A method for constructing a vaginal relaxation assessment model based on multi-source data

By combining electromyography sensors and voyeuroscope technology to construct a three-dimensional dynamic vaginal model, analyze vaginal movement and tension characteristics, and use Naive Bayes algorithm to construct a vaginal relaxation evaluation model, solving the problem of inaccurate vaginal relaxation status and motion analysis in traditional methods. , achieving higher analysis accuracy and accuracy.

CN118430825BActive Publication Date: 2024-12-03BEIDO YUNCHUANG (BEIJING) TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202410832435.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-26
Publication Date
2024-12-03
Estimated Expiration
2044-06-26

AI Technical Summary

Technical Problem

The traditional method of vaginal relaxation evaluation model construction based on multi-source data has problems such as inaccurate analysis of vaginal relaxation status and inaccurate analysis of vaginal movement.

Method used

The electromyography sensor is used to collect the internal electromyography signal of the vagina, and obtain high-resolution images through the built-in camera module of the voyeuroscope, build a three-dimensional dynamic vagina model with real-time activity, analyze the movement pattern and the changes in contraction intensity, and evaluate the degree of vaginal wall tension adjustment and Relaxation characteristics, a vaginal relaxation evaluation model was constructed by combining Naive Bayes algorithm.

Benefits of technology

Improves the accuracy of vaginal relaxation status analysis and the accuracy of vaginal motion analysis, providing a fast, automated evaluation tool that reduces the subjectivity and error of manual evaluation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118430825B_ABST
    Figure CN118430825B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of vaginal relaxation models, and in particular to a method for constructing a vaginal relaxation assessment model based on multi-source data. The method comprises the following steps: collecting internal vaginal electromyographic signals to obtain internal vaginal electromyographic signals; collecting image data inside the vagina and constructing a real-time active three-dimensional vaginal model to obtain a real-time active three-dimensional dynamic vaginal model; analyzing the regularity of vaginal movement on the real-time active three-dimensional dynamic vaginal model and analyzing the change of vaginal contraction intensity and evaluating the degree of vaginal wall tension regulation to obtain vaginal wall tension regulation degree data; analyzing the relaxation characteristics of the vaginal wall tension regulation degree data and calculating the vaginal elastic modulus, as well as performing correlation analysis and constructing a vaginal relaxation assessment model to obtain a vaginal relaxation assessment model. The present invention makes the model more perfect by improving the construction technology of the vaginal relaxation model.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vaginal relaxation models, and in particular to a method for constructing a vaginal relaxation assessment model based on multi-source data. Background Art

[0002] With the rapid development of data science and artificial intelligence technology, the evaluation model based on multi-source data has gradually become an effective solution. This method constructs a comprehensive evaluation model by integrating multiple types of data, including patients' physiological indicators (such as electromyography and pressure sensor data), imaging data (such as MRI or ultrasound images), and subjective evaluations such as patients' quality of life questionnaires. This multi-source data method can not only improve the objectivity and accuracy of the evaluation, but also provide a more comprehensive understanding of the user's specific situation. For example, by analyzing electromyography and pressure data, the functional status of vaginal muscles can be quantified; combined with imaging data, detailed information on vaginal structure can be provided; and the user's self-evaluation can reflect the actual impact of symptoms on quality of life. However, the traditional method of constructing a vaginal relaxation evaluation model based on multi-source data has the problems of inaccurate analysis of vaginal relaxation status and inaccurate analysis of vaginal movement. Summary of the invention

[0003] Based on this, it is necessary to provide a method for constructing a vaginal relaxation assessment model based on multi-source data to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for constructing a vaginal relaxation assessment model based on multi-source data is provided, the method comprising the following steps:

[0005] Step S1: using an electromyographic sensor to collect myoelectric signals inside the vagina to obtain myoelectric signals inside the vagina; using a camera module built into a spectacled scope to collect image data inside the vagina to obtain an image inside the vagina; constructing a real-time active three-dimensional vaginal model based on the image inside the vagina to obtain a real-time active three-dimensional dynamic vaginal model;

[0006] Step S2: analyzing the regularity of vaginal movement on the real-time active three-dimensional dynamic vaginal model to obtain vaginal movement regularity data; analyzing the change of vaginal contraction intensity on the vaginal movement regularity data to obtain vaginal contraction intensity change data; evaluating the degree of vaginal wall tension regulation based on the vaginal contraction intensity change data to obtain vaginal wall tension regulation degree data;

[0007] Step S3: performing relaxation characteristic analysis on the vaginal wall tension adjustment degree data to obtain vaginal relaxation characteristic data; performing vaginal elastic modulus calculation based on the vaginal relaxation characteristic data to obtain vaginal elastic modulus data; performing adhesion buffer force analysis based on the vaginal elastic modulus data to obtain elastic adhesion buffer force data;

[0008] Step S4: performing correlation analysis on the vaginal wall tension regulation degree data according to the elastic adhesion buffer force data to obtain adhesion buffer-tension regulation correlation data; constructing a vaginal relaxation evaluation model for the adhesion buffer-tension regulation correlation data using the naive Bayes algorithm to obtain a vaginal relaxation evaluation model.

[0009] The present invention uses an electromyographic sensor (EMG) to measure the myoelectric signals inside the vagina. These signals are electrophysiological data generated when the vaginal muscles are active, providing detailed electrical activity data of the vaginal muscles, which can capture the contraction and relaxation patterns of the muscles and help evaluate the functional status of the vaginal muscles. Especially for individuals who need rehabilitation or functional improvement, a vaginoscope with a camera function is used to collect high-resolution images of the inside of the vagina, providing detailed images of the internal structure of the vagina, which helps doctors or researchers observe and analyze the anatomical features of the vagina. Accurate internal images can be obtained without surgery or other invasive operations. By processing the collected image data, a three-dimensional dynamic model that reflects the real-time activity of the vagina is constructed, which not only provides a static anatomical structure, but also can show the dynamic changes of the vagina in different states, providing more detailed information for complex movements or structural changes. Intuitive understanding supports further research. By analyzing the three-dimensional dynamic vaginal model, the movement patterns of the vagina in various activities (such as contraction and relaxation) are determined, and the movement patterns and patterns of the vagina in different situations are identified, which provides a basis for subsequent functional evaluation. By analyzing the dynamic changes of vaginal electromyographic signals and three-dimensional models, the intensity changes of vaginal contractions are evaluated, and quantitative data of vaginal contraction intensity are provided, which helps to evaluate the strength and function of vaginal muscles. Based on the data of vaginal contraction intensity changes, the tension regulation ability of the vaginal wall under different conditions is evaluated, which helps to understand the response and regulation ability of the vaginal wall under different pressure or movement conditions, which is crucial in the case of rehabilitation or strengthening of vaginal muscle function; by analyzing the tension regulation data of the vaginal wall under different states, its relaxation characteristics are evaluated. This involves analyzing the recovery speed and degree of vaginal muscles and tissues after contraction, accurately evaluating the recovery ability of vaginal tissues after various pressures or movements, which is very important for diagnosing vaginal relaxation and related dysfunctions. Using relaxation characteristic data, the elastic modulus of vaginal tissue is calculated. The elastic modulus reflects the elastic response of the tissue when it is subjected to stress, that is, the resistance to deformation, and provides a quantitative indicator of the elasticity of the vaginal tissue, which helps to evaluate its functionality. It analyzes the adhesion buffering capacity of the vagina when it is subjected to external forces (such as physical contact or pressure) and when the vagina is peristaltic, that is, its ability to resist external forces and return to its original state, and helps to evaluate the structural integrity and durability of the vaginal tissue under different stress conditions. The relationship between the elastic adhesion buffering force data and the tension regulation degree data of the vagina under different conditions is analyzed by statistical methods, which provides a multi-dimensional understanding of the function of the vagina. Combining the elastic and tension characteristics, it can more comprehensively evaluate the condition of the vagina. Based on the correlation data, the Naive Bayes algorithm is used to construct a vaginal relaxation assessment model. This model can predict and evaluate the degree of vaginal relaxation, providing an automated tool that can quickly and accurately evaluate the degree of vaginal relaxation and reduce the subjectivity and errors of manual evaluation.Therefore, the present invention is an improvement on a traditional method for constructing a vaginal relaxation evaluation model based on multi-source data, which solves the problems of inaccurate analysis of vaginal relaxation state and inaccurate analysis of vaginal movement in a traditional method for constructing a vaginal relaxation evaluation model based on multi-source data, and improves the precision of analysis of vaginal relaxation state and the accuracy of analysis of vaginal movement.

[0010] Preferably, step S1 comprises the following steps:

[0011] Step S11: using an electromyographic sensor to collect internal vaginal electromyographic signals to obtain internal vaginal electromyographic signals;

[0012] Step S12: collecting image data of the interior of the vagina through the built-in camera module of the spectacles to obtain an image of the interior of the vagina;

[0013] Step S13: performing image exposure optimization processing on the image of the interior of the vagina to obtain an exposure-optimized image of the interior of the vagina;

[0014] Step S14: constructing a real-time active three-dimensional vaginal model based on the optimized exposure image of the vagina interior to obtain a real-time active three-dimensional dynamic vaginal model.

[0015] The present invention uses electromyographic sensors (EMG) to collect electrophysiological signals from the muscles inside the vagina. These signals reflect the contraction and relaxation activities of the vaginal muscles. By directly collecting electrophysiological signals, the activity patterns of the vaginal muscles can be accurately captured, providing real-time muscle function data, supporting dynamic functional evaluation and monitoring, and using a spectacles with built-in camera modules to collect high-resolution images of the inside of the vagina. These images capture the anatomical structure of the interior of the vagina. The high-resolution images provide detailed visual information of the interior of the vagina, which helps doctors or researchers to make precise observations and analyses. The collected images of the interior of the vagina are optimized for exposure to improve image clarity and contrast and eliminate areas that are too dark or too bright. Exposure optimization ensures that image details are clearly visible and reduces the loss of visual information caused by uneven lighting or improper exposure. Clear optimized images improve the accuracy of subsequent three-dimensional model construction, ensuring that the model can more accurately reflect the actual vaginal structure and dynamic changes. Using high-quality image data, a three-dimensional model is created that can reflect vaginal activity in real time, making the model more realistic and accurate in dynamic evaluation. The dynamic model not only provides static anatomical information, but also can show the dynamic changes of the vagina in various physiological activities, providing a basis for in-depth functional and structural analysis.

