Woman uterus disease auxiliary diagnosis and evaluation system based on magnetic resonance imaging

Abstract

The invention relates to the technical field of uterus image segmentation, in particular to a woman uterus disease auxiliary diagnosis and evaluation system based on magnetic resonance imaging. After a uterus image of a patient is obtained and a uterine cavity area is positioned, clustering clusters are divided according to pixel gray values by using a clustering algorithm to correspond to different uterus areas. And generating an endometrial region descriptor and screening out an endometrial region by combining the pixel number distribution of each region, the gray features and the shape difference between the endometrial region descriptor and the uterine cavity. Performing gray scale segmentation on the region to extract a high-signal region, and calculating an internal descriptor according to internal edge distribution and morphological characteristics of the high-signal region; meanwhile, an external descriptor is generated based on edge pixel neighborhood gray scale difference. And finally fusing internal and external descriptors to realize auxiliary diagnosis of uterine diseases. According to the method, the uterus image with obvious uterine cancer staging development tendency can be screened out, and a more accurate diagnosis effect is provided for related personnel.

Description

Technical Field

[0001] This invention relates to the technical field of uterine image segmentation, and more specifically to an auxiliary diagnostic and assessment system for uterine diseases in women based on magnetic resonance imaging. Background Technology

[0002] In the field of diagnosing uterine diseases in women, medical images acquired through magnetic resonance imaging (MRI) are crucial for clinical diagnosis. As a non-invasive and high-resolution imaging technique, MRI can clearly present the anatomical details and pathological changes of the uterus and surrounding tissues. With the development of image processing technologies, the application of combining relevant algorithms with MRI for assisted diagnosis is becoming increasingly widespread. This combination further enhances the value of MRI in diagnosing uterine diseases. The combined algorithms can provide a clearer representation of key anatomical structures in MRI, improving image readability and lesion specificity.

[0003] Uterine cancer, as a type of uterine disease in women, has received considerable attention from professionals in the field. Determining the stage of uterine cancer is crucial for planning subsequent interventions and methods. However, current technologies, particularly traditional edge detection algorithms, cannot accurately segment the anatomical structures of the uterus in a pathological state of uterine cancer, hindering the analysis of uterine MRI images. Expected The pathological changes of uterine cancer in the early stages were misdiagnosed. Summary of the Invention

[0004] To address the issue that traditional edge detection algorithms cannot accurately segment the anatomical structures of the uterus in a pathological state of uterine cancer, which leads to the limitations of using uterine magnetic resonance images... Expected To address the technical problem of misdiagnosis in the pathological changes of uterine cancer during its early stages, this invention aims to provide a magnetic resonance imaging (MRI)-based auxiliary diagnostic and assessment system for uterine diseases in women. The specific technical solution is as follows: A magnetic resonance imaging (MRI)-based auxiliary diagnostic and assessment system for uterine diseases in women, comprising: an image acquisition module for acquiring images of the patient's uterus; a region description module for acquiring the uterine cavity region within the uterine image; using a clustering algorithm to cluster all pixels in the uterine image based on their grayscale values ​​to obtain all clusters; acquiring the uterine region corresponding to the pixels in each cluster; and based on the pixel quantity distribution and grayscale distribution within each uterine region, and the... The endometrial region descriptor for each uterine region is obtained by considering the shape feature differences between the endometrial and uterine cavity regions. All uterine regions are then filtered based on these endometrial region descriptors to obtain the endometrial regions. The endometrial regions are then segmented based on grayscale values ​​to obtain all high-signal regions within the endometrial regions. An internal descriptor for each high-signal region within the endometrial regions is obtained based on its edge distribution and shape features. An external descriptor for each edge pixel within the endometrial regions is obtained based on the grayscale differences within a preset neighborhood. An auxiliary diagnostic module is used to assist in the diagnosis of uterine diseases based on the internal and external descriptors of the endometrial regions.

