Intelligent ophthalmology nursing monitoring system and method thereof

Abstract

The invention discloses an intelligent ophthalmology nursing monitoring system and method, and relates to the technical field of medical equipment, the system comprises the following components: a flexible sensing module, a data processing module, an artificial intelligence analysis module and a wireless communication module; key data of intraocular pressure, ocular surface humidity and blinking frequency are collected in real time through the flexible sensing module, efficient preprocessing is conducted through the data processing module, multi-source data are subjected to fusion analysis through the artificial intelligence analysis module by applying a deep learning algorithm, an eye health model can be established, and therefore the eye health condition is accurately judged, and the eye health effect is improved. According to the system and the method, eye diseases such as glaucoma can be early warned, the real-time monitoring and accurate analysis capability is beneficial to finding eye problems in time, the treatment success rate of the eye diseases is improved, and risks and losses caused by disease deterioration are reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent ophthalmic monitoring and management, and particularly relates to an intelligent ophthalmic nursing monitoring system and a method thereof. Background Art

[0002] With the acceleration of the modern life rhythm and the popularization of digital devices, eye health problems have become increasingly prominent. In particular, the incidence of eye diseases such as glaucoma and dry eye has been increasing year by year. These eye diseases not only seriously affect the quality of life of patients, but may also lead to serious consequences such as vision loss. Therefore, it is particularly important to monitor eye health in real time and manage it effectively.

[0003] Traditional eye health monitoring technologies have many deficiencies. First of all, traditional monitoring means are mostly single-point monitoring, such as monitoring intraocular pressure or ocular surface humidity alone, lacking a comprehensive assessment of the eye health status. Secondly, the data processing and analysis capabilities of traditional monitoring methods are limited, and it is difficult to perform in-depth fusion analysis on multi-source data, thus affecting the accurate judgment of the eye health status. Most importantly, traditional monitoring methods cannot achieve remote interaction and real-time data transmission. Patients cannot timely understand their own health status, and doctors cannot timely obtain the eye health data of patients, thus affecting the early diagnosis and treatment of diseases.

[0004] In summary, traditional eye health monitoring technologies have obvious deficiencies in real-time monitoring, comprehensive assessment, data processing, and remote interaction, and are difficult to meet the needs of modern eye health management. Therefore, it is particularly important to develop an intelligent ophthalmic nursing monitoring system and a method thereof. Summary of the Invention

[0005] The purpose of the present invention is to make up for the deficiencies of the existing technology, and provide an intelligent ophthalmic nursing monitoring system and a method thereof. It can make up for the deficiencies of traditional technologies by integrating flexible sensing technology, data processing technology, artificial intelligence analysis technology, and wireless communication technology, and provide a more comprehensive, accurate, and efficient solution for eye health management.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: On the one hand, an intelligent ophthalmic nursing monitoring system, which includes the following components: a flexible sensing module, a data processing module, an artificial intelligence analysis module, and a wireless communication module;

[0007] The flexible sensing module: real-time collects intraocular pressure, ocular surface humidity, and blink frequency data, uses a piezoresistive sensor to monitor intraocular pressure, a capacitive sensor to monitor ocular surface humidity, and monitors the blink frequency through a photoelectric sensor;

[0008] The data processing module: preliminarily processes the data collected by the flexible sensing module, including filtering and amplification, and then transmits the processed data to the artificial intelligence analysis module;

[0009] The artificial intelligence analysis module: uses deep learning algorithms to perform fusion analysis on multi-source data, establishes an eye health model, judges the eye health status, and gives early warnings for eye diseases such as glaucoma;

[0010] The wireless communication module: transmits the data processed by the data processing module and the analysis results of the artificial intelligence analysis module to the remote terminal through wireless communication technology to achieve remote interaction.

