is determined. Then, the largest pixel is extended along a direction perpendicular to the endometrial skeleton line to obtain the intersection point of the extended line and the endometrial contour. At this point, the measurement point of the endometrial thickness is obtained. Then, the Euclidean distance between the intersection point of the extended line and the endometrial contour is calculated, and the scale factor of the longitudinal section image of the uterus is introduced. The Euclidean distance between the intersection point of the extended line and the endometrial contour is divided by the scale factor of the longitudinal section image of the uterus to obtain the actual endometrial thickness.

[0084] The terminal records all point data along the endometrial skeleton line. Based on this data, a straight line is fitted to obtain the fitted line, i.e., the endometrial centerline. The slope and intercept of the fitted line are then determined, representing the values ​​of k and b in the expression y = kx + b for the endometrial centerline. The intersection point of the uterine body contour and the endometrial contour is obtained, yielding the first intersection point. This first intersection point is then substituted into the endometrial centerline to determine the intersection point on the other side of the uterine body contour, yielding the second intersection point. Figure 6As shown, the second intersection point is located on the side of the uterus where the cervix is ​​not open. Finally, the Euclidean distance between the first and second intersection points is calculated. The Euclidean distance between the intersection point of the extended line and the endometrial contour is divided by the scale factor of the longitudinal section image of the uterus to obtain the actual length of the major axis of the uterine body.

[0085] When the outline of the uterine body and the outline of the endometrium do not intersect, the intersection point of the endometrial centerline and the opening side of the uterine body outline is determined, resulting in the first intersection point. Then, the endometrial centerline is extended towards the closed side of the uterine body outline to obtain the intersection point of the endometrial centerline and the uterine body outline, resulting in the second intersection point. Finally, the Euclidean distance between the first and second intersection points is calculated. Dividing this Euclidean distance between the intersection point of the extended line and the endometrial outline by the scale factor of the uterine longitudinal section image yields the actual major axis length of the uterine body.

[0086] This embodiment can accurately determine the apex of the long axis of the uterine body based on the endometrial skeleton line, the outline of the uterine body, and the outline of the endometrium, and then accurately calculate the length of the long axis of the uterine body based on the apex of the long axis of the uterine body.

[0087] In an optional embodiment, the method further includes: determining the long diameter of the uterine body based on the first intersection point and the second intersection point; drawing a perpendicular line to the long diameter of the uterine body based on the long diameter of the uterine body, determining the longest perpendicular line formed by the perpendicular line to the long diameter of the uterine body and the outline of the uterine body, and obtaining the anteroposterior diameter of the uterine body; and calculating the length of the anteroposterior diameter of the uterine body.

[0088] Specifically, the straight line between the first and second intersection points is the long axis of the uterine body. Then, since the long axis of the uterus is perpendicular to its anteroposterior diameter, drawing a perpendicular line from the long axis of the uterine body will yield multiple perpendicular lines of varying lengths. Then, as... Figure 7 As shown, the longest perpendicular line formed by the longest diameter of the uterine body and the outline of the uterine body is determined to obtain the anteroposterior diameter of the uterine body; the length of the anteroposterior diameter of the uterine body is calculated, and the length of the anteroposterior diameter of the uterine body is divided by the scale multiple of the longitudinal section image of the uterus to obtain the actual anteroposterior diameter of the uterine body.

[0089] Furthermore, in practical applications, since the major axis of the uterus is perpendicular to its anteroposterior diameter, the slope of the major axis of the uterine body can be determined. Then, based on the slope of the major axis, the slope of the straight line containing the anteroposterior diameter of the uterine body can be determined. Next, all points in the uterine body contour are substituted with the slope information to obtain the two points with the maximum Euclidean distance. Finally, the Euclidean distance between the two points with the maximum Euclidean distance is calculated, and this distance is divided by the scale factor of the uterine longitudinal section image to obtain the anteroposterior diameter of the uterine body.

