System for evaluating curative effect of polycystic ovarian syndrome based on B ultrasonic image

Abstract

The invention discloses a polycystic ovarian syndrome curative effect evaluation system based on a B ultrasonic image, and relates to the technical field of polycystic ovarian syndrome curative effect evaluation, the system comprises a curative effect data acquisition module, a feature extraction module, a curative effect level monitoring module, a curative effect feedback evaluation module and an execution module, the data acquisition module is used for acquiring B-ultrasonic data, hormone data and physiological data of a polycystic ovarian syndrome patient, the feature extraction module is used for preprocessing the acquired data and carrying out feature extraction on the preprocessed B-ultrasonic data, and the execution module is used for updating the curative effect level of the polycystic ovarian syndrome patient through the curative effect feedback evaluation model and sending out a corresponding signal. According to the system, the curative effect data acquisition technology, the feature extraction technology, the multi-model construction technology and the modern information technology are closely combined, and real-time and comprehensive monitoring on the curative effect of the polycystic ovarian syndrome of a patient is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of evaluating the efficacy of polycystic ovary syndrome, and particularly to a system for evaluating the efficacy of polycystic ovary syndrome based on B-ultrasound images. Background Art

[0002] Polycystic ovary syndrome is a common gynecological endocrine disease that seriously affects women's health and fertility. In clinical treatment, it is extremely crucial to accurately evaluate the efficacy of polycystic ovary syndrome. The traditional diagnosis and efficacy evaluation of polycystic ovary syndrome are not comprehensive enough. Clinicians often make subjective judgments based on experience, resulting in differences in evaluation results and affecting the accurate formulation and adjustment of treatment plans. There are limitations, with strong subjectivity in symptoms and difficulty in intuitively reflecting the real changes in ovarian morphology and function. B-ultrasound imaging technology has brought new opportunities for the evaluation of polycystic ovary syndrome. Therefore, it is urgent to develop a standardized and scientific system for evaluating the efficacy of polycystic ovary syndrome based on B-ultrasound images to improve the evaluation accuracy and provide better medical services for patients;

[0003] Although there have been great advancements in the prior art in the direction of evaluating the efficacy of polycystic ovary syndrome, there are still some problems to be optimized. It is difficult for the prior art to combine B-ultrasound images and various other efficacy evaluation factors to comprehensively evaluate the efficacy of polycystic ovary syndrome, resulting in improper monitoring of the patient's treatment process and poor treatment effects of polycystic ovary syndrome in patients. Summary of the Invention

[0004] To achieve the above objectives, the present invention is realized through the following technical solutions: A system for evaluating the efficacy of polycystic ovary syndrome based on B-ultrasound images, including an efficacy data acquisition module, a feature extraction module, an efficacy level monitoring module, an efficacy feedback evaluation module, and an execution module, wherein each module is communicatively connected;

[0005] The efficacy data acquisition module collects the efficacy data of polycystic ovary syndrome patients, including B-ultrasound data, hormone data, and physiological data, providing data support for the realization of the functions of subsequent modules;

[0006] The feature extraction module preprocesses the collected efficacy data and extracts features from the preprocessed B-ultrasound data;

[0007] The efficacy level monitoring module analyzes the efficacy of polycystic ovary syndrome based on the B-ultrasound data after feature extraction, and further obtains the efficacy level of polycystic ovary syndrome patients by combining the multiple linear regression algorithm;

[0008] The efficacy feedback evaluation module uses the preprocessed hormone data and physiological data to respectively evaluate the efficacy of polycystic ovary syndrome patients and constructs an efficacy feedback evaluation model;

[0009] The execution module updates the efficacy level of patients with polycystic ovary syndrome based on the output result of the efficacy feedback evaluation model and sends out corresponding signals, solving the problem that it is difficult to comprehensively evaluate the efficacy of polycystic ovary syndrome by combining B-ultrasound images and various other efficacy evaluation factors in the prior art.

[0010] A further improvement of the technical solution of the present invention lies in: in the efficacy data acquisition module, the process of collecting the efficacy data of patients with polycystic ovary syndrome includes:

[0011] Deploy different types of acquisition devices to collect B-ultrasound data, hormone data and physiological data of patients with polycystic ovary syndrome. Among them, the acquisition devices include a three-dimensional color Doppler ultrasound instrument, an automatic chemiluminescence immunoassay analyzer, a thermometer and a height and weight measuring instrument;

[0012] The B-ultrasound data is the three-dimensional B-ultrasound image of the patient's ovary; the hormone data includes the testosterone concentration, estradiol concentration, luteinizing hormone concentration and follicle-stimulating hormone concentration of the patient; the physiological data includes the patient's body temperature and BMI value;

[0013] Use a three-dimensional color Doppler ultrasound instrument to collect the three-dimensional B-ultrasound image of the ovary of patients with polycystic ovary syndrome; use an automatic chemiluminescence immunoassay analyzer to collect the testosterone concentration, estradiol concentration, luteinizing hormone concentration and follicle-stimulating hormone concentration of patients with polycystic ovary syndrome; use a thermometer to collect the body temperature of patients with polycystic ovary syndrome; use a height and weight measuring instrument to collect the BMI value of patients with polycystic ovary syndrome.

[0014] A further improvement of the technical solution of the present invention lies in: in the feature extraction module, the process of preprocessing the collected efficacy data and extracting features from the preprocessed B-ultrasound data includes:

[0015] Perform data cleaning and data standardization processing on the collected testosterone concentration, estradiol concentration, luteinizing hormone concentration, follicle-stimulating hormone concentration, body temperature and BMI value of the ovary of patients with polycystic ovary syndrome, and perform noise removal and image enhancement processing on the three-dimensional B-ultrasound image of the ovary of patients with polycystic ovary syndrome;

[0016] Use the measurement tool in the three-dimensional color Doppler ultrasound instrument to measure the length, width, thickness, major axis and minor axis of the ovary through the three-dimensional B-ultrasound image of the ovary of patients with polycystic ovary syndrome;

[0017] Through the volume formula, the volume formula is , where V is the volume of the ovary, and x, y and z are the length, width and thickness of the ovary respectively;

[0018] Through the ellipticity formula, the ellipticity formula is , where \(a\) is the major axis and \(b\) is the minor axis, and the ellipticity of the ovary is calculated;

[0019] By using an automatic counting algorithm, count the number of follicles; turn on the color Doppler function of the three-dimensional color ultrasound instrument to obtain the blood flow signal of the ovary, and then extract the blood flow velocity of the blood flow signal in the ovary to obtain the three-dimensional B-ultrasound image feature data of the ovary;

[0020] Integrate the three-dimensional B-ultrasound image feature data, hormone data and physiological data of patients with polycystic ovary syndrome to generate a comprehensive efficacy evaluation data set, and divide the comprehensive efficacy evaluation data set into a training set and a test set.