[0005] Furthermore, the method for obtaining the endometrial region descriptor includes: obtaining the endometrial region descriptor according to the endometrial region descriptor calculation formula, the endometrial region descriptor calculation formula being as follows: ; In the formula, Endometrial region descriptors for each uterine region; This indicates the number of pixels in each uterine region; Represents the first of each uterine region The grayscale value of each pixel; The slope of the midline of the smallest circumscribed rectangle of the uterine cavity region; This represents the slope of the midline of the smallest bounding rectangle for each uterine region; Represents the absolute value function; This represents the normalization function.

[0006] Furthermore, the method for obtaining the endometrial region includes: among all uterine regions, the uterine region with the largest endometrial region descriptor is taken as the endometrial region.

[0007] Furthermore, the method for obtaining the high-signal region includes: performing threshold segmentation on the endometrial region, and taking all regions composed of pixels with gray values ​​greater than the gray value threshold as all high-signal regions in the endometrial region.

[0008] Further, the method for obtaining the internal descriptor includes: establishing a Cartesian coordinate system with the upper left corner of the uterine image as the origin; taking the edge pixel closest to the origin in each high-signal region as the proximal endpoint and the edge pixel farthest from the origin in each high-signal region as the distal endpoint; calculating the distance between the proximal endpoint and the distal endpoint as the extension length of the high-signal region; taking the line segment between the proximal endpoint and the distal endpoint as the extension line segment of the high-signal region; and obtaining the internal descriptor according to the internal descriptor calculation formula, which is shown below: ; In the formula, Internal descriptors representing the endometrial region; Indicates the number of high-signal regions in the endometrium; Indicates the first [unit / item] in the endometrial region The extension length of a high-signal region; Indicates the first [unit / item] in the endometrial region The number of edge pixels in a high-signal region; Indicates the first [unit / item] in the endometrial region The first high signal area The shortest distance from each edge pixel to the extended line segment.

[0009] Furthermore, the method for obtaining the external descriptor includes: obtaining the external descriptor according to the external descriptor calculation formula, the external descriptor calculation formula being as follows: ; In the formula, External descriptors representing the endometrial region; This indicates the number of edge pixels in the endometrial region; This indicates the number of other pixels within a preset neighborhood of each edge pixel in the endometrial region. Indicates the first [unit / item] in the endometrial region The grayscale value of each edge pixel; Indicates the first [unit / item] in the endometrial region Within the preset neighborhood of the nth edge pixel The grayscale values ​​of the other pixels; This represents a step function. If the result inside the parentheses is greater than 0, the function value is 1; if the result inside the parentheses is less than 0, the function value is 0. This represents an exponential function with the natural constant as its base.

[0010] A method for auxiliary diagnosis and assessment of uterine diseases in women based on magnetic resonance imaging (MRI) includes: acquiring a patient's uterine image; acquiring the uterine cavity region within the uterine image; using a clustering algorithm to cluster all pixels in the uterine image based on their grayscale values ​​to obtain all clusters; acquiring the uterine region corresponding to the pixels in each cluster; obtaining an endometrial region descriptor for each uterine region based on the pixel quantity distribution and grayscale distribution within each uterine region, as well as the shape feature differences between each uterine region and the uterine cavity region; filtering all uterine regions based on the endometrial region descriptors to obtain the endometrial region; segmenting the endometrial region based on grayscale values ​​to obtain all high-signal regions within the endometrial region; obtaining an internal descriptor for the endometrial region based on the edge distribution and shape features of each high-signal region within the endometrial region; obtaining an external descriptor for the endometrial region based on the grayscale differences within a preset neighborhood of each edge pixel within the endometrial region; and using the internal and external descriptors of the endometrial region for auxiliary diagnosis of uterine diseases.

[0011] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described magnetic resonance imaging-based auxiliary diagnostic and assessment system for uterine diseases in women.

[0012] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, performs the steps of the magnetic resonance imaging-based auxiliary diagnostic and assessment system for uterine diseases in women as described above.

