Differential Analysis System and Method for Traditional Chinese Medicine Diabetic Retinopathy Based on Artificial Intelligence

Abstract

The present invention discloses a differential analysis system and method for traditional Chinese medicine diabetic retinopathy based on artificial intelligence, which relates to the technical field of artificial intelligence and includes a real-time adjustment end, a differential analysis end, and a formula generation end; the real-time adjustment end is used to detect the retinal texture of diabetic patients in real time and automatically generate the retinal image texture information of diabetic patients through an image converter; the differential analysis end is used to receive the retinal texture comparison results of diabetic patients in real time; the formula generation end is used to receive the retinal lesion difference data in real time and based on the standard traditional Chinese medicine formulas set for different lesions according to the retinal lesion difference data. The differential analysis system and method for traditional Chinese medicine diabetic retinopathy based on artificial intelligence can scale or enlarge the retinal comparison image in real time according to the size of the retinas of different patients, improve the accuracy of the differential analysis of diabetic patients' retinas, and improve the accuracy of the differential analysis of diabetic patients' retinas.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence, and particularly to a differential analysis system and method for diabetic retinopathy of traditional Chinese medicine based on artificial intelligence. Background Art

[0002] Diabetic retinopathy is a common chronic complication of diabetes, mainly affecting the microvasculature of the fundus, leading to vision loss or even blindness. In traditional Chinese medicine, diabetic retinopathy is usually referred to as "consumptive thirst with cataract". Traditional Chinese medicine believes that the occurrence of diabetic retinopathy is closely related to factors such as qi and yin deficiency, liver and kidney deficiency, yin and yang deficiency, and collateral stasis caused by long-term diabetes. These factors can lead to microcirculation disorders in the eyes, causing characteristic changes in the retina such as retinal hemorrhage and exudation, macular edema, and proliferative retinal detachment, seriously affecting the visual function and quality of life of patients.

[0003] Publication No. CN112237415A discloses an early detection method for diabetic retinopathy based on OCTA image texture analysis. Through mathematical statistics methods, the differences in various characteristic parameters in the control group and ROC analysis are obtained to obtain the boundary values; the characteristic parameters of unknown samples are obtained through step four and compared with the boundary values in step five, and then the unknown samples are classified according to the comparison results. The present invention can quickly and accurately judge the retinopathy situation of diabetic patients, and accurately give corresponding traditional Chinese medicine treatment formulas according to individual differences. The whole process is through a non-invasive method, which will not cause discomfort and trauma to patients.

[0004] After searching the above patents, it is found that there are some deficiencies in the differential analysis of diabetic retinopathy of traditional Chinese medicine: 1. Due to individual differences, there are differences in the eye parameters of different patients. When analyzing diabetic retinopathy based on OCTA image texture analysis, image noise reduction and texture feature comparison are performed, and it is impossible to scale or enlarge the retinal comparison image in real time according to the size of the patient's retina, resulting in limitations in the differential analysis of diabetic retinopathy in different patients and affecting the convenience of differential analysis of diabetic retinopathy; 2. Since the differential analysis of diabetic retinopathy is based on real-time comparison of retinal textures, but due to the different retinal fundus structures of different patients, it is impossible to analyze the lesion differences in real time according to the texture structures at different positions and angles of the retinas of diabetic patients, affecting the accuracy of differential analysis of diabetic patients' retinas; 3. When analyzing the differences in diabetic patients' retinas, although the differential analysis results of retinal lesions can be evaluated and data can be generated, it is impossible to automatically generate a traditional Chinese medicine formula for treating diabetic patients according to the differential analysis data, resulting in limited differential analysis schemes for diabetic retinopathy.

[0005] Therefore, a differential analysis system and method for traditional Chinese medicine diabetic retinopathy based on artificial intelligence are proposed to solve the above problems. Summary of the Invention

[0006] The main object of the present invention is to provide a differential analysis system and method for traditional Chinese medicine diabetic retinopathy based on artificial intelligence to solve the problems raised in the above background.