[0090] In an optional embodiment, feature segmentation of the ultrasound cross-sectional image of the uterus and adnexa to obtain the contour of the uterus and adnexa includes: feature segmentation of the transverse cross-sectional image of the uterus to obtain the contour of the uterine body and the contour of the endometrium; extraction of the skeleton line of the contour of the uterus and adnexa, and measurement based on the extracted skeleton line and the contour of the uterus and adnexa includes: extraction of the skeleton line of the endometrial contour in the contour of the uterus and adnexa to obtain the endometrial skeleton line; generation of a horizontal line parallel to the endometrial skeleton line based on the endometrial skeleton line; determination of the longest straight line formed by the horizontal line and the contour of the uterine body to obtain the transverse diameter of the uterine body; and calculation of the length of the transverse diameter of the uterine body.

[0091] Among them, ultrasound images of the uterus and adnexa include cross-sectional images of the uterus, such as... Figure 8 The image shown is a schematic diagram of a cross-sectional view of the uterus.

[0092] Specifically, such as Figure 9 As shown, the terminal performs feature segmentation on the uterine cross-sectional image based on an instance segmentation model to obtain the uterine body contour and endometrial contour. The image containing the uterine body and endometrial contours is then converted into a binary image. The `cv2.findContours()` function is called to obtain the contour point set. The contour point set corresponding to the maximum value in the contour region is selected to obtain the uterine body and endometrial contour point sets. Then, the `cv2.fillPoly()` function is called to fill the uterine body and endometrial contour point sets into a matrix of the same size as the original uterine cross-sectional image, with all pixel values ​​set to 0, resulting in a mask image. Finally, median smoothing filtering is used to smooth the boundaries of the mask image, resulting in a clear uterine body and endometrial contour.

[0093] Subsequently, the terminal uses a skeleton line extraction algorithm to extract the skeleton line from the endometrial contour within the uterine adnexa, obtaining the endometrial skeleton line. The terminal records all point data on the endometrial skeleton line and performs straight line fitting based on this data to obtain the fitted line of the endometrial skeleton line, i.e., the endometrial centerline. The slope and intercept of the fitted line are then determined. Based on the slope of the endometrial centerline, horizontal lines parallel to the endometrial skeleton line are generated; for example... Figure 10 As shown, determine the longest straight line formed by the horizontal line and the outline of the uterine body to obtain the transverse diameter of the uterine body; calculate the length of the transverse diameter of the uterine body, and divide the length of the transverse diameter of the uterine body by the scale multiple of the longitudinal section image of the uterus to obtain the actual transverse diameter length of the uterine body.

[0094] In an optional embodiment, performing feature segmentation on the ultrasound cross-sectional image of the uterine adnexa to obtain the contour of the uterine adnexa includes: performing feature segmentation on the cross-sectional image of the ovary to obtain the ovarian contour and the contour of the largest follicle; measuring based on the contour of the uterine adnexa includes: calculating the distance based on the ovarian contour to determine the longest distance between the ovarian contours, obtaining the ovarian long axis and the length of the ovarian long axis; calculating the distance based on the contour of the largest follicle to determine the longest distance between the contours of the largest follicles, obtaining the length of the largest follicle and the length of the largest follicle.

[0095] Among them, the ultrasound cross-sectional images of the uterus and adnexa include cross-sectional images of the ovaries, such as... Figure 11 The image shown is a schematic diagram of a cross-sectional view of the ovary.

[0096] Specifically, such as Figure 12 As shown, the terminal performs feature segmentation on the ovarian cross-sectional image based on an instance segmentation model to obtain the ovarian contour and follicle contour. It then performs IOU (Intersection over Union, loss function) comparison on each follicle contour to determine the largest follicle contour. The images containing the ovarian contour and the largest follicle contour are converted into binary images, and then the `cv2.findContours()` function is called to obtain the contour point set. The contour point set corresponding to the maximum value in the contour region is then selected to obtain the ovarian contour and the largest follicle contour point set.

