Polycystic ovary syndrome therapeutic effect evaluation system based on B-ultrasound images

Abstract

The invention discloses a polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images, which relates to the technical field of disease monitoring and evaluation, and includes a comprehensive evaluation platform, wherein the comprehensive evaluation platform is communicatively connected with an image acquisition module, a follicle detection module, a follicle tracking module, an ovarian morphology analysis module, an efficacy evaluation module, and a report generation module, wherein electrical signals are connected between the modules. The polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images performs time series analysis on B-ultrasound images of patients that have been reviewed for many times, tracks the growth of follicles and changes in ovarian morphology, thereby evaluating the effect of the treatment plan, uses the growth curve change rate and morphological change index to provide doctors with quantitative indicators, assists in understanding the treatment effect, and can effectively achieve quantitative evaluation of the treatment effect by measuring and recording the follicle count and ovarian morphology data of patients at different time points during the treatment process.

Description

Technical Field

[0001] The invention relates to the technical field of disease monitoring and evaluation, and in particular to a polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images. Background Art

[0002] Polycystic ovary syndrome is a common reproductive endocrine metabolic disease. Patients often show symptoms such as infrequent menstruation or amenorrhea, infertility, hirsutism, and obesity. One of the main characteristics of the disease is that both ovaries show polycystic enlargement, the ovaries are enlarged, the number of follicles increases but they are immature, and they cannot form dominant follicles and ovulate normally. B-ultrasound, as a non-invasive, convenient and economical examination method, plays an important role in the diagnosis of polycystic ovary syndrome. Through B-ultrasound examination, the morphology, size, number and size of follicles of the ovaries can be clearly observed, providing an important basis for the diagnosis of polycystic ovary syndrome.

[0003] In the prior art, follicle counting is one of the important indicators for evaluating polycystic ovary syndrome. However, when counting, the counting process is easily disturbed by the distribution and morphology of the follicles, resulting in counting errors. Moreover, for patients who undergo multiple follow-up examinations, it is inconvenient to observe the changing trend of ovarian morphology over time to evaluate the treatment effect. Therefore, how to analyze the distribution of follicles in the ovaries, improve the accuracy of follicle counting, and the changes in ovarian morphology over time are the problems we need to solve. For this purpose, a polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images is proposed. Summary of the invention

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: a polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images, comprising a comprehensive evaluation platform, wherein the comprehensive evaluation platform is communicatively connected with an image acquisition module, a follicle detection module, a follicle tracking module, an ovarian morphology analysis module, an efficacy evaluation module and a report generation module, wherein the modules are electrically connected with each other;

[0005] The image acquisition module is used to acquire high-quality ovarian B-ultrasound images from the B-ultrasound device, and perform pre-processing operations of noise reduction, enhancement, and artifact removal on the acquired ovarian B-ultrasound images to ensure the clarity and contrast of the images, so as to improve the image quality and facilitate subsequent feature extraction and recognition;

[0006] The follicle detection module automatically detects follicles in B-ultrasound images based on a deep learning algorithm, and extracts their position and size information, thereby realizing automatic identification and counting of follicles, providing an accurate number of follicles, and reducing errors caused by manual intervention and subjective judgment;

[0007] The follicle tracking module tracks the changes of the same follicle at different time points for patients who undergo multiple reexaminations, observes the growth and development process of the follicle, and evaluates the growth rate and maturation of the follicle;

[0008] The ovarian morphology analysis module is used to analyze the overall morphology, size, and contour characteristics of the ovaries, and evaluate their changes, providing quantitative indicators of changes in ovarian morphology over time, and providing a basis for evaluating treatment effects;

[0009] The efficacy evaluation module performs time series analysis on the B-ultrasound images of the patient's multiple reexaminations, analyzes the changing trends of follicle distribution and ovarian morphology in combination with the changes in follicle count and ovarian morphology, and evaluates the therapeutic effect of polycystic ovary syndrome;

[0010] The report generation module is used to perform data statistics and analysis on the results of follicle counting and ovarian morphology analysis, generate an evaluation report, and intuitively display the changes in follicle count and ovarian morphology in the form of charts and curves, which is convenient for doctors to understand and judge.