[0013] This invention offers the following advantages: It acquires images of the patient's uterus; the initial FIGO staging of uterine cancer begins with the endometrial region. As the cancer progresses, the lesion spreads outward from the endometrial region, gradually occupying surrounding normal tissue and anatomical structures. Therefore, it is necessary to first determine the endometrial region. The difference between stage I and stage II lies in whether the lesion has diffused to the cervix. As the diffusion area expands, higher signal nodule regions appear in the endometrial region, and these nodules tend to extend towards the cervix. When high signal nodules extending towards the cervix are generated in the endometrial region, and the surrounding low signal nodule bands are weak, the tendency to transition from stage I to stage II is more pronounced. Therefore, the internal and external grayscale characteristics of the endometrial region are described. Based on the internal and external descriptors of the endometrial region, uterine diseases are diagnosed more effectively. This invention can screen uterine images with a clear tendency for uterine cancer staging, providing more accurate diagnostic results for relevant personnel. Attached Figure Description

[0014] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a block diagram of a magnetic resonance imaging-based auxiliary diagnostic and assessment system for uterine diseases in women, provided as an embodiment of the present invention. Detailed Implementation

[0016] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a magnetic resonance imaging-based auxiliary diagnostic and assessment system for uterine diseases in women proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0018] The following description, in conjunction with the accompanying drawings, details a specific scheme for an auxiliary diagnostic and assessment system for uterine diseases in women based on magnetic resonance imaging, provided by the present invention.

[0019] Please see Figure 1 This invention illustrates an embodiment of a magnetic resonance imaging-based auxiliary diagnostic and assessment system for uterine diseases in women. The system includes an image acquisition module 101, a region description module 102, and an auxiliary diagnostic module 103. The steps are as follows: Image acquisition module 101: Acquire images of the patient's uterus.

[0020] This invention is primarily applied to the scenario of assisting in the staging diagnosis of uterine cancer in patients. Therefore, the first step is to use specialized equipment to acquire the uterine images required for subsequent operations. The specific operation steps are as follows: First, the patient is asked to moderately fill their bladder to push aside the intestines and improve the display of pelvic organs. Then, the patient is asked to lie supine with their legs naturally straight and stable. A high-resolution pelvic phased array coil or body coil is used to ensure coverage of the entire pelvic region. First, the coronal and transverse positioning images are acquired. Then, the sagittal scanning plane is adjusted in the coronal plane to make it parallel to the long axis of the uterus, ensuring complete display of the uterus and adjacent structures. During the scanning process, the patient must remain still to avoid motion artifacts. After the scanning process is completed, MRI images of the patient's pelvic area are acquired.

[0021] Since the range of MRI pixel values ​​far exceeds the 0-255 range of a standard 8-bit grayscale image, normalization processing is required to map the original signal intensity to 256 gray levels. During the conversion process, the system retains the spatial resolution of the original data. In the final grayscale image, different tissues exhibit different gray levels according to the relaxation characteristics of their sequences, thus completing the conversion from the MRI image to the corresponding grayscale image, thereby obtaining the uterine image required for subsequent operations.

[0022] Region Description Module 102: Acquires the uterine cavity region from the uterine image; uses a clustering algorithm to cluster all pixels in the uterine image based on their grayscale values ​​to obtain all clusters; acquires the uterine region corresponding to the pixels in each cluster; obtains the endometrial region descriptor for each uterine region based on the pixel quantity distribution and grayscale distribution within each uterine region, as well as the shape feature differences between each uterine region and the uterine cavity region; filters all uterine regions based on the endometrial region descriptors to obtain the endometrial region; segments the endometrial region based on grayscale values ​​to obtain all high-signal regions within the endometrial region; obtains the internal descriptor of the endometrial region based on the edge distribution and shape features of each high-signal region within the endometrial region; obtains the external descriptor of the endometrial region based on the grayscale differences within the preset neighborhood of each edge pixel within the endometrial region.

[0023] Because the uterus's anatomical structure includes the innermost endometrium, followed by the junctional zone, myometrium, and serosa, these regions appear as follows in uterine images: the endometrium shows high signal intensity, the junctional zone shows low signal intensity, the myometrium shows intermediate signal intensity, and the serosa shows low signal intensity. The FIGO staging of uterine cancer begins with determining the location of the endometrium. As the cancer progresses, the lesion spreads outward from the endometrial region, gradually encroaching on surrounding normal tissue and anatomical structures. Therefore, it is necessary to first determine the endometrial region. Clustering algorithms are better adapted to regions with uneven gray levels, thus automatically identifying regions of different tissues. Therefore, in this embodiment of the invention, clustering is employed. Clustering algorithm.

