Vision screening data analysis system based on image technology
By using an image-based vision screening data analysis system that combines image processing and a logistic regression model, the efficiency and accuracy issues of existing eye disease screening methods have been resolved, enabling efficient and accurate eye health assessment and diagnosis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-10
- Publication Date
- 2026-03-13
AI Technical Summary
Existing methods for screening eye diseases rely on doctors' experience and visual observation, which is time-consuming, laborious, and has a high rate of misdiagnosis. Traditional equipment cannot comprehensively and accurately assess eye health.
The vision screening data analysis system based on image technology includes an image data acquisition module, a patient assessment module, a vision data acquisition module, and a logistic regression module. It acquires eye images through a high-definition camera and combines fuzzy inference and logistic regression models to quantify the complexity of eye lesions and vision data, and generate classification evaluation coefficients.
It improves the efficiency and accuracy of eye disease identification, reduces the misdiagnosis rate, provides a comprehensive eye health assessment, and assists doctors in making diagnoses.
Smart Images
Figure CN121662361A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data analysis technology, and more specifically, to a vision screening data analysis system based on image technology. Background Technology
[0002] With the development of image processing technology and machine learning algorithms, automated eye health assessment systems have become possible. These technologies can extract rich information from eye images and combine them with the patient's vision data for more comprehensive analysis and assessment. In modern medical systems, early detection and accurate diagnosis of eye diseases are key to improving treatment outcomes and preventing vision loss. However, most existing methods for screening eye diseases rely on doctors' experience and visual observation, which is not only time-consuming and laborious but also easily affected by human factors, leading to a high rate of misdiagnosis. In addition, traditional ophthalmic examination equipment usually only provides limited detection functions and cannot comprehensively and accurately assess the patient's eye health status.
[0003] To address the aforementioned shortcomings, a technical solution is provided. Summary of the Invention
[0004] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a vision screening data analysis system based on image technology to solve the problems mentioned in the background art.
[0005] To achieve the above objectives, the present invention provides the following technical solution: A vision screening data analysis system based on image technology includes an image data acquisition module, a patient assessment module, a vision data acquisition module, and a logistic regression module, with signal connections between the modules; The image data acquisition module is used to acquire patient eye images using a high-definition camera eye detection device, acquire patient eye data through image technology, and determine the credibility of historical cases through fuzzy reasoning based on the patient's historical medical records. The patient assessment module is used to perform weighted analysis of the disease complexity coefficient, the disease complexity coefficient, and the historical disease reliability coefficient to generate a disease assessment coefficient, which quantifies the reliability of eye images in judging the disease. The vision data acquisition module is used to perform vision tests on patients using high-precision vision testing instruments and obtain quantitative results of patients' vision data. The logistic regression module is used to combine the quantitative results of disease assessment coefficients and patient vision data for analysis, and selectively label patients through a logistic regression model.
[0006] In a preferred embodiment, acquiring patient eye data using imaging technology includes: By comparing features in an image with known diseases in a lesion feature database, a disease complexity coefficient for determining the patient's eye is obtained. The logic for obtaining the disease complexity coefficient is as follows: The number of different regions of the eye is obtained and labeled as n, where n = 1, 2, 3, ..., N, and N is a positive integer. Region similarity is obtained by comparing different regions with known diseases in the lesion feature database, and the region similarity is labeled as follows: Obtain the overall similarity of the eyes and label the overall similarity of the eyes as: Where i = 1, 2, 3, ..., I, I is a positive integer, and i is the ID of a known disease in the lesion feature database. The formula for calculating the overall similarity of the eye is: ; Set a region similarity threshold and an overall similarity threshold. Compare the region similarity scores with the region similarity thresholds to obtain the number of scores exceeding the region similarity thresholds, and mark the number of scores exceeding the region similarity thresholds as: The overall similarity is compared with an overall similarity threshold to obtain the number of values greater than the overall similarity threshold, and these values are marked as follows: ; Set standard values for the number of values exceeding the regional similarity threshold and the overall similarity threshold, obtain the number deviations of values exceeding the regional similarity threshold and the overall similarity threshold, and calculate the disease complexity coefficient using the formula: ;in, The complexity coefficient of the disease. The number of deviations exceeding the overall similarity threshold. The number of deviations exceeding the region similarity threshold; By comparing the grayscale values of different regions of the eye with the normal baseline grayscale values, the abnormal grayscale change coefficient of the patient is obtained. The logic for obtaining the abnormal grayscale change coefficient is as follows: obtain the normal baseline grayscale values of different regions of the eye, and mark the normal baseline grayscale values of different regions of the eye as follows: The grayscale values of different regions of the actual patient's eye were obtained and labeled as follows: ; The differences between the gray values of different regions of the actual patient's eye and the normal baseline gray values were obtained, and these differences were denoted as: ,in, ; The mean and standard deviation of the differences between the gray values of different regions of the actual patient's eye and the normal baseline gray values were obtained, and these mean and standard deviations were denoted as: and ,in, , ; The abnormal grayscale variation coefficient is calculated using the following formula: ;in, This is the coefficient for abnormal grayscale variation.
