An ophthalmic disease image diagnosis system based on a big data model

Abstract

This invention discloses an ophthalmic disease image diagnosis system based on a big data model, relating to the field of disease image diagnosis technology. The system comprises an ophthalmic disease multi-dimensional diagnostic data acquisition module, an ophthalmic disease multi-dimensional diagnostic data processing module, an ophthalmic disease comprehensive data analysis module, and an ophthalmic diagnosis execution module. The ophthalmic disease multi-dimensional diagnostic data acquisition module collects basic information of diabetic patients, blood glucose data, fundus images, images of reflected and scattered light from the patient's eyes, and eye movement data. The ophthalmic disease multi-dimensional diagnostic data processing module processes the collected data for subsequent analysis. The ophthalmic disease comprehensive data analysis module analyzes the processed data and transmits abnormal signals to the ophthalmic diagnosis execution module. The ophthalmic diagnosis execution module receives abnormal signals, takes appropriate measures to process them, and sends the signals back to the client, where medical personnel make diagnoses based on the actual condition.

Description

Technical Field

[0001] This invention relates to the field of disease image diagnosis technology, specifically to an ophthalmic disease image diagnosis system based on a big data model. Background Technology

[0002] This ophthalmic disease image diagnosis system based on a big data model integrates advanced medical image processing, big data analysis, and artificial intelligence technologies. The system uses high-precision image acquisition equipment to obtain eye images, improves image quality through image preprocessing techniques to facilitate subsequent analysis, and then uses big data analysis techniques to deeply mine massive amounts of eye image data, revealing the potential relationship between disease characteristics and images. Simultaneously, by combining machine learning algorithms and deep learning models, the system can automatically identify and classify lesions in eye images, achieving accurate diagnosis. Furthermore, the system uses big data models to predict disease development trends, providing doctors with personalized treatment suggestions. Overall, this system, relying on powerful data processing capabilities and advanced AI technology, greatly improves the accuracy and efficiency of ophthalmic disease diagnosis, bringing revolutionary changes to the field of ophthalmology. Despite significant progress in existing big data model-based ophthalmic disease image diagnostic systems, several challenges remain: patients with diabetes increase the complexity of ophthalmic diagnosis, as this disease may interfere with the characteristics of ophthalmic images, making diagnosis more difficult; patients wearing contact lenses or forgetting to remove them during diagnosis can affect the results; and patients moving their eyes or closing their eyes during diagnosis can lead to inaccurate results. Summary of the Invention

[0003] To address the aforementioned issues, this invention provides an ophthalmic disease image diagnosis system based on a big data model, comprising an ophthalmic disease multi-dimensional diagnostic data acquisition module, an ophthalmic disease multi-dimensional diagnostic data processing module, an ophthalmic disease comprehensive data analysis module, and an ophthalmic diagnosis execution module. The ophthalmic disease multi-dimensional diagnostic data acquisition module obtains basic information of diabetic patients through doctor consultation and data entry, acquires blood glucose data using a blood glucose meter, acquires fundus images of different stages of diabetes using a fundus camera, acquires images of reflected and scattered light from the patient's eyes using a slit lamp, and acquires the movement trajectory and angular velocity of the patient's eyeballs during the diagnostic process using an eye tracker. The ophthalmic disease multi-dimensional diagnostic data processing module processes blood glucose data, obtains a blood glucose data bar chart for patients, uses a multimodal deep learning model to obtain the diagnostic pattern of diabetic patients, performs feature point matching and parameter adjustment on the images of reflected and scattered light from the eye, maps the eyeball in the form of a curve on a two-dimensional coordinate system to obtain an eyeball motion trajectory diagram, and summarizes the eyeball motion angular velocity in a table to obtain an eyeball motion angular velocity table. The comprehensive data analysis module for ophthalmic diseases uses blood glucose data, eye images, and eye movement trajectories to determine the ophthalmic diagnostic mode, contact lens wearing status, and eye movement status, respectively. The ophthalmic diagnostic execution module receives abnormal signals, mobilizes the system for processing, and transmits them to the client for medical personnel to diagnose and intervene.