[0021] A further improvement of the technical solution of the present invention lies in that: the efficacy level monitoring module, based on the B-ultrasound data after feature extraction, the process of evaluating the efficacy of polycystic ovary syndrome includes:

[0022] Extract the historical three-dimensional B-ultrasound image feature data of the patient's ovary from the database. The historical three-dimensional B-ultrasound image feature data of the ovary includes the historical ovary volume, historical ovary ellipticity, historical number of follicles in the ovary, and historical blood flow velocity of the blood flow signal in the ovary. The database is the hospital internal database, and this database includes the historical three-dimensional B-ultrasound image feature data of the patient's ovary;

[0023] Respectively set the normal range, moderate abnormality range and severe abnormality range of the ovary volume, ovary ellipticity, number of follicles in the ovary, and historical blood flow velocity of the blood flow signal in the ovary, and assign 9 points, 6 points and 3 points to different levels of abnormality ranges respectively. Among them, the normal range is 9 points, the moderate abnormality range is 6 points, and the severe abnormality range is 3 points;

[0024] The normal range, moderate abnormality range and severe abnormality range of the ovary volume are 4 - 8 、8 - 12 and greater than 12 ;

[0025] The normal range of the ovary ellipticity is 1.5 - 2.5, the moderate abnormality range of the ovary ellipticity is 1.1 - 1.5 or 2.5 - 2.9, and the severe abnormality range of the ovary ellipticity is less than 1.1 or greater than 2.9;

[0026] The normal range, moderate abnormality range and severe abnormality range of the number of follicles in the ovary are less than 12, 12 - 18 and greater than 18 respectively;

[0027] The normal range, moderate abnormality range and severe abnormality range of the historical blood flow velocity of the blood flow signal in the ovary are 10 - 20 cm / s, 20 - 30 cm / s and greater than 30 cm / s respectively;

[0028] Based on the historical ovarian volume, historical ovarian ellipticity, historical number of follicles in the ovary, and historical blood flow velocity of the blood flow signal in the ovary of several patients, their corresponding scores are obtained respectively, and the mean values of the characteristic scores of several patients are integrated and corresponding weights are assigned to them. Among them, based on the historical ovarian volume of several polycystic ovary syndrome patients, the score of the historical ovarian volume is obtained, the mean value of the historical ovarian volume score is calculated, and a weight is assigned to it; based on the historical ovarian ellipticity of several polycystic ovary syndrome patients, the score of the historical ovarian ellipticity is obtained, the mean value of the historical ovarian ellipticity score is calculated, and a weight is assigned to it; based on the historical number of follicles in the ovary of several polycystic ovary syndrome patients, the score of the historical number of follicles in the ovary is obtained, the mean value of the historical number of follicles in the ovary score is calculated, and a weight is assigned to it; based on the historical blood flow velocity of the blood flow signal in the ovary of several polycystic ovary syndrome patients, the score of the historical blood flow velocity of the blood flow signal in the ovary is obtained, and the mean value of the historical blood flow velocity score of the blood flow signal in the ovary is calculated;

[0029] Using the mean value of the historical ovarian volume score and its weight, the mean value of the historical ovarian ellipticity score and its weight, the mean value of the historical number of follicles in the ovary score and its weight, and the mean value of the blood flow velocity score of the blood flow signal in the ovary and its weight, the ovarian B-ultrasound evaluation coefficient is calculated. The calculation formula is as follows:

[0030]

[0031] Among them, is the ovarian B-ultrasound evaluation coefficient, , , and are the weights of the mean value of the historical ovarian volume score, the mean value of the historical ovarian ellipticity score, the mean value of the historical number of follicles in the ovary score, and the mean value of the blood flow velocity score of the blood flow signal in the ovary respectively, , , and are the mean value of the historical ovarian volume score, the mean value of the historical ovarian ellipticity score, the mean value of the historical number of follicles in the ovary score, and the mean value of the blood flow velocity score of the blood flow signal in the ovary respectively.

[0032] A further improvement of the technical solution of the present invention lies in that: in the efficacy level monitoring module, the process of obtaining the efficacy level of polycystic ovary syndrome patients includes:

[0033] Extract the ovarian three-dimensional B-ultrasound image feature data from the comprehensive efficacy evaluation data set;

[0034] Using the ovarian three-dimensional B-ultrasound image feature data in the training set, combining the ovarian B-ultrasound evaluation coefficient and the neural network algorithm, taking the ovarian three-dimensional B-ultrasound image feature data as the input and the ovarian B-ultrasound evaluation coefficient as the output, learning the non-linear relationship between the ovarian three-dimensional B-ultrasound image feature data and the ovarian B-ultrasound evaluation coefficient, and training the B-ultrasound treatment effect evaluation model;

[0035] Input the ovarian three-dimensional B-ultrasound image feature data in the test set into the B-ultrasound treatment effect evaluation model. The B-ultrasound treatment effect evaluation model outputs the ovarian B-ultrasound evaluation coefficient. Compare the output result of the B-ultrasound treatment effect evaluation model with the actual ovarian B-ultrasound evaluation coefficient, evaluate the performance of the B-ultrasound treatment effect evaluation model, optimize the B-ultrasound treatment effect evaluation model, and deploy the optimized B-ultrasound treatment effect evaluation model into the system;

[0036] Combined with the current ovarian three-dimensional B-ultrasound image feature data of the patient, the B-ultrasound treatment effect evaluation model outputs the corresponding ovarian B-ultrasound evaluation coefficient, and determines the treatment effect level and the corresponding value of the ovarian B-ultrasound evaluation coefficient.

[0037] A further improvement of the technical solution of the present invention lies in that: in the treatment effect level monitoring module, the process of determining the treatment effect level and the corresponding value includes:

[0038] When the ovarian B-ultrasound evaluation coefficient is between 7 and 10 points, the corresponding polycystic ovary syndrome patient has a high treatment effect;

[0039] When the ovarian B-ultrasound evaluation coefficient is between 5 and 7 points, the corresponding polycystic ovary syndrome patient has a medium treatment effect;

[0040] When the ovarian B-ultrasound evaluation coefficient is between 0 and 5 points, the corresponding polycystic ovary syndrome patient has a low treatment effect, so as to obtain the treatment effect level of the polycystic ovary syndrome patient.