[0007] In a preferred embodiment, the credibility of the patient's historical medical records is determined by fuzzy reasoning, including: The reliability coefficient of a patient's historical medical history is determined by examining the patient's examination records. The logic for obtaining the reliability coefficient is as follows: based on the patient's historical medical records, the historical medical records are compared with known diseases in the lesion feature database to determine whether a known disease in the lesion feature database exists in the historical medical records. If not, the reliability coefficient of the historical medical history is 0; if it exists, the following steps are included: The system obtains the duration of the patient's illness, the severity of the illness, and the interval between examinations. The duration of the patient's illness refers to the length of time from the onset of the illness to the present, the severity of the illness refers to the severity of the condition, and the interval between examinations refers to the length of time from the last examination to the present. The duration of the patient's illness, the severity of the illness, and the interval between examinations are used as fuzzy variables, and a fuzzy set is defined. The credibility is used as the output, and fuzzy rules are defined. The specific input values are mapped to fuzzy sets, and fuzzy inference is performed to calculate the membership degree of the output fuzzy set. The credibility fuzzy set obtained by fuzzy inference is converted into specific numerical values to obtain the historical disease credibility coefficient.
[0008] In a preferred embodiment, quantifying the reliability of eye images in assessing a condition includes: A weighted analysis is performed on the disease complexity coefficient, the disease complexity index, and the historical disease reliability coefficient to generate a disease assessment coefficient. The formula for calculating the disease assessment coefficient is as follows: ;in, This is the disease assessment coefficient. The historical disease reliability coefficient. , , This refers to the proportional coefficients of the disease complexity coefficient, the disease complexity factor, and the historical disease reliability coefficient. , , Greater than 0.
[0009] In a preferred embodiment, the analysis combines the quantitative results of the disease assessment coefficient and the patient's visual acuity data, including: High-precision vision testing instruments are used to test patients' vision, and the vision data is standardized and the quantitative results of the vision data are divided into 0 to 10. By setting thresholds for disease assessment coefficients and visual acuity data, quantitative results of patients' disease assessment coefficients and visual acuity data are obtained. A classification assessment coefficient is then generated using a logistic regression model. The formula for calculating the classification assessment coefficient is as follows: ;in, For classification evaluation coefficients, The threshold for the disease assessment coefficient. For visual acuity data threshold, Here, b represents the quantitative result of the patient's visual acuity data, and b is the intercept in the linear regression model. , These are the weight coefficients of the features in the Logistic regression model.
[0010] In a preferred embodiment, patients are selectively labeled using a logistic regression model, including: A classification assessment coefficient threshold is set, and the classification assessment coefficient is compared with the classification assessment coefficient threshold. If the classification assessment coefficient is greater than the classification assessment coefficient threshold, a labeled signal is generated. Patients with labeled signals are diagnosed by an expert team or more experienced doctors. If the classification assessment coefficient is less than the classification assessment coefficient threshold, no labeled signal is generated, and the patient's eye treatment is handled according to the normal examination procedure.