[0016] The beneficial effect of the present invention is to use an electromyographic sensor (EMG) to measure the electromyographic signals inside the vagina. These signals are electrophysiological data generated when the vaginal muscles are active, providing detailed electrical activity data of the vaginal muscles, which can capture the contraction and relaxation patterns of the muscles and help evaluate the functional status of the vaginal muscles, especially for individuals who need rehabilitation or functional improvement. A vaginoscope with a camera function is used to collect high-resolution images of the inside of the vagina, providing detailed images of the internal structure of the vagina, which helps doctors or researchers observe and analyze the anatomical features of the vagina. Accurate internal images can be obtained without surgery or other invasive operations. By processing the collected image data, a three-dimensional dynamic model that reflects the real-time activity of the vagina is constructed, which not only provides a static anatomical structure, but also can show the dynamic changes of the vagina in different states, providing more detailed information for complex movements or structural changes. Intuitive understanding supports further research. By analyzing the three-dimensional dynamic vaginal model, the movement patterns of the vagina in various activities (such as contraction and relaxation) are determined, and the movement patterns and patterns of the vagina in different situations are identified, which provides a basis for subsequent functional evaluation. By analyzing the dynamic changes of vaginal electromyographic signals and three-dimensional models, the intensity changes of vaginal contractions are evaluated, and quantitative data of vaginal contraction intensity are provided, which helps to evaluate the strength and function of vaginal muscles. Based on the data of vaginal contraction intensity changes, the tension regulation ability of the vaginal wall under different conditions is evaluated, which helps to understand the response and regulation ability of the vaginal wall under different pressure or movement conditions, which is crucial in the case of rehabilitation or strengthening of vaginal muscle function; by analyzing the tension regulation data of the vaginal wall under different states, its relaxation characteristics are evaluated. This involves analyzing the recovery speed and degree of vaginal muscles and tissues after contraction, accurately evaluating the recovery ability of vaginal tissues after various pressures or movements, which is very important for diagnosing vaginal relaxation and related dysfunctions. Using relaxation characteristic data, the elastic modulus of vaginal tissue is calculated. The elastic modulus reflects the elastic response of the tissue when it is subjected to stress, that is, the resistance to deformation, and provides a quantitative indicator of the elasticity of the vaginal tissue, which helps to evaluate its functionality. It analyzes the adhesion buffering capacity of the vagina when it is subjected to external forces (such as physical contact or pressure) and when the vagina is peristaltic, that is, its ability to resist external forces and return to its original state, and helps to evaluate the structural integrity and durability of the vaginal tissue under different stress conditions. The relationship between the elastic adhesion buffering force data and the tension regulation degree data of the vagina under different conditions is analyzed by statistical methods, which provides a multi-dimensional understanding of the function of the vagina. Combining the elastic and tension characteristics, it can more comprehensively evaluate the condition of the vagina. Based on the correlation data, the Naive Bayes algorithm is used to construct a vaginal relaxation assessment model. This model can predict and evaluate the degree of vaginal relaxation, providing an automated tool that can quickly and accurately evaluate the degree of vaginal relaxation and reduce the subjectivity and errors of manual evaluation.Therefore, the present invention is an improvement on a traditional method for constructing a vaginal relaxation evaluation model based on multi-source data, which solves the problems of inaccurate analysis of vaginal relaxation state and inaccurate analysis of vaginal movement in a traditional method for constructing a vaginal relaxation evaluation model based on multi-source data, and improves the precision of analysis of vaginal relaxation state and the accuracy of analysis of vaginal movement. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of the steps of a method for constructing a vaginal relaxation assessment model based on multi-source data;

[0018] Figure 2 for Figure 1 Detailed implementation steps of step S2 in FIG.

[0019] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0020] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0022] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0023] To achieve this, please refer to Figure 1 to Figure 2 , a method for constructing a vaginal relaxation assessment model based on multi-source data, the method comprising the following steps:

[0024] Step S1: using an electromyographic sensor to collect myoelectric signals inside the vagina to obtain myoelectric signals inside the vagina; using a camera module built into a spectacled scope to collect image data inside the vagina to obtain an image inside the vagina; constructing a real-time active three-dimensional vaginal model based on the image inside the vagina to obtain a real-time active three-dimensional dynamic vaginal model;

[0025] Step S2: analyzing the regularity of vaginal movement on the real-time active three-dimensional dynamic vaginal model to obtain vaginal movement regularity data; analyzing the change of vaginal contraction intensity on the vaginal movement regularity data to obtain vaginal contraction intensity change data; evaluating the degree of vaginal wall tension regulation based on the vaginal contraction intensity change data to obtain vaginal wall tension regulation degree data;

[0026] Step S3: performing relaxation characteristic analysis on the vaginal wall tension adjustment degree data to obtain vaginal relaxation characteristic data; performing vaginal elastic modulus calculation based on the vaginal relaxation characteristic data to obtain vaginal elastic modulus data; performing adhesion buffer force analysis based on the vaginal elastic modulus data to obtain elastic adhesion buffer force data;

[0027] Step S4: performing correlation analysis on the vaginal wall tension regulation degree data according to the elastic adhesion buffer force data to obtain adhesion buffer-tension regulation correlation data; constructing a vaginal relaxation evaluation model for the adhesion buffer-tension regulation correlation data using the naive Bayes algorithm to obtain a vaginal relaxation evaluation model.

[0028] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a method for constructing a vaginal relaxation assessment model based on multi-source data of the present invention. In this example, the method for constructing a vaginal relaxation assessment model based on multi-source data includes the following steps:

[0029] Step S1: using an electromyographic sensor to collect myoelectric signals inside the vagina to obtain myoelectric signals inside the vagina; using a camera module built into a spectacled scope to collect image data inside the vagina to obtain an image inside the vagina; constructing a real-time active three-dimensional vaginal model based on the image inside the vagina to obtain a real-time active three-dimensional dynamic vaginal model;

[0030] In the embodiment of the present invention, a miniaturized myoelectric sensor (such as a surface electromyographic sensor, sEMG) suitable for use inside the vagina is selected. The sensor should have the characteristics of high sensitivity and low noise, and is suitable for detecting the electrical activity of the vaginal wall muscles. The sensor is disinfected to ensure sterility during use. The myoelectric sensor is attached to a designed spectacles or a special probe so that it can be in close contact with the vaginal wall after being inserted into the vagina. The spectacles or probes are inserted into the vagina so that the sensor is in contact with the vaginal wall. The myoelectric signal acquisition is started, and the myoelectric signal amplifier and acquisition equipment (such as a data acquisition card or a wireless data transmission module) are used to amplify the myoelectric signal and transmit it to a computer or data processing system. The acquisition parameters (such as sampling frequency, gain, etc.) are adjusted to obtain high-quality myoelectric signals. The collected raw myoelectric signals are preprocessed, including filtering (low-pass filtering, high-pass filtering, band-pass filtering, etc.) and denoising, removing interference such as baseline drift and power supply noise, and using envelope detection and other methods to extract the characteristics of the signal to obtain clearer muscle activity information; a small high-definition camera suitable for use inside the vagina is selected to ensure that clear images can be captured even under low light conditions. The camera should have a suitable focal length and field of view to cover the key areas inside the vagina. The camera should be fixed in the built-in position of the speculum so that it can stably capture images of the inside of the vagina after being inserted into the vagina. Ensure that the power supply and data transmission interface (such as USB or wireless transmission module) of the camera are connected normally and set to real-time video output mode. Insert the speculum and adjust the position so that the camera is aimed at the inside of the vagina. The camera is used to capture the image data of the inside of the vagina in real time and transmit the image data to a computer or data processing system. Adjust the exposure, white balance and focus parameters of the camera to ensure the best image quality; pre-process the collected images of the inside of the vagina, such as grayscale conversion, denoising and contrast enhancement, for subsequent image processing and three-dimensional reconstruction. Use feature extraction techniques (such as edge detection and feature point detection) to extract key features in the image. Use a multi-view registration algorithm to register and splice the multiple frames of images collected continuously to obtain a panoramic view of the inside of the vagina. Use image registration technology (such as SIFT, SURF, etc.) and optimization algorithms to correct the displacement and rotation of images under different viewing angles, and apply 3D reconstruction algorithms (such as structured light, stereo vision or shape from motion algorithms) to build a 3D geometric model of the inside of the vagina. Use the multi-view images of the camera to generate point cloud data or voxel models of the inside of the vagina through triangulation or volume reconstruction technology. Use continuously collected images and electromyographic signals to dynamically update the 3D model to reflect the real-time activities and changes of the vaginal wall. Calculate the deformation and muscle activity status of the model in real time, and enhance the dynamic expression of the model by fusing electromyographic signals and 3D image data. Use 3D rendering technology to visualize the constructed real-time dynamic 3D vaginal model. Provide an interactive interface that allows users to rotate, zoom and view different angles and details of the model.

[0031] Step S2: analyzing the regularity of vaginal movement on the real-time active three-dimensional dynamic vaginal model to obtain vaginal movement regularity data; analyzing the change of vaginal contraction intensity on the vaginal movement regularity data to obtain vaginal contraction intensity change data; evaluating the degree of vaginal wall tension regulation based on the vaginal contraction intensity change data to obtain vaginal wall tension regulation degree data;

[0032] In an embodiment of the present invention, key feature points in a real-time active three-dimensional dynamic vaginal model are tracked. These points can usually be geometric vertices of the model or marked anatomical areas (such as specific positions of the vaginal wall). A motion tracking algorithm (such as an optical flow method or a feature point matching algorithm) is used to calculate the motion trajectory of these feature points in a time series. The extracted motion trajectory is subjected to time series analysis to identify the periodic motion patterns of the vaginal muscles (such as contraction and relaxation cycles). Frequency domain analysis (such as Fourier transform) and time domain analysis (such as autocorrelation analysis) are used to evaluate the frequency and amplitude of vaginal motion. The motion data at different time points are integrated to generate complete vaginal motion law data, the dynamic motion of the vaginal model is quantified, the displacement amplitude and speed of the feature points are calculated, the contraction amplitude and frequency of the vaginal wall at different time points are obtained, and the direction and intensity of the contraction motion are determined by vector analysis of the motion of the feature points. Using the movement amplitude data, calculate the intensity change of vaginal contraction, including parameters such as maximum contraction force, average contraction force and contraction frequency. Combined with the amplitude characteristics in the electromyographic signal, further refine the contraction intensity analysis to obtain more accurate vaginal contraction intensity change data, analyze the temporal trend of contraction intensity, identify possible abnormal changes or regular changes, apply statistical analysis (such as regression analysis and trend analysis) to quantify the fluctuation and change pattern of contraction intensity, and define the evaluation indicators of vaginal wall tension, such as tension adjustment range, tension change rate and maximum tension value, based on the vaginal contraction intensity change data. By comparing the model tension under static and dynamic conditions, determine the key parameters for evaluating the degree of vaginal wall tension adjustment. Perform mechanical simulation on the real-time active three-dimensional model to calculate the vaginal wall tension at different time points and motion states. Use finite element analysis or other biomechanical models to simulate the mechanical response of the vaginal wall during contraction and relaxation, comprehensively evaluate the vaginal contraction intensity change data, analyze the tension adjustment ability of the vaginal wall under different motion states, combine the tension evaluation indicators to quantify the degree of vaginal wall tension adjustment, and generate data on the degree of vaginal wall tension adjustment.