[0041] A further improvement of the technical solution of the present invention lies in that: the treatment effect feedback evaluation module, the process of evaluating the treatment effect of polycystic ovary syndrome patients using the preprocessed hormone data includes:

[0042] Calculate the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration respectively, and integrate the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration into the comprehensive treatment effect evaluation data set;

[0043] Using the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration in the training set, and combining with the multi-layer neural network algorithm, taking the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration as inputs, and taking the hormone evaluation coefficient as the output, to learn the linear relationship between the ratio of testosterone concentration to estradiol concentration, the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration and the hormone evaluation coefficient, and train the hormone feedback evaluation model;

[0044] Input the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration in the test set into the hormone feedback evaluation model, optimize the hormone feedback evaluation model by adjusting the intercept term and regression coefficients of the hormone feedback evaluation model, obtain the final hormone feedback evaluation model, and combine with the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration of the patient currently, and output the hormone evaluation coefficient;

[0045] The expression of the hormone feedback evaluation model is:

[0046]

[0047] Wherein, is the hormone evaluation coefficient, is the ratio of testosterone concentration to estradiol concentration, is the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration, and are the regression coefficient of the ratio of testosterone concentration to estradiol concentration and the regression coefficient of the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration respectively, is the intercept term, is the error term;

[0048] A further improvement of the technical solution of the present invention lies in: the efficacy feedback evaluation module, the process of evaluating the efficacy of polycystic ovary syndrome patients by using the preprocessed physiological data includes:

[0049] Extract the BMI value and body temperature of polycystic ovary syndrome patients in the comprehensive efficacy evaluation dataset;

[0050] Through the BMI value and body temperature of polycystic ovary syndrome patients in the training set, and combining with the multiple linear regression algorithm, taking the BMI value and body temperature of polycystic ovary syndrome patients as inputs, and taking the physiological evaluation coefficient as the output, to learn the linear relationship between the BMI value and body temperature of polycystic ovary syndrome patients and the physiological evaluation coefficient, and train the physiological feedback evaluation model;

[0051] Input the BMI value and body temperature of patients with polycystic ovary syndrome in the test set into the physiological feedback evaluation model. Optimize the physiological feedback evaluation model by adjusting the intercept term and regression coefficients of the physiological feedback evaluation model, deploy the physiological feedback evaluation model to the system, and combine the current BMI value and body temperature of patients with polycystic ovary syndrome to output the physiological evaluation coefficient;

[0052] The expression of this physiological feedback evaluation model is:

[0053]

[0054] Among them, is the physiological evaluation coefficient, is the intercept term, and are the regression coefficients of the BMI value and body temperature of patients with polycystic ovary syndrome respectively, is the BMI value of patients with polycystic ovary syndrome, is the body temperature of patients with polycystic ovary syndrome, is the error term;

[0055] A further improvement of the technical solution of the present invention lies in: for the efficacy feedback evaluation module, the construction process of the efficacy feedback evaluation model includes:

[0056] Assign weights to the hormone evaluation coefficient and the physiological evaluation coefficient respectively, and use the weights to calculate the efficacy feedback evaluation coefficient. The calculation formula is as follows:

[0057]

[0058] Among them, is the efficacy feedback evaluation coefficient, and are the weights of the hormone evaluation coefficient and the physiological evaluation coefficient respectively, and are the hormone evaluation coefficient and the physiological evaluation coefficient respectively;

[0059] Integrate the efficacy feedback evaluation coefficient into the comprehensive efficacy evaluation dataset. Adopt the hormone data, physiological data and efficacy feedback evaluation coefficient of patients with polycystic ovary syndrome in the training set, and combine the convolutional neural network algorithm. Use the hormone data and physiological data of patients with polycystic ovary syndrome as inputs and the efficacy feedback evaluation coefficient as the output to learn the non-linear relationship between the hormone data, physiological data and the efficacy feedback evaluation coefficient, and train the efficacy feedback evaluation model;

[0060] Input the hormone data and physiological data of polycystic ovary syndrome patients in the test set into the efficacy feedback evaluation model. Compare the efficacy feedback evaluation coefficient output by the efficacy feedback evaluation model with the actual efficacy feedback evaluation coefficient to evaluate the performance of the efficacy feedback evaluation model, adjust the parameters of the efficacy feedback evaluation model, optimize the efficacy feedback evaluation model, deploy the optimized efficacy feedback evaluation model into the system, and combine the current hormone data and physiological data of the patients to output the corresponding efficacy feedback evaluation coefficient.

[0061] A further improvement of the technical solution of the present invention lies in that: the process of the execution module updating the efficacy level of polycystic ovary syndrome patients through the efficacy feedback evaluation model and sending out corresponding signals includes:

[0062] Based on the efficacy feedback evaluation coefficient output by the efficacy feedback evaluation model, when the efficacy feedback evaluation coefficient is between 0.7 and 1, the corresponding polycystic ovary syndrome patient has a high efficacy; when the efficacy feedback evaluation coefficient is between 0.4 and 0.7, the corresponding polycystic ovary syndrome patient has a medium efficacy; when the efficacy feedback evaluation coefficient is lower than 0.4, the corresponding polycystic ovary syndrome patient has a low efficacy, and obtain the feedback efficacy level of the polycystic ovary syndrome patient.

[0063] When the efficacy level of the polycystic ovary syndrome patient evaluated based on B-ultrasound data is the same as the feedback efficacy level of the polycystic ovary syndrome patient, do not update the efficacy level of the polycystic ovary syndrome patient evaluated based on B-ultrasound data; when the efficacy level of the polycystic ovary syndrome patient evaluated based on B-ultrasound data is different from the feedback efficacy level of the polycystic ovary syndrome patient, use the feedback efficacy level of the polycystic ovary syndrome patient to update the efficacy level of the polycystic ovary syndrome patient evaluated based on B-ultrasound data.

[0064] Set signal lights, which include red, yellow and green. Among them, the red signal light indicates that the polycystic ovary syndrome patient has a low efficacy, the yellow signal light indicates that the polycystic ovary syndrome patient has a medium efficacy, and the green signal light indicates that the polycystic ovary syndrome patient has a high efficacy. Use the set signal lights to send out corresponding signals according to polycystic ovary syndrome patients with different efficacy levels.