[0011] The technical effects and advantages of this invention are as follows: This invention first uses image processing technology to identify and analyze patient eye images, and quantifies the analysis results to determine the complexity of the patient's eye lesions. Secondly, it combines the patient's vision data and uses a logistic regression model to classify the patient. This helps to improve the efficiency and accuracy of eye lesion identification through automated image processing and machine learning models, and helps doctors provide a comprehensive eye health assessment by combining image features and vision data, thereby reducing the misdiagnosis rate. Attached Figure Description
[0012] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings; Figure 1 This is a schematic diagram of the structure of a vision screening data analysis system based on image technology according to the present invention. Detailed Implementation
[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example
[0014] Figure 1 A schematic diagram of the structure of a vision screening data analysis system based on image technology according to the present invention is provided, including an image data acquisition module, a patient assessment module, a vision data acquisition module, and a logistic regression module, with signal connections between the modules; The image data acquisition module is used to acquire patient eye images using a high-definition camera eye detection device, acquire patient eye data through image technology, and determine the credibility of historical cases through fuzzy reasoning based on the patient's historical medical records. The patient assessment module is used to perform weighted analysis of the disease complexity coefficient, the disease complexity coefficient, and the historical disease reliability coefficient to generate a disease assessment coefficient, which quantifies the reliability of eye images in judging the disease. The vision data acquisition module is used to perform vision tests on patients using high-precision vision testing instruments and obtain quantitative results of patients' vision data. The logistic regression module is used to combine the quantitative results of disease assessment coefficients and patient vision data for analysis, and selectively label patients through a logistic regression model.
[0015] The system is equipped with a high-definition camera to capture high-resolution images of the eye. The captured images are preprocessed, including noise reduction, contrast enhancement, and normalization, to improve image quality. Useful features are extracted from the preprocessed images, and the extracted features are compared with a feature library of common eye diseases to calculate similarity and identify possible diseases.
[0016] The features of the eye images include the texture, color, curvature, and transparency of the eye. A lesion feature database for different eye diseases is established. By comparing the features in the images with known diseases in the lesion feature database, the possible diseases in the patient's eyes can be identified.
[0017] It should be noted that visual examination of the eyes can help screen for various eye problems as a first step in disease screening, including conjunctivitis, dry eye syndrome, eyelid problems (such as meibomian gland dysfunction), and changes in iris color. These problems may require further professional examination, but visual examination of the eyes can serve as a first step to determine whether further examination is needed.
[0018] By using image processing techniques to segment eye images, including edge detection, image enhancement, and filtering, high-resolution eye images captured by high-definition cameras can be divided into regions. Using image processing techniques to segment eye images captured by high-definition cameras has several advantages, including: Precise localization of eye structures: Edge detection technology can accurately locate the boundaries of eye structures, such as the pupil, iris, and cornea, thereby enabling precise segmentation of eye images and helping to obtain features of different regions of the eye image.
[0019] Enhancing Image Details: Image enhancement and filtering techniques can improve image contrast and clarity, enhance details in eye images, help to more accurately analyze and diagnose eye diseases, reduce noise and interference in images, and improve the accuracy and stability of subsequent processing steps.
[0020] Similarity analysis of different regions of the eye is used to diagnose diseases. This includes using techniques such as edge detection and morphological processing to extract shape and structural features of different regions of the eye, comparing the shape and structural features of different regions to assess whether there are abnormal changes or deformities, and using methods such as gray-level co-occurrence matrix (GLCM) and local binary pattern (LBP) to extract texture features of different regions of the eye. The extracted texture features are compared and similarity is calculated to identify possible abnormalities or diseases in different regions.
[0021] By comparing features in an image with known diseases in a lesion feature database, a disease complexity coefficient for determining the patient's eye is obtained. The logic for obtaining the disease complexity coefficient is as follows: The number of different regions of the eye is obtained and labeled as n, where n = 1, 2, 3, ..., N, and N is a positive integer. Region similarity is obtained by comparing different regions with known diseases in the lesion feature database, and the region similarity is labeled as follows: Obtain the overall similarity of the eyes and label the overall similarity of the eyes as: Where i = 1, 2, 3, ..., I, I is a positive integer, and i is the ID of a known disease in the lesion feature database. The formula for calculating the overall similarity of the eye is: ; Set a region similarity threshold and an overall similarity threshold. Compare the region similarity scores with the region similarity thresholds to obtain the number of scores exceeding the region similarity thresholds, and mark the number of scores exceeding the region similarity thresholds as: The overall similarity is compared with an overall similarity threshold to obtain the number of values greater than the overall similarity threshold, and these values are marked as follows: ; It should be noted that regional similarity represents the similarity between different regions of the eye and a certain disease, while overall similarity represents the similarity between the entire eye and a certain disease. The number of different regions of the eye is usually set by ophthalmologists according to specific research or clinical needs, and is determined based on the quality of eye images and eye structure. When the similarity between a certain region of the eye and a certain disease is greater than the regional similarity threshold, it indicates that the patient is more likely to have that disease. A higher overall similarity of the eye indicates a higher probability of having the disease.