[0024] Since the uterus is shaped like an inverted pear in the sagittal plane, the uterine cavity region in the uterine image can be obtained based on the physiological shape characteristics of the uterus.

[0025] because Clustering algorithms are better suited to regions with uneven gray levels, thus automatically identifying regions of different tissues. Therefore, all pixels in the uterine image are clustered to obtain all clusters, and the region containing the pixels in each cluster is considered as a uterine region. It should be noted that... Clustering algorithms are well-known techniques in the field and are not limited here.

[0026] Preferably, in one embodiment of the present invention, the method for obtaining the endometrial region descriptor includes: obtaining the endometrial region descriptor according to the endometrial region descriptor calculation formula, wherein the endometrial region descriptor calculation formula is as follows: ; In the formula, Endometrial region descriptors for each uterine region; This indicates the number of pixels in each uterine region; Represents the first of each uterine region The grayscale value of each pixel; The slope of the midline of the smallest circumscribed rectangle of the uterine cavity region; This represents the slope of the midline of the smallest bounding rectangle for each uterine region; Represents the absolute value function; This represents the normalization function.

[0027] In the endometrial region descriptor calculation formula, under normal conditions, the endometrium naturally follows the uterine cavity, slightly wider at the fundus and gradually narrowing towards the internal cervical os. However, the occurrence of endometrial cancer causes abnormal thickening in the area where the endometrium was originally located. Visually, this appears as a diffusely expanded region, but its signal strength remains close to that of the normal endometrial region, thus it tends to be a high-signal area. High signal strength corresponds to higher pixel grayscale values ​​in a grayscale image. Therefore, the average grayscale value of pixels in the uterine region is... The higher the number of pixels in the uterine region The more [a certain number of lines], the more likely the uterine region is to be an endometrial region. Because the endometrium is tightly attached to the inner wall of the uterine cavity, its growth and development are restricted by the morphology and structure of the uterine cavity. Therefore, the orientation of the endometrial region is similar to the edge shape of the uterine cavity; that is, the closer the slope of the midline of the smallest circumscribed rectangle of the uterine cavity region is to the slope of the midline of the smallest circumscribed rectangle of the uterine region, the more likely it is to be an endometrial region. The smaller the area, the more likely it is to be the endometrial region.

[0028] It should be noted that the methods for obtaining the midline slope of the minimum bounding rectangle of the uterine cavity region and the midline slope of the minimum bounding rectangle of the uterine region can be directly obtained by existing technologies, and will not be elaborated here.

[0029] Preferably, in one embodiment of the present invention, the method for obtaining the endometrial region includes: among all uterine regions, the uterine region with the largest endometrial region descriptor is taken as the endometrial region.

[0030] In the FIGO staging system, the difference between stage I and stage II lies in whether the lesion has diffused to the cervix. As the diffuse area expands, nodular areas with higher signal intensity appear in the endometrial region, and these nodular areas tend to extend towards the cervical os. This is because the endometrium and cervical canal are connected through the isthmus of the uterus, forming a natural cavity from the uterine cavity to the cervix. The cervical stroma is rich in blood vessels and lymphatic vessels, providing a physical channel for lesion spread. Furthermore, the pressure in the uterine cavity is higher than that in the cervix, making it easier for lesion cells to flow towards the cervical os with secretions or tissue fluid. In addition, the junctional zone surrounding the endometrial region becomes increasingly weak due to the influence of lesion diffusion. Therefore, when high-signal nodules extending towards the cervix are generated in the endometrial region, and the surrounding low-signal nodular zone is weak, the tendency to transition from stage I to stage II is more pronounced. Therefore, in this embodiment of the invention, the internal and external grayscale characteristics of the endometrial region are described.