[0033] Step S3: performing relaxation characteristic analysis on the vaginal wall tension adjustment degree data to obtain vaginal relaxation characteristic data; performing vaginal elastic modulus calculation based on the vaginal relaxation characteristic data to obtain vaginal elastic modulus data; performing adhesion buffer force analysis based on the vaginal elastic modulus data to obtain elastic adhesion buffer force data;

[0034] In an embodiment of the present invention, the data of the degree of adjustment of the tension of the vaginal wall is sorted and preprocessed to ensure data quality and consistency. According to the previous evaluation results, an appropriate time point or state is selected for relaxation characteristic analysis, and the collected data is used to establish a mathematical model of the relaxation behavior of the vaginal wall, such as an exponential decay model or an elastic relaxation model. Through mathematical fitting or curve fitting technology, the relaxation characteristics of the vagina under different stress states are analyzed, and the key parameters of the relaxation model, such as the relaxation time constant or relaxation rate, are extracted to describe the relaxation behavior of the vagina under the action of external force. According to the relaxation characteristic parameters, vaginal relaxation characteristic data are generated, and based on the relaxation characteristic data, the stress-strain relationship of the vaginal wall under different tensile or compressive conditions is analyzed. Using elastic theory or material mechanics model, the elastic modulus calculation formula of the vaginal wall is derived, and the elastic modulus of the vagina under a specific strain is calculated according to the stress-strain relationship and experimental data. Through numerical calculation or analytical methods, the elastic modulus data of the vagina is obtained, and an adhesion buffer model between the vaginal wall and surrounding tissues is established. Taking into account the adhesion and friction characteristics between tissues, the biomechanical model or contact mechanics model is used to describe the adhesion behavior of the vagina under motion and pressure, and the stress distribution and force transfer mechanism of the vagina under external force are analyzed, especially the adhesion buffer effect under dynamic motion or deformation. The contact pressure, stress distribution and displacement response of the vaginal wall surface, as well as the mechanical properties of the adhesion area, are calculated. Based on the model calculation and experimental data, detailed data on the elastic adhesion buffer force of the vagina are generated, the force data are analyzed, and the adhesion buffer performance and force response characteristics of the vagina under different conditions are evaluated.

[0035] Step S4: performing correlation analysis on the vaginal wall tension regulation degree data according to the elastic adhesion buffer force data to obtain adhesion buffer-tension regulation correlation data; constructing a vaginal relaxation evaluation model for the adhesion buffer-tension regulation correlation data using the naive Bayes algorithm to obtain a vaginal relaxation evaluation model.

[0036] In an embodiment of the present invention, elastic adhesion buffer force data and vaginal wall tension regulation degree data are collated and prepared to ensure data quality and consistency, determine the time resolution or condition of the data to ensure the validity of the correlation analysis, and use statistical methods (such as Pearson correlation coefficient) or non-parametric methods (such as Spearman rank correlation coefficient) to analyze the correlation between elastic adhesion buffer force data and vaginal tension regulation degree data. Calculate the correlation coefficient and evaluate its significance, determine the degree of association between the two, and generate an adhesion buffer-tension regulation correlation data set according to the correlation analysis results, including the correlation coefficient and the statistical significance level of the correlation, and select the naive Bayes algorithm as the basis for constructing a vaginal relaxation assessment model, considering its advantages in processing multivariate relationships and probability inference. Design the model structure, determine the input variables (adhesion buffer-tension regulation correlation data) and the output variables (vaginal relaxation assessment results), pre-process the input data, including feature selection, missing value processing, data standardization or normalization, etc., to ensure that the data meets the requirements of the naive Bayes algorithm, use the labeled data set to train the naive Bayes model, and learn the probability distribution and conditional probability in the data. Use cross-validation or holdout set methods to verify the performance and generalization ability of the model, adjust the model parameters to improve the prediction accuracy, evaluate the constructed vaginal relaxation assessment model, and analyze its prediction ability and model robustness. Optimize and adjust the model according to the evaluation results to improve its effect and applicability in practical applications. Apply the trained vaginal relaxation assessment model to new data sets to predict and evaluate the vaginal relaxation state. Analyze the model output results, provide quantitative indicators and explanations for vaginal relaxation assessment, explain the model output results, explain the relationship between elastic adhesion buffer force and vaginal wall tension regulation, and provide clinical or research application suggestions for vaginal relaxation state. Discuss the limitations of the model and future improvement directions to improve the accuracy and practicality of vaginal relaxation assessment.

[0037] The present invention uses an electromyographic sensor (EMG) to measure the myoelectric signals inside the vagina. These signals are electrophysiological data generated when the vaginal muscles are active, providing detailed electrical activity data of the vaginal muscles, which can capture the contraction and relaxation patterns of the muscles and help evaluate the functional status of the vaginal muscles. Especially for individuals who need rehabilitation or functional improvement, a vaginoscope with a camera function is used to collect high-resolution images of the inside of the vagina, providing detailed images of the internal structure of the vagina, which helps doctors or researchers observe and analyze the anatomical features of the vagina. Accurate internal images can be obtained without surgery or other invasive operations. By processing the collected image data, a three-dimensional dynamic model that reflects the real-time activity of the vagina is constructed, which not only provides a static anatomical structure, but also can show the dynamic changes of the vagina in different states, providing more detailed information for complex movements or structural changes. Intuitive understanding supports further research. By analyzing the three-dimensional dynamic vaginal model, the movement patterns of the vagina in various activities (such as contraction and relaxation) are determined, and the movement patterns and patterns of the vagina in different situations are identified, which provides a basis for subsequent functional evaluation. By analyzing the dynamic changes of vaginal electromyographic signals and three-dimensional models, the intensity changes of vaginal contractions are evaluated, and quantitative data of vaginal contraction intensity are provided, which helps to evaluate the strength and function of vaginal muscles. Based on the data of vaginal contraction intensity changes, the tension regulation ability of the vaginal wall under different conditions is evaluated, which helps to understand the response and regulation ability of the vaginal wall under different pressure or movement conditions, which is crucial in the case of rehabilitation or strengthening of vaginal muscle function; by analyzing the tension regulation data of the vaginal wall under different states, its relaxation characteristics are evaluated. This involves analyzing the recovery speed and degree of vaginal muscles and tissues after contraction, accurately evaluating the recovery ability of vaginal tissues after various pressures or movements, which is very important for diagnosing vaginal relaxation and related dysfunctions. Using relaxation characteristic data, the elastic modulus of vaginal tissue is calculated. The elastic modulus reflects the elastic response of the tissue when it is subjected to stress, that is, the resistance to deformation, and provides a quantitative indicator of the elasticity of the vaginal tissue, which helps to evaluate its functionality. It analyzes the adhesion buffering capacity of the vagina when it is subjected to external forces (such as physical contact or pressure) and when the vagina is peristaltic, that is, its ability to resist external forces and return to its original state, and helps to evaluate the structural integrity and durability of the vaginal tissue under different stress conditions. The relationship between the elastic adhesion buffering force data and the tension regulation degree data of the vagina under different conditions is analyzed by statistical methods, which provides a multi-dimensional understanding of the function of the vagina. Combining the elastic and tension characteristics, it can more comprehensively evaluate the condition of the vagina. Based on the correlation data, the Naive Bayes algorithm is used to construct a vaginal relaxation assessment model. This model can predict and evaluate the degree of vaginal relaxation, providing an automated tool that can quickly and accurately evaluate the degree of vaginal relaxation and reduce the subjectivity and errors of manual evaluation.Therefore, the present invention is an improvement on a traditional method for constructing a vaginal relaxation evaluation model based on multi-source data, which solves the problems of inaccurate analysis of vaginal relaxation state and inaccurate analysis of vaginal movement in a traditional method for constructing a vaginal relaxation evaluation model based on multi-source data, and improves the precision of analysis of vaginal relaxation state and the accuracy of analysis of vaginal movement.

[0038] Preferably, step S1 comprises the following steps:

[0039] Step S11: using an electromyographic sensor to collect internal vaginal electromyographic signals to obtain internal vaginal electromyographic signals;

[0040] Step S12: collecting image data of the interior of the vagina through the built-in camera module of the spectacles to obtain an image of the interior of the vagina;

[0041] Step S13: performing image exposure optimization processing on the image of the interior of the vagina to obtain an exposure-optimized image of the interior of the vagina;

[0042] Step S14: constructing a real-time active three-dimensional vaginal model based on the optimized exposure image of the vagina interior to obtain a real-time active three-dimensional dynamic vaginal model.

[0043] In an embodiment of the present invention, a highly sensitive surface electromyographic sensor (sEMG) suitable for use inside the vagina is selected, and it is ensured to meet the safety and sanitation standards of medical equipment. The electromyographic sensor is fixed on a designed speculum or a special probe so that it can contact the key muscle area of ​​the vaginal wall, and the contact between the sensor and the vaginal wall is checked to ensure close contact to obtain high-quality signals. The sensor is connected to the electromyographic signal amplifier and the data acquisition system, and a suitable sampling frequency (such as 1000 Hz or higher) is set to capture rapidly changing electromyographic signals. The equipment is calibrated, and the gain and filter settings are adjusted to reduce interference from external noise. After inserting the speculum, the electromyographic signals inside the vagina are collected and the signal quality is monitored. The collected electromyographic signals are recorded and saved to a computer or other data storage device for subsequent analysis. The collected raw electromyographic signals are filtered (such as low-pass filtering, high-pass filtering, and band-pass filtering) to remove baseline drift and power supply noise. Useful information of the electromyographic signal, such as the amplitude and frequency of muscle activity, is extracted through envelope detection and feature extraction technology. A high-definition camera suitable for use inside the vagina is selected, with autofocus and low-light performance. Fix the camera in the built-in position of the speculum to ensure that it can stably capture the interior of the vagina after insertion. Connect the camera to the image acquisition system or computer, adjust the exposure, white balance and resolution settings of the camera to ensure the best image quality, and calibrate the field of view and focal length of the camera to cover the key areas of the interior of the vagina. Record and save high-quality interior vaginal image data for subsequent processing and analysis. Preprocess the acquired interior vaginal images, such as grayscale conversion and noise removal, to improve the basic quality of the image. Use methods such as histogram equalization to enhance the contrast of the image. Use automatic exposure adjustment algorithms, such as adaptive histogram equalization (CLAHE) or gamma correction to optimize the brightness and contrast of the image. Perform local exposure adjustment on different areas to ensure uniform exposure of the entire interior vaginal image. Evaluate the quality of the image after exposure optimization and check whether it meets the requirements of 3D model construction. If necessary, further adjust the exposure optimization parameters to achieve the best image quality. Output and save the interior vaginal image after exposure optimization for subsequent 3D model construction. Perform feature extraction on the exposure optimized image to identify the key structures and edges of the interior of the vagina. Use image segmentation algorithms to separate the vaginal wall and other anatomical structures in the image. Use multiple frames of continuous images for multi-view registration, and stitch images from different perspectives into a complete view of the vaginal interior through feature matching and geometric correction. Apply image stitching algorithms (such as global registration and local optimization) to generate seamless three-dimensional views. Use three-dimensional reconstruction techniques (such as structured light or stereo vision technology) to convert two-dimensional images into three-dimensional geometric models. Generate point cloud data or surface models of the vaginal interior to reflect the three-dimensional morphology of the vaginal wall. Combine electromyographic signal data with the three-dimensional model to update the dynamic deformation of the vaginal wall in real time.Use dynamic simulation or shape from motion (SfM) technology to reflect the morphological changes of the vagina in different motion states. Use 3D rendering technology to display a real-time 3D dynamic vaginal model. Provide an interactive interface that allows users to view, rotate and analyze different angles and details of the model.

