Hysteroscope uterine angle visual reference finite element analysis method and system

Abstract

The invention provides a hysteroscope uterine angle visual reference finite element analysis method and system, and relates to the technical field of hysteroscopes, and the method comprises the steps: obtaining image data in the uterus through a hysteroscope, and carrying out the graying processing of a color image in the image data in the uterus, and obtaining a gray image; performing edge detection on the grayscale image, identifying edge contours of the uterus and the uterine horn, and extracting uterine horn point candidate positions according to geometric features of the uterine horn, including curvature change, angle sharpness or specific positions on the edge contours; screening and verifying the candidate positions, and determining a final uterine angular point position; and establishing a geometric coordinate system corresponding to the image coordinate system by taking the final uterine angular point position as a reference, and automatically identifying and determining two uterine angular point coordinates in the uterus. According to the method, the uterus image is obtained through the hysteroscope, the uterus angular point is identified through processing, the geometric model is established, finite element analysis is executed, and the uterus structural characteristics are accurately evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of hysteroscopy, and in particular to a hysteroscopy uterine angle visual reference finite element analysis method and system. Background Art

[0002] Some traditional hysteroscopic image processing methods ignore the rich detail information in color images during grayscale processing, which leads to a decrease in the quality of grayscale images and affects the accuracy of subsequent edge detection and uterine corner point recognition.

[0003] Secondly, some existing edge detection technologies have limited ability to recognize the edge contours of the uterus and uterine cornua, especially when faced with complex internal uterine structures and image noise. Some of them find it difficult to accurately extract candidate locations of uterine cornua points, which not only increases the difficulty of subsequent screening and verification work, but may also introduce errors and affect the determination of the final uterine cornua point location. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a hysteroscopic uterine angle visual benchmark finite element analysis method and system, reduce the interference of human factors, and improve the accuracy and efficiency of evaluation.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows: In a first aspect, a hysteroscopic uterine angle visual benchmark finite element analysis method is provided, the method comprising: Acquiring internal image data of the uterus by using a hysteroscope, and graying a color image in the internal image data of the uterus to obtain a gray image; Perform edge detection on the grayscale image to identify the edge contours of the uterus and uterine horns, and extract candidate positions of uterine horn points based on the geometric features of the uterine horns, including curvature changes, sharp angles, or specific positions on the edge contours; Screen and verify the candidate positions to determine the final palace corner point position; Taking the final uterine horn point position as a reference, a geometric coordinate system corresponding to the image coordinate system is established to automatically identify and determine the coordinates of the two uterine horn points inside the uterus; Taking the palace corner point coordinates and the length of the horizontal bar as the basis, calculate the actual distance value between the two palace corner points; According to the coordinates of the uterine horn points, the actual distance between the two uterine horn points and the internal image data of the uterus, a geometric model of the uterus and the uterine horn is constructed; Based on the geometric model of the uterus and uterine horns, finite element analysis was performed to evaluate the stress distribution, strain state, and deformation degree of the uterine structure.

[0006] Furthermore, edge detection is performed on the grayscale image to identify the edge contours of the uterus and uterine horns, and candidate positions of uterine horns are extracted based on the geometric features of the uterine horns, including curvature changes, sharp angles, or specific positions on the edge contours, including: Perform edge detection on the grayscale image to identify the edges of the uterus and uterine cornu to obtain edge detection results; According to the edge detection results, the contours of the uterus and uterine horns are extracted; Traversing the contours of the uterus and uterine horns, and setting judgment conditions according to the geometric features of the uterine horns, including curvature changes, sharp angles, or specific positions on the contours, to obtain an image marked with candidate positions of uterine horn points and a list of candidate positions; Each point in the candidate position list is marked again to obtain a final marked image, and the candidate position of the palace corner point is extracted from the final marked image.

[0007] Further, edge detection is performed on the grayscale image to identify the edges of the uterus and uterine horns to obtain edge detection results, including: Traverse each pixel in the grayscale image, calculate the grayscale value change of the pixels around each pixel, and generate the grayscale value change information of each pixel; According to the preset threshold, determine whether each pixel is an edge point, and obtain the edge point determination result; A binary image is generated according to the edge point judgment result, in which white pixels represent detected edge points and black pixels represent non-edge areas, and the binary image is used as the edge detection result.

[0008] Furthermore, the actual distance between two palace corner points is calculated as follows: The actual distance between two palace corner points is calculated as follows: ; in, Indicates the actual distance between two palace corner points; , It represents the horizontal coordinate of the palace corner point in the geometric coordinate system; , It represents the ordinate of the palace corner point in the geometric coordinate system; , Respectively expressed in Direction and Scaling factor in direction; represents the overall scaling factor; , Respectively expressed in Direction and The offset in direction; , Respectively expressed in Direction and Noise term in the direction.

[0009] Furthermore, according to the uterine horn point coordinates, the actual distance value between the two uterine horn points and the internal image data of the uterus, a geometric model of the uterus and the uterine horn is constructed, including: According to the coordinates of the uterine horn points and the actual distance between the two uterine horn points, the geometric shapes of the uterus and uterine horns, including lines and arcs, are constructed; Combine the geometric shapes of the uterus and uterine horns, including lines and arcs, to construct the initial geometric outlines of the uterus and uterine horns; Adjusting the shape and size of the initial geometric contours of the uterus and uterine horns to obtain the adjusted geometric contours of the uterus and uterine horns; According to the adjusted geometric contours of the uterus and uterine horns, a geometric model of the uterus and uterine horns is constructed.