[0004] Furthermore, the process by which the ophthalmic disease multivariate diagnostic data acquisition module acquires basic information, blood glucose data, and fundus images of diabetic patients at different stages of diabetes includes: A1. Consult with doctors about diabetic patients and enter their names and ages into the ophthalmic disease image diagnosis system to obtain basic information about diabetic patients; A2. Collect a blood sample from the patient. The blood glucose meter test strip contains glucose oxidase. When the blood sample comes into contact with the test strip, hydrogen peroxide is generated. The hydrogen peroxide undergoes an oxidation reaction on the electrode to generate an electric current. The blood glucose meter has a built-in sensor and microprocessor. The sensor captures the current, and the microprocessor first converts the analog current signal into a digital signal. Then, after calculation and calibration, the blood glucose data of the diabetic patient is obtained. A3. The fundus camera receives light from the patient's fundus and transmits it through the lens to the sensor of the fundus camera. Using digital image processing technology, the captured fundus image is converted into a digital signal. The image processing system integrated on the fundus camera is used to process the digital signal into a fundus image of the patient. By collecting fundus images from different patients multiple times, fundus images of diabetic patients at different stages of the disease can be obtained.

[0005] Furthermore, the process by which the ophthalmic disease multivariate diagnostic data acquisition module acquires images of reflected and scattered light from the patient's eye includes: The slit lamp uses a high-brightness light source and a magnifying glass system to focus light onto the patient's eye structure. By appropriately adjusting the angle, brightness, and magnification of the slit lamp light source, images of reflected and scattered light from the patient's eye are obtained, which describe the characteristic data of the patient wearing contact lenses.

[0006] Furthermore, the ophthalmic disease multi-dimensional diagnostic data acquisition module acquires the patient's eye movement trajectory and angular velocity during the diagnostic process, including: Based on eye-tracking technology, the eye tracker emits infrared light to the patient's eyeball. The infrared light is reflected back to the eye tracker by the eyeball. The built-in sensor of the eye tracker collects the infrared light and calculates the movement trajectory and angular velocity of the eyeball based on the changes in the position and angle of the reflected light.

[0007] Furthermore, the process by which the ophthalmic disease multivariate diagnostic data processing module acquires the patient's blood glucose data bar chart and the diagnostic pattern for diabetic patients from the ophthalmic disease image diagnostic system includes: B1. Establish a two-dimensional coordinate system with the horizontal axis representing time and the vertical axis representing blood glucose concentration. Plot the patient's blood glucose data in the coordinate system in the form of bars to obtain a bar chart of blood glucose data for diabetic patients. B2. Perform preprocessing operations such as denoising, contrast enhancement, and size standardization on fundus images. Establish a data model on the data platform through a cloud database, define data tables, fields, and data types, and use ETL tools to import basic information of diabetic patients, blood glucose data, and fundus image data into the same data platform. B3. Use a convolutional neural network to extract lesion feature data from fundus images. This convolutional neural network includes a convolutional layer, a pooling layer, and a fully connected layer. The convolutional layer extracts local features from fundus image data, the pooling layer samples the local features output by the convolutional layer and retains important feature data, and the fully connected layer flattens the features extracted by the convolutional layer and the pooling layer. B4. By using feature fusion technology, the basic information of diabetic patients, blood glucose data and extracted fundus lesion feature data are fused to form diabetic ophthalmology feature fusion data; B5. Based on transfer learning, the ophthalmic disease image diagnosis system sets the source domain as the diabetic ophthalmic feature fusion dataset and the target domain as the diabetic patient eye diagnosis result dataset, and transfers the knowledge learned from the source domain to the target domain through transfer learning. The model is evaluated using a test set, the accuracy of the model is calculated, and the model is iteratively optimized based on the evaluation results to obtain an ophthalmic disease image diagnosis system with built-in diabetic patient diagnosis mode.