[0011] Further, the piezoresistive intraocular pressure sensor in the flexible sensing module uses a flexible piezoresistive material. When measuring the intraocular pressure, based on the piezoresistive effect of the material, the change in its resistance ΔR and the applied pressure P satisfy the relationship: ΔR = α·P β ·L / A, where α and β are characteristic parameters obtained by conducting a large number of experimental tests on the piezoresistive material in a simulated intraocular pressure environment and fitting using the least squares method. L is the length of the sensor's sensitive element, A is the cross-sectional area of the sensitive element, and the dimensions of the sensitive element are optimized according to the physiological structure characteristics of the eye measurement position to ensure accurate intraocular pressure measurement while maximizing wearing comfort. For the capacitive sensor used to measure the eye surface humidity, the functional relationship between its capacitance value C and the eye surface humidity H is established through experiments as C = γ·H + δ, where γ and δ are coefficients obtained by performing linear regression analysis on a large number of measurement data in different humidity environments. The photoelectric sensor for measuring the blink frequency uses an optical signal recognition algorithm to determine the blink frequency by calculating the number of times the light intensity changes per unit time, and its frequency resolution can reach ±0.1 times / minute, capable of accurately capturing each blink action.

[0012] Even further, the data processing module uses an adaptive filtering algorithm to preprocess the collected raw data. For the input one-dimensional data sequence {x 1 ,x 2 ,…,x n}, first calculate the absolute value of the difference between each data point and its adjacent data points to form a difference matrix D, and then assign an adaptive weight W i to each data point according to the difference matrix D. The calculation formula is Among them, k represents the number of adjacent data points participating in the weight calculation, and its optimal value is determined through experiments to balance the filtering effect and computational complexity. ∈ is a regulation factor, which is optimized through a large amount of data testing and is used to adjust the sensitivity of the weight to data differences. Then, median filtering is performed on the weighted data to effectively remove noise interference. In the signal amplification link, a programmable gain amplifier based on feedback regulation is used, and its gain G is automatically adjusted according to the dynamic range of the input signal. The adjustment formula is where max(|x|) and min(|x|) are the maximum and minimum values of the absolute value of the input signal respectively, is an empirical correction coefficient determined through experiments to ensure that the amplitude of the output signal is within a range suitable for subsequent processing.

[0013] Furthermore, the multi-source data fusion algorithm adopted by the artificial intelligence analysis module deeply fuses the intraocular pressure data P, the ocular surface humidity data H, and the blink frequency data F. First, the data of each data source is normalized. Let the normalized intraocular pressure data be P', the ocular surface humidity data be H', and the blink frequency data be F'. The fused data S is calculated through the formula where ω 1 、ω 2 、ω 3 are weight coefficients, which are determined through training on a large amount of clinical ocular health data and disease data and are iteratively optimized using the particle swarm optimization algorithm to maximize the accuracy of the fused data in judging the ocular health status. θ 1 、θ 2 、θ 3 are exponential parameters determined through the analysis of different data characteristics and combined with clinical experience, and are used to adjust the contribution degree of the data of each data source in the fusion process. Based on the fused data S, the disease early warning algorithm adopts an improved Bayesian probability model. By calculating P(D|S), where D represents the disease state and S is the fused data, the probability of the occurrence of ocular diseases is evaluated to achieve accurate early warning.

[0014] Furthermore, the wireless communication module adopts an intelligent switching hybrid communication mode. At close range, Bluetooth low energy technology is preferentially used to transmit data to reduce system power consumption. During the Bluetooth transmission process, the modulation and coding method is dynamically adjusted according to the signal strength indication. When RSSI > A 1 , the high-order modulation and coding method M 1 is adopted to improve the data transmission rate. When A 2 < RSSI ≤ A 1 , it is switched to the medium-order modulation and coding method M 2 . When RSSI ≤ A 2 , the low-order modulation and coding method M 3, A 1 and A 2 are RSSI thresholds determined through a large number of communication experiments in different environments, M 1 , M 2 , M 3 are different modulation and coding schemes, and their specific parameters are determined according to communication standards and experimental optimization. In a long-distance communication scenario, it automatically switches to the 4G / 5G communication network and uses network slicing technology to allocate slices for medical data transmission. The resource allocation parameters of the slices are adjusted in real time through a dynamic optimization algorithm based on traffic prediction. This algorithm uses an exponential smoothing prediction model to predict future data traffic based on historical data and current data traffic change trends, and then dynamically adjusts the slice resources to ensure the stability and real-time nature of data transmission.