[0065] The beneficial effects of the present invention are as follows: In the polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images of the present invention, compared with the traditional polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images, the efficacy data acquisition technology, feature extraction technology, multi-model construction technology in the system of the present invention are closely combined with modern information technology to accurately capture the B-ultrasound data, hormone data and physiological data of patients with polycystic ovary syndrome, and then obtain the ovarian B-ultrasound evaluation coefficient, hormone evaluation coefficient, physiological evaluation coefficient and efficacy feedback evaluation coefficient, achieving real-time and comprehensive monitoring of the patient's condition and treatment effect. By constructing a B-ultrasound efficacy evaluation model, a hormone feedback evaluation model, a physiological feedback evaluation model and an efficacy feedback evaluation model, various types of data are analyzed and processed, and the polycystic ovary syndrome efficacy evaluation level based on B-ultrasound images is refined, solving the problem that it is difficult for the prior art to comprehensively evaluate the efficacy of polycystic ovary syndrome by combining B-ultrasound images and various other efficacy evaluation factors, resulting in improper monitoring of the patient's treatment process and poor treatment effect of the patient's polycystic ovary syndrome. It ensures that the method in the present invention can refine the dynamic monitoring standard for the polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images within a more accurate range, making the monitored data more accurate indicators under the same conditions. The research and application of this method significantly enhance the degree of intelligence in the polycystic ovary syndrome efficacy evaluation process based on B-ultrasound images. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0067] Figure 1 It is a block diagram of the polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0069] Such as Figure 1As shown in the figure, the present invention provides a polycystic ovary syndrome efficacy evaluation system based on B-ultrasound images, including an efficacy data acquisition module, a feature extraction module, an efficacy level monitoring module, an efficacy feedback evaluation module, and an execution module. Among them, each module is communicatively connected;

[0070] The efficacy data acquisition module collects the efficacy data of polycystic ovary syndrome patients, including B-ultrasound data, hormone data, and physiological data, providing data support for the implementation of the subsequent module functions;

[0071] The feature extraction module preprocesses the collected efficacy data and extracts features from the preprocessed B-ultrasound data;

[0072] The efficacy level monitoring module analyzes the efficacy of polycystic ovary syndrome based on the B-ultrasound data after feature extraction, and combines the multiple linear regression algorithm to obtain the efficacy level of polycystic ovary syndrome patients;

[0073] The efficacy feedback evaluation module uses the preprocessed hormone data and physiological data to evaluate the efficacy of polycystic ovary syndrome patients respectively, and constructs an efficacy feedback evaluation model;

[0074] The execution module updates the efficacy level of polycystic ovary syndrome patients through the output result of the efficacy feedback evaluation model and issues corresponding signals, solving the problem that it is difficult to comprehensively evaluate the efficacy of polycystic ovary syndrome by combining B-ultrasound images and various other efficacy evaluation factors in the prior art.

[0075] Preferably, for the efficacy data acquisition module, the acquisition process of the efficacy data of polycystic ovary syndrome patients includes:

[0076] Deploy different types of acquisition devices to collect B-ultrasound data, hormone data, and physiological data of polycystic ovary syndrome patients. Among them, the acquisition devices include a three-dimensional color Doppler ultrasound instrument, an automatic chemiluminescence immunoassay analyzer, a thermometer, and a height and weight measuring instrument;

[0077] The B-ultrasound data is the three-dimensional B-ultrasound image of the patient's ovary; the hormone data includes the testosterone concentration, estradiol concentration, luteinizing hormone concentration, and follicle-stimulating hormone concentration of the patient; the physiological data includes the patient's body temperature and BMI value;

[0078] Use a three-dimensional color Doppler ultrasound instrument to collect the three-dimensional B-ultrasound image of the ovary of polycystic ovary syndrome patients; collect the testosterone concentration, estradiol concentration, luteinizing hormone concentration, and follicle-stimulating hormone concentration of polycystic ovary syndrome patients through an automatic chemiluminescence immunoassay analyzer; collect the body temperature of polycystic ovary syndrome patients using a thermometer; collect the BMI value of polycystic ovary syndrome patients using a height and weight measuring instrument.

[0079] Preferably, the process of the feature extraction module preprocessing the collected efficacy data and extracting features from the preprocessed B-ultrasound data includes:

[0080] Performing data cleaning and data standardization on the collected testosterone concentration, estradiol concentration, luteinizing hormone concentration, follicle-stimulating hormone concentration, body temperature and BMI value of the ovaries of polycystic ovary syndrome patients, and performing noise removal and image enhancement on the three-dimensional B-ultrasound images of the ovaries of polycystic ovary syndrome patients;

[0081] Using the measurement tool in the three-dimensional color Doppler ultrasound instrument, measuring the length, width, thickness, major axis and minor axis of the ovary through the three-dimensional B-ultrasound image of the ovary of polycystic ovary syndrome patients;

[0082] Through the volume formula, the volume formula is , where V is the volume of the ovary, and x, y and z are the length, width and thickness of the ovary respectively;

[0083] Through the ellipticity formula, the ellipticity formula is , where a is the major axis and b is the minor axis, to calculate the ellipticity of the ovary;

[0084] Counting the number of follicles through an automatic counting algorithm; turning on the color Doppler function of the three-dimensional color Doppler ultrasound instrument to obtain the blood flow signal of the ovary, and then extracting the blood flow velocity of the blood flow signal in the ovary to obtain the three-dimensional B-ultrasound image feature data of the ovary;

[0085] Integrating the three-dimensional B-ultrasound image feature data, hormone data and physiological data of the ovaries of polycystic ovary syndrome patients to generate a comprehensive efficacy evaluation data set, and dividing the comprehensive efficacy evaluation data set into a training set and a test set.