[0008] Furthermore, the ophthalmic disease multivariate diagnostic data processing module performs feature point matching and parameter adjustment on images of reflected and scattered light from the eye, including: Denoising and image enhancement are performed on images of reflected and scattered light from the eye. Feature point matching is used to extract feature points from the image, match these feature points, estimate the transformation model parameters between images using the matched feature points, and adjust the image based on these parameters to ensure spatial consistency between images.

[0009] Furthermore, the process by which the multivariate diagnostic data processing module for ophthalmic diseases acquires the eye movement trajectory map and the eye movement angular velocity table includes: Establish a two-dimensional coordinate system with the horizontal axis representing time and the vertical axis representing eye position. Map the eye positions corresponding to the collected eye movement trajectories onto this two-dimensional coordinate system to obtain an eye movement trajectory map. Create an Excel spreadsheet with the unit set to degrees per second, starting from 0s, and record the eye movement angular velocity per second to obtain an eye movement angular velocity table.

[0010] Furthermore, the comprehensive ophthalmic disease data analysis module, through the processes of determining the ophthalmic diagnostic mode, contact lens wearing status, and eye movement trajectory using blood glucose data, eye images, and eye movement trajectories, includes: By analyzing the patient's blood glucose data, when the patient's fasting blood glucose is greater than or equal to 7 mmol / L and postprandial blood glucose is greater than or equal to 11.1 mmol / L, the collected blood glucose data is higher than the critical value for blood glucose concentration in diabetes, and it is determined that the patient has diabetes. The signal is then wirelessly transmitted to the ophthalmology diagnostic execution module. Using the processed images of reflected and scattered light from the eye, when the reflected and scattered light images are regular and flat, it indicates that the corneal surface is smooth and the patient is not wearing contact lenses; when the reflected and scattered light images are irregular, it indicates that the patient is wearing contact lenses. This signal is then wirelessly transmitted to the ophthalmic diagnostic execution module. By observing the distribution of the patient's eye movement trajectory within the examination range of the ophthalmic testing equipment using the patient's eye movement trajectory diagram, a smooth and continuous trajectory indicates normal eye movement. Obvious jumps, pauses, and irregular movements indicate abnormalities in eye movement. Simultaneously, using an eye movement angular velocity table, a normal diagnostic eye movement angular velocity is set at 10° / s. When the patient's eye movement angular velocity is below 10° / s, eye movement will not affect the diagnostic results; when the patient's eye movement angular velocity is above 10° / s, eye movement will affect the diagnostic results. These abnormal movement signals are then transmitted to the ophthalmic diagnostic execution module.

[0011] Furthermore, the ophthalmic diagnostic execution module receives abnormal signals, initiates system processing, and transmits them to the client for medical personnel to diagnose and intervene. This process includes: The ophthalmic diagnosis execution module receives a signal from the analysis module indicating that the patient is wearing contact lenses, and transmits the signal to the client. Medical personnel receive the signal and remind the patient to remove the contact lenses before proceeding with the diagnosis. The ophthalmology diagnostic execution module receives abnormal eye movement signals from the analysis module and transmits them to the client. Upon receiving the signal, medical personnel adjust the ambient light intensity for the diagnostic environment and remind and guide the patient to maintain stable eye movements during the diagnostic process.

[0012] The beneficial effects of this invention: Compared with traditional ophthalmic disease image diagnostic systems, the ophthalmic disease image diagnostic technology in the method described in this invention is closely integrated with modern information technology. By integrating optical imaging technology, blood glucose monitoring technology, eye-tracking technology, and big data analysis, it achieves real-time and comprehensive monitoring of the ophthalmic disease image diagnostic system. This system can accurately collect patient blood glucose data, fundus image data of diabetic patients at different disease stages, image data of reflected and scattered light from the patient's eye, and dynamic changes in the eye's motion trajectory and angular velocity. By using the acquired patient blood glucose data bar chart and the diabetic patient diagnostic mode of the ophthalmic disease image diagnostic system, it monitors the patient's blood glucose data and, when the blood glucose data reaches a certain level... The system automatically switches to a diabetic patient ophthalmic diagnostic mode based on standard diabetic criteria, resolving the interference of diabetic patients with ophthalmic diagnosis. By acquiring high-quality images of reflected and scattered light from the eye, it monitors whether the patient is wearing contact lenses, reminding medical personnel to remind the patient to remove them before diagnosis, thus resolving the interference of contact lenses on ophthalmic diagnostic results. Furthermore, by acquiring eye movement trajectory and angular velocity maps, it monitors the patient's eye movement amplitude, reminding doctors to adjust environmental conditions and guiding patients to control eye movement, thus resolving the interference of eye movement on ophthalmic diagnosis. This ensures that ophthalmic diagnosis has more accurate indicators under the same conditions as traditional ophthalmic disease image diagnostic systems. The development and application of this system significantly enhances the accuracy of ophthalmic disease image diagnostic systems. Attached Figure Description