[0007] To achieve the above object, the technical solution adopted by the present invention is: a differential analysis system and method for traditional Chinese medicine diabetic retinopathy based on artificial intelligence, including a real-time adjustment terminal, a differential analysis terminal, and a formula generation terminal;

[0008] The real-time adjustment terminal is used to detect the retinal texture of diabetic patients in real time, automatically generate the retinal image texture information of diabetic patients through an image converter, set the standard retinal texture, and scale or enlarge the retinal comparison texture in real time according to the size of the patient's retina, move the retinal texture in real time to overlap with the standard retinal texture, and judge whether the color of the retinal texture is abnormal in real time according to the color difference;

[0009] The differential analysis terminal is used to receive the retinal texture comparison results of diabetic patients in real time, adjust the texture structures at different positions and different angles of the retinal texture comparison of diabetic patients in real time, track the retinal texture comparison position and comparison angle in real time, and analyze the lesion differences to detect whether the patient has retinal lesions in real time;

[0010] The formula generation terminal is used to receive the retinal lesion difference data in real time, and based on the standard traditional Chinese medicine formulas set for different lesions according to the retinal lesion difference data, automatically generate a traditional Chinese medicine formula report for treating diabetic patients, and visually display the traditional Chinese medicine formula report combined with the evaluation data on the computer page in real time, realizing the differential analysis of traditional Chinese medicine diabetic retinopathy based on artificial intelligence.

[0011] The real-time adjustment terminal includes a data detection module, an information acquisition module, and an adjustment tracking module;

[0012] The data detection module includes a retinal texture unit and a multi-position recording unit;

[0013] The retinal texture unit is used to capture the retinal texture data of diabetic patients in real time through Eye print ID and record it in real time through a data recorder;

[0014] The multi-position recording unit is used to set multiple positions for capturing the retinal texture of diabetic patients and record the corresponding positions of the multiple captured retinal textures of diabetic patients in real time through a data recorder.

[0015] The information acquisition module includes a standard data unit and an image generation unit;

[0016] The standard data unit is used to set standard retinal texture information according to the retinal data of normal objects;

[0017] The image generation unit is used to generate an image of the retinal texture of a diabetic patient captured at multiple positions in real time through an image generator, so as to obtain the patient's retinal texture image information.

[0018] The adjustment and tracking module includes a zooming unit, a retinal comparison unit, a real-time movement unit, and a color judgment unit;

[0019] The zooming unit is used to reduce or enlarge the texture comparison range of the retinal texture image information and the standard retinal texture information in real time through touch screen technology;

[0020] The retinal comparison unit is used to compare the retinal texture information in real time in combination with different positions, and the comparison formula is as follows:

[0021] ;

[0022] Wherein, represents the comparison data value of the retinal texture information at different positions, represents the standard value of the retinal texture information at the corresponding moment, represents the base value of the retinal texture information at the corresponding moment, represents the base data of the retinal texture information at the corresponding position tracked, represents the standard data of the retinal texture information at the corresponding position;

[0023] If the calculated = 0, it is determined that the operation of reducing or enlarging the tracked retinal texture is executed normally. If ≠ 0, it is determined that the operation of reducing or enlarging the tracked retinal texture is executed abnormally;

[0024] The real-time movement unit is used to adjust the comparison position and the comparison position of the retinal texture information in real time through touch screen technology;

[0025] The color judgment unit is used to calculate the color difference value △E between the retinal texture information and set a color difference threshold. The calculation formula of △E is as follows:

[0026] △E = [(△L)^2+(△a)^2+(△b )^2]^0.5;

[0027] Among them, △L, △a, and △b respectively represent the differences between two colors on the three axes of L, a, and b. △E represents the difference value between the retinal texture information and the standard retinal texture information. The difference value is calculated by subtracting the color difference threshold. If the result of the difference calculation is greater than or equal to 0.5 or less than or equal to -0.5, it indicates that the retinal texture information has abnormal retinopathy. If the result of the difference calculation is less than 0.5 and greater than -0.5, it indicates that the retinal texture information has normal retinopathy.

[0028] The difference analysis terminal includes a comparison adjustment module, a real-time comparison module, and a lesion tracking module;

[0029] The comparison adjustment module includes a different position unit and a different texture unit;

[0030] The different position unit is used to calculate the difference value between the retinal texture information and the standard retinal texture information by combining different positions;

[0031] The different texture unit is used to calculate the difference value between the retinal texture information and the standard retinal texture information by combining different texture routes.