[0097] Then, as Figure 13 As shown, distance calculations are performed based on the ovarian contour point set. The Euclidean distance between each pair of points in the set is calculated. The straight line between the two points with the largest Euclidean distance is determined as the ovarian major axis, and its length is calculated. This length is then divided by the scale factor of the ovarian cross-sectional image to obtain the actual ovarian major axis length. Similarly, distance calculations are performed based on the maximum follicle contour point set. The Euclidean distance between each pair of points in the maximum follicle contour point set is calculated. The straight line between the two points with the largest Euclidean distance is determined as the maximum follicle major axis, and its length is calculated. This length is then divided by the scale factor of the ovarian cross-sectional image to obtain the actual maximum follicle major axis length.

[0098] In an optional embodiment, the method further includes: drawing a perpendicular line from the ovarian long axis to the ovarian long axis, determining the longest perpendicular line from the ovarian long axis formed by the perpendicular line and the ovarian contour, obtaining the ovarian short axis, and calculating the length of the ovarian short axis; drawing a perpendicular line from the long axis of the largest follicle to the follicle long axis, determining the longest perpendicular line from the follicle long axis formed by the perpendicular line of the largest follicle and the follicle contour, obtaining the largest follicle short axis, and calculating the length of the largest follicle short axis.

[0099] Specifically, the minor and major axes of the ovary are perpendicular to each other. Drawing a perpendicular line from the major axis of the ovary will yield multiple perpendicular lines of varying lengths. Then, as... Figure 14 As shown, the longest perpendicular line to the ovarian major axis, formed by the perpendicular line to the ovarian contour, is the ovarian minor axis. The length of the ovarian minor axis is calculated by dividing it by the scale factor of the ovarian cross-sectional image. Similarly, the length of the actual minor axis of the largest follicle is calculated.

[0100] Furthermore, in practical applications, since the minor and major axes of the ovary are perpendicular to each other, the normal to the major axis of the ovary can be used to traverse the set of points on the ovarian contour to obtain the endpoint corresponding to the largest Euclidean distance. This endpoint is then divided by the scale factor of the ovarian cross-sectional image to obtain the actual length of the minor axis of the ovary. Similarly, the actual length of the minor axis of the largest follicle can be calculated.

[0101] In an optional embodiment, feature segmentation of the ultrasound cross-sectional image of the uterine adnexa to obtain the contour of the uterine adnexa includes: feature segmentation of the longitudinal cross-sectional image of the cervix to obtain the cervical contour and the cervical endometrial contour; measurement based on the contour of the uterine adnexa includes: determining the internal cervical os and the external cervical os based on the cervical endometrial contour; determining the cervical major diameter and the length of the cervical major diameter based on the internal cervical os and the external cervical os; drawing a perpendicular line to the cervical major diameter based on the cervical major diameter, determining the longest perpendicular line to the cervical major diameter formed by the perpendicular line to the cervical contour, obtaining the anteroposterior diameter of the cervix, and calculating the length of the anteroposterior diameter of the cervix.

[0102] Among them, the ultrasound cross-sectional images of the uterus and adnexa include longitudinal cross-sectional images of the cervix.

[0103] Specifically, such as Figure 15 As shown, the terminal performs feature segmentation on the longitudinal section image of the cervix based on the instance segmentation model to obtain the cervical contour and the cervical endometrial contour. The contour point information of the cervical contour and the cervical endometrial contour are converted into a binary image, and then the contour extraction function is called to re-obtain the contour point set and mask image of the cervical contour and the cervical endometrial contour. Median smoothing is used to smooth the boundary of the mask image to obtain the smoothed cervical contour and the cervical endometrial contour.

[0104] For the cervical endometrial contour point set, the region on the left side of the image is selected. Using the left smallest inscribed rectangle, the center point of the left smallest inscribed rectangle is fitted to obtain the internal cervical os. Similarly, in the region on the right side of the image, the center point of the right smallest inscribed rectangle is fitted to obtain the external cervical os. Figure 16As shown, the straight line between the internal and external cervical os is the cervical long axis. Calculate the Euclidean distance between the internal and external cervical os, and divide this distance by the scale factor of the cervical longitudinal section image to obtain the actual cervical long axis length. Then, as... Figure 17 As shown, a perpendicular line is drawn from the cervical long diameter line. The longest perpendicular line from the cervical long diameter line to the cervical contour is determined, and the anteroposterior diameter of the cervix is ​​obtained. The length of the anteroposterior diameter of the cervix is ​​calculated. The length of the anteroposterior diameter of the cervix is ​​divided by the scale factor of the cervical longitudinal section image to obtain the actual length of the anteroposterior diameter of the cervix.