[0013] Figure 1 This is a block diagram of an ophthalmic disease image diagnosis system based on a big data model. Detailed Implementation

[0014] 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.

[0015] like Figure 1 As shown, the present invention provides a technical solution: an ophthalmic disease image diagnosis system based on a big data model, including an ophthalmic disease multi-dimensional diagnostic data acquisition module, an ophthalmic disease multi-dimensional diagnostic data processing module, an ophthalmic disease comprehensive data analysis module, and an ophthalmic diagnosis execution module; The multi-dimensional diagnostic data acquisition module for ophthalmic diseases obtains basic information about diabetic patients through doctor consultation and data entry, acquires blood glucose data using a blood glucose meter, acquires fundus images at different stages of diabetes using a fundus camera, acquires images of reflected and scattered light from the patient's eyes using a slit lamp, and acquires the patient's eye movement trajectory and angular velocity during the diagnostic process using an eye tracker. The ophthalmic disease multi-dimensional diagnostic data processing module processes blood glucose data, obtains a bar chart of patients' blood glucose data, uses a multimodal deep learning model to obtain the diagnostic pattern of diabetic patients, performs feature point matching and parameter adjustment on the images of reflected and scattered light from the eye, maps the eyeball in the form of a curve on a two-dimensional coordinate system to obtain an eyeball motion trajectory map, and summarizes the eyeball motion angular velocity in a table to obtain an eyeball motion angular velocity table. The comprehensive data analysis module for ophthalmic diseases uses blood glucose data, eye images, and eye movement trajectories to determine ophthalmic diagnostic modes, contact lens wearing status, and eye movement status, respectively. The ophthalmology diagnostic execution module receives abnormal signals, mobilizes the system for processing, and transmits them to the client for medical personnel to diagnose and intervene.

[0016] The ophthalmic disease multi-dimensional diagnostic data acquisition module acquires basic information, blood glucose data, and fundus images of diabetic patients at different stages of diabetes. The process includes: A1. Consult with doctors about diabetic patients and enter their names and ages into the ophthalmic disease image diagnosis system to obtain basic information about diabetic patients; A2. Collect a blood sample from the patient. The blood glucose meter test strip contains glucose oxidase. When the blood sample comes into contact with the test strip, hydrogen peroxide is generated. The hydrogen peroxide undergoes an oxidation reaction on the electrode to generate an electric current. The blood glucose meter has a built-in sensor and microprocessor. The sensor captures the current, and the microprocessor first converts the analog current signal into a digital signal. Then, after calculation and calibration, the blood glucose data of the diabetic patient is obtained. A3. The fundus camera receives light from the patient's fundus and transmits it through the lens to the sensor of the fundus camera. Using digital image processing technology, the captured fundus image is converted into a digital signal. The image processing system integrated on the fundus camera is used to process the digital signal into a fundus image of the patient. By collecting fundus images from different patients multiple times, fundus images of diabetic patients at different stages of the disease can be obtained.

[0017] The process of acquiring images of reflected and scattered light from a patient's eye using the multivariate diagnostic data acquisition module for ophthalmic diseases includes: The slit lamp uses a high-brightness light source and a magnifying glass system to focus light onto the patient's eye structure. By appropriately adjusting the angle, brightness, and magnification of the slit lamp light source, images of reflected and scattered light from the patient's eye are obtained, which describe the characteristic data of the patient wearing contact lenses.