[0032] The real-time comparison module includes a position comparison unit and an angle comparison unit;

[0033] The position comparison unit is used to calculate the comparison difference value of the retinal texture information at different positions in real time. The calculation formula is as follows:

[0034] S = D_min - d_max, where S represents the comparison difference value of the retinal texture information at different positions, D_min represents the minimum diameter of the position comparison, and d_max represents the maximum diameter of the position comparison;

[0035] The angle comparison unit is used to calculate in real time whether there is an angle deviation in the comparison of the retinal texture information at different angles. The standard retinal texture angle at the corresponding position is set as the center point. The calculation method is as follows:

[0036] △θ = | - β|, where △θ represents the angle deviation that appears in the retinal texture information at different angles, represents the current angle of the corresponding retinal texture information comparison, and β represents the standard angle of the corresponding retinal texture information comparison. If △θ is not equal to 0, it indicates that the retinopathy is abnormal. If △θ is equal to 0, the retinopathy is normal.

[0037] The lesion tracking module includes a lesion judgment unit and a lesion warning unit;

[0038] The lesion judgment unit is used to comprehensively analyze the retinopathy by combining the comparison difference value of the retinal texture information at different positions and the angle deviation that appears in the retinal texture information at different angles;

[0039] The lesion warning unit is used to report a voice alarm issued by the system when the comparison difference value of the retinal texture information at different positions is not equal to 0 and the angular deviation of the retinal texture information at different angles is not equal to 0, record it in real time through a data recorder, and capture abnormal images of retinal lesions in real time through a camera.

[0040] The formula generation end includes a data receiving module, a traditional Chinese medicine formula module, a data evaluation module, and a merging and display module;

[0041] The data receiving module is used to receive retinal lesion difference data in real time through a data receiver;

[0042] The traditional Chinese medicine formula module includes a standard formula unit and a lesion matching unit, and records in real time through a data recorder the difference value of the comparison of the retinal texture information at different positions and the angular deviation of the retinal texture information at different angles;

[0043] The standard formula unit is used to set traditional Chinese medicine formulas with different dosages and different treatment cycles according to diabetic retinopathy;

[0044] The lesion matching unit is used to match traditional Chinese medicine formulas in real time according to the difference value of the comparison of the retinal texture information at different positions and the angular deviation of the retinal texture information at different angles, and in combination with different dosages and different treatment cycles.

[0045] The data evaluation module is used to automatically generate a traditional Chinese medicine formula report for treating diabetic patients according to the difference value of the comparison of the retinal texture information at different positions and the angular deviation of the retinal texture information at different angles;

[0046] The merging and display module is used to visually display the traditional Chinese medicine formula in real time through a computer, and simultaneously display the results of the difference analysis data of traditional Chinese medicine diabetic retinopathy during the visual display.

[0047] A method for differential analysis of traditional Chinese medicine diabetic retinopathy based on artificial intelligence includes the following steps:

[0048] Step 1: Configure the IP address information of the remote control area server for differential analysis;

[0049] Step 2: Enter the real-time adjustment end, automatically generate a retinal image by detecting the retinal texture of a diabetic patient in real time, set a standard retinal texture, and scale or enlarge the retinal comparison in real time according to the size of the patient's retina, move in real time according to the retinal texture, and judge in real time whether the color of the retinal texture is abnormal;

[0050] Step 3: Enter the difference analysis terminal. Through the retinal images, adjust the texture structures at different positions and angles of the retinal texture comparison of diabetic patients in real time to analyze the lesion differences in real time, and detect in real time whether the patient has retinal lesions;

[0051] Step 4: Enter the formula generation terminal. Automatically generate a traditional Chinese medicine formula for treating diabetic patients through the retinal lesion difference data, and automatically visualize and display the traditional Chinese medicine formula in combination with the evaluation data.