[0022] Set standard values for the number of values exceeding the regional similarity threshold and the overall similarity threshold, obtain the number deviations of values exceeding the regional similarity threshold and the overall similarity threshold, and calculate the disease complexity coefficient using the formula: ;in, The complexity coefficient of the disease. The number of deviations exceeding the overall similarity threshold. The number of deviations is greater than the region similarity threshold.
[0023] As the formula shows, the higher the disease complexity coefficient, the more likely the patient may have corresponding lesions or pathological features, and the more lesions there may be. It is not easy to distinguish which specific lesion it is, and direct diagnosis using image technology may lead to a decrease in accuracy.
[0024] To ensure consistent light sources and minimize the impact of external lighting conditions on grayscale values, grayscale values under normal conditions can be obtained. This specifically includes: Use specially designed light source equipment, such as standardized ring lights, uniform illumination systems, or other medical-grade lighting equipment, to ensure consistent brightness and color temperature of the light source, and to ensure consistent lighting conditions for each image acquisition.
[0025] We collected a large number of eye images from healthy individuals under the same light source conditions to ensure the diversity and representativeness of the samples. We used filtering techniques (such as Gaussian filtering and median filtering) to remove noise from the images, improve image quality, and obtain gray values of different regions of the eye.
[0026] By comparing the grayscale values of different regions of the eye with the normal baseline grayscale values, the abnormal grayscale change coefficient of the patient is obtained. The logic for obtaining the abnormal grayscale change coefficient is as follows: obtain the normal baseline grayscale values of different regions of the eye, and mark the normal baseline grayscale values of different regions of the eye as follows: The grayscale values of different regions of the actual patient's eye were obtained and labeled as follows: ; The differences between the gray values of different regions of the actual patient's eye and the normal baseline gray values were obtained, and these differences were denoted as: ,in, ; The mean and standard deviation of the differences between the gray values of different regions of the actual patient's eye and the normal baseline gray values were obtained, and these mean and standard deviations were denoted as: and ,in, , ; The abnormal grayscale variation coefficient is calculated using the following formula: ;in, This is the coefficient for abnormal grayscale variation.
[0027] As can be seen from the formula, the larger the abnormal gray value change coefficient, the greater the degree to which the gray value of the area deviates from the normal baseline gray value, which may indicate a significant lesion or abnormality. This suggests that the patient's eye may have a greater hidden danger, and further investigation is needed to determine the cause of the abnormality.
[0028] The reliability coefficient of a patient's historical medical history is determined by examining the patient's examination records. The logic for obtaining the reliability coefficient is as follows: based on the patient's historical medical records, the historical medical records are compared with known diseases in the lesion feature database to determine whether a known disease in the lesion feature database exists in the historical medical records. If not, the reliability coefficient of the historical medical history is 0; if it exists, the following steps are included: The system obtains the duration of the patient's illness, the severity of the illness, and the interval between examinations. The duration of the patient's illness refers to the length of time from the onset of the illness to the present, the severity of the illness refers to the severity of the condition, and the interval between examinations refers to the length of time from the last examination to the present. The duration of the patient's illness, the severity of the illness, and the interval between examinations are used as fuzzy variables, and a fuzzy set is defined. The credibility is used as the output, and fuzzy rules are defined. For example, if the duration of illness is long, the severity of illness is severe, and the interval between examinations is short, then the reliability is high. If the duration of illness is short or the severity of illness is mild, the credibility is low. If the duration of illness is moderate, the severity of illness is moderate, and the interval between examinations is long, then the confidence level is moderate.
[0029] The specific input values are mapped to fuzzy sets, and fuzzy inference is performed to calculate the membership degree of the output fuzzy set. The credibility fuzzy set obtained by fuzzy inference is converted into specific numerical values to obtain the historical disease credibility coefficient.
[0030] It should be noted that defuzzification is usually performed using methods such as weighted average method and the highest membership degree rule to obtain the final historical disease confidence coefficient. The higher the historical disease confidence coefficient, the more support and reference the patient's case can provide to the doctor, and the higher the accuracy of diagnosis.
[0031] The disease complexity coefficient, the disease reliability coefficient, and the historical disease reliability coefficient are weighted and analyzed to generate the disease assessment coefficient. The formula for calculating the disease assessment coefficient is as follows: ;in, This is the disease assessment coefficient. The historical disease reliability coefficient. , , This refers to the proportional coefficients of the disease complexity coefficient, the disease complexity factor, and the historical disease reliability coefficient. , , Greater than 0.