[0044] The present invention uses electromyographic sensors (EMG) to collect electrophysiological signals from the muscles inside the vagina. These signals reflect the contraction and relaxation activities of the vaginal muscles. By directly collecting electrophysiological signals, the activity patterns of the vaginal muscles can be accurately captured, providing real-time muscle function data, supporting dynamic functional evaluation and monitoring, and using a spectacles with built-in camera modules to collect high-resolution images of the inside of the vagina. These images capture the anatomical structure of the interior of the vagina. The high-resolution images provide detailed visual information of the interior of the vagina, which helps doctors or researchers to make precise observations and analyses. The collected images of the interior of the vagina are optimized for exposure to improve image clarity and contrast and eliminate areas that are too dark or too bright. Exposure optimization ensures that image details are clearly visible and reduces the loss of visual information caused by uneven lighting or improper exposure. Clear optimized images improve the accuracy of subsequent three-dimensional model construction, ensuring that the model can more accurately reflect the actual vaginal structure and dynamic changes. Using high-quality image data, a three-dimensional model is created that can reflect vaginal activity in real time, making the model more realistic and accurate in dynamic evaluation. The dynamic model not only provides static anatomical information, but also can show the dynamic changes of the vagina in various physiological activities, providing a basis for in-depth functional and structural analysis.

[0045] Preferably, step S2 comprises the following steps:

[0046] Step S21: analyzing the regularity of vaginal movement on the real-time active three-dimensional dynamic vaginal model to obtain vaginal movement regularity data;

[0047] Step S22: mapping the vaginal movement law data to the vaginal movement law myoelectric signal according to the vaginal internal myoelectric signal to obtain the vaginal movement law myoelectric signal;

[0048] Step S23: performing spectrum conversion on the vaginal movement regular electromyographic signal to obtain a spectrum diagram of the vaginal movement electromyographic signal;

[0049] Step S24: analyzing the change of vaginal contraction intensity on the vaginal movement law data according to the vaginal movement electromyographic signal spectrum to obtain vaginal contraction intensity change data;

[0050] Step S25: Evaluate the degree of vaginal wall tension regulation based on the vaginal contraction intensity variation data and the vaginal movement electromyographic signal spectrum to obtain vaginal wall tension regulation degree data.

[0051] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0052] Step S21: analyzing the regularity of vaginal movement on the real-time active three-dimensional dynamic vaginal model to obtain vaginal movement regularity data;

[0053] In an embodiment of the present invention, in a real-time active three-dimensional dynamic vaginal model, key feature points or areas of the vaginal wall (such as geometric vertices or anatomical feature points of the model) are selected, and a motion tracking algorithm (such as optical flow method, SIFT feature matching) is applied to track the motion trajectory of these feature points in continuous frames. The position changes of the feature points in the time series are recorded to form complete motion trajectory data. The key parameters of the motion trajectory, such as velocity, acceleration and displacement amplitude, are calculated. The motion trajectory data are analyzed in the frequency domain using Fourier transform or wavelet transform to identify the frequency components and periodicity of the motion. Autocorrelation analysis or peak detection technology is used to identify the changing patterns of the motion period and amplitude. The motion data of different feature points are integrated into a comprehensive motion law data.

[0054] Step S22: mapping the vaginal movement law data to the vaginal movement law myoelectric signal according to the vaginal internal myoelectric signal to obtain the vaginal movement law myoelectric signal;

[0055] In an embodiment of the present invention, the collected electromyographic signals are filtered to remove power frequency noise and baseline drift. The electromyographic signals are normalized to ensure that the signal amplitude is within a uniform range. The time axis of the electromyographic signals and the real-time active three-dimensional dynamic vaginal model is calibrated to ensure synchronization of the two data. Based on the vaginal motion law data, the electromyographic signals are mapped to corresponding time periods and feature points to form a time series of motion law electromyographic signals. Corresponding electromyographic signal data is assigned to each feature point or motion trajectory segment. The motion law data and the corresponding electromyographic signal data are fused together to form comprehensive vaginal motion law electromyographic signal data.

[0056] Step S23: performing spectrum conversion on the vaginal movement regular electromyographic signal to obtain a spectrum diagram of the vaginal movement electromyographic signal;

[0057] In the embodiment of the present invention, FFT is applied to the electromyographic signal data of the movement pattern and converted to the frequency domain to obtain the spectrum of the signal. Wavelet transform can be used for more detailed time-frequency analysis to retain the local characteristics of time and frequency, extract the key frequency components and energy distribution in the spectrum, identify the main electromyographic activity frequency bands, mark the main peaks and frequency components in the spectrum diagram, and form a spectrum diagram of the vaginal movement electromyographic signal.

[0058] Step S24: analyzing the change of vaginal contraction intensity on the vaginal movement law data according to the vaginal movement electromyographic signal spectrum to obtain vaginal contraction intensity change data;

[0059] In the embodiment of the present invention, the power spectral density (PSD) is extracted from the spectrum, the power of different frequency components is analyzed, and characteristic parameters such as frequency peak, total power, dominant frequency, etc. are calculated to reflect the strength and activity of the electromyographic signal. The frequency components in the spectrum are correlated with the vaginal movement law data to determine the electromyographic activity level under different movement states. Based on the frequency components and power data, the change in the contraction intensity of the vagina in different movement cycles is mapped, and the change in the intensity of the vaginal contraction is calculated, including the maximum contraction force, the average contraction force and the change amplitude. The change trend and periodic change of the contraction intensity during the entire movement process are analyzed to generate the vaginal contraction intensity change data.

[0060] Step S25: Evaluate the degree of vaginal wall tension regulation based on the vaginal contraction intensity variation data and the vaginal movement electromyographic signal spectrum to obtain vaginal wall tension regulation degree data.

[0061] In an embodiment of the present invention, based on the vaginal contraction intensity variation data and the spectrum diagram, key parameters for evaluating vaginal wall tension regulation, such as tension regulation range and response speed, are defined, evaluation criteria for tension regulation degree, such as maximum allowable tension, recovery time and stability, are determined, a mechanical model of the vaginal wall is established, and the tension variation of the vagina under different motion and contraction conditions is simulated. Based on the biomechanical model, the mechanical response and tension distribution of the vaginal wall under different contraction intensities are calculated, and the tension regulation ability of the vaginal wall under different motion states is analyzed in combination with the vaginal contraction intensity variation data. The effect of vaginal wall tension regulation is quantified, and detailed tension regulation degree data is generated. The vaginal contraction intensity variation data and tension regulation degree data are integrated to form a comprehensive evaluation report. The final vaginal wall tension regulation degree evaluation data is output for further clinical analysis or research.

[0062] The present invention extracts the movement law of the vagina in various physiological activities by analyzing the real-time active three-dimensional dynamic vagina model. This includes the morphological changes and movement patterns of the vagina in different states, provides detailed movement laws of the vagina under different activity conditions, helps to identify normal and abnormal movement patterns, obtains comprehensive movement data from the three-dimensional model, improves the understanding of the dynamic behavior of the vagina, maps the vagina movement law data with the myoelectric signals inside the vagina, and establishes the association between the two. This process helps to understand the relationship between myoelectric activity and movement patterns, combines movement laws and myoelectric signals, and realizes multi-source data fusion analysis of vaginal function. The combination of myoelectric signals and movement data can provide richer physiological information, improve the comprehensiveness and accuracy of analysis, and convert the mapped myoelectric signals into spectrum to obtain a spectrum diagram. This process analyzes the distribution of myoelectric signals in the frequency domain, reveals the muscle activity characteristics under different frequency components, and the spectrum diagram provides the energy distribution of myoelectric signals at different frequencies, which helps to identify the frequency domain characteristics of vaginal muscle activity. Spectral analysis can reveal hidden patterns and features that cannot be directly observed in the original signal. The information in the spectrum diagram is used to analyze the vaginal movement law data and evaluate the changes in the contraction intensity of the vaginal muscles. This can help understand the strength and function of muscles in different activities. Through spectrum information, it provides a quantitative assessment of the contraction strength of vaginal muscles, reveals the strength and functional status of muscles, and can track the changes in the strength of vaginal muscles in various activities, providing basic data for further analysis. Combined with the vaginal contraction strength change data and the electromyographic signal spectrum, the tension regulation ability of the vaginal wall is evaluated. This process evaluates the tension adaptation ability of the vaginal wall under different conditions. Based on multidimensional data, it evaluates the tension regulation ability of the vaginal wall under different states, reveals its response and adaptability under stress conditions, and provides data on the adaptability of the vaginal wall to different pressures and movements, which helps to understand the functional performance of the vagina under various physiological states.

[0063] Preferably, step S25 comprises the following steps:

[0064] Step S251: Plotting a contraction strength change curve for the vaginal contraction strength change data to obtain a vaginal contraction strength change curve;

[0065] Step S252: performing strength change trend analysis according to the vaginal contraction strength change curve to obtain strength change trend data;

[0066] Step S253: Calculating the contraction mechanical performance of the vaginal contraction strength change curve according to the strength change trend data to obtain vaginal contraction mechanical performance data;

[0067] Step S254: Calculating the vaginal contraction fatigue threshold based on the vaginal contraction mechanical efficiency data and the vaginal movement electromyographic signal spectrum to obtain vaginal contraction fatigue threshold data;

[0068] Step S255: Evaluate the degree of vaginal wall tension regulation according to the vaginal contraction fatigue critical data and the vaginal contraction mechanical efficiency data to obtain vaginal wall tension regulation degree data.