[0052] The present invention has the following beneficial effects:

[0053] 1. In the present invention, by setting up a real-time adjustment terminal, during the difference analysis of diabetic retinopathy in traditional Chinese medicine based on artificial intelligence, by automatically generating the texture information of the retinal images of diabetic patients and scaling or enlarging the retinal comparison texture in real time according to the size of the patient's retina, when performing the difference analysis of diabetic retinopathy pathology, it is possible to scale or enlarge the retinal comparison image in real time according to the size of different patients' retinas, reducing the limitations existing in the difference analysis of diabetic retinopathy for different patients and increasing the convenience of the difference analysis of diabetic retinopathy.

[0054] 2. In the present invention, by setting up a difference analysis terminal, during the difference analysis of diabetic retinopathy in traditional Chinese medicine based on artificial intelligence, by adjusting the texture structures at different positions and different angles of the retinal texture comparison of diabetic patients in real time, the comprehensiveness of diabetic retinopathy detection is increased, and by analyzing the lesion differences in real time according to the texture structures at different positions and different angles of the diabetic patient's retina, the accuracy of the difference analysis of the diabetic patient's retina can be further improved.

[0055] 3. In the present invention, by setting up a formula generation terminal, during the difference analysis of diabetic retinopathy in traditional Chinese medicine based on artificial intelligence, by automatically generating a traditional Chinese medicine formula for treating diabetic patients and visualizing and displaying the traditional Chinese medicine formula in combination with the evaluation data in real time on the computer page, not only can the difference analysis results of retinal lesions be evaluated and data be generated, but also a traditional Chinese medicine formula for treating diabetic patients can be automatically generated according to the difference analysis data, reducing the limitations existing in the difference analysis scheme of diabetic retinopathy. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic flow chart of the difference analysis method of diabetic retinopathy in traditional Chinese medicine based on artificial intelligence of the present invention;

[0057] Figure 2 It is a schematic diagram of the structure of the real-time adjustment terminal of the difference analysis system of diabetic retinopathy in traditional Chinese medicine based on artificial intelligence of the present invention;

[0058] Figure 3Schematic diagram of the architecture of the differential analysis terminal of the differential analysis system for traditional Chinese medicine diabetic retinopathy based on artificial intelligence of the present invention;

[0059] Figure 4 Schematic diagram of the architecture of the real-time adjustment terminal of the differential analysis system for traditional Chinese medicine diabetic retinopathy based on artificial intelligence of the present invention. Detailed implementation manners

[0060] To make the technical means, creative features, achieved purposes and functions of the present invention easy to understand, the present invention will be further described below in conjunction with specific implementation manners.

[0061] Embodiment 1

[0062] Please refer to Figures 1 to 2 As shown: The differential analysis system and method for traditional Chinese medicine diabetic retinopathy based on artificial intelligence include a real-time adjustment terminal, a differential analysis terminal and a formula generation terminal;

[0063] The real-time adjustment terminal is used to detect the retinal texture of diabetic patients in real time, automatically generate the retinal image texture information of diabetic patients through an image converter, set the standard retinal texture, and scale or enlarge the retinal comparison texture in real time according to the size of the patient's retina, move the retinal texture in real time to overlap with the standard retinal texture, and judge whether the color of the retinal texture is abnormal in real time according to the color difference;

[0064] The differential analysis terminal is used to receive the retinal texture comparison results of diabetic patients in real time, adjust the texture structures at different positions and different angles of the retinal texture comparison of diabetic patients in real time, track the retinal texture comparison position and comparison angle in real time, and analyze the lesion differences to detect whether the patient has retinal lesions in real time;

[0065] The formula generation terminal is used to receive the retinal lesion difference data in real time, automatically generate a traditional Chinese medicine formula report for treating diabetic patients based on the standard traditional Chinese medicine formulas set for different lesions according to the retinal lesion difference data, and visually display the traditional Chinese medicine formula report combined with the evaluation data on the computer page in real time to realize the differential analysis of traditional Chinese medicine diabetic retinopathy based on artificial intelligence.

[0066] The real-time adjustment terminal includes a data detection module, an information acquisition module and an adjustment tracking module;

[0067] The data detection module includes a retinal texture unit and a multi-position recording unit;

[0068] The retinal texture unit is used to capture the retinal texture data of diabetic patients in real time through Eye print ID and record it in real time through a data recorder;

[0069] The multi-position recording unit is used to set multiple positions for capturing the retinal texture of diabetic patients, and record the corresponding positions of the multiple captured retinal textures of diabetic patients in real time through a data recorder.