[0032] As can be seen from the formula, the smaller the historical disease confidence coefficient and the larger the disease complexity coefficient and disease complexity coefficient, the larger the disease assessment coefficient, indicating that the reliability of judging the disease through the eye images with higher resolution is lower. Conversely, the larger the historical disease confidence coefficient and the smaller the disease complexity coefficient and disease complexity coefficient, the smaller the disease assessment coefficient, indicating that the reliability of judging the disease through the eye images with higher resolution is higher.
[0033] High-precision vision testing instruments are used to test patients' vision. The vision data is standardized and the quantitative results of the vision data are divided into 0 to 10. The higher the quantitative result of the vision data, the better the vision data performance. Conversely, the lower the quantitative result of the vision data, the worse the vision data performance.
[0034] By setting thresholds for disease assessment coefficients and visual acuity data, quantitative results of patients' disease assessment coefficients and visual acuity data are obtained. A classification assessment coefficient is then generated using a logistic regression model. The formula for calculating the classification assessment coefficient is as follows: ;in, For classification evaluation coefficients, The threshold for the disease assessment coefficient. For visual acuity data threshold, Here, b represents the quantitative result of the patient's visual acuity data, and b is the intercept in the linear regression model. , These are the weight coefficients of the features in the Logistic regression model.
[0035] It should be noted that a higher classification assessment coefficient indicates a higher importance of the patient's condition, suggesting that the patient's eye lesions may be more severe and complex, requiring closer observation by the doctor in order to obtain a more accurate diagnosis.
[0036] A classification assessment coefficient threshold is set, and the classification assessment coefficient is compared with the classification assessment coefficient threshold. If the classification assessment coefficient is greater than the classification assessment coefficient threshold, a labeled signal is generated. Patients with labeled signals are diagnosed by an expert team or more experienced doctors. If the classification assessment coefficient is less than the classification assessment coefficient threshold, no labeled signal is generated, and the patient's eye treatment is handled according to the normal examination procedure.
[0037] This invention first uses image processing technology to identify and analyze patient eye images, and quantifies the analysis results to determine the complexity of the patient's eye lesions. Secondly, it combines the patient's vision data and uses a logistic regression model to classify the patient. This helps to improve the efficiency and accuracy of eye lesion identification through automated image processing and machine learning models, and helps doctors provide a comprehensive eye health assessment by combining image features and vision data, thereby reducing the misdiagnosis rate.
[0038] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.
[0039] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.
[0040] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.
[0041] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0042] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.
[0043] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0044] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0045] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A vision screening data analysis system based on image technology, characterized in that, It includes an image data acquisition module, a patient assessment module, a vision data acquisition module, and a logistic regression module, with signal connections between the modules; The image data acquisition module is used to acquire patient eye images using a high-definition camera eye detection device, acquire patient eye data through image technology, and determine the credibility of historical cases through fuzzy reasoning based on the patient's historical medical records. The patient assessment module is used to perform weighted analysis of the disease complexity coefficient, the disease complexity coefficient, and the historical disease reliability coefficient to generate a disease assessment coefficient, which quantifies the reliability of eye images in judging the disease. The vision data acquisition module is used to perform vision tests on patients using high-precision vision testing instruments and obtain quantitative results of patients' vision data. The logistic regression module is used to combine the quantitative results of disease assessment coefficients and patient vision data for analysis, and selectively label patients through a logistic regression model.