[0010] Furthermore, based on the geometric model of the uterus and uterine horns, finite element analysis was performed to evaluate the stress distribution, strain state, and deformation degree of the uterine structure, including: Define material properties for uterine tissue, including elastic modulus and Poisson's ratio parameters; Meshing the geometric model of the uterus and uterine horns, dividing the geometric model of the uterus and uterine horns into multiple small, interconnected units; Set boundary conditions and apply loads to the geometric model of the uterus and uterine horns, including simulating pressure changes or external forces in the uterus; The uterus and uterine horn geometric model is subjected to finite element analysis by the solver. During the analysis, the solver calculates the stress, strain and deformation of each unit according to the set boundary conditions and loads as well as the material properties of the uterine tissue to obtain finite element analysis result data, including stress distribution diagram, strain state diagram and deformation degree diagram; The stress distribution diagram, strain state diagram and deformation degree diagram were analyzed to evaluate the stress distribution, strain state and deformation degree of the uterine structure.

[0011] Furthermore, the stress distribution diagram, strain state diagram and deformation degree diagram are analyzed to evaluate the stress distribution, strain state and deformation degree of the uterine structure, including: Analyze the stress distribution diagram, strain state diagram and deformation degree diagram, identify high stress areas and low stress areas, stress gradient changes, strain and deformation direction, and obtain the analysis results of stress distribution, strain state and deformation degree diagram; Based on the analysis results of the stress distribution, strain state and deformation degree diagrams, a comprehensive assessment of the uterine structure is conducted to identify problem areas, including stress concentration, excessive strain or abnormal deformation, in order to assess the stress distribution, strain state and deformation degree of the uterine structure.

[0012] In a second aspect, a hysteroscopic uterine angle visual reference finite element analysis system comprises: The acquisition module is used to acquire the internal image data of the uterus by using the hysteroscope, and grayscale the color image in the internal image data of the uterus to obtain a grayscale image; perform edge detection on the grayscale image to identify the edge contours of the uterus and the uterine horns, and extract the candidate positions of the uterine horns according to the geometric features of the uterine horns, including curvature changes, sharp angles or specific positions on the edge contours; screen and verify the candidate positions to determine the final uterine horn point positions; The processing module is used to establish a geometric coordinate system corresponding to the image coordinate system based on the final uterine horn point position, automatically identify and determine the coordinates of the two uterine horn points inside the uterus; use the uterine horn point coordinates and the crossbar length as a reference to calculate the actual distance value between the two uterine horn points; construct a geometric model of the uterus and uterine horns based on the uterine horn point coordinates, the actual distance value between the two uterine horn points and the internal image data of the uterus; perform finite element analysis based on the geometric model of the uterus and uterine horns to evaluate the stress distribution, strain state and deformation degree of the uterine structure.

[0013] According to a third aspect, a computing device includes: one or more processors; The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method described.

[0014] In a fourth aspect, a computer-readable storage medium stores a program, and when the program is executed by a processor, the method described is implemented.

[0015] The above solution of the present invention includes at least the following beneficial effects: By obtaining the internal image data of the uterus through hysteroscopy and performing image processing techniques such as grayscale processing and edge detection, the edge contours of the uterus and uterine horns can be more accurately identified, thereby accurately extracting the position of the uterine horns. This helps to more accurately understand the morphology and structure of the uterus and improve the accuracy of diagnosis. It can automatically identify and determine the coordinates of the two uterine horns inside the uterus, and calculate the actual distance value between the two uterine horns. This realizes automated measurement, reduces the interference of human factors, and improves the accuracy and efficiency of measurement. At the same time, based on these coordinates and distance values, the geometric model of the uterus and uterine horns can be automatically constructed, providing a basis for subsequent finite element analysis.

[0016] Finite element analysis can be used to comprehensively evaluate the stress distribution, strain state, and degree of deformation of the uterine structure. This helps to gain a deeper understanding of the mechanical properties of the uterus under different physiological conditions. Accurate geometric models of the uterus and uterine horns and finite element analysis results can provide important references for surgical planning and navigation. Based on these results, doctors can develop more accurate surgical plans, reduce surgical risks, and improve surgical success rates. By processing and analyzing a large amount of uterine image data, the changing patterns of uterine morphology can be revealed, providing new perspectives and methods for uterine-related research. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of a hysteroscopic uterine angle visual benchmark finite element analysis method provided by an embodiment of the present invention.

[0018] Figure 2 It is a schematic diagram of a hysteroscopic uterine angle visual reference finite element analysis system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0020] like Figure 1 As shown, an embodiment of the present invention provides a hysteroscopic uterine angle visual benchmark finite element analysis method, the method comprising the following steps: Step 11, using a hysteroscope to obtain image data of the interior of the uterus, and gray-scale processing is performed on the color image in the image data of the interior of the uterus to obtain a gray-scale image; Step 12, performing edge detection on the grayscale image to identify the edge contours of the uterus and uterine horns, and extracting candidate positions of uterine horn points based on geometric features of the uterine horns, including curvature changes, sharp angles, or specific positions on the edge contours; Step 13, screening and verifying the candidate positions to determine the final palace corner point position; Step 14, using the final uterine horn point position as a reference, establishing a geometric coordinate system corresponding to the image coordinate system, and automatically identifying and determining the coordinates of the two uterine horn points inside the uterus; Step 15, using the palace corner point coordinates and the crossbar length as a reference, calculate the actual distance value between the two palace corner points; Step 16, constructing a geometric model of the uterus and uterine horns according to the uterine horn point coordinates, the actual distance value between the two uterine horn points and the internal image data of the uterus; Step 17, performing finite element analysis based on the geometric model of the uterus and uterine horns to evaluate the stress distribution, strain state and deformation degree of the uterine structure.