[0070] The information acquisition module includes a standard data unit and an image generation unit;

[0071] The standard data unit is used to set standard retinal texture information based on the retinal data of normal subjects;

[0072] The image generation unit is used to generate an image of the retinal texture of a diabetic patient captured at multiple positions in real time through an image generator to obtain the retinal texture image information of the patient.

[0073] The adjustment and tracking module includes a zooming unit, a retinal comparison unit, a real-time movement unit, and a color judgment unit;

[0074] The zooming unit is used to reduce or enlarge the texture comparison range between the retinal texture image information and the standard retinal texture information in real time through touch screen technology, set the standard retinal texture, and scale or enlarge the retinal comparison texture in real time according to the size of the patient's retina, move the retinal texture to overlap with the standard retinal texture in real time, and judge whether the color of the retinal texture is abnormal in real time according to the color difference;

[0075] The retinal comparison unit is used to compare the retinal texture information in real time at different positions. The comparison formula is as follows:

[0076] ;

[0077] Among them, represents the comparison data value of the retinal texture information at different positions, represents the standard value of the retinal texture information at the corresponding moment, represents the basic value of the retinal texture information at the corresponding moment, represents the basic data of the retinal texture information at the corresponding position tracked, represents the standard data of the retinal texture information at the corresponding position;

[0078] If the calculated = 0, it is determined that the operation of reducing or enlarging the tracked retinal texture is executed normally. If ≠ 0, it is determined that the operation of reducing or enlarging the tracked retinal texture is executed abnormally;

[0079] The real-time movement unit is used to adjust the comparison position and the comparison position of the retinal texture information in real time through touch screen technology;

[0080] The color judgment unit is used to calculate the retinal texture information and the color difference value △E, and set the color difference threshold. The calculation formula of △E is as follows:

[0081] △E = [(△L)^2 + (△a)^2 + (△b )^2]^0.5;

[0082] Among them, △L, △a, and △b respectively represent the differences between two colors on the three axes of L, a, and b. △E represents the difference value between the retinal texture information and the standard retinal texture information. The difference value is calculated with the color difference threshold. If the result of the difference calculation is greater than or equal to 0.5 or less than or equal to -0.5, it indicates that the retinal texture information has abnormal retinopathy. If the result of the difference calculation is less than 0.5 and greater than -0.5, it indicates that the retinal texture information has normal retinopathy. It can scale or enlarge the retinal comparison image in real time according to the size of the retina of different patients, increase the accuracy of sharing the differences in retinopathy, reduce the limitations in the differential analysis of diabetic retinopathy for different patients, and increase the convenience of the differential analysis of diabetic retinopathy.

[0083] Example Two

[0084] Please refer to Figure 3 as shown: Based on Example One, the differential analysis terminal includes a comparison adjustment module, a real-time comparison module, and a lesion tracking module;

[0085] The comparison adjustment module includes a different position unit and a different texture unit;

[0086] The different position unit is used to calculate the difference value between the retinal texture information and the standard retinal texture information by combining different positions;

[0087] The different texture unit is used to calculate the difference value between the retinal texture information and the standard retinal texture information by combining different texture routes, and analyze the lesion differences in real time according to the texture structures at different positions and angles of the retinas of diabetic patients, which can further improve the accuracy of the differential analysis of the retinas of diabetic patients.

[0088] The real-time comparison module includes a position comparison unit and an angle comparison unit;

[0089] The position comparison unit is used to calculate the comparison difference value of the retinal texture information at different positions in real time. The calculation formula is as follows:

[0090] S = D_min - d_max, where S represents the comparison difference value of the retinal texture information at different positions, D_min represents the minimum diameter of the position comparison, and d_max represents the maximum diameter of the position comparison;

[0091] The angle comparison unit is used to calculate in real time whether there is an angle deviation in the comparison of retinal texture information at different angles. The standard retinal texture angle at the corresponding position is set as the center point, and the calculation method is as follows:

[0092] △θ = | -β|, where △θ represents the angle deviation that appears in the retinal texture information at different angles, represents the current angle of the comparison of the corresponding retinal texture information, and β represents the standard angle of the comparison of the corresponding retinal texture information. If △θ is not equal to 0, it indicates abnormal retinal lesions. If △θ is equal to 0, the retinal lesions are normal.