[0011] Preferably, in the image acquisition module, the process of acquiring the ovarian B-ultrasound image includes:

[0012] Connect the image acquisition module to the B-ultrasound device to ensure the stability and real-time performance of data transmission, obtain high-quality ovarian B-ultrasound images from the B-ultrasound device, ensure that the B-ultrasound device is in good working condition, and adjust the scanning parameters to obtain the best image quality;

[0013] Use an image acquisition card to convert the ovarian B-ultrasound image into a digital format, ensure that the image is not distorted during the digitization process, maintain high resolution, and perform preprocessing operations on the ovarian B-ultrasound image. The preprocessing operations include noise reduction, image enhancement, and artifact removal steps. Among them, noise reduction is used to reduce random noise in the image, improve the signal-to-noise ratio of the image, make the image smoother, and reduce interference information. Image enhancement is used to improve the visual effect of the image, make the structure and characteristics of the ovary more prominent, the boundaries clearer, and the internal details more obvious, which is convenient for doctors to observe and diagnose. Artifact removal is used to eliminate artifacts in the image, improve the accuracy and reliability of the image, and avoid false information generated in the image due to equipment failure, improper operation, or patient physical factors. Interference with the image quality;

[0014] Integrate the preprocessed ovarian B-ultrasound image data, add a unique index identifier to the preprocessed image metadata, associate patient information and examination date, ensure the traceability of image data, and store the processed images in the data warehouse to provide doctors with a convenient image calling interface so that they can quickly access and view images and make diagnostic calls when needed.

[0015] Preferably, in the follicle detection module, the process of automatic follicle identification and counting includes:

[0016] A large number of B-ultrasound images containing follicles are collected as training data. The training data covers women of different age groups and different physiological cycles to ensure the generalization ability of the model. The collected B-ultrasound images are finely annotated to mark the location and size characteristics of each follicle to obtain a feature data set, which is divided into a training set and a test set. Different age groups include adolescence, childbearing age, and menopause, and different physiological cycles include menstruation, ovulation, and luteal phase.

[0017] The follicle recognition model was trained using the training set data combined with the convolutional neural network. The model parameters were adjusted through the back-propagation algorithm to minimize the difference between the predicted bounding box and the true bounding box. The trained follicle recognition model was evaluated and tested using the test set data. The performance of the model was optimized by adjusting the model's hyperparameters.

[0018] The follicle recognition model is used to automatically identify the follicles in the image, and the bounding box of each follicle is given. According to the coordinates and size of the bounding box, the specific position and size information of each follicle is further extracted, and the total number of follicles is automatically calculated based on the extracted follicle position and size information.

[0019] Preferably, in the follicle tracking module, the process of evaluating the growth rate and maturation of follicles includes:

[0020] Collect B-ultrasound images of patients at different time points to ensure that there is enough data to track the changes in follicles, and use the follicle recognition model of the follicle detection module to detect follicles in the B-ultrasound images at each time point;

[0021] Extract the location and size features of each follicle, match the features of the same follicle at different time points, apply the tracking algorithm of feature point matching to track the movement and changes of the follicle, and monitor the growth and development process of the follicle over time;

[0022] Based on the matching analysis of the position and size characteristics of the same follicle, the follicle matching index is obtained to evaluate the degree of change of the follicle and analyze the degree of deviation of the growth and development of the follicle from the baseline value;

[0023] Based on the tracking and evaluation results, a growth curve of the follicle size over time is drawn, and the rate of change of the growth curve is calculated to evaluate whether the growth rate of the follicle is normal and to intuitively display the development of the follicle;

[0024] Analyze the relationship between the follicle growth curve and treatment measures, and predict the maturation of follicles based on the follicle growth rate and historical data.

[0025] Preferably, the calculation expression of the follicle matching index is:

[0026] ;

[0027] in, is the follicle matching index, is the number of time points tracked, For the The radius of the follicle at each time point, is the base follicle radius, which is the average radius of normal follicle development. For the The follicle area at each time point, is the benchmark follicle area, which is the average area of ​​normal follicle development. The closer the value is to 1, the closer the growth and development of the follicle is to the baseline value, that is, the more normal the development is. If the value is far from 1, it means that the growth and development of the follicles deviate greatly from the baseline value, and further diagnosis and treatment may be required;

[0028] The calculation expression of the growth curve change rate is:

[0029] ;

[0030] in, is the rate of change of the growth curve, indicating time The growth curve change rate, For the The radius of the follicle at each time point, is the base follicle radius, is the number of time points tracked, is the adjustment coefficient, is the average value of follicle radius at all time points, is the standard deviation of the follicle radius at all time points, if , indicating that the growth rate of the follicle is faster than the baseline value, , indicating that the growth rate of the follicles is consistent with the baseline value. , indicating that the growth rate of the follicle is slow. If the growth rate of the follicle is accelerated, then It will gradually increase in size, indicating that the follicle is developing well. If the growth rate of the follicle remains stable, then It will be close to 1, indicating that the follicles are developing normally. If the growth rate of the follicles slows down, then It will gradually decrease, which may indicate that the development of follicles is blocked.