[0021] In an embodiment of the present invention, hysteroscopy can directly observe the internal situation of the uterus and obtain high-quality image data. Grayscale processing simplifies the image information and highlights the edges and contours. Edge detection can accurately identify the edge contours of the uterus and uterine horns, providing a basis for uterine horn point extraction and geometric model construction. By extracting the candidate positions of the uterine horns, screening and verifying them, it can be ensured that the final uterine horn point positions are accurate and reliable. This helps to improve the accuracy of subsequent geometric model construction and the accuracy of finite element analysis. Based on the final uterine horn point position, a geometric coordinate system corresponding to the image coordinate system is established, which can realize the automatic identification and determination of the coordinates of the two uterine horn points inside the uterus. By calculating the actual distance value between the two uterine horn points, important size parameters can be provided for geometric model construction and finite element analysis. This helps to have a more comprehensive understanding of the structure and size of the uterus and provide more accurate information for the diagnosis and treatment of diseases.

[0022] Based on the coordinates of the uterine horn points, the actual distance between the two uterine horn points, and the internal image data of the uterus, a geometric model of the uterus and uterine horns can be constructed. This helps doctors understand the three-dimensional morphology and structure of the uterus more intuitively. Through finite element analysis, the stress distribution, strain state, and deformation degree of the uterine structure can be comprehensively evaluated. This helps doctors have a deeper understanding of the mechanical properties of the uterus under different physiological states, providing stronger support for the diagnosis and treatment of diseases.

[0023] In a preferred embodiment of the present invention, the above step 11, using a hysteroscope to obtain the internal image data of the uterus, and graying the color image in the internal image data of the uterus to obtain the gray image, may include: Apply the grayscale algorithm to convert the color image and convert the color value of each pixel into a grayscale value. Taking the weighted average method as an example, according to the sensitivity of human vision to color, the three components of RGB are weighted averaged to obtain a grayscale image. The specific formula is: ;in , , are the values ​​of the red, green, and blue components respectively. Save the processed grayscale image in a suitable format (such as JPEG, PNG, etc.).

[0024] In a preferred embodiment of the present invention, the above step 12, performing edge detection on the grayscale image, identifying the edge contours of the uterus and the uterine horns, and extracting candidate positions of the uterine horns according to the geometric features of the uterine horns, including curvature changes, sharp angles, or specific positions on the edge contours, may include: Step 122, performing edge detection on the grayscale image to identify the edges of the uterus and uterine cornu to obtain edge detection results; Step 123, extracting the contours of the uterus and uterine horns according to the edge detection result; Step 124, traversing the contours of the uterus and uterine horns, and setting judgment conditions according to the geometric features of the uterine horns, including curvature changes, sharp angles, or specific positions on the contours, to obtain an image marked with candidate positions of uterine horn points and a list of candidate positions; Step 125, mark each point in the candidate position list again to obtain a final marked image, and extract the candidate position of the palace corner point from the final marked image.

[0025] In an embodiment of the present invention, the Canny edge detection algorithm is selected based on the characteristics of the grayscale image. According to the specific situation of the image, the parameters of the Canny algorithm are adjusted, such as the standard deviation of the Gaussian filter, the high and low thresholds, etc. The Canny edge detection algorithm is applied to the grayscale image to identify the edges of the uterus and the uterine horns. A binary image is generated, in which the edge portion is marked as white (or high brightness) and the background is black (or low brightness). Based on the edge detection results, a contour tracking algorithm is applied to extract the contours of the uterus and uterine horns from the edge detection results. An image containing only the contours of the uterus and uterine horns is generated, in which the contours are clearly marked and the background is black or low brightness. A series of judgment conditions are set according to the geometric features of the uterine horns, such as changes in curvature, sharp angles, or specific positions on the contours. These conditions can be calculations based on mathematical formulas, such as curvature, angles, etc., or threshold settings based on experience.

[0026] Traverse point by point along the extracted contour and apply the judgment condition to each point. When a point meets the judgment condition, mark it as a candidate position of the uterine horn point. The candidate position can be marked on the original image, the contour image, or the new image, and represented by different colors or shapes. Generate a list of all candidate positions, each element of the list contains the coordinates of the candidate position and other related information, and generate an image marked with the candidate position of the uterine horn point. According to the actual morphology and positional relationship of the uterine horn point, formulate screening criteria. These criteria can be based on constraints such as distance, angle, shape, etc., which are used to remove candidate positions that do not meet the characteristics of the actual uterine horn point. Filter the candidate position list to remove points that do not meet the criteria. Finally mark the filtered candidate positions on the original image, the contour image, or the new image. Use eye-catching colors or shapes to represent the final uterine horn point position. Get the final marked image, in which the two uterine horn point positions inside the uterus are clearly marked. Extract the final uterine horn point candidate position from the final marked image.

[0027] Assuming there is a grayscale image of the inside of the uterus obtained by hysteroscopy, the following is the recognition process: Select the Canny edge detection algorithm to perform edge detection on the grayscale image to obtain a binary image containing the edges of the uterus and uterine horns. Apply the contour tracking algorithm to extract the contours of the uterus and uterine horns to obtain an image containing only the contours. Find points on the contour where the curvature changes significantly and where the angles are sharp. Apply judgment conditions to each point. Mark the points that meet the conditions as candidate positions for the uterine horn points, and obtain a list of candidate positions and an image marked with candidate positions. The distance between two candidate positions should be within a certain range, and the angle formed should be close to the actual angle of the uterine horn. Apply the screening criteria to filter the candidate positions. Finally mark the filtered candidate positions on the original image. Extract the final uterine horn point candidate positions from the final marked image.