[0086] Preferably, the process of the efficacy level monitoring module evaluating the efficacy of polycystic ovary syndrome based on the B-ultrasound data after feature extraction includes:

[0087] Extracting the historical three-dimensional B-ultrasound image feature data of the ovaries of patients from the database. The historical three-dimensional B-ultrasound image feature data of the ovaries includes the historical ovarian volume, historical ovarian ellipticity, historical number of follicles in the ovaries, and historical blood flow velocity of the blood flow signal in the ovaries. The database is the hospital internal database, and the database includes the historical three-dimensional B-ultrasound image feature data of the ovaries of patients;

[0088] Respectively setting the normal range, moderately abnormal range and severely abnormal range of the ovarian volume, ovarian ellipticity, number of follicles in the ovaries, and historical blood flow velocity of the blood flow signal in the ovaries, and assigning 9 points, 6 points and 3 points to different levels of abnormal ranges respectively. Among them, the normal range is 9 points, the moderately abnormal range is 6 points, and the severely abnormal range is 3 points;

[0089] The normal range, moderately abnormal range, and severely abnormal range of the ovarian volume are 4 - 8 , 8 - 12 and greater than 12 ;

[0090] The normal range of the ovarian ellipticity is 1.5 - 2.5, the moderately abnormal range of the ovarian ellipticity is 1.1 - 1.5 or 2.5 - 2.9, and the severely abnormal range of the ovarian ellipticity is less than 1.1 or greater than 2.9;

[0091] The normal range, moderately abnormal range, and severely abnormal range of the number of follicles in the ovary are less than 12, 12 - 18, and greater than 18 respectively;

[0092] The normal range, moderately abnormal range, and severely abnormal range of the historical blood flow velocity of the blood flow signal in the ovary are 10 - 20 cm / s, 20 - 30 cm / s, and greater than 30 cm / s respectively;

[0093] Based on the historical ovarian volume, historical ovarian ellipticity, historical number of follicles in the ovary, and historical blood flow velocity of the blood flow signal in the ovary of a number of patients, the corresponding scores are obtained respectively, and the mean values of the characteristic scores of a number of patients are integrated to assign corresponding weights to them;

[0094] Among them, based on the historical ovarian volume of a number of polycystic ovary syndrome patients, the score of the historical ovarian volume is obtained, the mean value of the historical ovarian volume score is calculated, and a weight is assigned to it; based on the historical ovarian ellipticity of a number of polycystic ovary syndrome patients, the score of the historical ovarian ellipticity is obtained, the mean value of the historical ovarian ellipticity score is calculated, and a weight is assigned to it; based on the historical number of follicles in the ovary of a number of polycystic ovary syndrome patients, the score of the historical number of follicles in the ovary is obtained, the mean value of the historical number of follicles in the ovary score is calculated, and a weight is assigned to it; based on the historical blood flow velocity of the blood flow signal in the ovary of a number of polycystic ovary syndrome patients, the score of the historical blood flow velocity of the blood flow signal in the ovary is obtained, and the mean value of the historical blood flow velocity score of the blood flow signal in the ovary is calculated;

[0095] Using the mean value of the historical ovarian volume score and its weight, the mean value of the historical ovarian ellipticity score and its weight, the mean value of the historical number of follicles in the ovary score and its weight, and the mean value of the blood flow velocity score of the blood flow signal in the ovary and its weight, calculate the ovarian B - ultrasound evaluation coefficient, and its calculation formula is as follows:

[0096]

[0097] Among them, is the ovarian B - ultrasound evaluation coefficient, , , and are the weights of the historical average ovarian volume fraction, the historical average ovarian ellipticity fraction, the historical average number of follicles in the ovary fraction, and the historical average blood flow velocity fraction of the blood flow signal in the ovary, respectively. , , and are the historical average ovarian volume fraction, the historical average ovarian ellipticity fraction, the historical average number of follicles in the ovary fraction, and the historical average blood flow velocity fraction of the blood flow signal in the ovary, respectively.

[0098] Preferably, in the efficacy level monitoring module, the process of obtaining the efficacy level of patients with polycystic ovary syndrome includes:

[0099] Extracting the ovarian three-dimensional B-ultrasound image feature data from the comprehensive efficacy evaluation dataset;

[0100] Using the ovarian three-dimensional B-ultrasound image feature data in the training set, combining the ovarian B-ultrasound evaluation coefficient and the neural network algorithm, taking the ovarian three-dimensional B-ultrasound image feature data as the input and the ovarian B-ultrasound evaluation coefficient as the output, learning the non-linear relationship between the ovarian three-dimensional B-ultrasound image feature data and the ovarian B-ultrasound evaluation coefficient, and training the B-ultrasound efficacy evaluation model;

[0101] Inputting the ovarian three-dimensional B-ultrasound image feature data in the test set into the B-ultrasound efficacy evaluation model, the B-ultrasound efficacy evaluation model outputs the ovarian B-ultrasound evaluation coefficient, comparing the output result of the B-ultrasound efficacy evaluation model with the actual ovarian B-ultrasound evaluation coefficient, evaluating the performance of the B-ultrasound efficacy evaluation model, optimizing the B-ultrasound efficacy evaluation model, and deploying the optimized B-ultrasound efficacy evaluation model into the system;

[0102] Combining with the patient's current ovarian three-dimensional B-ultrasound image feature data, the B-ultrasound efficacy evaluation model outputs the corresponding ovarian B-ultrasound evaluation coefficient, and determines the efficacy level and the corresponding value of the ovarian B-ultrasound evaluation coefficient.

[0103] Preferably, in the efficacy level monitoring module, the process of determining the efficacy level and the corresponding value includes:

[0104] When the ovarian B-ultrasound evaluation coefficient is between 7 and 10 points, the corresponding patient with polycystic ovary syndrome has a high efficacy;

[0105] When the ovarian B-ultrasound evaluation coefficient is between 5 and 7 points, the corresponding patient with polycystic ovary syndrome has a medium efficacy;

[0106] When the ovarian B-ultrasound evaluation coefficient is between 0 and 5 points, the corresponding patient with polycystic ovary syndrome has a low efficacy, and thus the efficacy level of the patient with polycystic ovary syndrome is obtained.