[0015] Furthermore, the system is also equipped with a data storage module. This module adopts a distributed storage architecture and stores the collected raw data, processed data, and artificial intelligence analysis results on different storage nodes respectively. Data distribution among the storage nodes is carried out through the consistent hashing algorithm to ensure uniform data distribution, improve storage efficiency and reliability. For the raw data, an incremental storage strategy is adopted, that is, only the difference between the newly collected data and the previously stored data is stored. The storage formula is ΔD = D n -D n-1 , where D n is the data collected this time, and D n-1 is the data stored last time, effectively saving storage space. The processed data and analysis results are stored according to the time series, facilitating subsequent query, statistics, and in-depth analysis. The data storage module also has a data backup function, regularly backing up key data to cloud storage, and the backup period can be flexibly adjusted in the system settings according to user needs to ensure data security and recoverability.

[0016] Furthermore, the remote terminals are divided into doctor terminals and patient terminals. The software of the doctor terminals integrates data analysis and diagnostic assistance functions. It can receive the data and analysis results transmitted by the system and display the change trends of eye health indicators through an intuitive visual interface. In terms of diagnostic assistance, an intelligent diagnostic algorithm combining case-based reasoning and rule-based reasoning is adopted. First, similar cases are matched in the case library according to the input data. A large amount of clinical case data is stored in the case library. The most similar case is determined by calculating the case similarity. When the similarity exceeds the set threshold, the diagnostic results and treatment plans of the similar cases are referred to. If the similarity does not reach the threshold, reasoning is carried out according to the preset diagnostic rules. The diagnostic rules are summarized from the experience of ophthalmology experts and the results of clinical research. In this way, comprehensive and accurate diagnostic suggestions and personalized care plan references are provided for doctors. The software of the patient terminals focuses on health information display and reminder functions. Patients can view their own eye health indicator data and analysis results in real time. When the indicators exceed the normal range, the software automatically sends reminder messages in multiple ways such as sound, vibration, and pop-up windows. The reminder threshold can be personalized by doctors in the system according to the specific conditions of patients, which is convenient for patients to timely understand their own health status.

[0017] Furthermore, the system has an automatic calibration function. The calibration process is divided into regular calibration and calibration triggered by anomalies. Regular calibration is automatically started according to the preset time period. An analog intraocular pressure, ocular surface humidity, and blink frequency signal are generated by the built-in high-precision standard signal source and input into the flexible sensing module. Then, the measurement results of the system are compared with the standard signal values, and the least squares method is used to fit the error curve. The key parameters of the sensor are adjusted according to the error curve. Calibration triggered by anomalies is automatically triggered when the system detects abnormal fluctuations in the data. The basis for judging abnormal fluctuations is based on statistical principles. When the data deviates from the mean by more than k times the standard deviation, k is determined through statistical analysis of a large amount of experimental data, and the abnormal calibration process is started. The automatic calibration function effectively ensures the long-term stable operation of the system and the accuracy and reliability of the measurement data.

[0018] On the other hand, an embodiment of the present invention provides an intelligent ophthalmic care monitoring method, which specifically includes the following steps:

[0019] S1. Flexible sensing step: Real-time collect data on intraocular pressure, ocular surface humidity, and blink frequency. A piezoresistive sensor is used to monitor intraocular pressure, a capacitive sensor is used to monitor ocular surface humidity, and a photoelectric sensor is used to monitor blink frequency;

[0020] S2. Data processing step: Preliminarily process the data collected in the flexible sensing step, including filtering and amplification, and then transmit the processed data to the artificial intelligence analysis module;

[0021] S3. Artificial intelligence analysis step: Apply deep learning algorithms to perform fusion analysis on multi-source data, establish an eye health model, judge the eye health status, and conduct early warnings for eye diseases such as glaucoma;

[0022] S4. Wireless communication step: Transmit the data processed by the data processing module and the analysis results of the artificial intelligence analysis module to the remote terminal through wireless communication technology to achieve remote interaction. Compared with the prior art, the present invention has the following beneficial effects:

[0023] (1) The present invention can collect key data such as intraocular pressure, ocular surface humidity, and blink frequency in real time through the flexible sensing module, and after efficient preprocessing by the data processing module, the artificial intelligence analysis module uses deep learning algorithms to perform fusion analysis on multi-source data, and an eye health model can be established, so as to accurately judge the eye health status and conduct early warnings for eye diseases such as glaucoma. This ability of real-time monitoring and precise analysis helps to detect eye problems in a timely manner, improve the treatment success rate of eye diseases, and reduce the risks and losses brought by disease deterioration.