[0105] In practical applications, since the major axis of the cervix is ​​perpendicular to its anteroposterior diameter, the normal of the major axis of the cervix can be used to traverse the set of cervical contour points to obtain the endpoint corresponding to the largest Euclidean distance. Then, divide by the scale factor of the cervical longitudinal section image to obtain the actual length of the anteroposterior diameter of the cervix.

[0106] If the internal and external cervical os cannot be determined during the calculation of the cervical major diameter, a method can be used to measure the length of an arc-shaped curve along the cervical endometrial contour. Based on the contour point set and mask image of the cervical endometrial contour, a skeleton line of the cervical endometrial line is obtained using distance variation, and the data of all points on the skeleton line are recorded. The skeleton line is then fitted into an arc-shaped curve, and the circumference of the arc-shaped curve is calculated. Dividing this circumference by the scale factor of the cervical longitudinal section image yields the actual cervical major diameter.

[0107] In an optional embodiment, the process of extracting the skeleton line of the uterine adnexa and measuring based on the extracted skeleton line and the uterine adnexa contour includes: obtaining a set of uterine adnexa contour points; generating a uterine adnexa contour mask based on the set of uterine adnexa contour points; and performing distance transformation and inversion operations on the uterine adnexa contour mask to obtain the uterine adnexa skeleton line.

[0108] Specifically, the image containing the uterine adnexa contour is converted into a binary image. Then, the `cv2.findContours()` function is called to obtain the contour point set. The contour point set corresponding to the maximum value in the contour region is selected to obtain the uterine adnexa contour point set. Then, the `cv2.fillPoly()` function is called to fill the uterine adnexa contour point set into a matrix with the same size as the original uterine adnexa ultrasound cross-sectional image and all pixel values ​​being 0, resulting in a uterine adnexa contour mask. Finally, median smoothing filtering is used to smooth the boundaries of the mask image to obtain the smoothed uterine adnexa contour.

[0109] The distance transformation function `cv2.distanceTransform()` is used to transform the uterine adnexa contour mask, obtaining a distance relationship matrix between each pixel and the background (pixels with a value of 0). The distance relationship values ​​are then sorted, and the index of the maximum value in each column is taken. A reversal operation is then performed, ensuring that the values ​​of points on the skeleton line are not 0, while the values ​​of all other points are 0, thus obtaining the uterine adnexa skeleton line.

[0110] To facilitate understanding of the technical solutions provided in the embodiments of this application, such as Figure 18 As shown, the ultrasound image processing method for the uterus and adnexa provided in this application embodiment is briefly described using a complete ultrasound image processing process for the uterus and adnexa:

[0111] Step 1801: Obtain ultrasound images of the uterus and adnexa. These ultrasound images include longitudinal and transverse sections of the uterus, sections of the ovaries, and longitudinal sections of the cervix.

[0112] Step 1802: Perform feature segmentation on the ultrasound cross-sectional image of the uterus and adnexa to obtain the contour of the uterus and adnexa. The contour of the uterus and adnexa includes the contour of the uterine body, the contour of the endometrium, the contour of the ovary, the contour of the largest follicle, the contour of the cervix, and the contour of the cervical endometrium.

[0113] Step 1803: If the ultrasound cross-sectional image of the uterus and adnexa is a first type image, then obtain the set of uterine adnexa contour points of the uterine adnexa contour.

[0114] Step 1804: Generate a uterine adnexa contour mask based on the uterine adnexa contour point set.

[0115] Step 1805: Perform distance transformation and inversion operations on the uterine adnexa contour mask to obtain the uterine adnexa skeleton line, and perform measurements based on the extracted skeleton line and the uterine adnexa contour.

[0116] Step 1806: If the ultrasound cross-sectional image of the uterus and adnexa is a type II image, then measurements are taken based on the contour of the uterus and adnexa.