[0107] Preferably, the process of the efficacy feedback evaluation module for evaluating the efficacy of patients with polycystic ovary syndrome using the preprocessed hormone data includes:

[0108] Calculate the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration respectively, and integrate the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration into the comprehensive efficacy evaluation dataset;

[0109] Using the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration in the training set, combined with the multi-layer neural network algorithm, taking the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration as inputs and the hormone evaluation coefficient as the output, learn the linear relationship between the ratio of testosterone concentration to estradiol concentration, the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration and the hormone evaluation coefficient, and train the hormone feedback evaluation model;

[0110] Input the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration in the test set into the hormone feedback evaluation model, optimize the hormone feedback evaluation model by adjusting the intercept term and regression coefficient of the hormone feedback evaluation model, obtain the final hormone feedback evaluation model, and combine the ratio of the patient's current testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration to output the hormone evaluation coefficient;

[0111] The expression of the hormone feedback evaluation model is:

[0112]

[0113] Wherein, is the hormone evaluation coefficient, is the ratio of testosterone concentration to estradiol concentration, is the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration, and are the regression coefficients of the ratio of testosterone concentration to estradiol concentration and the regression coefficient of the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration respectively, is the intercept term, is the error term;

[0114] Preferably, the process of the efficacy feedback evaluation module for evaluating the efficacy of patients with polycystic ovary syndrome using the preprocessed physiological data includes:

[0115] Extract the BMI value and body temperature of patients with polycystic ovary syndrome from the comprehensive efficacy evaluation dataset;

[0116] Using the BMI values and body temperatures of polycystic ovary syndrome patients in the training set, and combining with the multiple linear regression algorithm, taking the BMI values and body temperatures of polycystic ovary syndrome patients as inputs and the physiological evaluation coefficient as the output, learning the linear relationship between the BMI values and body temperatures of polycystic ovary syndrome patients and the physiological evaluation coefficient, and training the physiological feedback evaluation model;

[0117] Input the BMI values and body temperatures of polycystic ovary syndrome patients in the test set into the physiological feedback evaluation model, optimize the physiological feedback evaluation model by adjusting the intercept term and regression coefficients of the physiological feedback evaluation model, deploy the physiological feedback evaluation model into the system, and combine with the current BMI values and body temperatures of polycystic ovary syndrome patients to output the physiological evaluation coefficient;

[0118] The expression of this physiological feedback evaluation model is:

[0119]

[0120] Among them, is the physiological evaluation coefficient, is the intercept term, and are the regression coefficients of the BMI value and body temperature of polycystic ovary syndrome patients respectively, is the BMI value of polycystic ovary syndrome patients, is the body temperature of polycystic ovary syndrome patients, is the error term;

[0121] Preferably, for the efficacy feedback evaluation module, the construction process of the efficacy feedback evaluation model includes:

[0122] Assign weights to the hormone evaluation coefficient and the physiological evaluation coefficient respectively, and calculate the efficacy feedback evaluation coefficient using the weights. The calculation formula is as follows:

[0123]

[0124] Among them, is the efficacy feedback evaluation coefficient, and are the weights of the hormone evaluation coefficient and the physiological evaluation coefficient respectively, and are the hormone evaluation coefficient and the physiological evaluation coefficient respectively;

[0125] Integrate the efficacy feedback evaluation coefficient into the comprehensive efficacy evaluation dataset. Use the hormone data, physiological data of polycystic ovary syndrome patients in the training set, and the efficacy feedback evaluation coefficient. Combine with the convolutional neural network algorithm. Take the hormone data and physiological data of polycystic ovary syndrome patients as input, and the efficacy feedback evaluation coefficient as output. Learn the non-linear relationship between hormone data, physiological data and the efficacy feedback evaluation coefficient, and train the efficacy feedback evaluation model;

[0126] Input the hormone data and physiological data of polycystic ovary syndrome patients in the test set into the efficacy feedback evaluation model. Compare the efficacy feedback evaluation coefficient output by the efficacy feedback evaluation model with the actual efficacy feedback evaluation coefficient. Evaluate the performance of the efficacy feedback evaluation model, adjust the parameters of the efficacy feedback evaluation model, optimize the efficacy feedback evaluation model, deploy the optimized efficacy feedback evaluation model into the system, and combine with the current hormone data and physiological data of the patient to output the corresponding efficacy feedback evaluation coefficient.

[0127] Preferably, the process of the execution module updating the efficacy level of polycystic ovary syndrome patients and sending out corresponding signals through the efficacy feedback evaluation model includes:

[0128] Based on the efficacy feedback evaluation coefficient output by the efficacy feedback evaluation model, when the efficacy feedback evaluation coefficient is between 0.7 and 1, the corresponding polycystic ovary syndrome patient has a high efficacy; when the efficacy feedback evaluation coefficient is between 0.4 and 0.7, the corresponding polycystic ovary syndrome patient has a medium efficacy; when the efficacy feedback evaluation coefficient is lower than 0.4, the corresponding polycystic ovary syndrome patient has a low efficacy, and obtain the feedback efficacy level of polycystic ovary syndrome patients;

[0129] When the efficacy level of polycystic ovary syndrome patients evaluated based on B-ultrasound data is the same as the feedback efficacy level of polycystic ovary syndrome patients, do not update the efficacy level of polycystic ovary syndrome patients evaluated based on B-ultrasound data; when the efficacy level of polycystic ovary syndrome patients evaluated based on B-ultrasound data is different from the feedback efficacy level of polycystic ovary syndrome patients, use the feedback efficacy level of polycystic ovary syndrome patients to update the efficacy level of polycystic ovary syndrome patients evaluated based on B-ultrasound data;

[0130] Set signal lights, which include red, yellow and green. Among them, the red signal light indicates that the polycystic ovary syndrome patient has a low efficacy, the yellow signal light indicates that the polycystic ovary syndrome patient has a medium efficacy, and the green signal light indicates that the polycystic ovary syndrome patient has a high efficacy. Use the set signal lights to send out corresponding signals according to polycystic ovary syndrome patients with different efficacy levels.

[0131] First, the B-ultrasound data, hormone data, and physiological data of polycystic ovary syndrome patients are collected through a three-dimensional color Doppler ultrasound instrument, an automated chemiluminescence immunoassay analyzer, a thermometer, and a height and weight measuring instrument, respectively. The collected data are preprocessed and feature-extracted, and the processed data are integrated into a comprehensive efficacy evaluation dataset. Secondly, based on the B-ultrasound data after feature extraction, the efficacy of polycystic ovary syndrome is evaluated, the ovarian B-ultrasound evaluation coefficient is calculated, a B-ultrasound efficacy evaluation model is constructed, and then the efficacy level of polycystic ovary syndrome patients is obtained. Immediately afterwards, the hormone data after preprocessing are used to evaluate the efficacy of polycystic ovary syndrome patients, a hormone feedback evaluation model is constructed, and then the hormone evaluation coefficient is obtained. Then, the physiological data after preprocessing are used to evaluate the efficacy of polycystic ovary syndrome patients, a physiological feedback evaluation model is constructed, and then the physiological evaluation coefficient is obtained. Then, the hormone evaluation coefficient and the physiological evaluation coefficient are combined to calculate the efficacy feedback evaluation coefficient, and a efficacy feedback evaluation model is constructed. Finally, through the efficacy feedback evaluation model, the efficacy level of polycystic ovary syndrome patients is updated, and corresponding signals are sent out.