[0028] In another preferred embodiment of the present invention, the above step 122, performing edge detection on the grayscale image to identify the edges of the uterus and uterine horns to obtain edge detection results, may include: Step 1223, traverse each pixel in the grayscale image, calculate the grayscale value change of the pixels around each pixel, and generate grayscale value change information of each pixel; Step 1224, judging whether each pixel is an edge point according to a preset threshold value, and obtaining an edge point judgment result; Step 1225, generating a binary image according to the edge point determination result, in which white pixels represent detected edge points and black pixels represent non-edge areas, and using the binary image as the edge detection result.

[0029] In the embodiment of the present invention, an image processing algorithm (such as Sobel operator) is used to traverse each pixel in the grayscale image. The Sobel operator is a gradient filter used to calculate the gradient value of each pixel in the image.

[0030] For each pixel, calculate the grayscale value change of the surrounding pixels. This is achieved through convolution operation, that is, using a specific filter (such as the gradient filter of the Sobel operator) to convolve with the image. The convolution kernel of the Sobel operator usually includes two components, horizontal and vertical, which are used to calculate the gradient of the pixel in the horizontal and vertical directions respectively. Specifically, for a pixel point (x, y) in the grayscale image, its horizontal gradient and the vertical gradient It can be calculated by the following formula: ; ;in, , ..., Represents pixel points The pixel values ​​in the surrounding 3x3 neighborhood. The edge point is determined based on the preset threshold T, and the gradient value of each pixel is determined whether it exceeds the threshold. If the gradient value ( and If the square root of the sum of the squares of , i.e. the gradient amplitude) exceeds the threshold T, the pixel is considered to be an edge point; otherwise, it is a non-edge point. Specifically, for the pixel point , its gradient amplitude It can be calculated by the following formula: ; Then, Compare with the threshold T to determine whether the pixel is an edge point. Generate a binary image based on the edge point judgment result. In the binary image, white pixels represent detected edge points (pixels whose gradient values ​​exceed the threshold), and black pixels represent non-edge areas (pixels whose gradient values ​​do not exceed the threshold). Through the binary image, the edge contours in the image can be clearly seen.

[0031] Take a grayscale image as an example, assuming that the size of the grayscale image is 512 512 pixels, using the Sobel operator for edge detection. The Sobel operator is a gradient filter, and its convolution kernel usually includes two components in the horizontal direction and the vertical direction. By convolving the convolution kernel of the Sobel operator with the grayscale image, the gradient value of each pixel in the horizontal direction and the vertical direction can be calculated. Set a threshold (such as 100) to determine whether the gradient value of each pixel exceeds the threshold. If the gradient value of a pixel is greater than 100, the pixel is considered to be an edge point; otherwise, it is a non-edge point. Based on the edge point judgment result, a 512x512 binary image is generated. In the binary image, white pixels represent detected edge points (pixels with gradient values ​​greater than 100), and black pixels represent non-edge areas (pixels with gradient values ​​less than or equal to 100).

[0032] By traversing each pixel in the grayscale image and calculating the grayscale value changes of the surrounding pixels, the edge position of the uterus and uterine horns can be accurately located. This precise edge positioning helps to improve the accuracy of subsequent image analysis. Based on the edge point judgment results, a binary image can be generated, in which white pixels represent the detected edge points and black pixels represent non-edge areas. The binary image obtained by edge detection can more easily extract feature information such as the shape and size of the uterus and uterine horns. The edge detection results are presented in the form of binary images, providing an intuitive visualization effect, which helps doctors more accurately judge the morphological structure of the uterus and uterine horns and whether there are abnormalities. The edge detection algorithm can be integrated into the automated detection system to realize functions such as automatic traversal of grayscale images, edge point judgment, and binary image generation, thereby improving detection efficiency and accuracy.

[0033] In a preferred embodiment of the present invention, the above step 13, screening and verifying the candidate positions to determine the final uterine horn point position; the above step 14, using the final uterine horn point position as a reference, establishing a geometric coordinate system corresponding to the image coordinate system, and automatically identifying and determining the coordinates of the two uterine horn points inside the uterus, may include: The grayscale image is edge detected by image processing algorithms (such as Sobel operator, Canny edge detection, etc.) to obtain a preliminary set of edge points. According to the anatomical structure characteristics of the uterus and uterine horns, certain screening conditions (such as shape, size, position, etc.) are set to screen candidate uterine horn point positions from the edge point set. For each candidate position, the image features (such as grayscale value, texture, shape, etc.) around it are extracted.

[0034] The extracted features were classified using the support vector machine (SVM) algorithm. The extracted feature vector was used as input to construct the SVM classifier. The known anatomical data of the uterus and uterine horns were used as training samples to train the SVM classifier. The features of the candidate positions were extracted and input into the trained SVM classifier for classification to screen out more likely uterine horn locations. The convolutional neural network (CNN) algorithm was used for automatic recognition of uterine horn points. A CNN model suitable for uterine horn point recognition was constructed and trained using a large amount of uterine and uterine horn image data. The grayscale image was automatically analyzed and the two uterine horn points inside the uterus were identified using the trained CNN model. According to the position of the identified uterine horn points in the image, combined with the setting of the geometric coordinate system, its coordinate position in the geometric coordinate system was determined.