[0117] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0118] Based on the same inventive concept, this application also provides a uterine adnexa ultrasound image processing apparatus for implementing the aforementioned uterine adnexa ultrasound image processing method. The solution provided by this apparatus is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the uterine adnexa ultrasound image processing apparatus provided below can be found in the limitations of the uterine adnexa ultrasound image processing method described above, and will not be repeated here.

[0119] In one embodiment, such as Figure 19 As shown, a uterine adnexa ultrasound image processing device is provided, comprising: an acquisition module 1902, a segmentation module 1904, a first measurement module 1906, and a second measurement module 1908, wherein:

[0120] Acquisition module 1902 is used to acquire ultrasound cross-sectional images of the uterine adnexa;

[0121] The segmentation module 1904 is used to perform feature segmentation on the ultrasound cross-sectional image of the uterus and adnexa to obtain the contour of the uterus and adnexa.

[0122] The first measurement module 1906 is used to extract the skeleton line of the uterine adnexa if the type of the uterine adnexa is the first type, and to measure based on the extracted skeleton line and the uterine adnexa outline to obtain the measurement data of the first type of uterine adnexa.

[0123] The second measurement module 1908 is used to measure the outline of the uterine adnexa if the type of the uterine adnexa is the second type, and to obtain the measurement data of the second type of uterine adnexa.

[0124] The first type of uterine adnexa includes a longitudinal section of the uterus and a transverse section of the uterus, while the second type of uterine adnexa includes a section of the ovary and a longitudinal section of the cervix.

[0125] In one embodiment, the ultrasound cross-sectional image of the uterine adnexa includes a longitudinal cross-sectional image of the uterus; the segmentation module 1904 is further used to perform feature segmentation on the longitudinal cross-sectional image of the uterus to obtain the contour of the uterine body and the contour of the endometrium; the first measurement module 1906 is further used to extract the skeleton line of the endometrial contour in the contour of the uterine adnexa to obtain the endometrial skeleton line; and measurements are performed based on the endometrial skeleton line, the contour of the uterine body, and the contour of the endometrium to obtain measurement data of the uterus and cervix.

[0126] In one embodiment, the first measurement module 1906 is further configured to: determine the largest pixel in the image corresponding to the endometrial skeleton line; extend the largest pixel along a direction perpendicular to the endometrial skeleton line to obtain the intersection of the extended line and the endometrial contour; calculate the Euclidean distance between the intersection of the extended line and the endometrial contour to obtain the endometrial thickness; perform fitting processing based on the endometrial skeleton line to obtain the endometrial centerline; obtain the intersection of the uterine body contour and the endometrial contour to obtain the first intersection point; determine the intersection of the endometrial centerline and the uterine body contour to obtain the second intersection point; and calculate the Euclidean distance between the first intersection point and the second intersection point to obtain the major axis length of the uterine body.

[0127] In one embodiment, the first measurement module 1906 is further configured to determine the long diameter of the uterine body based on the first intersection point and the second intersection point; draw a perpendicular line to the long diameter of the uterine body based on the long diameter of the uterine body, determine the longest perpendicular line formed by the perpendicular line to the long diameter of the uterine body and the outline of the uterine body, and obtain the anteroposterior diameter of the uterine body; and calculate the length of the anteroposterior diameter of the uterine body.

[0128] In one embodiment, the ultrasound cross-sectional image of the uterine adnexa includes a transverse cross-sectional image of the uterus; the segmentation module 1904 is further used to perform feature segmentation on the transverse cross-sectional image of the uterus to obtain the outline of the uterine body and the outline of the endometrium; the first measurement module 1906 is further used to extract the skeleton line of the endometrial outline in the outline of the uterine adnexa to obtain the endometrial skeleton line; based on the endometrial skeleton line, a horizontal line parallel to the endometrial skeleton line is generated; the longest straight line formed by the horizontal line and the outline of the uterine body is determined to obtain the transverse diameter of the uterine body; and the length of the transverse diameter of the uterine body is calculated.