[0069] In an embodiment of the present invention, the vaginal contraction intensity change data is preliminarily cleaned to remove outliers and noise. Ensure that the time axis of the vaginal contraction intensity change data is synchronized with the time axis of the electromyographic signal. Apply a suitable fitting method (such as polynomial fitting, spline interpolation, etc.) to fit discrete contraction intensity data points into a continuous change curve. Use smoothing techniques (such as moving average or Savitzky-Golay filter) to eliminate high-frequency noise in the curve to ensure the smoothness and accuracy of the curve. Set the time axis and intensity axis of the curve, and indicate the units of each key time point and intensity change. Draw a vaginal contraction intensity change curve in a two-dimensional coordinate system to show the change of contraction intensity over time. Use time domain analysis methods, such as slope calculation and peak detection, to analyze the overall trend and local changes of the curve. Use frequency domain analysis methods, such as Fourier transform, to identify the periodic components in the contraction intensity change curve, calculate the slope of different intervals of the curve, and determine the rate of increase or decrease of contraction intensity. Mark the obvious rising, falling and stable stages in the curve, and analyze the change characteristics of different stages. Extract key trend features from the change curve, such as rise time, fall time and peak hold time. Organize the extracted feature data to form intensity change trend data for subsequent analysis. According to the principles of biomechanics, define the key parameters of vaginal contraction mechanical effectiveness, such as power, energy consumption and efficiency. Determine the formula and method for calculating the mechanical effectiveness of vaginal contraction, usually based on the contraction intensity change curve and the spectrum data of the electromyographic signal. Calculate the instantaneous power and average power according to the speed and amplitude of the contraction intensity change. Integrate the contraction power curve and calculate the energy consumption during the entire contraction process. Combine the power spectrum data of the electromyographic signal to evaluate the efficiency and effectiveness of the contraction. According to the theory of biomechanics and electromyography, define the key parameters of vaginal contraction fatigue, such as fatigue limit, fatigue endurance and fatigue recovery time.Fatigue analysis methods, such as stress-strain curve and energy dissipation model, are used to extract fatigue-related frequency features from the EMG signal spectrum, such as the attenuation of high-frequency components and the increase of low-frequency components. Combined with the contraction mechanical efficiency data, the effect of fatigue on contraction power and energy efficiency is analyzed. The time point or intensity at which the vagina reaches fatigue criticality during continuous contraction is calculated through contraction intensity and efficiency data. The fatigue endurance of the vagina at different intensities and frequencies is evaluated, and its maximum tolerable contraction load is determined. The calculated fatigue critical time, intensity and endurance data are sorted out to form vaginal contraction fatigue critical data. Appropriate evaluation indicators (such as tension adjustment range, adjustment response speed, and fatigue recovery ability) are selected to evaluate the tension adjustment degree of the vaginal wall. The vaginal contraction mechanical efficiency data is combined with the fatigue critical data to form complete tension adjustment evaluation data. Multifactor analysis methods are used to evaluate the performance of tension adjustment degree under different conditions. According to the contraction intensity and fatigue data, the adjustment range and maximum adjustment capacity of the vaginal wall tension are calculated. The response speed of the vaginal wall under different intensity changes is analyzed to determine its adjustment efficiency and ability. Based on the analysis results, an evaluation report on the adjustment degree of the vaginal wall tension is generated, including indicators such as adjustment ability, response speed and fatigue endurance.

[0070] The present invention draws a contraction strength change curve of the vagina at different time points based on the vaginal contraction strength change data. The curve reflects the change in the strength of the vaginal muscles in a specific time period. The contraction strength change curve clearly shows the dynamic strength change of the vaginal muscles, provides intuitive visual data, and can easily identify the change pattern and trend of the contraction strength through the curve, laying the foundation for subsequent analysis, analyzing the contraction strength change curve, identifying the trend of strength change, including the pattern of increase, decrease or fluctuation, revealing the change trend of vaginal contraction strength in different time periods, and helping to understand its performance under different conditions. According to the change trend, different contraction modes can be classified to provide a reference for further functional evaluation, and the mechanical performance of the vaginal contraction strength change curve is calculated using the strength change trend data. This includes evaluating the energy consumption and efficiency of each contraction, calculating the mechanical efficiency of the vaginal muscles during contraction, providing a quantitative evaluation of muscle function and efficiency, analyzing the efficiency of energy utilization during contraction, revealing the working efficiency of muscles at different intensities, and calculating the fatigue critical point of the vagina during continuous contraction, that is, the point at which the muscle begins to fatigue under certain conditions, based on the contraction mechanical efficiency data and the electromyographic signal spectrum. Provide fatigue critical data on the vaginal muscles under long-term contraction or high-intensity activities to help understand their endurance and stamina. Identifying fatigue critical points can help formulate reasonable activities or training plans to avoid excessive fatigue. Combined with vaginal contraction fatigue critical data and contraction mechanical efficiency data, a comprehensive evaluation of the tension regulation degree of the vaginal wall is conducted. This process evaluates the adaptability and adjustment ability of the vaginal wall under different intensities and fatigue conditions, provides comprehensive tension regulation data of the vaginal wall under different pressures and motion conditions, reveals its functional performance in a stress environment, and understands the tolerance and adaptability of the vaginal wall in long-term or high-intensity activities by evaluating the tension regulation degree.

[0071] Preferably, calculating the vaginal contraction fatigue threshold based on the vaginal contraction mechanical performance data and the vaginal movement electromyographic signal spectrum includes the following steps:

[0072] Conducting contraction force symmetry analysis on vaginal contraction mechanical performance data to obtain contraction force symmetry data;

[0073] The vaginal contraction efficiency attenuation is calculated based on the contraction force symmetry data and the vaginal contraction mechanical efficiency data to obtain the vaginal contraction efficiency attenuation data;

[0074] The muscle activity decrease gradient is evaluated on the spectrum of vaginal movement electromyographic signals to obtain muscle activity decrease gradient data;

[0075] The vaginal contraction fatigue critical value is calculated based on the vaginal contraction efficiency attenuation data and the muscle activity decline gradient data to obtain the vaginal contraction fatigue critical data.

[0076] In an embodiment of the present invention, the mechanical performance data of vaginal contraction is obtained, including force measurement at different time points or under different contraction conditions, and the definition of contraction force symmetry is determined. Usually, the contraction force on both sides or multiple measurement points is compared to see if it is consistent, and statistical methods (such as mean value, standard deviation) are used to evaluate the symmetry data. The contraction force symmetry data is obtained from the previous step, including the relationship between the maximum force, average force, torque, etc. and time, and the vaginal contraction efficiency is defined. Based on the maximum force and duration, an attenuation model is established, and the collected performance data is used to fit the attenuation model to obtain the attenuation constant, and the characteristics of the spectrum are analyzed, such as power spectrum density, to define the muscle activity decline gradient, the decline rate of the high-frequency component in the spectrum, and the muscle activity decline gradient data is obtained by dividing the high-frequency component by the time change; based on the attenuation of vaginal contraction efficiency and the decline of muscle activity, the fatigue criticality is defined. Usually, the time point when the contraction efficiency drops to a certain threshold or the activity drops to a certain level is used to obtain the vaginal contraction fatigue criticality data.

[0077] The present invention analyzes whether the mechanical efficiency of the muscles on both sides of the vagina during contraction is symmetrical. This involves comparing muscle mechanical efficiency data at different time points or in different contraction cycles. Through symmetry analysis, it is possible to identify whether there is an asymmetric force distribution in the muscle during contraction, which helps to understand the coordination of muscle function, identify and correct asymmetric contraction force, and help optimize the movement performance and functional coordination of the vagina. Combined with the contraction force symmetry data and mechanical efficiency data, the contraction efficiency attenuation of the vagina during continuous contraction is calculated. This evaluates the degree of reduction in muscle efficiency in long-term or repeated contractions, provides efficiency change data of vaginal muscles in different time periods, helps monitor its efficiency attenuation in continuous activities, understands the attenuation of contraction efficiency, helps evaluate the continuous performance of the vagina in long-term or high-intensity activities, analyzes the frequency components in the electromyographic signal spectrum, and evaluates the decreasing gradient of muscle activity in different time periods. This includes identifying changes in high-frequency and low-frequency components, reflecting the gradual decline in muscle activity, providing detailed spectral data on changes in vaginal muscle activity, helping to track its decline in activity during sustained contraction, and through gradient assessment, early identification of the trend of declining muscle activity, predicting a potential state of fatigue, and combining contraction efficiency decay data and muscle activity decline gradient data to calculate the fatigue critical point of the vagina during sustained contraction or high-intensity activities. This critical point is the moment when the vaginal muscles begin to fatigue significantly. By integrating multi-dimensional data, an accurate fatigue critical point is provided to help understand the fatigue limit of the vagina under different conditions, evaluate the durability and endurance of the vaginal muscles, and provide data on its performance in high-intensity or long-term activities.

[0078] Preferably, step S3 comprises the following steps:

[0079] Step S31: performing baseline tension calculation on the vaginal wall tension adjustment degree data to obtain vaginal wall baseline tension data;

[0080] Step S32: performing relaxation characteristic analysis on the vaginal wall tension adjustment degree data according to the vaginal wall baseline tension data to obtain vaginal relaxation characteristic data;

[0081] Step S33: Calculating the vaginal elastic modulus according to the vaginal relaxation characteristic data and the vaginal contraction strength change data to obtain vaginal elastic modulus data;

[0082] Step S34: performing adhesion buffer force analysis on the vaginal elastic modulus data according to the vaginal relaxation characteristic data to obtain elastic adhesion buffer force data.

[0083] In the embodiment of the present invention, the vaginal wall tension regulation data in the non-contracted state is obtained, usually by measuring the pressure distribution or tension in the static state, ensuring that the data is collected under uniform conditions, such as temperature, humidity and posture, using filtering technology to remove noise in the data, and commonly used methods include low-pass filtering and mean filtering. Smoothing algorithms, such as moving average method or Gaussian smoothing, are applied to reduce random fluctuations in the data. The average value of tension is calculated in the static data as the baseline tension, and the baseline tension is used as a reference point for subsequent relaxation and elasticity analysis; the vaginal wall tension regulation data at different time points or different contraction states is obtained to ensure that the time series of the data is complete in order to analyze the dynamic changes of tension, remove the long-term trend in the data, usually use differential or high-pass filtering, normalize the data, and facilitate comparison and analysis, calculate the time constant for the tension to recover from the contraction state to the baseline tension, and obtain the relaxation index by calculating the relaxation time constant, reflecting the speed of the vaginal wall from tension to relaxation, and collect the mechanical data of the vagina during contraction, including the tension and displacement during contraction. The relaxation characteristic data obtained from step S32. Integrate the contraction intensity data and displacement data to construct a force-displacement relationship curve. Tension is converted into displacement data, usually through mechanical models or experimental data. Apply Hooke's law in elastic mechanics to calculate the elastic modulus, extract stress and strain data from the force-displacement curve, and calculate the elastic modulus. Build a model to describe the viscoelastic behavior of the vaginal wall, such as the Maxwell model or the Kelvin-Voigt model. Study the buffering effect of the vaginal wall under different stress and relaxation conditions through simulation or numerical analysis. Analyze how adhesion and elastic effects are distributed and transmitted on the vaginal wall. Use numerical simulation to further test the characteristics of adhesion buffering force.