[0035] In a preferred embodiment of the present invention, in the above step 15, the actual distance value between the two palace corner points is calculated by taking the palace corner point coordinates and the crossbar length as a reference; the actual distance value calculation formula between the two palace corner points is: ; in, Indicates the actual distance between two palace corner points; , It represents the horizontal coordinate of the palace corner point in the geometric coordinate system; , It represents the ordinate of the palace corner point in the geometric coordinate system; , represents the error correction term; Represents the proportional error correction term.

[0036] In the embodiment of the present invention, the coordinates of the two palace corner points in the geometric coordinate system are obtained by image processing technology (such as edge detection, contour extraction and feature point recognition), which are respectively recorded as and .

[0037] Determine the error correction term based on actual conditions and These correction items are used to correct errors caused by image acquisition, processing or coordinate system setting. Similarly, the scale error correction item is determined according to the actual situation. This correction term is used to adjust the scale error caused by image scaling, lens distortion, or inconsistent measurement units. Substitute the obtained coordinate values, error correction term, and scale error correction term into the formula and calculate to obtain the actual distance between the two palace corner points. .

[0038] Assume that the coordinates of the two palace corner points have been obtained through image processing technology: Coordinates of Palace Corner Point 1: ; Coordinates of palace corner point 2: .

[0039] At the same time, the error correction term and the proportional error correction term are determined: ; . Proportional error correction term: . Substituting these values ​​into the formula, calculate the coordinate difference and add the error correction term: ; .

[0040] Compute the sum of squares: ; ;Sum of squares = + = . Take the square root and multiply by the proportional error correction term: Actual distance Therefore, the actual distance between the two palace corners is approximately unit.

[0041] Through the error correction term and , can correct errors caused by image acquisition, processing or coordinate system setting, thereby improving the accuracy of measurement. , the proportional error caused by image scaling, lens distortion or inconsistent measurement units can be further adjusted to make the measurement result closer to the true value. Each parameter (coordinate value, error correction item, proportional error correction item) is set according to the actual situation. By substituting each parameter into the formula, the actual distance value can be directly calculated. The calculated actual distance value can provide objective reference information, which helps them to more accurately evaluate the structure and status of the uterus and make more appropriate decisions.

[0042] In a preferred embodiment of the present invention, the above step 16, constructing a geometric model of the uterus and uterine horns according to the uterine horn point coordinates, the actual distance value between the two uterine horn points and the internal image data of the uterus, may include: Step 166, constructing the geometric shapes of the uterus and uterine horns, including lines and arcs, according to the uterine horn point coordinates and the actual distance between the two uterine horn points; Step 167, combining the geometric shapes of the uterus and uterine horns, including lines and arcs, to construct an initial geometric outline of the uterus and uterine horns; Step 168, adjusting the shape and size of the initial geometric contours of the uterus and uterine horns to obtain adjusted geometric contours of the uterus and uterine horns; Step 169, constructing a geometric model of the uterus and uterine horns according to the adjusted geometric contours of the uterus and uterine horns.

[0043] In the embodiment of the present invention, according to the palace corner point coordinates and The actual distance between the two palace corners , a line segment connecting two uterine horn points can be drawn. Assuming that the shape of the uterus is approximately an ellipse or other suitable geometric shape, the coordinates of the uterine horn points and the actual distance values ​​can be used as references to draw the approximate outline of the uterus. This can be done by drawing arcs or other curves, which should be as close to the actual shape of the uterus as possible. The drawn line segments and arcs (or other curves) are combined to form the initial geometric outline of the uterus and uterine horns. This outline should be a closed figure that can roughly represent the position and shape of the uterus and uterine horns. According to the internal image data of the uterus, the shape and size of the initial outline are adjusted. This can be achieved by moving, scaling or rotating the points on the outline so that the outline fits the actual shape of the uterus and uterine horns more closely. During the adjustment process, the feature points, edges or textures in the internal image of the uterus can be referred to to ensure that the adjusted outline has sufficient accuracy. After the shape and size adjustment is completed, the adjusted geometric outline of the uterus and uterine horns is used as the basis of the geometric model. The outline is converted into a three-dimensional geometric model using computer-aided design (CAD) software. This model can include information such as the uterine cavity, the position and shape of the uterine horns.

[0044] Assume that the following data has been obtained: Palace corner point coordinates: ; The actual distance between two palace corner points: (This value is calculated using the actual distance between the two uterine horns mentioned above.) The uterus is elliptical, with the uterine horns located at both ends of the ellipse.

[0045] Follow these steps to build the geometric model of the uterus and uterine horns: Draw a line segment connecting two palace corner points in the coordinate system, that is, connect the points and Point .

[0046] Assuming that the uterus is an ellipse, the major and minor axes of the ellipse can be estimated based on the coordinates of the uterine horn points and the actual distance values. Draw an ellipse so that the two uterine horn points are located at both ends of the ellipse, and the shape and size of the ellipse are consistent with the internal image data of the uterus. Combine the drawn line segments and the ellipse to form the initial geometric contours of the uterus and uterine horns. According to the detailed information in the internal image data of the uterus, adjust the initial contour so that the contour is more in line with the actual shape of the uterus and uterine horns. For example, you can adjust the curvature of the ellipse, move the points on the contour, or change the size of the contour. Use CAD software or other tools to convert the adjusted contour into a three-dimensional geometric model. This model should be able to accurately represent information such as the uterine cavity, the position and shape of the uterine horns, etc.