[0093] The lesion tracking module includes a lesion judgment unit and a lesion warning unit;

[0094] The lesion judgment unit is used to analyze retinal lesions by combining the difference value of the comparison of retinal texture information at different positions and the angle deviation that appears in the retinal texture information at different angles;

[0095] The lesion warning unit is used to report to the system to issue a voice alarm when the difference value of the comparison of retinal texture information at different positions is not equal to 0 and the angle deviation of the retinal texture information at different angles is not equal to 0, record in real time through a data recorder, and capture abnormal retinal lesion pictures in real time through a camera. By analyzing the lesion differences in real time according to the texture structures at different positions and different angles of the retinas of diabetic patients, the accuracy of the analysis of retinal differences in diabetic patients can be further improved.

[0096] Embodiment III

[0097] Please refer to Figure 4 as shown: Based on Embodiment I, the formula generation end includes a data receiving module, a traditional Chinese medicine formula module, a data evaluation module, and a merging and display module;

[0098] The data receiving module is used to receive retinal lesion difference data in real time through a data receiver;

[0099] The traditional Chinese medicine formula module includes a standard formula unit and a lesion matching unit, and records in real time through a data recorder the difference value of the comparison of retinal texture information at different positions and the angle deviation of the retinal texture information at different angles;

[0100] The standard formula unit is used to set traditional Chinese medicine formulas with different dosages and different treatment cycles according to diabetic retinopathy;

[0101] The lesion matching unit is used to match traditional Chinese medicine formulas in real time according to the difference value of the comparison of retinal texture information at different positions and the angle deviation of the retinal texture information at different angles, and in combination with different dosages and different treatment cycles.

[0102] The data evaluation module is used to automatically generate a traditional Chinese medicine formula report for treating diabetic patients based on the difference values obtained from comparing retinal texture information at different positions and the angular deviation of retinal texture information at different angles;

[0103] The combined display module is used to visually display the traditional Chinese medicine formula in real time through a computer, and simultaneously display the results of the differential analysis of diabetic retinopathy in traditional Chinese medicine during the visual display. It records in real time through a data recorder and captures abnormal images of retinopathy in real time through a camera. When conducting differential analysis of the retinas of diabetic patients, it can not only evaluate the results of the differential analysis of retinopathy and generate data, but also automatically generate a traditional Chinese medicine formula for treating diabetic patients based on the differential analysis data, reducing the limitations of the differential analysis scheme for diabetic retinopathy.

[0104] In the present invention, there is a system and method for differential analysis of traditional Chinese medicine diabetic retinopathy based on artificial intelligence. When operating this system, first configure the IP address information of the differential analysis remote control area server; enter the real-time adjustment end, automatically generate a retinal image by detecting the retinal texture of diabetic patients in real time, set a standard retinal texture, and scale or enlarge the retinal comparison in real time according to the size of the patient's retina, move the retina in real time according to the retinal texture, and judge in real time whether the color of the retinal texture is abnormal. By detecting the retinal texture of diabetic patients in real time and automatically generating the retinal image texture information of diabetic patients through an image converter, set a standard retinal texture, scale or enlarge the retinal comparison texture in real time according to the size of the patient's retina, overlap the real-time moved retinal texture with the standard retinal texture, and judge in real time whether the color of the retinal texture is abnormal according to the color difference, so that when performing differential analysis of diabetic retinopathy, it is possible to scale or enlarge the retinal comparison image in real time according to the size of different patients' retinas, increase the accuracy of differential analysis of retinal lesions, reduce the limitations existing in differential analysis of diabetic retinopathy for different patients, and increase the convenience of differential analysis of diabetic retinopathy; enter the differential analysis end, analyze the lesion differences in real time through the retinal image and adjust the texture structures at different positions and different angles of the retinal texture comparison of diabetic patients in real time, detect in real time whether the patient has retinal lesions, receive the retinal texture comparison results of diabetic patients in real time, and adjust the texture structures at different positions and different angles of the retinal texture comparison of diabetic patients in real time, analyze the lesion differences in real time, be able to detect in real time whether the patient has retinal lesions, increase the comprehensiveness of diabetic retinopathy detection, and analyze the lesion differences in real time according to the texture structures at different positions and different angles of the diabetic patient's retina, which can further improve the accuracy of differential analysis of the diabetic patient's retina; enter the formula generation end, automatically generate a traditional Chinese medicine formula for treating diabetic patients through the retinal lesion difference data, and visually display the traditional Chinese medicine formula combined with the evaluation data automatically. By receiving the retinal lesion difference data in real time and based on the standard traditional Chinese medicine formula set for different lesions according to the retinal lesion difference data, automatically generate a traditional Chinese medicine formula for treating diabetic patients, and visually display the traditional Chinese medicine formula combined with the evaluation data on the computer page in real time, so that when the system performs differential analysis of the diabetic patient's retina, it can not only evaluate the differential analysis results of retinal lesions and generate data, but also automatically generate a traditional Chinese medicine formula for treating diabetic patients according to the differential analysis data, reducing the limitations of the differential analysis scheme for diabetic retinopathy.