[0018] The multi-dimensional diagnostic data acquisition module for ophthalmic diseases acquires the patient's eye movement trajectory and angular velocity during the diagnostic process, including: Based on eye-tracking technology, the eye tracker emits infrared light to the patient's eyeball. The infrared light is reflected back to the eye tracker by the eyeball. The built-in sensor of the eye tracker collects the infrared light and calculates the movement trajectory and angular velocity of the eyeball based on the changes in the position and angle of the reflected light.

[0019] The ophthalmic disease multi-dimensional diagnostic data processing module, in its process of acquiring patient blood glucose data bar charts and the diagnostic model for diabetic patients from the ophthalmic disease image diagnostic system, includes: B1. Establish a two-dimensional coordinate system with the horizontal axis representing time and the vertical axis representing blood glucose concentration. Plot the patient's blood glucose data in the coordinate system in the form of bars to obtain a bar chart of blood glucose data for diabetic patients. B2. Perform preprocessing operations such as denoising, contrast enhancement, and size standardization on fundus images. Establish a data model on the data platform through a cloud database, define data tables, fields, and data types, and use ETL tools to import basic information of diabetic patients, blood glucose data, and fundus image data into the same data platform. B3. Use a convolutional neural network to extract lesion feature data from fundus images. This convolutional neural network includes a convolutional layer, a pooling layer, and a fully connected layer. The convolutional layer extracts local features from fundus image data, the pooling layer samples the local features output by the convolutional layer and retains important feature data, and the fully connected layer flattens the features extracted by the convolutional layer and the pooling layer. B4. By using feature fusion technology, the basic information of diabetic patients, blood glucose data and extracted fundus lesion feature data are fused to form diabetic ophthalmology feature fusion data; B5. Based on transfer learning, the ophthalmic disease image diagnosis system sets the source domain as the diabetic ophthalmic feature fusion dataset and the target domain as the diabetic patient eye diagnosis result dataset, and transfers the knowledge learned from the source domain to the target domain through transfer learning. The model is evaluated using a test set, the accuracy of the model is calculated, and the model is iteratively optimized based on the evaluation results to obtain an ophthalmic disease image diagnosis system with built-in diabetic patient diagnosis mode.

[0020] The multivariate diagnostic data processing module for ophthalmic diseases involves feature point matching and parameter adjustment of images of reflected and scattered light from the eye, including: Denoising and image enhancement are performed on images of reflected and scattered light from the eye. Feature point matching is used to extract feature points from the image, match these feature points, estimate the transformation model parameters between images using the matched feature points, and adjust the image based on these parameters to ensure spatial consistency between images.

[0021] The ophthalmic disease multi-dimensional diagnostic data processing module acquires eye movement trajectory maps and eye movement angular velocity tables through the following processes: Establish a two-dimensional coordinate system with the horizontal axis representing time and the vertical axis representing eye position. Map the eye positions corresponding to the collected eye movement trajectories onto this two-dimensional coordinate system to obtain an eye movement trajectory map. Create an Excel spreadsheet with the unit set to degrees per second, starting from 0s, and record the eye movement angular velocity per second to obtain an eye movement angular velocity table.

[0022] The comprehensive data analysis module for ophthalmic diseases uses blood glucose data, eye images, and eye movement trajectories to determine the ophthalmic diagnostic mode, contact lens wearing status, and eye movement patterns, respectively. By analyzing the patient's blood glucose data, when the patient's fasting blood glucose is greater than or equal to 7 mmol / L and postprandial blood glucose is greater than or equal to 11.1 mmol / L, the collected blood glucose data is higher than the critical value for blood glucose concentration in diabetes, and it is determined that the patient has diabetes. The signal is then wirelessly transmitted to the ophthalmology diagnostic execution module. Using the processed images of reflected and scattered light from the eye, when the reflected and scattered light images are regular and flat, it indicates that the corneal surface is smooth and the patient is not wearing contact lenses; when the reflected and scattered light images are irregular, it indicates that the patient is wearing contact lenses. This signal is then wirelessly transmitted to the ophthalmic diagnostic execution module. By observing the distribution of the patient's eye movement trajectory within the examination range of the ophthalmic testing equipment using the patient's eye movement trajectory diagram, a smooth and continuous trajectory indicates normal eye movement. Obvious jumps, pauses, and irregular movements indicate abnormalities in eye movement. Simultaneously, using an eye movement angular velocity table, a normal diagnostic eye movement angular velocity is set at 10° / s. When the patient's eye movement angular velocity is below 10° / s, eye movement will not affect the diagnostic results; when the patient's eye movement angular velocity is above 10° / s, eye movement will affect the diagnostic results. These abnormal movement signals are then transmitted to the ophthalmic diagnostic execution module.