[0047] By using the coordinates of the uterine angle points and the actual distance values, the position of the uterine angle can be accurately located, thereby ensuring that the constructed geometric model matches the real uterus and uterine angle in spatial position. Combined with the internal image data of the uterus, the shape and size of the model can be further refined to make it closer to the real uterine structure. The constructed geometric model provides an intuitive three-dimensional visualization effect, which can more clearly observe and understand the structure and morphology of the uterus and uterine angle. The geometric model can provide valuable reference information for medical research and diagnosis. For example, when studying problems such as uterine dysplasia and uterine angle pregnancy, the model can help doctors more accurately evaluate and analyze the condition. By constructing a personalized uterine and uterine angle geometric model based on the specific data of each patient, more accurate personalized medical treatment can be achieved. The constructed geometric model can easily perform data analysis and processing, such as calculating the volume, surface area and other parameters of the uterus, or analyzing the angle and position relationship of the uterine angle.

[0048] In a preferred embodiment of the present invention, the above step 17, performing finite element analysis based on the geometric model of the uterus and uterine horns to evaluate the stress distribution, strain state and deformation degree of the uterine structure, may include: Step 171, defining material properties for the uterine tissue, including elastic modulus and Poisson's ratio parameters; Meshing the geometric model of the uterus and uterine horns, dividing the geometric model of the uterus and uterine horns into multiple small, interconnected units; Step 172, setting boundary conditions and applying loads to the geometric model of the uterus and uterine horns, including simulating pressure changes in the uterus or external forces; Step 173, performing finite element analysis on the geometric model of the uterus and uterine horns through a solver. During the analysis, the solver calculates the stress, strain and deformation of each unit according to the set boundary conditions and loads and the material properties of the uterine tissue to obtain finite element analysis result data, including a stress distribution diagram, a strain state diagram and a deformation degree diagram; Step 174, analyzing the stress distribution diagram, the strain state diagram, and the deformation degree diagram to evaluate the stress distribution, strain state, and deformation degree of the uterine structure.

[0049] In an embodiment of the present invention, material properties are defined for uterine tissue, including elastic modulus (indicating the ease with which a material can produce elastic deformation when subjected to stress) and Poisson's ratio parameter (indicating the proportional relationship between lateral deformation and longitudinal deformation of a material when subjected to stress). The geometric model of the uterus and uterine horns is meshed to divide the model into multiple small, interconnected units. These units are the basic calculation units of finite element analysis. By calculating the stress, strain, and deformation of each unit, the stress distribution, strain state, and degree of deformation of the entire model can be obtained. Boundary conditions are set for the geometric model of the uterus and uterine horns, such as fixing certain parts of the model to simulate the fixed state of the uterus in the body. Loads are applied, including simulating pressure changes in the uterus (such as pressure caused by fetal growth, increased amniotic fluid, etc.) or external forces (such as external forces on the abdomen).

[0050] Finite element analysis is performed on the geometric model of the uterus and uterine horns through the solver. The solver calculates the stress, strain, and deformation of each unit based on the set boundary conditions and loads as well as the material properties of the uterine tissue. During the analysis, the solver iteratively solves the set of equations until a converged solution is obtained. After obtaining the finite element analysis result data, stress distribution diagrams, strain state diagrams, and deformation degree diagrams can be generated. The stress distribution diagrams, strain state diagrams, and deformation degree diagrams are analyzed to evaluate the stress distribution, strain state, and deformation degree of the uterine structure. For example, it can be observed which areas have greater stress, which areas have more obvious deformation, and whether the strain state is within a safe range.

[0051] Suppose you want to evaluate the stress distribution, strain state, and deformation degree of a pregnant woman's uterus during the second trimester. Follow these steps: Define appropriate elastic modulus and Poisson's ratio parameters for uterine tissue according to medical literature or experimental data. For example, assume that the elastic modulus is 10 kPa and the Poisson's ratio is 0.45. Use finite element analysis software to mesh the geometric model of the uterus and uterine horns. To ensure calculation accuracy and efficiency, you can select appropriate mesh density and unit type. For example, you can use tetrahedral units for meshing and ensure that the mesh density is high in key areas (such as uterine horns, uterine walls, etc.). Set boundary conditions to fix the uterine fundus to simulate its fixed state in the body. Apply loads to simulate the pressure changes in the uterus. For example, you can assume that the pressure in the uterus gradually increases due to fetal growth and increased amniotic fluid. Use the solver to perform finite element analysis on the geometric model of the uterus and uterine horns. During the analysis, the solver calculates the stress, strain, and deformation of each unit and generates the corresponding result data. Generate stress distribution diagrams, strain state diagrams, and deformation degree diagrams based on the finite element analysis result data. Observing these diagrams, we can find that: the stress is large in the uterine fundus and uterine horn area; the uterine wall has undergone a certain deformation; the overall strain state is within a safe range but needs to be monitored, etc. This information can provide valuable reference for evaluating the uterine health of pregnant women and formulating appropriate medical measures.

[0052] Finite element analysis can be used to obtain the stress distribution, strain state, and deformation degree of the uterus under specific conditions. The data provided by finite element analysis can help doctors more accurately assess the health of the uterus and make more appropriate medical decisions. For example, in uterine surgery planning, doctors can use the results of finite element analysis to predict the deformation and stress distribution of the uterus during surgery, thereby optimizing the surgical plan and reducing surgical risks. By building a personalized uterine and uterine horn geometric model for each patient and performing finite element analysis, we can achieve more accurate personalized medicine. By simulating different physiological and pathological conditions, researchers can explore the relationship between the mechanical properties of the uterus and the occurrence and development of the disease, providing new ideas and methods for the prevention and treatment of the disease.