[0031] Preferably, in the ovarian morphology analysis module, the process of evaluating the changes in ovarian morphology includes:

[0032] Extracting B-ultrasound images of ovaries from the data warehouse, and using image segmentation algorithms to separate the ovaries from the background, so as to accurately identify the boundaries of the ovaries;

[0033] The edge detection algorithm is used to extract the outline of the ovary and obtain its two-dimensional shape. Medical image processing software is used to measure the size of the ovary, record the length, width, and thickness parameters of the ovary, form quantitative indicators, analyze the edge outline of the ovary, observe whether the overall shape of the ovary is regular, determine whether there is any abnormal shape, identify irregular and polycystic changes, and analyze the abnormal characteristics of the ovarian morphology;

[0034] The ovarian morphology data and characteristic data at different time points were entered into the database, and the ovarian morphology characteristic data of the patients at different time points were compared and analyzed to evaluate the changes in ovarian morphology over time and the trend of morphological changes;

[0035] The morphological change index was calculated based on the degree of change in contour features and the growth rate of ovarian size to quantitatively evaluate the changes in ovarian morphology and its growth rate.

[0036] Preferably, the calculation expression of the morphological change index is:

[0037] ;

[0038] in, is the morphological change index, is the number of time points tracked, For the The ovarian area at each time point, is the baseline ovarian area, which is the average area of ​​a normal ovary. For the The square root of the ovarian area at each time point, is the square root of the baseline ovarian area, is the coefficient of variation of the ovarian radius, that is, the ratio of the standard deviation of the ovarian radius to the mean value of the ovarian radius. is the weight coefficient of the coefficient of variation of the ovarian radius, used to adjust right The degree of influence The closer the value is to 1, the closer the growth and morphological changes of the ovary are to the normal physiological range, that is, the more normal the development is.

[0039] Preferably, in the efficacy evaluation module, the process of evaluating the therapeutic effect of polycystic ovary syndrome includes:

[0040] Traverse the B-ultrasound images and follicle count data of patients at different time points during the treatment process in the data warehouse, and enter the follicle count and ovarian morphology data at different time points into the database, perform time series comparison, and analyze the changing trends of follicle distribution and ovarian morphology;

[0041] According to the changes in follicle counts, the change rate of the follicle growth curve is calculated to evaluate the growth rate and development of the follicles. According to the changes in ovarian morphological parameters, the morphological change index is calculated to quantify the degree of change in ovarian morphology.

[0042] The diagnosis and treatment evaluation coefficient was calculated based on the change rate of the follicle growth curve and the morphological change index, and the patient's status before and after diagnosis and treatment was compared and judged. The changing trends of the follicles and ovaries before and after diagnosis and treatment were analyzed to further evaluate the treatment effect of PCOS.

[0043] Preferably, the calculation expression of the diagnosis and treatment evaluation coefficient is:

[0044] ;

[0045] in, is the diagnosis and treatment evaluation coefficient, is the number of time points tracked, For the The change rate of the follicle growth curve at each time point, is the rate of change of the baseline growth curve, which is determined based on the average rate of change of normal follicle development. For the The morphological change index at each time point, is the weight coefficient, which is used to adjust the relative importance of the growth curve change rate and morphological change index in the evaluation coefficient. Its value range is between 0 and 1. The closer the value is to 1, the better the treatment effect is, and the closer the follicle growth and morphological changes are to the normal physiological state. The smaller the value, the worse the treatment effect, and the greater the deviation between the follicle growth and morphological changes and the normal physiological state. The square sum function is used to emphasize the deviation between the growth curve change rate at each time point and the baseline value. The square root of the sum result is taken to obtain an average growth curve change rate. The product function is used to emphasize the morphological change index at each time point, and the product result is taken. square root to obtain an average morphological change index.

[0046] Preferably, in the report generation module, the generation process of the evaluation report includes:

[0047] Traverse the different time points of the patient's treatment process from the data warehouse, collect the data of follicle count, ovarian morphological parameters, growth curve change rate and morphological change index, and sort the data in chronological order;

[0048] Perform statistical analysis on the collected data and generate charts based on the statistical results to show the changing trends of follicle counts and ovarian morphology. At the same time, draw follicle growth curves and ovarian morphology change curves to show the changes over time, helping doctors understand the dynamic changes of follicle and ovarian morphology;

[0049] Mark the data points with abnormal changes in the chart, and provide a text description of the treatment effect and trend evaluation results in combination with the diagnosis and treatment evaluation coefficient;

[0050] Integrate all information to generate an evaluation report that describes the results of follicle counting and ovarian morphology analysis, and output the report in the form of an electronic or paper document for the doctor's reference and decision-making.