[0084] The present invention performs baseline calculation on the vaginal wall tension adjustment degree data to determine the natural tension state of the vaginal wall under no external force or standard conditions. This is equivalent to determining the basic tension of the vaginal wall in a relaxed state. The baseline tension calculation provides the natural tension state of the vaginal wall without the influence of external force, which is the benchmark for subsequent tension analysis. Identifying the baseline state of the vaginal wall helps to understand its natural state and equilibrium point under different physiological conditions. According to the baseline tension data, the tension adjustment of the vaginal wall under different conditions is analyzed for relaxation characteristics. This includes evaluating the tension change and relaxation degree of the vaginal wall at different time points or activity states, analyzing the relaxation characteristics of the vaginal wall under different conditions, helping to understand its relaxation and recovery ability under stress conditions, providing data on the tension adjustment and adaptability of the vaginal wall under different activity states, helping to understand its functional dynamics, and calculating the elastic modulus of the vaginal wall according to the vaginal relaxation characteristic data and the vaginal contraction strength change data. The elastic modulus is a parameter that describes the relationship between stress and strain of a material within the elastic range. It is used to evaluate the elasticity and hardness of the vaginal wall, provide elastic modulus data of the vaginal wall, reveal its elasticity and hardness under different stress conditions, and help understand its physical properties. Through the elastic modulus, the hardness and softness of the vaginal wall are analyzed to provide key data for functional and comfort evaluation. Combined with the vaginal relaxation characteristic data, the vaginal elastic modulus data is used for adhesion buffering force analysis. This process evaluates the force and buffering capacity of the vaginal wall under adhesion and external force, and provides data on the force distribution and buffering capacity of the vaginal wall under adhesion and external force conditions, which helps to understand its performance under complex stress conditions, analyze the buffering and adaptability of the vaginal wall in a stress environment, and reveal its tolerance in long-term or high-intensity activities.

[0085] Preferably, step S34 includes the following steps:

[0086] Step S341: dividing the vaginal elastic modulus data into vaginal elasticity difference regions according to the vaginal relaxation characteristic data to obtain vaginal elasticity difference region data;

[0087] Step S342: performing peristalsis simulation of different elastic regions on the vaginal elasticity difference region data according to the vaginal relaxation characteristic data to obtain elastic region peristalsis simulation data;

[0088] Step S343: performing linkage force analysis of different vaginal regions on the elastic region peristalsis simulation data to obtain vaginal peristalsis linkage force data;

[0089] Step S344: performing coupling force gradient loss calculation according to the vaginal peristaltic coupling force data and the vaginal elasticity difference area data to obtain peristaltic coupling force gradient loss data;

[0090] Step S345: Perform adhesion buffer force analysis based on the vaginal peristaltic coupling force data and the peristaltic coupling force gradient loss data to obtain elastic adhesion buffer force data.

[0091] In the embodiment of the present invention, according to the difference in elastic modulus of different parts of the vagina, different elastic regions, such as high elastic region and low elastic region, are divided, and the critical value and demarcation point of elastic difference are determined using statistical methods or methods based on data distribution, and each region is marked as high elastic, low elastic or medium elastic region for subsequent analysis and use. Based on the data of vaginal elastic difference region, the peristaltic motion characteristics of the vagina in different regions are simulated, and the elastic region data marked in step S341 are used to select a suitable peristaltic model, such as a biomechanical model or a computational fluid dynamics model, and the boundary conditions and initial conditions of the model are set, such as peristaltic frequency, amplitude, etc., and an appropriate numerical simulation algorithm is selected, such as finite element method or finite volume method, and the simulation program is run to obtain peristaltic motion simulation data in different vaginal regions. The key motion parameters in the simulation process, such as displacement, velocity and stress, are recorded, and the linkage dynamic characteristics of vaginal peristalsis in different elastic regions are analyzed, and the vaginal peristalsis simulation data obtained from step S342 are obtained. The vaginal elastic difference region data obtained from step S341. A linkage dynamic model is established to describe the interaction between different regions of the vagina. The force transmission, stress distribution and mutual influence between different regions in the simulation process are analyzed. Extract the magnitude, direction and change trend of the linkage force in different regions. Use charts and curves to show the changes in mechanical parameters and compare the linkage forces in different regions. Calculate the reduction of the peristaltic linkage force gradient based on the vaginal peristaltic linkage force data and the elastic difference area data. Define the gradient of the peristaltic linkage force, usually the rate of change of force in space or time. Use mathematical methods or differential methods to calculate the gradient of the force. Establish a model to describe the gradient reduction, considering the reduction effect of the elastic region. Apply the model to calculate the gradient reduction of the peristaltic linkage force in different elastic regions. Analyze the reduction data to understand the degree of influence of different elastic regions on the peristaltic force gradient. Evaluate the reduction effect of the elastic region during the peristalsis process and explain the physiological significance behind the mechanical phenomenon. Based on the peristaltic linkage force data, analyze the adhesion buffering force characteristics of the vaginal wall in the elastic region. Select an appropriate viscoelastic model to describe the force behavior of the vaginal wall during the peristalsis process. Set the parameters of the model according to experimental data or literature. Use numerical methods to simulate the force distribution and transmission of the vaginal wall during the peristalsis process. Record key parameters in the force simulation process, such as stress, strain and displacement. Analyze the mechanical properties of the vagina under different peristaltic conditions, including adhesion and cushioning effects. Explain the impact of mechanical phenomena on vaginal health and function.

[0092] The present invention uses vaginal relaxation characteristic data to analyze vaginal elastic modulus data and divides the vaginal wall into regions with different elastic characteristics. These regions may exhibit different physical properties due to different elasticity and relaxation levels. Dividing regions with different elastic characteristics helps to understand in detail the physical characteristics and functional performance of the vaginal wall in each region, providing a basis for further personalized functional evaluation and processing, and being able to formulate more accurate strategies for different regions. Based on the vaginal relaxation characteristic data, peristaltic simulation is performed on the divided different elastic regions. This simulation aims to understand the dynamic behavior and changes of different regions during contraction and relaxation. Through simulation, the dynamic changes of different elastic regions during contraction and relaxation are intuitively demonstrated. Peristaltic simulation helps evaluate the performance of each region in actual dynamic motion, provides key data for functional analysis, analyzes the peristaltic simulation data of the elastic region, and evaluates the linkage behavior and interaction of different vaginal regions during peristalsis. This includes identifying which areas are coordinated or influence each other during movement, revealing how different areas coordinate and influence each other during peristalsis, providing a deeper understanding of the overall function of the vagina, analyzing the coordination ability of various parts of the vagina during movement, helping to identify functional disorders and optimization directions, and calculating the gradient loss of linkage force in different vaginal areas during peristalsis based on vaginal peristaltic linkage force data and elastic difference area data. This evaluates the changes and losses of linkage force between different areas, provides the changes in linkage force between various areas of the vagina during peristalsis, helps understand its energy and force transmission in dynamic movement, identifies and quantifies the loss of linkage force, and reveals the functional loss of the vagina during long-term or high-intensity exercise. Combining vaginal peristaltic linkage force data and peristaltic linkage force gradient loss data, adhesion buffering force analysis is performed. This evaluates the force and buffering capacity of the vaginal wall under adhesion and external force in different areas, provides detailed force and buffering capacity data of the vaginal wall under adhesion and external force conditions, helps understand its performance in complex stress environments, and through precise analysis, reveals the buffering and adaptability of the vaginal wall under high stress conditions, providing a more refined functional assessment.

[0093] Preferably, the calculation of the gradient loss of the coupling force according to the vaginal peristaltic coupling force data and the vaginal elasticity difference area data comprises the following steps:

[0094] Conducting refined force field analysis on vaginal peristaltic linkage dynamics data to obtain linkage refined force field data;

[0095] Performing morphological and structural difference analysis on the data of vaginal elasticity difference areas to obtain morphological and structural difference data;

[0096] Calculate the mean square deviation of vaginal wall thickness for the vaginal elasticity difference area data according to the morphological structure difference data, and obtain the mean square deviation data of vaginal wall thickness;

[0097] The linkage force gradient loss calculation is performed on the mean square error data of vaginal wall thickness according to the linkage refined force field data to obtain the peristaltic linkage force gradient loss data.

[0098] In an embodiment of the present invention, the vaginal peristaltic linkage power data obtained from step S343. The peristaltic linkage power data is spatially analyzed to obtain the distribution of forces in different regions. A force field model is established based on actual data, taking into account the changes in the magnitude and direction of forces in different regions. The specific values ​​and distribution characteristics of the forces in each region are calculated. The key parameters of the force field analysis, such as the maximum force, the minimum force, the average force, etc. are recorded. The regional data of vaginal elasticity differences obtained from step S341. The morphological and structural features of different regions are compared, including shape, size, surface characteristics, etc. The main morphological and structural differences between each region are determined. The morphological and structural difference types of each region are marked, such as shape irregularity, surface smoothness, etc. The morphological and structural difference data obtained from step 2. The wall thickness data in each region is obtained based on medical imaging or measured data. For the wall thickness data in each region, its mean square error is calculated, and the loss of the peristaltic linkage power gradient is calculated based on the linkage refined force field data and the mean square error data of the vaginal wall thickness. A model describing gradient loss is established, taking into account the influence of the vaginal wall thickness on the force field. Combine the force field data and wall thickness mean square error data to calculate the reduction of the peristaltic dynamic gradient. Analyze the reduction data to evaluate the reduction of the peristaltic dynamic gradient in different regions. Evaluate the influence of vaginal wall structure on peristaltic behavior and explore its application value in physiology and clinical practice.

[0099] The present invention conducts in-depth analysis of vaginal peristaltic linkage power data and generates linkage refined force field data. This includes careful calculation and visualization of the force field distribution of different regions of the vagina during peristalsis, providing detailed force field distribution of the vagina during peristalsis, helping to understand the stress state of each region, helping to reveal the mechanical behavior of different regions of the vagina during peristalsis, especially how to interact and transmit force, analyzing the data of the vaginal elastic difference region, identifying and quantifying the morphological and structural differences in different regions. This includes comparison of the geometric shape, size and elastic properties of each region. Revealing the morphological and structural differences in different elastic regions helps to understand their influence on peristalsis and force. Basic data is provided for further analysis of the functions and performances of different regions during peristalsis. Using the morphological and structural difference data, the mean square error of the vaginal wall thickness is calculated for the vaginal elastic difference region. This evaluates the uniformity and variation of the wall thickness in different regions. Providing the wall thickness uniformity data in different regions helps to understand its tolerance and adaptability under stress conditions. Revealing the thickness variation of the vaginal wall in different regions provides key data for further force and buffer analysis. Combined with the linkage refined force field data and vaginal wall thickness mean square error data, the peristaltic linkage force gradient loss is calculated. This analyzes the changes and losses of linkage force between different regions. It provides the changes in linkage force between different regions of the vagina during peristalsis, helping to understand the force transmission and loss mechanism. Evaluate the performance loss and impairment of the vaginal wall during dynamic peristalsis, providing data support for functional optimization.

[0100] Preferably, step S4 comprises the following steps:

[0101] Step S41: normalizing the elastic adhesion buffer force data and the vaginal wall tension adjustment degree data respectively to obtain adhesion buffer force normalized data and vaginal wall tension adjustment normalized data respectively;

[0102] Step S42: performing correlation analysis on vaginal wall tension regulation normalized data according to adhesion buffer force normalized data to obtain adhesion buffer-tension regulation correlation data;

[0103] Step S43: Using the naive Bayes algorithm to construct a vaginal relaxation evaluation model for the adhesion buffer-tension regulation correlation data to obtain a vaginal relaxation evaluation model.