[0129] In one embodiment, the ultrasound cross-sectional image of the uterus and adnexa includes an ovarian cross-sectional image; the segmentation module 1904 is further used to perform feature segmentation on the ovarian cross-sectional image to obtain the ovarian contour and the contour of the largest follicle; the second measurement module 1908 is further used to perform distance calculation based on the ovarian contour to determine the longest distance between the ovarian contours and obtain the ovarian long axis and the length of the ovarian long axis; and to perform distance calculation based on the contour of the largest follicle to determine the longest distance between the contours of the largest follicles and obtain the long axis of the largest follicle and the length of the largest follicle.

[0130] In one embodiment, the second measurement module 1908 is further configured to: draw a perpendicular line from the ovarian long axis to the ovarian long axis; determine the longest perpendicular line from the ovarian long axis formed by the perpendicular line from the ovarian long axis and the ovarian contour; obtain the ovarian short axis; and calculate the length of the ovarian short axis. Additionally, it is configured to: draw a perpendicular line from the long axis of the largest follicle to the follicle long axis; determine the longest perpendicular line from the follicle long axis formed by the perpendicular line from the largest follicle long axis and the follicle contour; obtain the short axis of the largest follicle; and calculate the length of the short axis of the largest follicle.

[0131] In one embodiment, the ultrasound cross-sectional image of the uterine adnexa includes a longitudinal cross-sectional image of the cervix; the segmentation module 1904 is further used to perform feature segmentation on the longitudinal cross-sectional image of the cervix to obtain the cervical contour and the cervical endometrial contour; the second measurement module 1908 is further used to determine the internal cervical os and the external cervical os based on the cervical endometrial contour; determine the cervical major diameter and the length of the cervical major diameter based on the internal cervical os and the external cervical os; draw a perpendicular line to the cervical major diameter based on the cervical major diameter, determine the longest perpendicular line to the cervical major diameter formed by the perpendicular line to the cervical contour, obtain the anteroposterior diameter of the cervix, and calculate the length of the anteroposterior diameter of the cervix.

[0132] In one embodiment, the first measurement module 1906 is further configured to acquire a set of uterine adnexal contour points; generate a uterine adnexal contour mask based on the set of uterine adnexal contour points; and perform distance transformation and inversion operations on the uterine adnexal contour mask to obtain the uterine adnexal skeleton line.

[0133] Each module in the aforementioned uterine adnexa ultrasound image processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0134] In one embodiment, a uterine adnexa ultrasound image processing device is provided. This device can be a terminal, and its internal structure diagram can be as follows: Figure 20As shown, the uterine adnexa ultrasound image processing device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a uterine adnexa ultrasound image processing method. The display screen of the uterine adnexa ultrasound image processing device can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad located on the outer casing of the device, or an external keyboard, touchpad, or mouse.

[0135] Those skilled in the art will understand that Figure 20 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the uterine adnexal ultrasound image processing device to which the present application is applied. A specific uterine adnexal ultrasound image processing device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0136] In one embodiment, a uterine adnexa ultrasound image processing device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0137] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0138] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0139] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0140] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0141] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0142] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for processing ultrasound images of the uterus and adnexa, characterized in that, The method includes: Acquire ultrasound cross-sectional images of the uterus and adnexa; the ultrasound cross-sectional images of the uterus and adnexa include longitudinal cross-sectional images of the uterus; The ultrasound cross-sectional image of the uterus and adnexa is segmented to obtain the outline of the uterus and adnexa; If the ultrasound cross-sectional image of the uterine adnexa is a first type image, then the skeleton line of the uterine adnexa is extracted, and measurements are performed based on the extracted skeleton line and the uterine adnexa contour. If the ultrasound cross-sectional image of the uterus and adnexa is a type II image, then measurements are performed based on the outline of the uterus and adnexa. The first type of image includes a longitudinal section image of the uterus and a transverse section image of the uterus, and the second type of image includes a section image of the ovary and a longitudinal section image of the cervix. The step of segmenting the ultrasound cross-sectional image of the uterus and adnexa to obtain the contour of the uterus and adnexa includes: The longitudinal section image of the uterus is segmented to obtain the outline of the uterine body and the outline of the endometrium. Extracting the skeleton lines of the uterine adnexa and performing measurements based on the extracted skeleton lines and the uterine adnexa contour includes: The endometrial skeleton line is extracted from the uterine adnexa contour to obtain the endometrial skeleton line. Determine the largest pixel in the image corresponding to the endometrial skeleton line; The largest pixel is extended along a direction perpendicular to the endometrial skeleton line to obtain the intersection of the extended line and the endometrial contour. The endometrial thickness is obtained by calculating the Euclidean distance between the intersection of the extended line and the endometrial contour. The endometrial centerline is obtained by fitting the endometrial skeleton line. Obtain the intersection point of the uterine body contour and the endometrial contour to obtain the first intersection point; The intersection point of the endometrial centerline and the outline of the uterine body is determined to obtain the second intersection point; Calculate the Euclidean distance between the first intersection point and the second intersection point to obtain the major axis length of the uterine body.