[0053] In another preferred embodiment of the present invention, the above step 174, analyzing the stress distribution diagram, the strain state diagram and the deformation degree diagram to evaluate the stress distribution, strain state and deformation degree of the uterine structure, may include: Step 1745, analyzing the stress distribution diagram, the strain state diagram, and the deformation degree diagram, identifying high stress areas and low stress areas, stress gradient changes, strain and deformation direction magnitudes, so as to obtain analysis results of the stress distribution, strain state, and deformation degree diagram; Step 1746, based on the analysis results of the stress distribution, strain state and deformation degree diagram, a comprehensive assessment is performed on the uterine structure to identify problem areas, including stress concentration, excessive strain or abnormal deformation, so as to assess the stress distribution, strain state and deformation degree of the uterine structure.

[0054] In the embodiment of the present invention, by observing the stress distribution diagram, it is possible to clearly see which areas of the uterus are subject to greater stress (high stress areas) and which areas are subject to less stress (low stress areas). The high stress areas may be weak points or vulnerable areas in the uterine structure and require special attention.

[0055] The stress gradient indicates the rate of change of stress in the uterine structure. By observing the stress distribution diagram, the changes in the stress gradient can be analyzed to understand whether the stress is evenly distributed in the uterus and whether there is a sudden stress change. The strain state diagram and deformation degree diagram can be used to understand the deformation of the uterus under stress. By observing the direction and magnitude of the strain and deformation, it can be determined whether the uterus has undergone abnormal deformation or distortion, and whether these deformations have affected the function of the uterus. Based on the analysis results of the stress distribution, strain state and deformation degree diagrams, a comprehensive assessment of the uterine structure is performed. Identify problem areas, such as stress concentration, excessive strain or abnormal deformation, which may be potential risk points in the uterine structure.

[0056] Suppose you are evaluating the stress distribution, strain state, and deformation degree of a pregnant woman's uterus in the third trimester. The following is the analysis process: By observing the stress distribution diagram, it is found that the stress on the uterine fundus and uterine horn area is relatively large, while the stress on the middle part of the uterine wall is relatively small. This indicates that the uterine fundus and uterine horn area may be weak points in the uterine structure and need special attention. By observing the stress distribution diagram, we found that the stress has a large gradient change in the uterine fundus and uterine horn area, while it is relatively uniform in the middle part of the uterine wall. This indicates that when the uterus is under stress, the fundus and uterine horn area may be more prone to stress concentration and damage. Through the strain state diagram and deformation degree diagram, it is found that the uterus has undergone significant deformation in the late pregnancy, especially in the uterine fundus and uterine horn area. The strain and deformation directions in these areas are mainly directed to the outside of the uterus, and the degree of deformation is large. This indicates that the uterus is under great pressure in the late pregnancy, and the deformation of these areas needs to be paid attention to. According to the above analysis results, a comprehensive assessment of the uterine structure found that the uterine fundus and uterine horn area are potential risk points. These areas may have problems such as stress concentration, excessive strain or abnormal deformation, and appropriate medical measures need to be taken for intervention and monitoring. For example, pregnant women can be advised to reduce activities, avoid external force impact on the abdomen, and have regular prenatal examinations to monitor changes in the uterus.

[0057] like Figure 2As shown, an embodiment of the present invention further provides a hysteroscopic uterine angle visual reference finite element analysis system 20, comprising: The acquisition module 21 is used to acquire the internal image data of the uterus by using a hysteroscope, and grayscale the color image in the internal image data of the uterus to obtain a grayscale image; perform edge detection on the grayscale image to identify the edge contours of the uterus and the uterine horns, and extract the candidate positions of the uterine horns according to the geometric features of the uterine horns, including curvature changes, sharp angles or specific positions on the edge contours; screen and verify the candidate positions to determine the final uterine horn point positions; The processing module 22 is used to establish a geometric coordinate system corresponding to the image coordinate system based on the final uterine horn point position, automatically identify and determine the coordinates of the two uterine horn points inside the uterus; use the uterine horn point coordinates and the crossbar length as a reference to calculate the actual distance value between the two uterine horn points; construct a geometric model of the uterus and uterine horns based on the uterine horn point coordinates, the actual distance value between the two uterine horn points and the internal image data of the uterus; perform finite element analysis based on the geometric model of the uterus and uterine horns to evaluate the stress distribution, strain state and deformation degree of the uterine structure.

[0058] It should be noted that the system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0059] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0060] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0061] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A hysteroscopic uterine angle visual benchmark finite element analysis method, characterized in that: The method comprises: Acquiring internal image data of the uterus by using a hysteroscope, and graying a color image in the internal image data of the uterus to obtain a gray image; Perform edge detection on the grayscale image to identify the edge contours of the uterus and uterine horns, and extract candidate positions of uterine horn points based on the geometric features of the uterine horns, including curvature changes, sharp angles, or specific positions on the edge contours; Screen and verify the candidate positions to determine the final palace corner point position; Taking the final uterine horn point position as a reference, a geometric coordinate system corresponding to the image coordinate system is established to automatically identify and determine the coordinates of the two uterine horn points inside the uterus; Taking the palace corner point coordinates and the length of the horizontal bar as the basis, calculate the actual distance value between the two palace corner points; According to the coordinates of the uterine horn points, the actual distance between the two uterine horn points and the internal image data of the uterus, a geometric model of the uterus and the uterine horn is constructed; Based on the geometric model of the uterus and uterine horns, finite element analysis was performed to evaluate the stress distribution, strain state, and deformation degree of the uterine structure.