[0023] The ophthalmology diagnostic execution module receives abnormal signals, mobilizes the system for processing, and transmits them to the client for medical personnel to diagnose and intervene. The process includes: The ophthalmic diagnosis execution module receives a signal from the analysis module indicating that the patient is wearing contact lenses, and transmits the signal to the client. Medical personnel receive the signal and remind the patient to remove the contact lenses before proceeding with the diagnosis. The ophthalmology diagnostic execution module receives abnormal eye movement signals from the analysis module and transmits them to the client. Upon receiving the signal, medical personnel adjust the ambient light intensity for the diagnostic environment and remind and guide the patient to maintain stable eye movements during the diagnostic process.

[0024] First, the system inputs basic patient information, collects blood glucose data using a blood glucose meter, acquires fundus images of patients at different disease stages using a fundus camera, collects reflected and scattered light images of the patient's eyes using a slit lamp, and collects eye movement trajectories and angular velocities using an eye tracker. Second, the multi-dimensional diagnostic data processing module for ophthalmic diseases processes the above data to generate a histogram of blood glucose data for diabetic patients, a diagnostic mode for diabetic patients in the ophthalmic disease image diagnosis system, high-quality images of reflected and scattered light from the eyes, eye movement trajectory diagrams, and eye movement angular velocity tables. Then, the comprehensive data analysis module for ophthalmic diseases performs in-depth analysis of the processed data and sends abnormal signals to the ophthalmic diagnosis execution module. Finally, the ophthalmic diagnosis execution module receives the signals from the comprehensive data analysis module for ophthalmic diseases, activates the ophthalmic disease image diagnosis system to switch modes and take countermeasures, and transmits the signals to the client for medical personnel to intervene, thereby improving the accuracy of the entire ophthalmic diagnosis process.

[0025] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. An ophthalmic disease image diagnosis system based on a big data model, comprising an ophthalmic disease multivariate diagnostic data acquisition module, an ophthalmic disease multivariate diagnostic data processing module, an ophthalmic disease comprehensive data analysis module, and an ophthalmic diagnosis execution module, characterized in that: The ophthalmic disease multi-dimensional diagnostic data acquisition module obtains basic information of diabetic patients through doctor consultation and data entry, acquires blood glucose data using a blood glucose meter, acquires fundus images of different stages of diabetes using a fundus camera, acquires images of reflected and scattered light from the patient's eyes using a slit lamp, and acquires the movement trajectory and angular velocity of the patient's eyeballs during the diagnostic process using an eye tracker. The ophthalmic disease multi-dimensional diagnostic data processing module processes blood glucose data, obtains a blood glucose data bar chart for patients, uses a multimodal deep learning model to obtain the diagnostic pattern of diabetic patients, performs feature point matching and parameter adjustment on the images of reflected and scattered light from the eye, maps the eyeball in the form of a curve on a two-dimensional coordinate system to obtain an eyeball motion trajectory diagram, and summarizes the eyeball motion angular velocity in a table to obtain an eyeball motion angular velocity table. The comprehensive data analysis module for ophthalmic diseases uses blood glucose data, eye images, and eye movement trajectories to determine the ophthalmic diagnostic mode, contact lens wearing status, and eye movement status, respectively. The ophthalmic diagnostic execution module receives abnormal signals, mobilizes the system for processing, and transmits them to the client for medical personnel to diagnose and intervene.