[0105] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and what is described in the above embodiments and the specification is only to illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements fall within the scope of the present invention claimed. The scope of protection claimed for the present invention is defined by the appended claims and their equivalents.

Claims

1. An artificial intelligence-based differential analysis system for diabetic retinopathy in traditional Chinese medicine, characterized in that The system includes a real-time adjustment terminal, a difference analysis terminal, and a formula generation terminal; The real-time adjustment terminal is used to detect the retinal texture of diabetic patients in real time, automatically generate the retinal image texture information of diabetic patients through an image converter, set the standard retinal texture, and scale or enlarge the retinal comparison texture in real time according to the size of the patient's retina, move the retinal texture in real time to overlap with the standard retinal texture, and judge whether the color of the retinal texture is abnormal in real time according to the color difference; The difference analysis terminal is used to receive the retinal texture comparison results of diabetic patients in real time, adjust the texture structures at different positions and angles of the retinal texture comparison of diabetic patients in real time, track the retinal texture comparison position and comparison angle in real time, and analyze the lesion differences to detect whether the patient has retinal lesions in real time; The formula generation terminal is used to receive the retinal lesion difference data in real time, and based on the standard traditional Chinese medicine formulas set for different lesions according to the retinal lesion difference data, automatically generate a traditional Chinese medicine formula report for treating diabetic patients, and visually display the traditional Chinese medicine formula report combined with the evaluation data on the computer page in real time, realizing the difference analysis of traditional Chinese medicine diabetic retinopathy based on artificial intelligence; The formula generation terminal includes a data reception module, a traditional Chinese medicine formula module, a data evaluation module, and a merging and display module; The data reception module is used to receive the retinal lesion difference data in real time through a data receiver; The traditional Chinese medicine formula module includes a standard formula unit and a lesion matching unit, and records the difference values of the retinal texture information comparison at different positions and the angular deviations of the retinal texture information at different angles in real time through a data recorder; The standard formula unit is used to set traditional Chinese medicine formulas with different dosages and different treatment cycles according to diabetic retinopathy; The lesion matching unit is used to match the traditional Chinese medicine formula in real time according to the difference values of the retinal texture information comparison at different positions and the angular deviations of the retinal texture information at different angles, combined with different dosages and different treatment cycles; The data evaluation module is used to automatically generate a traditional Chinese medicine formula report for treating diabetic patients according to the difference values of the retinal texture information comparison at different positions and the angular deviations of the retinal texture information at different angles; The merging and display module is used to visually display the traditional Chinese medicine formula on the computer in real time, and simultaneously display the difference analysis data results of traditional Chinese medicine diabetic retinopathy during the visual display.

2. The system according to claim 1, wherein: The real-time adjustment terminal includes a data detection module, an information acquisition module, and an adjustment tracking module; The data detection module includes a retinal texture unit and a multi-position recording unit; The retinal texture unit is used to capture the retinal texture data of diabetic patients in real time through Eye print ID and record it in real time through a data recorder; The multi-position recording unit is used to set multiple positions for capturing the retinal texture of diabetic patients and record the corresponding positions of capturing the retinal texture of diabetic patients in real time through a data recorder.