[0104] In the embodiment of the present invention, the elastic adhesion buffer force data and the vaginal wall tension regulation data are normalized so as to unify the scale for analysis, and the elastic adhesion buffer force data obtained in step S345 and the vaginal wall tension regulation degree data obtained in step S31 are obtained. The data is scaled to the range of [0, 1], and the selected normalization method is applied to process the elastic adhesion buffer force data to obtain normalized data, and the same normalization method is applied to process the vaginal wall tension regulation data to obtain normalized data, and the processed elastic adhesion buffer force normalized data is saved, and the processed vaginal wall tension regulation normalized data is saved; through correlation analysis, the relationship between the adhesion buffer force and the vaginal wall tension regulation is studied, and the linear correlation between the two sets of data is calculated. If the data is not a linear relationship, the Spearman rank correlation coefficient is used to calculate the correlation coefficient between the adhesion buffer force data and the tension regulation data, and the correlation is tested for significance to determine whether the correlation has Statistical significance, using the naive Bayes algorithm, to build a vaginal relaxation assessment model based on the correlation data, the correlation analysis data obtained from step S42, the data is divided into a training set and a test set, usually 80% of the data is used as the training set and 20% as the test set, the features that contribute most to the vaginal relaxation assessment are selected, usually based on the significant features in the correlation data, a suitable data distribution hypothesis is selected, the accuracy of the model is verified using the test set data, and indicators such as the classification accuracy and confusion matrix are calculated using the naive Bayes algorithm of Gaussian distribution or multinomial distribution, the naive Bayes model is trained using the training set data, the conditional probability of the features is calculated, and the model parameters and feature selection are adjusted according to the evaluation results to optimize the model performance.

[0105] The present invention normalizes elastic adhesion buffer force data and vaginal wall tension adjustment degree data to obtain adhesion buffer force normalized data and vaginal wall tension adjustment normalized data, respectively. Normalization is intended to convert data of different scales and units into comparable standardized forms for further analysis and modeling. Through normalization, data of different dimensions are converted into dimensionless standardized data, which is convenient for direct comparison and analysis. Standardized data can reduce the complexity in subsequent calculations, improve calculation efficiency and accuracy of results. The adhesion buffer force normalized data was used to perform correlation analysis on the vaginal wall tension regulation normalized data to obtain the adhesion buffer-tension regulation correlation data. The correlation analysis aims to reveal the relationship between the two sets of data and determine the degree and pattern of correlation between them. Through correlation analysis, the relationship between elastic adhesion buffer force and vaginal wall tension regulation is revealed, which provides data support for understanding the mechanical and functional properties of the vagina. The correlation data provides key input features for subsequent evaluation models, improving the accuracy and reliability of the model. The adhesion buffer-tension regulation correlation data is processed using the naive Bayes algorithm to construct a vaginal relaxation evaluation model. The naive Bayes algorithm trains and evaluates the model by calculating the probabilistic relationship between each input feature and the output variable. The naive Bayes algorithm has efficient computing power when processing large amounts of data and features, and can quickly build and optimize the model. The constructed model can effectively evaluate the relaxation state of the vagina under different stresses and conditions, providing a powerful tool for functional analysis and optimization.

[0106] Preferably, step S43 includes the following steps:

[0107] Step S431: performing feature dimension reduction on the adhesion buffer-tension adjustment correlation data to obtain adhesion buffer-tension adjustment correlation dimension reduction data;

[0108] Step S432: using a preset correlation weight influence model to calculate the data partition ratio of the adhesion buffer-tension regulation correlation dimension reduction data to obtain correlation partition ratio data;

[0109] Step S433: dividing the adhesion buffer-tension regulation correlation dimension reduction data into data sets according to the correlation division ratio data, and obtaining an adhesion-tension correlation training set and an adhesion-tension correlation test set respectively;

[0110] Step S434: constructing an initial vaginal relaxation assessment model for the adhesion-tension correlation training set using a naive Bayes algorithm to obtain an initial vaginal relaxation assessment model;

[0111] Step S435: performing regularization processing on the adhesion-tension correlation test set to obtain an adhesion-tension regularized test set;

[0112] Step S436: Performing model testing on the initial vaginal relaxation assessment model according to the adhesion-tension regularization test set to obtain a vaginal relaxation assessment model.

[0113] In an embodiment of the present invention, the correlation data obtained from step S42 is often used for dimensionality reduction, and the variance of the retained data is maximized by converting to a new coordinate system. Data centralization. Calculate the covariance matrix. Calculate the eigenvalues ​​and eigenvectors of the covariance matrix. Select the eigenvectors corresponding to the first k largest eigenvalues ​​to form a new feature space. Map the data to the new feature space. Calculate the intra-class scatter matrix and the inter-class scatter matrix. Calculate the eigenvalues ​​and eigenvectors. Select the eigenvector corresponding to the largest eigenvalue to form a new feature space. Map the data to the new feature space. Used for dimensionality reduction and visualization of high-dimensional data to retain local structure. Utilize the preset correlation weight influence model to calculate the partition ratio of the data set to ensure the representativeness of the training set and the test set. Based on the business rules or prior knowledge, the dimensionality reduction data obtained from step S431 is used to construct a correlation weight influence model, define the importance of each feature, assign a weight to each feature to indicate its importance to the data partition, and calculate the partition ratio of the data set according to the weight and the distribution characteristics of the data. For example, a weighted average or other statistical method is used to obtain the division ratio of the training set and the test set, such as 80% of the data is used for training and 20% of the data is used for testing. The calculated division ratio data is recorded and saved, and the data is divided into a training set and a test set according to the calculated division ratio. The reduced dimension data obtained from step S431. The division ratio data obtained from step S432. According to the ratio data, the reduced dimension data is randomly or systematically divided into a training set and a test set. Ensure that each data set after division represents the overall distribution characteristics and maintains the balance of categories and features. The naive Bayes algorithm is used to construct an initial vaginal relaxation assessment model based on the training set data. From the training set data obtained in step S433, a suitable naive Bayes model (such as Gaussian Bayes, polynomial Bayes) is selected according to the distribution characteristics of the data. The naive Bayes model is trained using the training set data to estimate the conditional probability of the feature, and the trained initial model is saved for subsequent testing and optimization. The test set data is regularized to reduce overfitting and improve the generalization ability of the model. From the test set data obtained in step S433, the influence of unimportant features is reduced by adding a constraint on the sum of the absolute values ​​of the features. Select L1 or L2 regularization method according to the requirements of data and model. Regularize the test set data. Regularize the test set data. Regularized test set data obtained from step S435. Initial model obtained from step S434. Use the initial model to predict the regularized test set, calculate the performance indicators of the model, such as accuracy, precision, recall, F1 score, adjust the hyperparameters of the model (such as regularization strength, conditional probability estimation method, etc.) according to the test results, retrain the model with the adjusted parameters, improve the model performance, and save the optimized model as the final evaluation tool.

[0114] The present invention performs feature dimensionality reduction on adhesion buffer-tension regulation correlation data, reduces data dimensions, and obtains adhesion buffer-tension regulation correlation reduced dimensionality data. Commonly used dimensionality reduction techniques include principal component analysis (PCA), linear discriminant analysis (LDA), etc. The purpose is to retain the main features of the data and reduce redundant information. By reducing the dimensionality, the data structure is simplified, the complexity and time of model calculation are reduced, the analysis efficiency is improved, the redundancy and noise in the data are removed, and the accuracy and robustness of the subsequent model are improved. The preset correlation weight influence model is used to calculate the data partition ratio of the adhesion buffer-tension regulation correlation reduced dimensionality data to obtain the correlation partition ratio data. The correlation weight influence model is used to evaluate and allocate the importance and influence of different features in the data set partitioning. By calculating the correlation weight, the distribution of features in the data set is optimized, the effect of model training and testing is improved, the rationality and scientificity of the data set partitioning are ensured, and a reliable data basis is provided for the training and verification of the model. According to the correlation partition ratio data, The adhesion buffer-tension regulation correlation dimension reduction data is divided into a data set to obtain an adhesion-tension correlation training set and an adhesion-tension correlation test set. The data is usually divided into a training set and a test set according to a certain ratio (such as 70:30 or 80:20). The data set is reasonably divided into a training set and a test set to prevent the model from overfitting during the training process, improve the generalization ability of the model, and ensure that the test data set can effectively evaluate the performance and stability of the model. The naive Bayes algorithm is used to construct an initial vaginal relaxation evaluation model for the adhesion-tension correlation training set to obtain an initial vaginal relaxation evaluation model. The naive Bayes algorithm constructs an initial evaluation model by calculating the probability relationship between each feature in the training set. The naive Bayes algorithm is simple to calculate and can quickly construct an initial model. It is suitable for processing large-scale data. The initial model provides basic vaginal relaxation evaluation capabilities and provides a basis for subsequent optimization and testing. The adhesion-tension correlation test set is regularized to standardize data features to obtain an adhesion-tension regularized test set. Regularization processing usually includes normalization or standardization to make data features conform to a certain range or distribution to facilitate model testing and evaluation. Regularization makes the test data have a consistent scale and distribution, which is convenient for comparison with training data and ensures the reliability of test results. Regularization processing can reduce the risk of overfitting of the model on test data and improve generalization ability. According to the adhesion-tension regularization test set, the initial vaginal relaxation assessment model is tested and verified to obtain the final vaginal relaxation assessment model. The accuracy, precision and stability of the model are tested, and the model is adjusted and optimized as necessary according to the test results. Through testing and verification, the performance of the model is evaluated, including accuracy, precision, recall and other indicators to ensure the actual application effect of the model. According to the test results, the model is optimized and adjusted to improve the performance of the model in practical applications.