[0024] (2) Through the wireless communication module of the present invention, the system can transmit the processed data and analysis results to the remote terminal in real time. Whether it is a doctor or a patient, they can view the eye health index data and analysis results anytime and anywhere. The doctor-side software integrates data analysis and diagnostic assistance functions, and can provide comprehensive and accurate diagnostic suggestions and personalized care plan references. The patient-side software focuses on health information display and reminder functions, which is convenient for patients to understand their own health status in a timely manner. This remote interaction and intelligent diagnosis mode not only saves the time and cost of patients' medical treatment, but also improves the coverage and efficiency of medical services. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0026] Figure 1 It is a flow operation diagram of an intelligent ophthalmic care monitoring system of the present invention;

[0027] Figure 2 It is a flow operation diagram of an intelligent ophthalmic care monitoring method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation manners, structures, features and their effects according to the present invention as follows.

[0029] Example 1 :

[0030] This embodiment describes a 60-year-old elderly person with a family history of glaucoma, who is classified as a high-risk population for glaucoma. In order to monitor the eye health status in real time and prevent the onset of glaucoma, he uses an intelligent ophthalmic care monitoring system for daily monitoring.

[0031] After getting up and before going to bed every day, the elderly person wears a device integrated with a flexible sensing module. The piezoresistive intraocular pressure sensor is attached to the eye measurement position to measure the intraocular pressure based on the piezoresistive effect. The change in its resistance ΔR and the applied pressure P satisfy The capacitive sensor monitors the eye surface humidity in real time. The relationship between its capacitance value C and the eye surface humidity H is C = γ·H + δ. The photoelectric sensor calculates the blinking frequency by identifying the number of times of light intensity change per unit time through an optical signal recognition algorithm.

[0032] The collected raw data is transmitted to the data processing module. This module uses an adaptive filtering algorithm. For the input one-dimensional data sequence x 1 , x 2 , …, x n , first calculate the difference matrix D, and then according to Assign adaptive weights to each data point. The weighted data is subjected to median filtering to remove noise. In the signal amplification link, the programmable gain amplifier based on feedback regulation automatically adjusts the gain according to Automatic gain adjustment.

[0033] The processed data is transmitted to the artificial intelligence analysis module. This module normalizes the intraocular pressure data P, the eye surface humidity data H, and the blinking frequency data F. The fused data Based on the fused data S, an improved Bayesian probability model is used to calculate P(D|S) to evaluate the glaucoma onset probability. If the onset probability exceeds the warning threshold, the system issues a warning.

[0034] At close range, the data is transmitted through Bluetooth Low Energy technology. If the signal strength indicator RSSI > A 1 , the high-order modulation coding method M 1 is adopted. If A 2 < RSSI ≤ A 1 , switch to the medium-order modulation coding method M 2 . If RSSI ≤ A 2 , the low-order modulation coding method M is adopted3 , when at a long distance, it automatically switches to the 4G / 5G network and uses network slicing technology to transmit data.

[0035] The raw data collected adopts an incremental storage strategy, ΔD = D n -D n-1 , the processed data and analysis results are stored in time series, and the key data is regularly backed up to the cloud.

[0036] The doctor's end receives the data and analysis results, and views the change trend of the elderly's eye health indicators through the visualization interface. If diagnosis is required, an algorithm combining case-based reasoning and rule-based reasoning is adopted. The elderly at the patient end can view the eye health indicators in real time and receive reminders when the indicators are abnormal. The reminder threshold is set by the doctor according to their condition.

[0037] The system is automatically calibrated regularly. An analog signal is generated by the built-in high-precision standard signal source and input into the flexible sensing module. The measurement result is compared with the standard value, and the least squares method is used to fit the error curve to adjust the sensor parameters. If the data shows abnormal fluctuations (deviating from the mean by more than k times the standard deviation), automatic abnormal calibration is triggered.

[0038] Example 2:

[0039] This example describes that in the ward of an ophthalmic hospital, there are multiple patients with eye diseases living at the same time. Doctors need to understand the patients' eye conditions in real time in order to adjust the treatment plan in a timely manner. The intelligent ophthalmic care monitoring system is applied to the ward to achieve centralized monitoring of patients.