2. The ophthalmic disease image diagnosis system based on a big data model according to claim 1, characterized in that: The ophthalmic disease multi-dimensional diagnostic data acquisition module acquires basic information, blood glucose data, and fundus images of diabetic patients at different stages of diabetes, including the following processes: A1. Consult with doctors about diabetic patients and enter their names and ages into the ophthalmic disease image diagnosis system to obtain basic information about diabetic patients; A2. Collect a blood sample from the patient. The blood glucose meter test strip contains glucose oxidase. When the blood sample comes into contact with the test strip, hydrogen peroxide is generated. The hydrogen peroxide undergoes an oxidation reaction on the electrode to generate an electric current. The blood glucose meter has a built-in sensor and microprocessor. The sensor captures the current, and the microprocessor first converts the analog current signal into a digital signal. Then, after calculation and calibration, the blood glucose data of the diabetic patient is obtained. A3. The fundus camera receives light from the patient's fundus and transmits it through the lens to the sensor of the fundus camera. Using digital image processing technology, the captured fundus image is converted into a digital signal. The image processing system integrated on the fundus camera is used to process the digital signal into a fundus image of the patient. By collecting fundus images from different patients multiple times, fundus images of diabetic patients at different stages of the disease can be obtained.

3. The ophthalmic disease image diagnosis system based on a big data model according to claim 2, characterized in that: The process by which the multivariate diagnostic data acquisition module for ophthalmic diseases acquires images of reflected and scattered light from the patient's eye includes: The slit lamp uses a high-brightness light source and a magnifying glass system to focus light onto the patient's eye structure. By appropriately adjusting the angle, brightness, and magnification of the slit lamp light source, images of reflected and scattered light from the patient's eye are obtained, which describe the characteristic data of the patient wearing contact lenses.

4. The ophthalmic disease image diagnosis system based on a big data model according to claim 3, characterized in that: The ophthalmic disease multi-dimensional diagnostic data acquisition module acquires the patient's eye movement trajectory and angular velocity during the diagnostic process, including: Based on eye-tracking technology, the eye tracker emits infrared light to the patient's eyeball. The infrared light is reflected back to the eye tracker by the eyeball. The built-in sensor of the eye tracker collects the infrared light and calculates the movement trajectory and angular velocity of the eyeball based on the changes in the position and angle of the reflected light.

5. The ophthalmic disease image diagnosis system based on a big data model according to claim 4, characterized in that: The process by which the ophthalmic disease multivariate diagnostic data processing module acquires a patient's blood glucose data bar chart and the diagnostic mode for diabetic patients from the ophthalmic disease image diagnostic system includes: B1. Establish a two-dimensional coordinate system with the horizontal axis representing time and the vertical axis representing blood glucose concentration. Plot the patient's blood glucose data in the coordinate system in the form of bars to obtain a bar chart of blood glucose data for diabetic patients. B2. Perform preprocessing operations such as denoising, contrast enhancement, and size standardization on fundus images. Establish a data model on the data platform through a cloud database, define data tables, fields, and data types, and use ETL tools to import basic information of diabetic patients, blood glucose data, and fundus image data into the same data platform. B3. Use a convolutional neural network to extract lesion feature data from fundus images. This convolutional neural network includes a convolutional layer, a pooling layer, and a fully connected layer. The convolutional layer extracts local features from fundus image data, the pooling layer samples the local features output by the convolutional layer and retains important feature data, and the fully connected layer flattens the features extracted by the convolutional layer and the pooling layer. B4. By using feature fusion technology, the basic information of diabetic patients, blood glucose data and extracted fundus lesion feature data are fused to form diabetic ophthalmology feature fusion data; B5. Based on transfer learning, the ophthalmic disease image diagnosis system sets the source domain as the diabetic ophthalmic feature fusion dataset and the target domain as the diabetic patient eye diagnosis result dataset, and transfers the knowledge learned from the source domain to the target domain through transfer learning. The model is evaluated using a test set, the accuracy of the model is calculated, and the model is iteratively optimized based on the evaluation results to obtain an ophthalmic disease image diagnosis system with built-in diabetic patient diagnosis mode.