[0115] The beneficial effect of the present invention is to use an electromyographic sensor (EMG) to measure the electromyographic signals inside the vagina. These signals are electrophysiological data generated when the vaginal muscles are active, providing detailed electrical activity data of the vaginal muscles, which can capture the contraction and relaxation patterns of the muscles and help evaluate the functional status of the vaginal muscles, especially for individuals who need rehabilitation or functional improvement. A vaginoscope with a camera function is used to collect high-resolution images of the inside of the vagina, providing detailed images of the internal structure of the vagina, which helps doctors or researchers observe and analyze the anatomical features of the vagina. Accurate internal images can be obtained without surgery or other invasive operations. By processing the collected image data, a three-dimensional dynamic model that reflects the real-time activity of the vagina is constructed, which not only provides a static anatomical structure, but also can show the dynamic changes of the vagina in different states, providing more detailed information for complex movements or structural changes. Intuitive understanding supports further research. By analyzing the three-dimensional dynamic vaginal model, the movement patterns of the vagina in various activities (such as contraction and relaxation) are determined, and the movement patterns and patterns of the vagina in different situations are identified, which provides a basis for subsequent functional evaluation. By analyzing the dynamic changes of vaginal electromyographic signals and three-dimensional models, the intensity changes of vaginal contractions are evaluated, and quantitative data of vaginal contraction intensity are provided, which helps to evaluate the strength and function of vaginal muscles. Based on the data of vaginal contraction intensity changes, the tension regulation ability of the vaginal wall under different conditions is evaluated, which helps to understand the response and regulation ability of the vaginal wall under different pressure or movement conditions, which is crucial in the case of rehabilitation or strengthening of vaginal muscle function; by analyzing the tension regulation data of the vaginal wall under different states, its relaxation characteristics are evaluated. This involves analyzing the recovery speed and degree of vaginal muscles and tissues after contraction, accurately evaluating the recovery ability of vaginal tissues after various pressures or movements, which is very important for diagnosing vaginal relaxation and related dysfunctions. Using relaxation characteristic data, the elastic modulus of vaginal tissue is calculated. The elastic modulus reflects the elastic response of the tissue when it is subjected to stress, that is, the resistance to deformation, and provides a quantitative indicator of the elasticity of the vaginal tissue, which helps to evaluate its functionality. It analyzes the adhesion buffering capacity of the vagina when it is subjected to external forces (such as physical contact or pressure) and when the vagina is peristaltic, that is, its ability to resist external forces and return to its original state, and helps to evaluate the structural integrity and durability of the vaginal tissue under different stress conditions. The relationship between the elastic adhesion buffering force data and the tension regulation degree data of the vagina under different conditions is analyzed by statistical methods, which provides a multi-dimensional understanding of the function of the vagina. Combining the elastic and tension characteristics, it can more comprehensively evaluate the condition of the vagina. Based on the correlation data, the Naive Bayes algorithm is used to construct a vaginal relaxation assessment model. This model can predict and evaluate the degree of vaginal relaxation, providing an automated tool that can quickly and accurately evaluate the degree of vaginal relaxation and reduce the subjectivity and errors of manual evaluation.Therefore, the present invention is an improvement on a traditional method for constructing a vaginal relaxation evaluation model based on multi-source data, which solves the problems of inaccurate analysis of vaginal relaxation state and inaccurate analysis of vaginal movement in a traditional method for constructing a vaginal relaxation evaluation model based on multi-source data, and improves the precision of analysis of vaginal relaxation state and the accuracy of analysis of vaginal movement.

[0116] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0117] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for constructing a vaginal relaxation assessment model based on multi-source data, characterized in that: The following steps are involved: Step S1: using an electromyographic sensor to collect internal vaginal electromyographic signals to obtain internal vaginal electromyographic signals; The camera module built into the spectacles is used to collect image data of the interior of the vagina to obtain an image of the interior of the vagina; a real-time active three-dimensional vaginal model is constructed based on the image of the interior of the vagina to obtain a real-time active three-dimensional dynamic vaginal model; Step S2: analyzing the regularity of vaginal movement on the real-time active three-dimensional dynamic vaginal model to obtain vaginal movement regularity data; Analyzing the change of vaginal contraction intensity on the vaginal movement law data to obtain the change data of vaginal contraction intensity; evaluating the degree of vaginal wall tension regulation based on the change data of vaginal contraction intensity to obtain the degree of vaginal wall tension regulation data; Step S3: performing relaxation characteristic analysis on the vaginal wall tension adjustment degree data to obtain vaginal relaxation characteristic data; performing vaginal elastic modulus calculation based on the vaginal relaxation characteristic data to obtain vaginal elastic modulus data; Perform adhesion buffer force analysis according to the vaginal elastic modulus data to obtain elastic adhesion buffer force data; wherein step S3 includes: Step S31: performing baseline tension calculation on the vaginal wall tension adjustment degree data to obtain vaginal wall baseline tension data; Step S32: performing relaxation characteristic analysis on the vaginal wall tension adjustment degree data according to the vaginal wall baseline tension data to obtain vaginal relaxation characteristic data; Step S33: Calculating the vaginal elastic modulus according to the vaginal relaxation characteristic data and the vaginal contraction strength change data to obtain vaginal elastic modulus data; Step S34: performing adhesion buffer force analysis on the vaginal elastic modulus data according to the vaginal relaxation characteristic data to obtain elastic adhesion buffer force data; wherein step S34 includes: Step S341: dividing the vaginal elastic modulus data into vaginal elasticity difference regions according to the vaginal relaxation characteristic data to obtain vaginal elasticity difference region data; Step S342: performing peristalsis simulation of different elastic regions on the vaginal elasticity difference region data according to the vaginal relaxation characteristic data to obtain elastic region peristalsis simulation data; Step S343: performing linkage force analysis of different vaginal regions on the elastic region peristalsis simulation data to obtain vaginal peristalsis linkage force data; Step S344: performing coupling force gradient loss calculation according to the vaginal peristaltic coupling force data and the vaginal elasticity difference area data to obtain peristaltic coupling force gradient loss data; Step S345: performing adhesion buffer force analysis according to the vaginal peristaltic coupling force data and the peristaltic coupling force gradient loss data to obtain elastic adhesion buffer force data; Step S4: performing correlation analysis on the vaginal wall tension regulation degree data according to the elastic adhesion buffer force data to obtain adhesion buffer-tension regulation correlation data; constructing a vaginal relaxation evaluation model for the adhesion buffer-tension regulation correlation data using the naive Bayes algorithm to obtain a vaginal relaxation evaluation model.

2. The method for constructing a vaginal relaxation assessment model based on multi-source data according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: using an electromyographic sensor to collect internal vaginal electromyographic signals to obtain internal vaginal electromyographic signals; Step S12: collecting image data of the interior of the vagina through the built-in camera module of the spectacles to obtain an image of the interior of the vagina; Step S13: performing image exposure optimization processing on the image of the interior of the vagina to obtain an exposure-optimized image of the interior of the vagina; Step S14: constructing a real-time active three-dimensional vagina model based on the optimized exposure image of the vagina interior to obtain a real-time active three-dimensional dynamic vagina model.

3. The method for constructing a vaginal relaxation assessment model based on multi-source data according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: analyzing the regularity of vaginal movement on the real-time active three-dimensional dynamic vaginal model to obtain vaginal movement regularity data; Step S22: mapping the vaginal movement law data to the vaginal movement law myoelectric signal according to the vaginal internal myoelectric signal to obtain the vaginal movement law myoelectric signal; Step S23: performing spectrum conversion on the vaginal movement regular electromyographic signal to obtain a spectrum diagram of the vaginal movement electromyographic signal; Step S24: analyzing the change of vaginal contraction intensity on the vaginal movement law data according to the vaginal movement electromyographic signal spectrum to obtain vaginal contraction intensity change data; Step S25: Evaluate the degree of vaginal wall tension regulation based on the vaginal contraction intensity variation data and the vaginal movement electromyographic signal spectrum to obtain vaginal wall tension regulation degree data.

4. The method for constructing a vaginal relaxation assessment model based on multi-source data according to claim 3, characterized in that: Step S25 includes the following steps: Step S251: Plotting a contraction strength change curve for the vaginal contraction strength change data to obtain a vaginal contraction strength change curve; Step S252: performing strength change trend analysis according to the vaginal contraction strength change curve to obtain strength change trend data; Step S253: Calculating the contraction mechanical performance of the vaginal contraction strength change curve according to the strength change trend data to obtain vaginal contraction mechanical performance data; Step S254: Calculating the vaginal contraction fatigue threshold based on the vaginal contraction mechanical efficiency data and the vaginal movement electromyographic signal spectrum to obtain vaginal contraction fatigue threshold data; Step S255: Evaluate the degree of vaginal wall tension regulation according to the vaginal contraction fatigue critical data and the vaginal contraction mechanical efficiency data to obtain vaginal wall tension regulation degree data.

5. The method for constructing a vaginal relaxation assessment model based on multi-source data according to claim 4, characterized in that: The calculation of vaginal contraction fatigue threshold based on vaginal contraction mechanical efficiency data and vaginal electromyographic signal spectrum includes the following steps: Conducting contraction force symmetry analysis on vaginal contraction mechanical performance data to obtain contraction force symmetry data; The vaginal contraction efficiency attenuation is calculated based on the contraction force symmetry data and the vaginal contraction mechanical efficiency data to obtain the vaginal contraction efficiency attenuation data; The muscle activity decrease gradient is evaluated on the spectrum of vaginal movement electromyographic signals to obtain muscle activity decrease gradient data; The vaginal contraction fatigue critical value is calculated based on the vaginal contraction efficiency attenuation data and the muscle activity decline gradient data to obtain the vaginal contraction fatigue critical data.

6. The method for constructing a vaginal relaxation assessment model based on multi-source data according to claim 1, characterized in that: The calculation of the gradient loss of the coupling force based on the vaginal peristaltic coupling force data and the vaginal elasticity difference area data includes the following steps: Conducting refined force field analysis on vaginal peristaltic linkage dynamics data to obtain linkage refined force field data; Performing morphological and structural difference analysis on the data of vaginal elasticity difference areas to obtain morphological and structural difference data; Calculate the mean square deviation of vaginal wall thickness for the vaginal elasticity difference area data according to the morphological structure difference data, and obtain the mean square deviation data of vaginal wall thickness; The linkage force gradient loss calculation is performed on the mean square error data of vaginal wall thickness according to the linkage refined force field data to obtain the peristaltic linkage force gradient loss data.

7. The method for constructing a vaginal relaxation assessment model based on multi-source data according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: normalizing the elastic adhesion buffer force data and the vaginal wall tension adjustment degree data respectively to obtain adhesion buffer force normalized data and vaginal wall tension adjustment normalized data respectively; Step S42: performing correlation analysis on vaginal wall tension regulation normalized data according to adhesion buffer force normalized data to obtain adhesion buffer-tension regulation correlation data; Step S43: Using the naive Bayes algorithm to construct a vaginal relaxation evaluation model for the adhesion buffer-tension regulation correlation data to obtain a vaginal relaxation evaluation model.

8. The method for constructing a vaginal relaxation assessment model based on multi-source data according to claim 7, characterized in that: Step S43 includes the following steps: Step S431: performing feature dimension reduction on the adhesion buffer-tension adjustment correlation data to obtain adhesion buffer-tension adjustment correlation dimension reduction data; Step S432: using a preset correlation weight influence model to calculate the data partition ratio of the adhesion buffer-tension regulation correlation dimension reduction data to obtain correlation partition ratio data; Step S433: dividing the adhesion buffer-tension regulation correlation dimension reduction data into data sets according to the correlation division ratio data, and obtaining an adhesion-tension correlation training set and an adhesion-tension correlation test set respectively; Step S434: constructing an initial vaginal relaxation assessment model for the adhesion-tension correlation training set using a naive Bayes algorithm to obtain an initial vaginal relaxation assessment model; Step S435: performing regularization processing on the adhesion-tension correlation test set to obtain an adhesion-tension regularized test set; Step S436: Performing model testing on the initial vaginal relaxation assessment model according to the adhesion-tension regularization test set to obtain a vaginal relaxation assessment model.

Citation Information

Patent Citations

  • Female pelvic floor dysfunction disease risk early warning model and construction method and system thereof

    CN114464322A

  • Private evaluation method and system for female vagina relaxation state

    CN117612724A

  • Pelvic floor muscle strength automatic evaluation and training system

    CN117831773A

  • Female pelvic floor dysfunction disease risk early warning model, and construction method and system therefor

    WO2023138418A1