[0040] After the patient is admitted to the hospital, the nurse will wear a monitoring device integrated with a flexible sensing module for each patient. These devices are adjusted individually according to the patient's age and eye condition to ensure the fitting degree of the sensor to the eye and the measurement accuracy. The piezoresistive intraocular pressure sensor, capacitive eye surface humidity sensor, and optoelectronic blink frequency sensor start to collect data in real time. For example, patient Xiao Wang is admitted to the hospital due to retinal lesions. After wearing the device, the piezoresistive intraocular pressure sensor monitors the intraocular pressure in real time. At a certain moment, the resistance change amount is measured, and through the intraocular pressure data is calculated. The capacitive sensor measures the eye surface humidity, and the eye surface humidity value is obtained according to C = γ·H + δ. The optoelectronic sensor accurately records Xiao Wang's blink frequency. When Xiao Wang's blink frequency abnormally increases due to eye discomfort, the sensor can capture the change in time.

[0041] The raw data collected from each patient is transmitted to the data processing module in the ward by wired or wireless means. This module uses an adaptive filtering algorithm to process a large amount of data. For the input data sequence of each patient, the difference matrix D is calculated respectively, and then according to Assign adaptive weights to each data point, and then perform median filtering to remove noise. In the signal amplification stage, a programmable gain amplifier based on feedback regulation automatically adjusts the gain according to to ensure that the processed data is clear and accurate, meeting the requirements of subsequent analysis.

[0042] The processed data is transmitted to the artificial intelligence analysis module, which processes the data of each patient separately. After normalizing the intraocular pressure data P, ocular surface humidity data H, and blink frequency data F, the fusion data is calculated according to Based on the fusion data S, an improved Bayesian probability model is used to calculate P(D|S) to evaluate the development and potential risks of eye diseases for each patient. For example, patient Zhang has glaucoma. Through continuous monitoring and analysis, the system predicts the disease development trend based on the changes in his data and issues an early warning in a timely manner if the predicted disease shows signs of deterioration.

[0043] In the ward, the data is first transmitted to the local server in the ward through Bluetooth Low Energy technology. If the Bluetooth signal strength is poor, such as the received signal strength indication (RSSI) being lower than the set threshold, it will automatically switch to the 4G / 5G network for transmission to the hospital's remote server. When the network environment in the ward is complex and the Bluetooth signal is interfered, the system quickly switches to the 4G / 5G network to ensure stable data transmission to the hospital's core data processing center.

[0044] The data storage module adopts a distributed storage architecture and is stored on different storage nodes of the hospital's internal server. The original data adopts an incremental storage strategy to save storage space. The processed data and analysis results are stored in time series for doctors to query the patient's historical data at any time. For example, if a doctor wants to view the intraocular pressure changes of patient Li in the past week, he can directly filter and query by time in the system. The data storage module also regularly backs up key data to the hospital's private cloud, with a backup period of once a day, to ensure data security and prevent data loss.

[0045] Doctors can centrally view the eye health data and analysis results of all patients in the ward through the doctor-side software in the doctor's office. The visual interface of the software displays the change trends of various indicators of each patient in the form of charts, curves, etc., facilitating doctors' comparative analysis. In terms of diagnostic assistance, doctors can use an intelligent diagnostic algorithm that combines case-based reasoning and rule-based reasoning to quickly provide diagnostic suggestions for patients. For example, when a doctor views the data of patient Zhao, the system matches similar cases in the case library. If the similarity is high, it refers to the diagnostic results of similar cases; if the similarity is low, it reasons based on the diagnostic rules.

[0046] The patient-side software is installed on the patient's mobile device (such as a mobile phone or tablet computer). The patient can view their own health data and receive reminders at any time. When a certain index of the patient exceeds the normal range, the system automatically sends a reminder message to remind the patient to inform the medical staff in a timely manner. The reminder threshold is set personalized in the system by the doctor according to the patient's specific condition.

[0047] The system is regularly calibrated at 3 am every day. An analog signal is generated by the built-in high-precision standard signal source and input into the flexible sensing module. The measurement result is compared with the standard value, and the error curve is fitted using the least squares method to adjust the sensor parameters. If the system detects abnormal fluctuations in the data of a certain patient (according to statistical principles, when the data deviates from the mean by more than k times the standard deviation, k is determined through statistical analysis of a large amount of experimental data), an abnormal calibration of the monitoring device for this patient is automatically triggered to ensure the reliability of the monitoring data.