2. The method according to claim 1, characterized in that, Also includes: Determine the long diameter of the uterine body based on the first intersection point and the second intersection point; Based on the long diameter of the uterine body, draw a perpendicular line to the long diameter of the uterine body, determine the longest perpendicular line formed by the perpendicular line to the long diameter of the uterine body and the outline of the uterine body, and obtain the anteroposterior diameter of the uterine body. Calculate the length of the anteroposterior diameter of the uterine body.

3. The method according to claim 1, characterized in that, The ultrasound images of the uterine adnexa include cross-sectional images of the uterus; The step of segmenting the ultrasound cross-sectional image of the uterus and adnexa to obtain the contour of the uterus and adnexa includes: Feature segmentation is performed on the cross-sectional image of the uterus to obtain the outline of the uterine body and the outline of the endometrium; Extracting the skeleton lines of the uterine adnexa and performing measurements based on the extracted skeleton lines and the uterine adnexa contour includes: The endometrial skeleton line is extracted from the uterine adnexa contour to obtain the endometrial skeleton line. Based on the endometrial skeleton line, a horizontal line parallel to the endometrial skeleton line is generated; The longest straight line formed by the horizontal line and the outline of the uterine body is determined to obtain the transverse diameter of the uterine body. Calculate the length of the transverse diameter of the uterine body.

4. The method according to claim 1, characterized in that, The ultrasound cross-sectional images of the uterus and adnexa include cross-sectional images of the ovaries; The step of segmenting the ultrasound cross-sectional image of the uterus and adnexa to obtain the contour of the uterus and adnexa includes: Feature segmentation was performed on the ovarian cross-sectional image to obtain the ovarian contour and the contour of the largest follicle; Measurements based on the described uterine adnexal contours include: Based on the ovarian contours, distance calculations are performed to determine the longest distance between the ovarian contours, thereby obtaining the ovarian major axis and the length of the ovarian major axis. Distance calculations are performed based on the maximum follicle contours to determine the longest distance between the maximum follicle contours, thereby obtaining the maximum follicle major axis and the maximum follicle major axis length.

5. The method according to claim 4, characterized in that, Also includes: Based on the ovarian long axis, draw a perpendicular line to the ovarian long axis, determine the longest perpendicular line to the ovarian long axis formed by the perpendicular line to the ovarian contour, obtain the ovarian short axis, and calculate the length of the ovarian short axis. Based on the longest diameter of the largest follicle, draw a perpendicular line to the longest diameter of the largest follicle, determine the longest perpendicular line to the longest diameter of the follicle formed by the perpendicular line to the follicle outline, obtain the shortest diameter of the largest follicle, and calculate the length of the shortest diameter of the largest follicle.