[0132] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. The polycystic ovary syndrome efficacy evaluation system based on B-ultrasound imaging includes an efficacy data acquisition module, a feature extraction module, an efficacy level monitoring module, an efficacy feedback evaluation module and an execution module, wherein: Each module is connected in communication, characterized by: The efficacy data collection module collects efficacy data of PCOS patients including B-ultrasound data, hormone data and physiological data; The feature extraction module preprocesses the collected therapeutic effect data and extracts features from the preprocessed B-ultrasound data; The efficacy level monitoring module analyzes the efficacy of polycystic ovary syndrome based on the B-ultrasound data after feature extraction and combines the multivariate linear regression algorithm to obtain the efficacy level of polycystic ovary syndrome patients; The efficacy feedback evaluation module uses the pre-processed hormone data and physiological data to evaluate the efficacy of patients with polycystic ovary syndrome and construct an efficacy feedback evaluation model; The execution module updates the efficacy level of the polycystic ovary syndrome patients through the output results of the efficacy feedback evaluation model and sends a corresponding signal.

2. The PCOS therapeutic effect evaluation system based on B-ultrasound imaging according to claim 1, characterized in that: The efficacy data collection module includes the following steps: Deploy different types of collection equipment to collect B-ultrasound data, hormone data and physiological data of patients with polycystic ovary syndrome, wherein the collection equipment includes a three-dimensional color ultrasound instrument, an automatic chemiluminescence immunoassay analyzer, a thermometer and a height and weight measuring instrument; The B-ultrasound data are three-dimensional B-ultrasound images of the patient's ovaries; the hormone data include the patient's testosterone concentration, estradiol concentration, luteinizing hormone concentration and follicle-stimulating hormone concentration; and the physiological data include the patient's body temperature and BMI value.

3. The PCOS therapeutic effect evaluation system based on B-ultrasound imaging according to claim 2, characterized in that: The feature extraction module preprocesses the collected therapeutic effect data and extracts features from the preprocessed B-ultrasound data, including: The collected data of testosterone concentration, estradiol concentration, luteinizing hormone concentration, follicle-stimulating hormone concentration, body temperature and BMI values ​​of the ovaries of patients with PCOS were cleaned and standardized, and the three-dimensional B-ultrasound images of the ovaries of patients with PCOS were subjected to noise removal and image enhancement. Using the measuring tools in the three-dimensional color ultrasound machine, the length, width, thickness, long diameter and short diameter of the ovaries of patients with PCOS were measured through three-dimensional B-ultrasound images of the ovaries. The volume formula is used to calculate the volume of the ovary; the ovality of the ovary is calculated by the ovality formula; the number of follicles is counted by the automatic counting algorithm; the color Doppler function of the three-dimensional color ultrasound instrument is turned on to obtain the blood flow signal of the ovary, and then the blood flow velocity of the blood flow signal in the ovary is extracted to obtain the characteristic data of the three-dimensional B-ultrasound image of the ovary; The ovarian three-dimensional B-ultrasound image feature data, hormone data and physiological data of patients with polycystic ovary syndrome were integrated to generate a comprehensive efficacy evaluation data set, which was then divided into a training set and a test set.

4. The PCOS therapeutic effect evaluation system based on B-ultrasound imaging according to claim 3 is characterized in that: The efficacy level monitoring module, based on the B-ultrasound data after feature extraction, evaluates the efficacy of polycystic ovary syndrome in the following process: Extract the patient's historical ovarian three-dimensional B-ultrasound image feature data from a database, the historical ovarian three-dimensional B-ultrasound image feature data including historical ovarian volume, historical ovarian ellipticity, historical number of ovarian follicles, and historical blood flow velocity of ovarian blood flow signals, the database being an internal database of the hospital, and the database including the patient's historical ovarian three-dimensional B-ultrasound image feature data; The normal range, moderately abnormal range, and severely abnormal range were set for the ovarian volume, ovarian ellipticity, number of ovarian follicles, and historical blood flow velocity of the ovarian blood flow signal, and 9 points, 6 points, and 3 points were assigned to the different levels of abnormal ranges, respectively; Based on the historical ovarian volume, historical ovarian ellipticity, historical number of ovarian follicles, and historical blood flow velocity of ovarian blood flow signals of several patients, the corresponding scores are obtained respectively, and the mean values ​​of the characteristic scores of several patients are combined to assign corresponding weights to them; The ovarian B-ultrasound evaluation coefficient was calculated using the mean of the historical ovarian volume fraction and its weight, the mean of the historical ovarian ellipticity fraction and its weight, the mean of the historical ovarian follicle number fraction and its weight, and the mean of the historical blood flow velocity fraction of the ovarian blood flow signal and its weight.

5. The PCOS therapeutic effect evaluation system based on B-ultrasound imaging according to claim 4 is characterized in that: In the efficacy level monitoring module, the process of obtaining the efficacy level of patients with polycystic ovary syndrome includes: Extract the ovarian three-dimensional B-ultrasound image feature data from the comprehensive efficacy evaluation data set; Using the ovarian three-dimensional B-ultrasound image feature data in the training set, combined with the ovarian B-ultrasound evaluation coefficient and the neural network algorithm, the ovarian three-dimensional B-ultrasound image feature data is used as input, and the ovarian B-ultrasound evaluation coefficient is used as output to learn the nonlinear relationship between the ovarian three-dimensional B-ultrasound image feature data and the ovarian B-ultrasound evaluation coefficient, and train the B-ultrasound efficacy evaluation model; The ovarian three-dimensional B-ultrasound image feature data in the test set is input into the B-ultrasound efficacy evaluation model, and the B-ultrasound efficacy evaluation model outputs the ovarian B-ultrasound evaluation coefficient. The output result of the B-ultrasound efficacy evaluation model is compared with the actual ovarian B-ultrasound evaluation coefficient, the performance of the B-ultrasound efficacy evaluation model is evaluated, and the B-ultrasound efficacy evaluation model is optimized. The optimized B-ultrasound efficacy evaluation model is deployed into the system, and the corresponding ovarian B-ultrasound evaluation coefficient is output in combination with the patient's current ovarian three-dimensional B-ultrasound image feature data, and the efficacy level and corresponding value of the ovarian B-ultrasound evaluation coefficient are determined.