6. The ophthalmic disease image diagnosis system based on a big data model according to claim 5, characterized in that: The ophthalmic disease multivariate diagnostic data processing module performs feature point matching and parameter adjustment on images of reflected and scattered light from the eye, including: Denoising and image enhancement are performed on images of reflected and scattered light from the eye. Feature point matching is used to extract feature points from the image, match these feature points, estimate the transformation model parameters between images using the matched feature points, and adjust the image based on these parameters to ensure spatial consistency between images.

7. The ophthalmic disease image diagnosis system based on a big data model according to claim 6, characterized in that: The ophthalmic disease multi-dimensional diagnostic data processing module acquires the eye movement trajectory diagram and the eye movement angular velocity table, including the following processes: Establish a two-dimensional coordinate system with the horizontal axis representing time and the vertical axis representing eye position. Map the eye positions corresponding to the collected eye movement trajectories onto this two-dimensional coordinate system to obtain an eye movement trajectory map. Create an Excel spreadsheet with the unit set to degrees per second, starting from 0s, and record the eye movement angular velocity per second to obtain an eye movement angular velocity table.

8. The ophthalmic disease image diagnosis system based on a big data model according to claim 7, characterized in that: The comprehensive data analysis module for ophthalmic diseases analyzes blood glucose data, eye images, and eye movement trajectories to determine the ophthalmic diagnostic mode, contact lens wearing status, and eye movement status, including the following processes: By analyzing the patient's blood glucose data, when the patient's fasting blood glucose is greater than or equal to 7 mmol / L and postprandial blood glucose is greater than or equal to 11.1 mmol / L, the collected blood glucose data is higher than the critical value for blood glucose concentration in diabetes, and it is determined that the patient has diabetes. The signal is then wirelessly transmitted to the ophthalmology diagnostic execution module. Using the processed images of reflected and scattered light from the eye, when the reflected and scattered light images are regular and flat, it indicates that the corneal surface is smooth and the patient is not wearing contact lenses; when the reflected and scattered light images are irregular, it indicates that the patient is wearing contact lenses. This signal is then wirelessly transmitted to the ophthalmic diagnostic execution module. By observing the distribution of the patient's eye movement trajectory within the examination range of the ophthalmic testing equipment using the patient's eye movement trajectory diagram, a smooth and continuous trajectory indicates normal eye movement. Obvious jumps, pauses, and irregular movements indicate abnormalities in eye movement. Simultaneously, using an eye movement angular velocity table, a normal diagnostic eye movement angular velocity is set at 10° / s. When the patient's eye movement angular velocity is below 10° / s, eye movement will not affect the diagnostic results; when the patient's eye movement angular velocity is above 10° / s, eye movement will affect the diagnostic results. These abnormal movement signals are then transmitted to the ophthalmic diagnostic execution module.

9. The ophthalmic disease image diagnosis system based on a big data model according to claim 8, characterized in that: The ophthalmic diagnostic execution module receives abnormal signals, processes them within the system, and transmits them to the client for medical personnel to diagnose and intervene. The process includes: The ophthalmology diagnosis execution module receives a signal from the ophthalmology disease comprehensive data analysis module that the patient's blood glucose concentration exceeds the critical value for diabetes. It automatically switches the ophthalmology disease image diagnosis system to the diabetes patient diagnosis mode and transmits the signal to the client. Medical personnel then make a diagnosis based on the actual situation. The ophthalmic diagnosis execution module receives a signal from the analysis module indicating that the patient is wearing contact lenses, and transmits the signal to the client. Medical personnel receive the signal and remind the patient to remove the contact lenses before proceeding with the diagnosis. The ophthalmology diagnostic execution module receives abnormal eye movement signals from the analysis module and transmits them to the client. Upon receiving the signal, medical personnel adjust the ambient light intensity for the diagnostic environment and remind and guide the patient to maintain stable eye movements during the diagnostic process.

Images