[0051] The present invention provides a polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images. It has the following beneficial effects:

[0052] 1. This PCOS efficacy evaluation system based on B-ultrasound images tracks the growth of follicles and changes in ovarian morphology through time series analysis of B-ultrasound images of patients reviewed multiple times, thereby evaluating the effectiveness of treatment plans. It uses the growth curve change rate and morphological change index to provide doctors with quantitative indicators to assist in understanding the treatment effect. By measuring and recording the follicle count and ovarian morphology data of patients at different time points during the treatment process, it can effectively achieve quantitative evaluation of the treatment effect.

[0053] 2. This PCOS efficacy evaluation system based on B-ultrasound images can timely detect potential problems in patients' condition changes, identify and take intervention measures in a timely manner through continuous monitoring and analysis of follicle counts and ovarian morphology data. By analyzing the patient's B-ultrasound image data, it can deeply understand the characteristics of the patient's follicle growth and ovarian morphology changes, and intuitively display the changes in follicle counts and ovarian morphology in the form of charts and curves, making complex medical data easy to understand and communicate, and facilitating the explanation of changing trends in the condition to patients, thereby improving treatment compliance and effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a module structure diagram of the polycystic ovary syndrome therapeutic effect evaluation system based on B-ultrasound images of the present invention;

[0055] Figure 2 A flow chart of the present invention for evaluating the growth rate and maturation of follicles;

[0056] Figure 3 This is a flow chart of the present invention for evaluating changes in ovarian morphology;

[0057] Figure 4 The present invention is a flow chart for evaluating the therapeutic effect of polycystic ovary syndrome. DETAILED DESCRIPTION

[0058] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0059] The first embodiment, as Figures 1 to 3 As shown, the present invention provides a technical solution: a polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images, comprising a comprehensive evaluation platform, wherein the comprehensive evaluation platform is communicatively connected with an image acquisition module, a follicle detection module, a follicle tracking module, an ovarian morphology analysis module, an efficacy evaluation module and a report generation module, wherein the modules are electrically connected with each other;

[0060] The image acquisition module is used to acquire high-quality ovarian B-ultrasound images from the B-ultrasound device, and perform pre-processing operations such as noise reduction, enhancement, and artifact removal on the acquired ovarian B-ultrasound images to ensure the clarity and contrast of the image, so as to improve the image quality and facilitate subsequent feature extraction and recognition. The image acquisition module is connected to the B-ultrasound device to ensure the stability and real-time performance of data transmission, acquire high-quality ovarian B-ultrasound images from the B-ultrasound device, ensure that the B-ultrasound device is in good working condition, adjust scanning parameters to obtain the best image quality, use an image acquisition card to convert the ovarian B-ultrasound images into a digital format, ensure that the images are not distorted during the digitization process, maintain high resolution, and perform pre-processing operations on the ovarian B-ultrasound images. The pre-processing operations include the steps of noise reduction, image enhancement, and artifact removal, wherein noise reduction is used to reduce the Random noise, improve the signal-to-noise ratio of the image, make the image smoother, reduce interference information, use image enhancement processing to improve the visual effect of the image, make the structure and characteristics of the ovary more prominent, the boundary clearer, and the internal details more obvious, which is convenient for doctors to observe and diagnose. Use artifact removal processing to eliminate artifacts in the image, improve the accuracy and reliability of the image, and avoid false information generated in the image due to equipment failure, improper operation or patient physical factors. Interference with image quality, integrate pre-processed ovarian B-ultrasound image data, add a unique index identifier to the pre-processed image metadata, associate patient information and examination date, ensure the traceability of image data, and store the processed images in the data warehouse to provide doctors with a convenient image call interface so that they can quickly access and view images when needed and make diagnostic calls;