6. The method according to claim 1, characterized in that, The ultrasound cross-sectional images of the uterine adnexa include longitudinal cross-sectional images of the cervix; The step of segmenting the ultrasound cross-sectional image of the uterus and adnexa to obtain the contour of the uterus and adnexa includes: Feature segmentation is performed on the longitudinal section image of the cervix to obtain the cervical contour and the cervical endometrial contour. Measurements based on the described uterine adnexal contours include: Based on the cervical endometrial contour, the internal cervical os and external cervical os are determined. Based on the internal cervical os and the external cervical os, determine the cervical long diameter line and the cervical long diameter length. Based on the cervical long diameter, draw a perpendicular line to the cervical long diameter, determine the longest perpendicular line to the cervical long diameter formed by the perpendicular line to the cervical outline, obtain the anteroposterior diameter of the cervix, and calculate the length of the anteroposterior diameter of the cervix.

7. The method according to claim 1, characterized in that, Extracting the skeleton lines of the uterine adnexa and performing measurements based on the extracted skeleton lines and the uterine adnexa contour includes: Obtain the set of uterine adnexa contour points of the uterine adnexa contour; Generate a mask image of the uterine adnexa based on the set of uterine adnexa contour points; The uterine adnexa contour mask is subjected to distance transformation and inversion operations to obtain the uterine adnexa skeleton line.

8. A device for processing ultrasound images of the uterus and adnexa, characterized in that, The device includes: The acquisition module is used to acquire ultrasound cross-sectional images of the uterus and adnexa; the ultrasound cross-sectional images of the uterus and adnexa include longitudinal cross-sectional images of the uterus; The segmentation module is used to perform feature segmentation on the ultrasound cross-sectional image of the uterine adnexa to obtain the outline of the uterine adnexa. The first measurement module is used to extract the skeleton line of the uterine adnexa if the type of the uterine adnexa is the first type, and to perform measurement based on the extracted skeleton line and the uterine adnexa contour to obtain the measurement data of the first type of uterine adnexa. The second measurement module is used to measure the outline of the uterine adnexa if the type of the uterine adnexa is the second type, and to obtain the measurement data of the second type of uterine adnexa. The first type of uterine adnexa includes a longitudinal section of the uterus and a transverse section of the uterus, while the second type of uterine adnexa includes a section of the ovary and a longitudinal section of the cervix. The segmentation module is also used to perform feature segmentation on the longitudinal section image of the uterus to obtain the outline of the uterine body and the outline of the endometrium. The first measurement module is further configured to extract the skeleton line of the endometrial contour in the uterine adnexa contour to obtain the endometrial skeleton line; determine the largest pixel in the image corresponding to the endometrial skeleton line; extend the largest pixel along a direction perpendicular to the endometrial skeleton line to obtain the intersection of the extended line and the endometrial contour; calculate the Euclidean distance between the intersection of the extended line and the endometrial contour to obtain the endometrial thickness; perform fitting processing based on the endometrial skeleton line to obtain the endometrial centerline; obtain the intersection of the uterine body contour and the endometrial contour to obtain the first intersection point; determine the intersection of the endometrial centerline and the uterine body contour to obtain the second intersection point; calculate the Euclidean distance between the first intersection point and the second intersection point to obtain the major axis length of the uterine body.

9. The apparatus according to claim 8, characterized in that, The first measurement module is also used for: Determine the long diameter of the uterine body based on the first intersection point and the second intersection point; Based on the long diameter of the uterine body, draw a perpendicular line to the long diameter of the uterine body, determine the longest perpendicular line formed by the perpendicular line to the long diameter of the uterine body and the outline of the uterine body, and obtain the anteroposterior diameter of the uterine body. Calculate the length of the anteroposterior diameter of the uterine body.

10. The apparatus according to claim 8, characterized in that, The ultrasound images of the uterine adnexa include cross-sectional images of the uterus; The segmentation module is also used to: perform feature segmentation on the cross-sectional image of the uterus to obtain the outline of the uterine body and the outline of the endometrium; The first measurement module is further configured to: extract the skeleton line of the endometrial contour in the uterine adnexa contour to obtain the endometrial skeleton line; Based on the endometrial skeleton line, a horizontal line parallel to the endometrial skeleton line is generated; The longest straight line formed by the horizontal line and the outline of the uterine body is determined to obtain the transverse diameter of the uterine body. Calculate the length of the transverse diameter of the uterine body.

Images