2. The hysteroscopic uterine angle visual reference finite element analysis method according to claim 1, characterized in that: Perform edge detection on the grayscale image to identify the edge contours of the uterus and uterine horns, and extract candidate positions of uterine horns based on the geometric features of the uterine horns, including curvature changes, sharp angles, or specific positions on the edge contours, including: Perform edge detection on the grayscale image to identify the edges of the uterus and uterine cornu to obtain edge detection results; According to the edge detection results, the contours of the uterus and uterine horns are extracted; Traversing the contours of the uterus and uterine horns, and setting judgment conditions according to the geometric features of the uterine horns, including curvature changes, sharp angles, or specific positions on the contours, to obtain an image marked with candidate positions of uterine horn points and a list of candidate positions; Each point in the candidate position list is marked again to obtain a final marked image, and the candidate position of the palace corner point is extracted from the final marked image.

3. The hysteroscopic uterine angle visual reference finite element analysis method according to claim 2, characterized in that: Perform edge detection on the grayscale image to identify the edges of the uterus and uterine horns to obtain edge detection results, including: Traverse each pixel in the grayscale image, calculate the grayscale value change of the pixels around each pixel, and generate the grayscale value change information of each pixel; According to the preset threshold, determine whether each pixel is an edge point, and obtain the edge point determination result; A binary image is generated according to the edge point judgment result, in which white pixels represent detected edge points and black pixels represent non-edge areas, and the binary image is used as the edge detection result.

4. The hysteroscopic uterine angle visual reference finite element analysis method according to claim 3, characterized in that: The actual distance between two palace corner points is calculated as follows: ； in, Indicates the actual distance between two palace corner points; , It represents the horizontal coordinate of the palace corner point in the geometric coordinate system; , It represents the ordinate of the palace corner point in the geometric coordinate system; , Respectively expressed in Direction and Scaling factor in direction; represents the overall scaling factor; , Respectively expressed in Direction and The offset in direction; , Respectively expressed in Direction and Noise term in the direction.

5. The hysteroscopic uterine angle visual reference finite element analysis method according to claim 4, characterized in that: According to the coordinates of the uterine horn points, the actual distance between the two uterine horn points and the internal image data of the uterus, a geometric model of the uterus and uterine horns is constructed, including: According to the coordinates of the uterine horn points and the actual distance between the two uterine horn points, the geometric shapes of the uterus and uterine horns, including lines and arcs, are constructed; Combine the geometric shapes of the uterus and uterine horns, including lines and arcs, to construct the initial geometric outlines of the uterus and uterine horns; Adjusting the shape and size of the initial geometric contours of the uterus and uterine horns to obtain the adjusted geometric contours of the uterus and uterine horns; According to the adjusted geometric contours of the uterus and uterine horns, a geometric model of the uterus and uterine horns is constructed.

6. The hysteroscopic uterine angle visual reference finite element analysis method according to claim 5, characterized in that: Based on the geometric model of the uterus and uterine horns, finite element analysis was performed to evaluate the stress distribution, strain state, and deformation degree of the uterine structure, including: Define material properties for uterine tissue, including elastic modulus and Poisson's ratio parameters; Meshing the geometric model of the uterus and uterine horns, dividing the geometric model of the uterus and uterine horns into multiple small, interconnected units; Set boundary conditions and apply loads to the geometric model of the uterus and uterine horns, including simulating pressure changes or external forces in the uterus; The uterus and uterine horn geometric model is subjected to finite element analysis by the solver. During the analysis, the solver calculates the stress, strain and deformation of each unit according to the set boundary conditions and loads as well as the material properties of the uterine tissue to obtain finite element analysis result data, including stress distribution diagram, strain state diagram and deformation degree diagram; The stress distribution diagram, strain state diagram and deformation degree diagram were analyzed to evaluate the stress distribution, strain state and deformation degree of the uterine structure.

7. The hysteroscopic uterine angle visual reference finite element analysis method according to claim 6, characterized in that: The stress distribution diagram, strain state diagram and deformation degree diagram are analyzed to evaluate the stress distribution, strain state and deformation degree of the uterine structure, including: Analyze the stress distribution diagram, strain state diagram and deformation degree diagram, identify high stress areas and low stress areas, stress gradient changes, strain and deformation direction, and obtain the analysis results of stress distribution, strain state and deformation degree diagram; Based on the analysis results of the stress distribution, strain state and deformation degree diagrams, a comprehensive assessment of the uterine structure is conducted to identify problem areas, including stress concentration, excessive strain or abnormal deformation, in order to assess the stress distribution, strain state and deformation degree of the uterine structure.

8. A hysteroscopic uterine angle visual reference finite element analysis system, the system implementing the method according to claim 1, characterized in that: include: An acquisition module is used to acquire the internal image data of the uterus by using a hysteroscope, and grayscale the color image in the internal image data of the uterus to obtain a grayscale image; Perform edge detection on the grayscale image to identify the edge contours of the uterus and uterine horns, and extract candidate uterine horn locations based on the geometric features of the uterine horns, including curvature changes, sharp angles, or specific locations on the edge contours; screen and verify the candidate locations to determine the final uterine horn location; The processing module is used to establish a geometric coordinate system corresponding to the image coordinate system based on the final uterine horn point position, automatically identify and determine the coordinates of the two uterine horn points inside the uterus; use the uterine horn point coordinates and the crossbar length as a reference to calculate the actual distance value between the two uterine horn points; construct a geometric model of the uterus and uterine horns based on the uterine horn point coordinates, the actual distance value between the two uterine horn points and the internal image data of the uterus; perform finite element analysis based on the geometric model of the uterus and uterine horns to evaluate the stress distribution, strain state and deformation degree of the uterine structure.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images