[0061] The follicle detection module automatically detects follicles in B-ultrasound images based on deep learning algorithms, extracts their location and size information, realizes automatic identification and counting of follicles, provides accurate follicle counts, reduces errors caused by manual intervention and subjective judgment, collects a large number of B-ultrasound images containing follicles as training data, and the training data covers women of different age groups and different physiological cycles to ensure the generalization ability of the model. The collected B-ultrasound images are finely annotated to mark the location and size characteristics of each follicle to obtain a feature data set, which is divided into a training set and a test set. Different age groups include adolescence, childbearing age, and menopause. Different physiological cycles include menstrual period, ovulation period and luteal period. The follicle recognition model is trained by combining the training set data with the convolutional neural network. The model parameters are adjusted by the back propagation algorithm to minimize the difference between the predicted bounding box and the true bounding box. The trained follicle recognition model is evaluated and tested by the test set data. The performance of the model is optimized by adjusting the hyperparameters of the model. The follicle recognition model is used to automatically identify the follicles in the image and give the bounding box of each follicle. According to the coordinates and size of the bounding box, the specific position and size information of each follicle is further extracted. According to the extracted follicle position and size information, the total number of follicles is automatically calculated.

[0062] The follicle tracking module tracks the changes of the same follicle at different time points for patients who have been reviewed multiple times, observes the growth and development process of the follicle, evaluates the growth rate and maturity of the follicle, collects B-ultrasound images of patients at different time points, ensures that there is enough data to track the changes of the follicles, uses the follicle recognition model of the follicle detection module to detect the follicles in the B-ultrasound images at each time point, extracts the position and size characteristics of each follicle, and matches the characteristics of the same follicle at different time points, and applies the tracking algorithm of feature point matching to track the movement and changes of the follicles, monitors the growth and development process of the follicles over time, obtains the follicle matching index based on the matching analysis of the position and size characteristics of the same follicle, evaluates the degree of change of the follicle, analyzes the degree of deviation of the growth and development of the follicle from the baseline value, draws the growth curve of the follicle size over time based on the tracking results and evaluation results, calculates the rate of change of the growth curve, so as to evaluate whether the growth rate of the follicle is normal, intuitively displays the development of the follicle, analyzes the relationship between the follicle growth curve and the treatment measures, and predicts the maturity of the follicle based on the growth rate of the follicle and historical data;

[0063] Furthermore, the calculation expression of the follicle matching index is:

[0064] ;

[0065] in, is the follicle matching index, is the number of time points tracked, For the The radius of the follicle at each time point, is the base follicle radius, which is the average radius of normal follicle development. For the The follicle area at each time point, is the benchmark follicle area, which is the average area of ​​normal follicle development. The closer the value is to 1, the closer the growth and development of the follicle is to the baseline value, that is, the more normal the development is. If the value is far from 1, it means that the growth and development of the follicles deviate greatly from the baseline value, and further diagnosis and treatment may be required;

[0066] The calculation expression of the growth curve change rate is:

[0067] ;

[0068] in, is the rate of change of the growth curve, indicating time The growth curve change rate, For the The radius of the follicle at each time point, is the base follicle radius, is the number of time points tracked, is the adjustment coefficient, is the average value of follicle radius at all time points, is the standard deviation of the follicle radius at all time points, if , indicating that the growth rate of the follicle is faster than the baseline value, , indicating that the growth rate of the follicles is consistent with the baseline value. , indicating that the growth rate of the follicle is slow. If the growth rate of the follicle is accelerated, then It will gradually increase in size, indicating that the follicle is developing well. If the growth rate of the follicle remains stable, then It will be close to 1, indicating that the follicles are developing normally. If the growth rate of the follicles slows down, then It will gradually decrease, which may indicate that the development of follicles is blocked;

[0069] The ovarian morphology analysis module is used to analyze the overall morphology, size, and contour characteristics of the ovaries, and evaluate their changes, provide quantitative indicators of ovarian morphology changes over time, and provide a basis for evaluating the treatment effect. The B-ultrasound image of the ovaries is extracted from the data warehouse, and the ovaries are separated from the background using an image segmentation algorithm to accurately identify the boundaries of the ovaries. The contours of the ovaries are extracted using an edge detection algorithm to obtain their two-dimensional morphology. The size of the ovaries is measured using medical image processing software, and the length, width, and thickness parameters of the ovaries are recorded to form quantitative indicators. The edge contours of the ovaries are analyzed, and the overall morphology of the ovaries is observed to be regular, and abnormal morphology is determined. Identify irregular and polycystic changes, analyze abnormal characteristics of ovarian morphology, enter ovarian morphology data and feature data at different time points into the database, compare and analyze the ovarian morphology feature data of patients at different time points, evaluate the changes in ovarian morphology over time and the trend of morphological changes, and calculate the morphological change index based on the degree of change in contour features and the growth rate of ovarian size to quantitatively evaluate the changes in ovarian morphology and its growth rate.