[0048] The above are only the preferred embodiments of the present invention, and do not impose any form of limitation on the present invention. Although the present invention has been disclosed above with the preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to the equivalent embodiments by using the above-disclosed technical content within the scope of the technical solution of the present invention. However, as long as it does not depart from the content of the technical solution of the present invention, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. An intelligent ophthalmic care monitoring system, characterized in that: The system includes the following components: a flexible sensing module, a data processing module, an artificial intelligence analysis module, and a wireless communication module: The flexible sensing module: Real-time collects intraocular pressure, ocular surface humidity, and blink frequency data. A piezoresistive sensor is used to monitor intraocular pressure, a capacitive sensor is used to monitor ocular surface humidity, and a photoelectric sensor is used to monitor blink frequency; The data processing module: Preliminarily processes the data collected by the flexible sensing module, including filtering and amplification, and then transmits the processed data to the artificial intelligence analysis module; The artificial intelligence analysis module: Uses deep learning algorithms to perform fusion analysis on multi-source data, establishes an eye health model, judges the eye health status, and conducts early warning for eye diseases such as glaucoma; The wireless communication module: Transmits the data processed by the data processing module and the analysis results of the artificial intelligence analysis module to a remote terminal through wireless communication technology to achieve remote interaction.

2. The intelligent ophthalmic care monitoring system according to claim 1, characterized in that: The piezoresistive intraocular pressure sensor in the flexible sensing module adopts flexible piezoresistive material. When measuring intraocular pressure, based on the piezoresistive effect of the material, the resistance change ΔR and the pressure P satisfy the relationship: ΔR = α·P β ·L / A, where α and β are characteristic parameters obtained by performing a large number of experimental tests on the piezoresistive material in a simulated intraocular pressure environment and fitting using the least squares method, L is the length of the sensor sensitive element, A is the cross-sectional area of ​​the sensitive element, and the size of the sensitive element is optimized according to the physiological structure characteristics of the eye measurement position. The capacitive sensor used to measure the humidity of the ocular surface has a functional relationship between its capacitance value C and the humidity H of the ocular surface established through experiments: C=γ·H+δ, γ and δ are coefficients obtained by linear regression analysis of a large number of measurement data under different humidity environments, and the photoelectric sensor for measuring the blinking frequency uses an optical signal recognition algorithm to determine the blinking frequency by calculating the number of light intensity changes per unit time.

3. The intelligent ophthalmic care monitoring system according to claim 1, characterized in that: The data processing module uses an adaptive filtering algorithm to pre-process the collected raw data. n }, first calculate the absolute value of the difference between each data point and its adjacent data points to form a difference matrix D, and then assign an adaptive weight W to each data point according to the difference matrix D i , the calculation formula is Where k represents the number of adjacent data points involved in the weight calculation, ∈ is an adjustment factor, and then the weighted data is median filtered to effectively remove noise interference. In the signal amplification link, a programmable gain amplifier based on feedback adjustment is used, and its gain G is automatically adjusted according to the dynamic range of the input signal. The adjustment formula is: Where max(|x|) and min(|x|) are the maximum and minimum absolute values ​​of the input signal, respectively, and φ is an empirical correction coefficient determined experimentally.

4. The intelligent ophthalmic care monitoring system according to claim 1, characterized in that: The multi-source data fusion algorithm adopted by the artificial intelligence analysis module deeply fuses the intraocular pressure data P, the ocular surface humidity data H, and the blink frequency data F. First, the data of each data source is normalized. The normalized intraocular pressure data is P', the ocular surface humidity data is H', and the blink frequency data is F'. The fused data S is calculated by the formula It is calculated that ω1, ω2, and ω3 are weight coefficients, θ1, θ2, and θ3 are index parameters determined by analyzing different data characteristics and combining clinical experience, which are used to adjust the contribution of each data source in the fusion process. Based on the fused data S, the disease warning algorithm adopts an improved Bayesian probability model to calculate P(D|S), where D represents the disease status and S is the fused data, to evaluate the probability of eye diseases and achieve accurate early warning.