3. The system according to claim 2, wherein: The information acquisition module includes a standard data unit and an image generation unit; The standard data unit is used to set standard retinal texture information according to the retinal data of normal objects; The image generation unit is used to generate an image of the retinal texture of a diabetic patient captured at multiple positions in real time through an image generator to obtain the retinal texture image information of the patient.

4. The system according to claim 3, characterized in that: The adjustment and tracking module includes a zooming and enlarging unit, a retinal comparison unit, a real-time movement unit, and a color judgment unit; The zooming and enlarging unit is used to reduce or enlarge the texture comparison range of the retinal texture image information and the standard retinal texture information in real time through touch screen technology; The retinal comparison unit is used to compare the retinal texture information in real time at different positions. The comparison formula is as follows: ； Among them, represents the comparison data value of retinal texture information at different positions, represents the standard value of retinal texture information at the corresponding moment, represents the base value of retinal texture information at the corresponding moment, represents the base data of the retinal texture information at the corresponding position traced, represents the standard data of the retinal texture information at the corresponding position; If the calculated = 0, it is determined that the tracked shrinking or expanding retinal texture operation is executed normally. If ≠ 0, it is determined that the tracked shrinking or expanding retinal texture operation is executed abnormally; The real-time movement unit is used to adjust the comparison position of the retinal texture information in real time through touch screen technology; The color judgment unit is used to calculate the color difference value △E between the retinal texture information, and set a color difference threshold. The calculation formula of △E is as follows: △E = [(△L)^2 + (△a)^2 + (△b )^2]^0.5; Where, △L, △a, and △b respectively represent the differences between two colors on the three axes of L, a, and b. △E represents the difference value between the retinal texture information and the standard retinal texture information. Calculate the difference between the difference value and the color difference threshold. If the result of the difference calculation is greater than or equal to 0.5 or less than or equal to -0.5, it means that the retinal texture information has abnormal retinal lesions. If the result of the difference calculation is less than 0.5 and greater than -0.5, it means that the retinal texture information has normal retinal lesions.

5. The system according to claim 1, wherein: The difference analysis terminal includes a comparison adjustment module, a real-time comparison module, and a lesion tracking module; The comparison adjustment module includes a different position unit and a different texture unit; The different position unit is used to calculate the difference value between the retinal texture information and the standard retinal texture information by combining different positions; The different texture unit is used to calculate the difference value between the retinal texture information and the standard retinal texture information by combining different texture routes.

6. The system according to claim 5, characterized in that: The real-time comparison module includes a position comparison unit and an angle comparison unit; The position comparison unit is used to calculate the comparison difference value of the retinal texture information at different positions in real time. The calculation formula is as follows: S = D_min - d_max, where S represents the comparison difference value of the retinal texture information at different positions, D_min represents the minimum diameter of the position comparison, and d_max represents the maximum diameter of the position comparison; The angle comparison unit is used to calculate in real time whether there is an angle deviation in the comparison of the retinal texture information at different angles. Set the standard retinal texture angle at the corresponding position as the center point. The calculation method is as follows: △θ = | -β|, where △θ represents the angular deviation of the retinal texture information at different angles, represents the current angle for comparison of the corresponding retinal texture information, β represents the standard angle for comparison of the corresponding retinal texture information. If △θ is not equal to 0, it indicates abnormal retinal lesions. If △θ is equal to 0, the retinal lesions are normal.

7. The system according to claim 6, wherein: The lesion tracking module includes a lesion judgment unit and a lesion warning unit; The lesion judgment unit is used to analyze retinal lesions by combining the comparison difference value of the retinal texture information at different positions and the angle deviation of the retinal texture information at different angles; The lesion warning unit is used to report to the system to issue a voice alarm when the comparison difference value of the retinal texture information at different positions is not equal to 0, and the angle deviation of the retinal texture information at different angles is not equal to 0. Record it in real time through a data recorder, and capture abnormal retinal lesion pictures in real time through a camera.

Images