[0070] Furthermore, the calculation expression of the morphological change index is:

[0071] ;

[0072] in, is the morphological change index, is the number of time points tracked, For the The ovarian area at each time point, is the baseline ovarian area, which is the average area of ​​a normal ovary. For the The square root of the ovarian area at each time point, is the square root of the baseline ovarian area, is the coefficient of variation of the ovarian radius, that is, the ratio of the standard deviation of the ovarian radius to the mean value of the ovarian radius. is the weight coefficient of the coefficient of variation of the ovarian radius, used to adjust right The degree of influence The closer the value is to 1, the closer the growth and morphological changes of the ovary are to the normal physiological range, that is, the more normal the development is. If the value is far from 1, it means that the growth and morphological changes of the ovary deviate greatly from the normal physiological range and further diagnosis and treatment may be needed;

[0073] The efficacy evaluation module conducts time series analysis on the B-ultrasound images of patients who have undergone multiple reexaminations, analyzes the changing trends of follicle distribution and ovarian morphology in combination with the changes in follicle count and ovarian morphology, and evaluates the therapeutic effect of PCOS;

[0074] The report generation module is used to collect data and analyze the results of follicle counting and ovarian morphology analysis, generate an evaluation report, and intuitively display the changes in follicle count and ovarian morphology in the form of charts and curves to facilitate doctors' understanding and judgment.

[0075] The second embodiment is based on the first embodiment. Figure 4 As shown in the efficacy evaluation module, the process of evaluating the treatment effect of polycystic ovary syndrome includes:

[0076] Traverse the B-ultrasound images and follicle count data of patients at different time points during the treatment process in the data warehouse, and enter the follicle count and ovarian morphology data at different time points into the database for time series comparison, analyze the change trend of follicle distribution and ovarian morphology, calculate the change rate of follicle growth curve according to the change of follicle count, evaluate the growth rate and development of follicles, and calculate the morphological change index according to the change of ovarian morphological parameters, quantify the degree of change of ovarian morphology, calculate the diagnosis and treatment evaluation coefficient by combining the change rate of follicle growth curve and morphological change index, compare and judge the status of patients before and after diagnosis and treatment, and analyze the change trend of follicles and ovaries over time before and after diagnosis and treatment, and further evaluate the treatment effect of polycystic ovary syndrome;

[0077] Furthermore, the calculation expression of the diagnosis and treatment evaluation coefficient is:

[0078] ;

[0079] in, is the diagnosis and treatment evaluation coefficient, is the number of time points tracked, For the The change rate of the follicle growth curve at each time point, is the rate of change of the baseline growth curve, which is determined based on the average rate of change of normal follicle development. For the The morphological change index at each time point, is the weight coefficient, which is used to adjust the relative importance of the growth curve change rate and morphological change index in the evaluation coefficient. Its value range is between 0 and 1. The closer the value is to 1, the better the treatment effect is, and the closer the follicle growth and morphological changes are to the normal physiological state. The smaller the value, the worse the treatment effect, and the greater the deviation between the follicle growth and morphological changes and the normal physiological state. The square sum function is used to emphasize the deviation between the growth curve change rate at each time point and the baseline value. The square root of the sum result is taken to obtain an average growth curve change rate. The product function is used to emphasize the morphological change index at each time point, and the product result is taken. square root to obtain an average morphological change index;

[0080] In the report generation module, the generation process of the evaluation report includes:

[0081] The data warehouse is traversed at different time points during the treatment of patients, and the data of follicle count, ovarian morphological parameters, growth curve change rate and morphological change index are collected. The data are sorted in chronological order, and statistical analysis is performed on the collected data. According to the statistical results, a chart is generated to show the changing trend of follicle count and ovarian morphology. At the same time, the follicle growth curve and ovarian morphological change curve are drawn to show the changes over time, so as to assist doctors in understanding the dynamic changes of follicle and ovarian morphology. The data points with abnormal changes are marked in the chart, and combined with the diagnosis and treatment evaluation coefficient, a text description of the treatment effect and trend evaluation results are provided. All information is integrated to generate an evaluation report, which describes the results of follicle count and ovarian morphological analysis, and the report is output in the form of electronic or paper documents for doctors' reference and decision-making. By performing time series analysis on the B-ultrasound images of patients reviewed multiple times, the growth of follicles and changes in ovarian morphology are tracked, so as to evaluate the effect of the treatment plan. The growth curve change rate and morphological change index are used to provide doctors with quantitative indicators to assist in understanding the treatment effect. By measuring and recording the follicle count and ovarian morphological data of patients at different time points during the treatment process, quantitative evaluation of the treatment effect can be effectively achieved.