[0031] Preferably, in one embodiment of the present invention, the method for obtaining high-signal regions includes: performing threshold segmentation on the endometrial region, and identifying all regions composed of pixels with gray values ​​greater than a gray value threshold as all high-signal regions within the endometrial region. It should be noted that the threshold segmentation algorithm is a well-known technique among those skilled in the art and will not be elaborated upon here.

[0032] Preferably, in one embodiment of the present invention, the method for obtaining internal descriptors includes: establishing a Cartesian coordinate system with the upper left corner of the uterine image as the origin, taking the edge pixel closest to the origin in each high-signal region as the proximal endpoint, and taking the edge pixel farthest from the origin in each high-signal region as the farthest endpoint.

[0033] Calculate the distance between the proximal endpoint and the distal endpoint as the extension length of the high-signal region; use the line segment between the proximal endpoint and the distal endpoint as the extension line segment of the high-signal region.

[0034] The internal descriptor is obtained based on the internal descriptor calculation formula, which is shown below: ; In the formula, Internal descriptors representing the endometrial region; Indicates the number of high-signal regions in the endometrium; Indicates the first [unit / item] in the endometrial region The extension length of a high-signal region; Indicates the first [unit / item] in the endometrial region The number of edge pixels in a high-signal region; Indicates the first [unit / item] in the endometrial region The first high signal area The shortest distance from each edge pixel to the extended line segment.

[0035] In the internal descriptor calculation formula, the length of the extended line segment The longer the line, the greater the average distance between all edge pixels and the extended line segment. The shorter the length, the more obvious the aspect ratio of the high-signal region, and the more obvious the extension of the high-signal region. In this case, the high-signal region is more likely to be a nodule region. Analyzing each high-signal region, the aspect ratio of the high-signal region in each region is relatively obvious, indicating that the internal descriptor of the endometrial region is larger.

[0036] Preferably, in one embodiment of the present invention, the method for obtaining an external descriptor includes: obtaining an external descriptor according to an external descriptor calculation formula, wherein the external descriptor calculation formula is as follows: ; In the formula, External descriptors representing the endometrial region; This indicates the number of edge pixels in the endometrial region; This indicates the number of other pixels within a preset neighborhood of each edge pixel in the endometrial region. Indicates the first [unit / item] in the endometrial region The grayscale value of each edge pixel; Indicates the first [unit / item] in the endometrial region Within the preset neighborhood of the nth edge pixel The grayscale values ​​of the other pixels; This represents a step function. If the result inside the parentheses is greater than 0, the function value is 1; if the result inside the parentheses is less than 0, the function value is 0. This represents an exponential function with the natural constant as its base.

[0037] In the external descriptor calculation formula, under normal anatomical conditions, the junctional zone is a low-signal tissue surrounding the endometrial region. However, as the lesion spreads, the junctional zone's coverage of the endometrial region becomes increasingly weak. Therefore, the difference between pixels in the neighborhood of the edge pixel is compared, i.e., the calculation... When the gray value of an edge pixel is less than its own and its gray value is close to the edge pixel's gray value, The more pixels there are, the more likely it is that the number of pixels of this type can be analyzed using a step function, and the grayscale value is smaller than itself and closer to the grayscale value. The more pixels there are, the more likely the spread of the lesion in the endometrial region has not weakened the signal of the binding zone, and the larger the external descriptor of the endometrial region will be.

[0038] Assisted Diagnosis Module 103: Provides assisted diagnosis of uterine diseases based on the internal and external descriptors of the endometrial region.

[0039] In one embodiment of the present invention, a larger internal descriptor of the endometrial region indicates a more pronounced staging development within the endometrial region, and a larger external descriptor indicates a more pronounced staging development outside the endometrial region. Therefore, the internal and external descriptors are added together, and this sum is used as the first sum. This first sum is then used as the tendency of the endometrial region's staging development in the uterine image, and the tendency of all uterine images is obtained. The maximum value of the tendency of all uterine images is used as the judgment value. All uterine images whose difference from the judgment value is less than the judgment threshold are screened out as images to be analyzed. Clinical assessment is performed using these images to quickly locate the staging development of uterine cancer and complete the auxiliary diagnostic work.