5. The intelligent ophthalmic care monitoring system according to claim 1, characterized in that: The wireless communication module adopts an intelligent switching hybrid communication mode. At short distances, it preferentially uses Bluetooth Low Energy technology to transmit data to reduce system power consumption. During Bluetooth transmission, the modulation and coding method is dynamically adjusted according to the signal strength indication. When RSSI > A1, the high-order modulation and coding method M1 is used. When A2 < RSSI ≤ A1, it switches to the medium-order modulation and coding method M2. When RSSI ≤ A2, the low-order modulation and coding method M3 is used. A1 and A2 are RSSI thresholds determined through a large number of communication experiments in different environments. M1, M2, and M3 are different modulation and coding methods. In long-distance communication scenarios, it automatically switches to the 4G / 5G communication network and uses network slicing technology to allocate slices for medical data transmission. The resource allocation parameters of the slices are adjusted in real time through a dynamic optimization algorithm based on traffic prediction. This algorithm uses an exponential smoothing prediction model to predict future data traffic based on historical data and the current data traffic change trend, and then dynamically adjusts the slice resources.

6. The intelligent ophthalmic care monitoring system according to claim 1, characterized in that: The system is also equipped with a data storage module, which adopts a distributed storage architecture to store the collected raw data, processed data and artificial intelligence analysis results on different storage nodes. The data is distributed between the storage nodes through a consistent hashing algorithm. For the raw data, an incremental storage strategy is adopted, that is, only the difference between the newly collected data and the last stored data is stored. The storage formula is ΔD = D n -D n-1 , where D n For the data collected this time, D n-1 The data stored last time can effectively save storage space. The processed data and analysis results are stored in time series to facilitate subsequent query, statistics and in-depth analysis. The data storage module also has a data backup function, which regularly backs up key data to cloud storage. The backup cycle can be flexibly adjusted in the system settings according to user needs.

7. The intelligent ophthalmic care monitoring system according to claim 1, characterized in that: The remote terminal is divided into a doctor's end and a patient's end. The doctor's end software integrates data analysis and diagnostic assistance functions, can receive the data and analysis results transmitted by the system, and displays the change trend of eye health indicators through an intuitive visual interface. In terms of diagnostic assistance, an intelligent diagnostic algorithm combining case-based reasoning and rule-based reasoning is adopted. First, similar cases are matched in the case library according to the input data. The case library stores a large amount of clinical case data, and the most similar case is determined by calculating the case similarity. When the similarity exceeds the set threshold, the diagnostic results and treatment plans of the similar cases are referred to. If the similarity does not reach the threshold, reasoning is carried out according to the pre-set diagnostic rules. The patient's end software focuses on health information display and reminder functions. Patients can view their own eye health indicator data and analysis results in real time. When the indicators exceed the normal range, the software automatically sends reminder messages in multiple ways such as sound, vibration, and pop-up windows. The reminder threshold can be personalized by the doctor according to the specific condition of the patient in the system.

8. The intelligent ophthalmic care monitoring system according to claim 1, characterized in that: The system has an automatic calibration function. The calibration process is divided into periodic calibration and abnormal trigger calibration. The periodic calibration is automatically started according to a preset time period. The simulated intraocular pressure, ocular surface humidity and blinking frequency signals are generated by the built-in high-precision standard signal source and input into the flexible sensing module. The system measurement results are then compared with the standard signal values. The error curve is fitted using the least squares method, and the key parameters of the sensor are adjusted according to the error curve. The abnormal trigger calibration is automatically triggered when the system detects abnormal fluctuations in the data. The judgment of abnormal fluctuations is based on statistical principles. When the data deviates from the mean by more than k times the standard deviation, k is determined through statistical analysis of a large amount of experimental data, and the abnormal calibration process is started.

9. An intelligent ophthalmic care monitoring method, based on an intelligent ophthalmic care monitoring system according to any one of claims 1 to 8, characterized in that: The specific steps of this method are: S1, flexible sensing step: real-time collection of intraocular pressure, ocular surface humidity, and blinking frequency data, using a piezoresistive sensor to monitor intraocular pressure, a capacitive sensor to monitor ocular surface humidity, and a photoelectric sensor to monitor blinking frequency; S2, data processing step: preliminary processing of the data collected by the flexible sensing step, including filtering and amplification, and then transmitting the processed data to the artificial intelligence analysis step; S3, AI analysis steps: Use deep learning algorithms to integrate and analyze multi-source data, establish an eye health model, determine eye health status, and provide early warning for eye diseases such as glaucoma; S4, wireless communication step: the data processed by the data processing step and the analysis results of the artificial intelligence analysis step are transmitted to the remote terminal through wireless communication technology to realize remote interaction.

Images