[0082] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. A polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images, including a comprehensive evaluation platform, characterized in that: The comprehensive evaluation platform is communicatively connected with an image acquisition module, a follicle detection module, a follicle tracking module, an ovarian morphology analysis module, an efficacy evaluation module, and a report generation module, wherein electrical signals are connected between the modules; The image acquisition module is used to acquire high-quality ovarian B-ultrasound images from the B-ultrasound device and perform preprocessing operations on the acquired ovarian B-ultrasound images. In the image acquisition module, the process of acquiring the ovarian B-ultrasound images includes: Connecting the image acquisition module to the B-ultrasound device to acquire the ovarian B-ultrasound image from the B-ultrasound device; Using an image acquisition card, converting the ovarian B-ultrasound image into a digital format, and performing a preprocessing operation on the ovarian B-ultrasound image, the preprocessing operation includes the steps of noise reduction, image enhancement and artifact removal; Integrate the preprocessed ovarian B-ultrasound image data, add a unique index identifier to the preprocessed image metadata, associate patient information and examination date, and store the processed images in a data warehouse; The follicle detection module automatically detects follicles in B-ultrasound images based on a deep learning algorithm, and extracts their position and size information to achieve automatic identification and counting of follicles. In the follicle detection module, the process of automatic identification and counting of follicles includes: B-ultrasound images containing follicles are collected as training data. The training data covers women of different age groups and different physiological cycles. The collected B-ultrasound images are annotated to mark the location and size characteristics of each follicle to obtain a feature data set, which is divided into a training set and a test set. Different age groups include adolescence, childbearing age, and menopause, and different physiological cycles include menstruation, ovulation, and luteal phase. The follicle recognition model was trained using the training set data combined with a convolutional neural network. The model parameters were adjusted through the back-propagation algorithm to minimize the difference between the predicted bounding box and the true bounding box. The trained follicle recognition model was evaluated and tested using the test set data. The follicle recognition model is used to automatically identify the follicles in the image, and the bounding box of each follicle is given. According to the coordinates and size of the bounding box, the specific position and size information of each follicle is further extracted, and the total number of follicles is automatically calculated according to the extracted follicle position and size information; The follicle tracking module tracks the changes of the same follicle at different time points for patients who have been reviewed multiple times, observes the growth and development process of the follicle, and evaluates the growth rate and maturation of the follicle. In the follicle tracking module, the process of evaluating the growth rate and maturation of the follicle includes: Collect B-ultrasound images of patients at different time points, and use the follicle recognition model of the follicle detection module to detect follicles in the B-ultrasound images at each time point; Extract the location and size features of each follicle, match the features of the same follicle at different time points, apply the tracking algorithm of feature point matching to track the movement and changes of the follicle, and monitor the growth and development process of the follicle over time; Based on the matching analysis of the position and size characteristics of the same follicle, the follicle matching index is obtained to evaluate the degree of change of the follicle and analyze the degree of deviation of the growth and development of the follicle from the baseline value; Based on the tracking and evaluation results, a growth curve of the follicle size over time is drawn, and the rate of change of the growth curve is calculated to evaluate whether the growth rate of the follicle is normal and to intuitively display the development of the follicle; The relationship between the follicle growth curve and the treatment measures is analyzed, and the maturation of the follicle is predicted based on the growth rate of the follicle and historical data. The calculation expression of the follicle matching index is: ; in, is the follicle matching index, is the number of time points tracked, For the The radius of the follicle at each time point, is the base follicle radius, which is the average radius of normal follicle development. For the The follicle area at each time point, is the benchmark follicle area, which is the average area of ​​normal follicle development. The closer the value is to 1, the closer the growth and development of the follicle is to the baseline value, that is, the more normal the development is; The calculation expression of the growth curve change rate is: ; in, is the rate of change of the growth curve, indicating time The growth curve change rate, For the The radius of the follicle at each time point, is the reference follicle radius, is the number of time points tracked, is the adjustment coefficient, is the average value of follicle radius at all time points, is the standard deviation of the follicle radius at all time points; The ovarian morphology analysis module is used to analyze the overall morphology, size, and contour characteristics of the ovaries and evaluate their changes; The efficacy evaluation module performs time series analysis on the B-ultrasound images of the patient's multiple reexaminations, analyzes the changing trends of follicle distribution and ovarian morphology in combination with the changes in follicle count and ovarian morphology, and evaluates the therapeutic effect of polycystic ovary syndrome; The report generation module is used to perform data statistics and analysis on the results of follicle counting and ovarian morphology analysis to generate an evaluation report.