[0040] It should be noted that the judgment threshold is set to 10%, and can be set by the user; no restrictions are imposed here.

[0041] In summary, the following steps are taken: 1) Obtain images of the patient's uterus; 2) Extract the uterine cavity region from the uterine image; 3) Utilize a clustering algorithm to cluster all pixels in the uterine image based on their grayscale values ​​to obtain all clusters; 4) Obtain the uterine region corresponding to the pixels in each cluster; 5) Obtain the endometrial region descriptor for each uterine region based on the pixel quantity distribution and grayscale distribution within each uterine region, as well as the shape feature differences between each uterine region and the uterine cavity region; 6) Filter all uterine regions based on the endometrial region descriptors to obtain the endometrial region; 7) Segment the endometrial region based on grayscale values ​​to obtain all high-signal regions within the endometrial region; 8) Obtain the internal descriptor of the endometrial region based on the edge distribution and shape features of each high-signal region within the endometrial region; 9) Obtain the external descriptor of the endometrial region based on the grayscale differences within a preset neighborhood of each edge pixel within the endometrial region; 10) Use the internal and external descriptors of the endometrial region to assist in the diagnosis of uterine diseases.

[0042] A second objective of one embodiment of the present invention is to provide a method for auxiliary diagnosis and assessment of uterine diseases in women based on magnetic resonance imaging. The method includes: acquiring a patient's uterine image; acquiring the uterine cavity region within the uterine image; using a clustering algorithm to cluster all pixels in the uterine image according to their grayscale values ​​to obtain all clusters; acquiring the uterine region corresponding to the pixels in each cluster; obtaining an endometrial region descriptor for each uterine region based on the pixel quantity distribution and grayscale distribution within each uterine region, as well as the shape feature differences between each uterine region and the uterine cavity region; filtering all uterine regions based on the endometrial region descriptors to obtain the endometrial region; segmenting the endometrial region according to grayscale values ​​to obtain all high-signal regions within the endometrial region; obtaining an internal descriptor for the endometrial region based on the edge distribution and shape features of each high-signal region within the endometrial region; obtaining an external descriptor for the endometrial region based on the grayscale differences within a preset neighborhood of each edge pixel within the endometrial region; and using the internal and external descriptors of the endometrial region for auxiliary diagnosis of uterine diseases.

[0043] A third objective of this invention is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the system described in steps S1-S3.

[0044] The fourth objective of this invention is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the system described in steps S1-S3.

[0045] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0046] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A system for auxiliary diagnosis and evaluation of uterine diseases in women based on magnetic resonance imaging, characterized in that, The system includes: an image acquisition module for acquiring images of the patient's uterus; a region description module for acquiring the uterine cavity region in the uterine image; using a clustering algorithm to cluster all pixels in the uterine image based on their grayscale values ​​to obtain all clusters; acquiring the uterine region corresponding to the pixels in each cluster; obtaining an endometrial region descriptor for each uterine region based on the pixel quantity distribution and grayscale distribution within each uterine region, as well as the shape feature differences between each uterine region and the uterine cavity region; filtering all uterine regions based on the endometrial region descriptors to obtain the endometrial region; segmenting the endometrial region based on grayscale values ​​to obtain all high-signal regions within the endometrial region; obtaining an internal descriptor for the endometrial region based on the edge distribution and shape features of each high-signal region within the endometrial region; obtaining an external descriptor for the endometrial region based on the grayscale differences within a preset neighborhood of each edge pixel within the endometrial region; and an auxiliary diagnosis module for assisting in the diagnosis of uterine diseases based on the internal and external descriptors of the endometrial region.

2. The assisted diagnostic and assessment system for uterine diseases in women based on magnetic resonance imaging according to claim 1, characterized in that, The method for obtaining the endometrial region descriptor includes: obtaining the endometrial region descriptor according to the endometrial region descriptor calculation formula, wherein the endometrial region descriptor calculation formula is as follows: ; In the formula, Endometrial region descriptors for each uterine region; This indicates the number of pixels in each uterine region; Represents the first of each uterine region The grayscale value of each pixel; The slope of the midline of the smallest circumscribed rectangle of the uterine cavity region; This represents the slope of the midline of the smallest bounding rectangle for each uterine region; Represents the absolute value function; This represents the normalization function.