2. The vision screening data analysis system based on image technology according to claim 1, characterized in that, Data on a patient's eyes is acquired through imaging technology, including: By comparing features in an image with known diseases in a lesion feature database, a disease complexity coefficient for determining the patient's eye is obtained. The logic for obtaining the disease complexity coefficient is as follows: The number of different regions of the eye is obtained and labeled as n, where n = 1, 2, 3, ..., N, and N is a positive integer. Region similarity is obtained by comparing different regions with known diseases in the lesion feature database, and the region similarity is labeled as follows: Obtain the overall similarity of the eyes and label the overall similarity of the eyes as: Where i = 1, 2, 3, ..., I, I is a positive integer, and i is the ID of a known disease in the lesion feature database. The formula for calculating the overall similarity of the eye is: ; Set a region similarity threshold and an overall similarity threshold. Compare the region similarity scores with the region similarity thresholds to obtain the number of scores exceeding the region similarity thresholds, and mark the number of scores exceeding the region similarity thresholds as: The overall similarity is compared with an overall similarity threshold to obtain the number of values greater than the overall similarity threshold, and these values are marked as follows: ; Standard values are set for the number of values exceeding the regional similarity threshold and the overall similarity threshold. The number deviations of values exceeding the regional similarity threshold and the overall similarity threshold are obtained. The disease complexity coefficient is then calculated using the following formula: ;in, The complexity coefficient of the disease. The number of deviations exceeding the overall similarity threshold. The number of deviations exceeding the region similarity threshold; By comparing the grayscale values of different regions of the eye with the normal baseline grayscale values, the abnormal grayscale change coefficient of the patient is obtained. The logic for obtaining the abnormal grayscale change coefficient is as follows: obtain the normal baseline grayscale values of different regions of the eye, and mark the normal baseline grayscale values of different regions of the eye as follows: The grayscale values of different regions of the actual patient's eye were obtained and labeled as follows: ; The differences between the gray values of different regions of the actual patient's eye and the normal baseline gray values were obtained, and these differences were denoted as: ,in, ; The mean and standard deviation of the differences between the gray values of different regions of the actual patient's eye and the normal baseline gray values were obtained, and these mean and standard deviations were denoted as: and ,in, , ; The abnormal grayscale variation coefficient is calculated using the following formula: ;in, This is the coefficient for abnormal grayscale variation.
3. The vision screening data analysis system based on image technology according to claim 1, characterized in that, Based on the patient's medical history, the reliability of the medical history is determined through fuzzy reasoning, including: The reliability coefficient of a patient's historical medical history is determined by examining the patient's examination records. The logic for obtaining the reliability coefficient is as follows: based on the patient's historical medical records, the historical medical records are compared with known diseases in the lesion feature database to determine whether a known disease in the lesion feature database exists in the historical medical records. If not, the reliability coefficient of the historical medical history is 0; if it exists, the following steps are included: The system obtains the duration of the patient's illness, the severity of the illness, and the interval between examinations. The duration of the patient's illness refers to the length of time from the onset of the illness to the present, the severity of the illness refers to the severity of the condition, and the interval between examinations refers to the length of time from the last examination to the present. The duration of the patient's illness, the severity of the illness, and the interval between examinations are used as fuzzy variables, and a fuzzy set is defined. The credibility is used as the output, and fuzzy rules are defined. The specific input values are mapped to fuzzy sets, and fuzzy inference is performed to calculate the membership degree of the output fuzzy set. The credibility fuzzy set obtained by fuzzy inference is converted into specific numerical values to obtain the historical disease credibility coefficient.
4. The vision screening data analysis system based on image technology according to claim 3, characterized in that, Quantifying the reliability of eye images in diagnosing the condition includes: A weighted analysis is performed on the disease complexity coefficient, the disease complexity index, and the historical disease reliability coefficient to generate a disease assessment coefficient. The formula for calculating the disease assessment coefficient is as follows: ;in, This is the disease assessment coefficient. The historical disease reliability coefficient. , , This refers to the proportional coefficients of the disease complexity coefficient, the disease complexity factor, and the historical disease reliability coefficient. , , Greater than 0.
5. The vision screening data analysis system based on image technology according to claim 1, characterized in that, The analysis combines the quantitative results of disease assessment coefficients and patient vision data, including: High-precision vision testing instruments are used to test patients' vision, and the vision data is standardized and the quantitative results of the vision data are divided into 0 to 10. By setting thresholds for disease assessment coefficients and visual acuity data, quantitative results of patients' disease assessment coefficients and visual acuity data are obtained. A classification assessment coefficient is then generated using a logistic regression model. The formula for calculating the classification assessment coefficient is as follows: ;in, For classification evaluation coefficients, The threshold for the disease assessment coefficient. For visual acuity data threshold, Here, b represents the quantitative result of the patient's visual acuity data, and b is the intercept in the linear regression model. , These are the weight coefficients of the features in the Logistic regression model.
6. The vision screening data analysis system based on image technology according to claim 5, characterized in that, Patients were selectively labeled using a logistic regression model, including: A classification assessment coefficient threshold is set, and the classification assessment coefficient is compared with the classification assessment coefficient threshold. If the classification assessment coefficient is greater than the classification assessment coefficient threshold, a labeled signal is generated. Patients with labeled signals are diagnosed by an expert team or more experienced doctors. If the classification assessment coefficient is less than the classification assessment coefficient threshold, no labeled signal is generated, and the patient's eye treatment is handled according to the normal examination procedure.