6. The PCOS therapeutic effect evaluation system based on B-ultrasound imaging according to claim 5, characterized in that: In the efficacy level monitoring module, the process of determining the efficacy level and the corresponding value includes: When the ovarian ultrasound evaluation coefficient is between 7 and 10 points, it corresponds to a high efficacy for patients with polycystic ovary syndrome; When the ovarian ultrasound evaluation coefficient is between 5 and 7 points, it corresponds to a moderate therapeutic effect for patients with polycystic ovary syndrome; When the ovarian B-ultrasound evaluation coefficient is between 0 and 5 points, it corresponds to a low efficacy for patients with polycystic ovary syndrome, thereby obtaining the efficacy level for patients with polycystic ovary syndrome.

7. The PCOS therapeutic effect evaluation system based on B-ultrasound imaging according to claim 6, characterized in that: The efficacy feedback evaluation module uses the pre-processed hormone data to evaluate the efficacy of patients with polycystic ovary syndrome, including: The ratios of testosterone concentration to estradiol concentration and the ratios of luteinizing hormone concentration to follicle-stimulating hormone concentration were calculated separately and integrated into the comprehensive efficacy evaluation data set; Using the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration in the training set, combined with the multivariate neural network algorithm, the ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration are used as input, and the hormone evaluation coefficient is used as output to learn the linear relationship between the ratio of testosterone concentration to estradiol concentration, the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration and the hormone evaluation coefficient, and train the hormone feedback evaluation model; The ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration in the test set are input into the hormone feedback evaluation model. The hormone feedback evaluation model is optimized by adjusting the intercept term and regression coefficient of the hormone feedback evaluation model to obtain the final hormone feedback evaluation model. The corresponding hormone evaluation coefficient is output based on the patient's current ratio of testosterone concentration to estradiol concentration and the ratio of luteinizing hormone concentration to follicle-stimulating hormone concentration.

8. The PCOS therapeutic effect evaluation system based on B-ultrasound imaging according to claim 7, characterized in that: The efficacy feedback evaluation module uses the pre-processed physiological data to evaluate the efficacy of patients with polycystic ovary syndrome, including: Extract BMI values ​​and body temperature of patients with PCOS in the comprehensive efficacy evaluation dataset; By training the BMI values ​​and body temperatures of patients with polycystic ovary syndrome in the training set, combined with the multivariate linear regression algorithm, the BMI values ​​and body temperatures of patients with polycystic ovary syndrome are used as input, and the physiological evaluation coefficient is used as output, the linear relationship between the BMI values ​​and body temperatures of patients with polycystic ovary syndrome and the physiological evaluation coefficient is learned, and the physiological feedback evaluation model is trained; The BMI value and body temperature of the PCOS patients in the test set were input into the physiological feedback assessment model. The physiological feedback assessment model was optimized by adjusting the intercept term and regression coefficient of the physiological feedback assessment model. The physiological feedback assessment model was deployed into the system, and the corresponding physiological assessment coefficient was output in combination with the patient's current BMI value and body temperature.

9. The PCOS therapeutic effect evaluation system based on B-ultrasound imaging according to claim 8, characterized in that: The therapeutic effect feedback evaluation module and the construction process of the therapeutic effect feedback evaluation model include: Assign weights to the hormone evaluation coefficient and the physiological evaluation coefficient respectively, calculate the efficacy feedback evaluation coefficient using the weights, and integrate the efficacy feedback evaluation coefficient into the comprehensive efficacy evaluation data set; The hormone data and physiological data of patients with polycystic ovary syndrome in the training set, as well as the efficacy feedback evaluation coefficient, are used in combination with the convolutional neural network algorithm. The hormone data and physiological data of patients with polycystic ovary syndrome are used as input, and the efficacy feedback evaluation coefficient is used as output. The nonlinear relationship between the hormone data, physiological data and the efficacy feedback evaluation coefficient is learned, and the efficacy feedback evaluation model is trained. The hormone data and physiological data of the PCOS patients in the test set are input into the efficacy feedback evaluation model, the efficacy feedback evaluation coefficient output by the efficacy feedback evaluation model is compared with the actual efficacy feedback evaluation coefficient, the performance of the efficacy feedback evaluation model is evaluated, the parameters of the efficacy feedback evaluation model are adjusted, the efficacy feedback evaluation model is optimized, the optimized efficacy feedback evaluation model is deployed into the system, and the corresponding efficacy feedback evaluation coefficient is output in combination with the current patient's hormone data and physiological data.

10. The PCOS therapeutic effect evaluation system based on B-ultrasound imaging according to claim 9, characterized in that: The execution module updates the efficacy level of the polycystic ovary syndrome patient through the efficacy feedback evaluation model and sends a corresponding signal, including: Based on the efficacy feedback evaluation coefficient output by the efficacy feedback evaluation model, when the efficacy feedback evaluation coefficient is between 0.7 and 1, it corresponds to high efficacy for patients with polycystic ovary syndrome; when the efficacy feedback evaluation coefficient is between 0.4 and 0.7, it corresponds to medium efficacy for patients with polycystic ovary syndrome; when the efficacy feedback evaluation coefficient is lower than 0.4, it corresponds to low efficacy for patients with polycystic ovary syndrome, and the feedback efficacy level of patients with polycystic ovary syndrome is obtained; When the efficacy level of PCOS patients evaluated based on B-ultrasound data is the same as the efficacy level reported by PCOS patients, the efficacy level of PCOS patients evaluated based on B-ultrasound data will not be updated; when the efficacy level of PCOS patients evaluated based on B-ultrasound data is different from the efficacy level reported by PCOS patients, the efficacy level reported by PCOS patients will be used to update the efficacy level of PCOS patients evaluated based on B-ultrasound data; A signal light is set, which includes red, yellow and green colors. The red signal light indicates that the patient with polycystic ovary syndrome has a low therapeutic effect, the yellow signal light indicates that the patient with polycystic ovary syndrome has a medium therapeutic effect, and the green signal light indicates that the patient with polycystic ovary syndrome has a high therapeutic effect. The set signal lights are used to send corresponding signals according to patients with polycystic ovary syndrome at different levels of therapeutic effect.

Images