3. The auxiliary diagnostic and assessment system for uterine diseases in women based on magnetic resonance imaging according to claim 1, characterized in that, The method for obtaining the endometrial region includes: among all uterine regions, the uterine region with the largest endometrial region descriptor is taken as the endometrial region.

4. The assisted diagnostic and assessment system for uterine diseases in women based on magnetic resonance imaging according to claim 1, characterized in that, The method for obtaining the high-signal region includes: performing threshold segmentation on the endometrial region, and taking all regions composed of pixels with gray values ​​greater than the gray threshold as all high-signal regions in the endometrial region.

5. The auxiliary diagnostic and assessment system for uterine diseases in women based on magnetic resonance imaging according to claim 1, characterized in that, The method for obtaining the internal descriptor includes: establishing a Cartesian coordinate system with the upper left corner of the uterine image as the origin; taking the edge pixel closest to the origin in each high-signal region as the proximal endpoint and the edge pixel farthest from the origin in each high-signal region as the distal endpoint; calculating the distance between the proximal endpoint and the distal endpoint as the extension length of the high-signal region; taking the line segment between the proximal endpoint and the distal endpoint as the extension line segment of the high-signal region; and obtaining the internal descriptor according to the internal descriptor calculation formula, which is shown below: ； In the formula, Internal descriptors representing the endometrial region; Indicates the number of high-signal regions in the endometrium; Indicates the first [unit / item] in the endometrial region The extension length of a high-signal region; Indicates the first [unit / item] in the endometrial region The number of edge pixels in a high-signal region; Indicates the first [unit / item] in the endometrial region The first high signal area The shortest distance from each edge pixel to the extended line segment.

6. The auxiliary diagnostic and assessment system for uterine diseases in women based on magnetic resonance imaging according to claim 1, characterized in that, The method for obtaining the external descriptor includes: obtaining the external descriptor according to an external descriptor calculation formula, wherein the external descriptor calculation formula is as follows: ; In the formula, External descriptors representing the endometrial region; This indicates the number of edge pixels in the endometrial region; This indicates the number of other pixels within a preset neighborhood of each edge pixel in the endometrial region. Indicates the first [unit / item] in the endometrial region The grayscale value of each edge pixel; Indicates the first [unit / item] in the endometrial region Within the preset neighborhood of the nth edge pixel The grayscale values ​​of the other pixels; This represents a step function. If the result inside the parentheses is greater than 0, the function value is 1; if the result inside the parentheses is less than 0, the function value is 0. This represents an exponential function with the natural constant as its base.

7. A method for auxiliary diagnosis and evaluation of uterine diseases in women based on magnetic resonance imaging, characterized in that, The method includes: acquiring a patient's uterine image; acquiring the uterine cavity region within the uterine image; using a clustering algorithm to cluster all pixels in the uterine image based on their grayscale values ​​to obtain all clusters; acquiring the uterine region corresponding to the pixels in each cluster; obtaining an endometrial region descriptor for each uterine region based on the pixel quantity distribution and grayscale distribution within each uterine region, as well as the shape feature differences between each uterine region and the uterine cavity region; filtering all uterine regions based on the endometrial region descriptors to obtain the endometrial regions; segmenting the endometrial regions based on their grayscale values ​​to obtain all high-signal regions within the endometrial regions; obtaining an internal descriptor for the endometrial regions based on the edge distribution and shape features of each high-signal region within the endometrial regions; obtaining an external descriptor for the endometrial regions based on the grayscale differences within a preset neighborhood of each edge pixel within the endometrial regions; and using the internal and external descriptors of the endometrial regions for auxiliary diagnosis of uterine diseases.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the magnetic resonance imaging-based auxiliary diagnostic and assessment system for uterine diseases in women as described in any one of claims 1 to 6.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the magnetic resonance imaging-based auxiliary diagnosis and evaluation system for uterine diseases in women as described in any one of claims 1 to 6.

Images