2. The PCOS therapeutic effect evaluation system based on B-ultrasound images according to claim 1, characterized in that: In the ovarian morphology analysis module, the process of evaluating the changes in ovarian morphology includes: Extracting B-ultrasound images of ovaries from the data warehouse and using image segmentation algorithms to separate the ovaries from the background; The edge detection algorithm is used to extract the outline of the ovary and obtain its two-dimensional shape. Medical image processing software is used to measure the size of the ovary, record the length, width, and thickness parameters of the ovary, form quantitative indicators, analyze the edge outline of the ovary, observe whether the overall shape of the ovary is regular, determine whether there is any abnormal shape, identify irregular and polycystic changes, and analyze the abnormal characteristics of the ovarian morphology; The ovarian morphology data and characteristic data at different time points were entered into the database, and the ovarian morphology characteristic data of the patients at different time points were compared and analyzed to evaluate the changes in ovarian morphology over time and the trend of morphological changes; The morphological change index was calculated based on the degree of change in contour features and the growth rate of ovarian size to quantitatively evaluate the changes in ovarian morphology and its growth rate.

3. The PCOS therapeutic effect evaluation system based on B-ultrasound images according to claim 2, characterized in that: The calculation expression of the morphological change index is: ; in, is the morphological change index, is the number of time points tracked, For the The ovarian area at each time point, is the baseline ovarian area, which is the average area of ​​a normal ovary. For the The square root of the ovarian area at each time point, is the square root of the baseline ovarian area, is the coefficient of variation of the ovarian radius, that is, the ratio of the standard deviation of the ovarian radius to the mean value of the ovarian radius. is the weight coefficient of the coefficient of variation of ovarian radius, used to adjust right The degree of influence, The closer the value is to 1, the closer the growth and morphological changes of the ovary are to the normal physiological range, that is, the more normal the development is.

4. The PCOS therapeutic effect evaluation system based on B-ultrasound images according to claim 3 is characterized in that: In the efficacy evaluation module, the process of evaluating the therapeutic effect of polycystic ovary syndrome includes: Traverse the B-ultrasound images and follicle count data of patients at different time points during the treatment process in the data warehouse, and enter the follicle count and ovarian morphology data at different time points into the database, perform time series comparison, and analyze the changing trends of follicle distribution and ovarian morphology; According to the changes in follicle counts, the change rate of the follicle growth curve is calculated to evaluate the growth rate and development of the follicles. According to the changes in ovarian morphological parameters, the morphological change index is calculated to quantify the degree of change in ovarian morphology. The diagnosis and treatment evaluation coefficient was calculated based on the change rate of the follicle growth curve and the morphological change index, and the patient's status before and after diagnosis and treatment was compared and judged. The changing trends of the follicles and ovaries before and after diagnosis and treatment were analyzed to further evaluate the treatment effect of PCOS.

5. The PCOS therapeutic effect evaluation system based on B-ultrasound images according to claim 4 is characterized in that: The calculation expression of the diagnosis and treatment evaluation coefficient is: ; in, is the diagnosis and treatment evaluation coefficient, is the number of time points tracked, For the The change rate of the follicle growth curve at each time point, is the rate of change of the baseline growth curve, which is determined based on the average rate of change of normal follicle development. For the The morphological change index at each time point, is the weight coefficient, The closer the value is to 1, the better the treatment effect is, and the closer the follicle growth and morphological changes are to the normal physiological state.

6. The PCOS therapeutic effect evaluation system based on B-ultrasound images according to claim 5, characterized in that: In the report generation module, the generation process of the evaluation report includes: Traverse the different time points of the patient's treatment process from the data warehouse, collect the data of follicle count, ovarian morphological parameters, growth curve change rate and morphological change index, and sort the data in chronological order; Perform statistical analysis on the collected data and generate charts based on the statistical results to show the changing trends of follicle counts and ovarian morphology. At the same time, draw follicle growth curves and ovarian morphology change curves to show the changes over time; Mark the data points with abnormal changes in the chart, and provide a text description of the treatment effect and trend evaluation results in combination with the diagnosis and treatment evaluation coefficient; All information is integrated into an evaluation report describing the results of follicle count and ovarian morphology analysis, and the report is output in the form of an electronic or paper document.

Images