Myopia prevention and control system and method based on multi-source data

Abstract

The application discloses a myopia prevention and control system and method based on multi-source data, and relates to the technical field of myopia prevention and control.The system comprises the following steps: collecting multi-source data through intelligent wearable devices, cameras and environmental sensors; preprocessing the collected multi-source data to form a structured time series data set; decomposing the structured time series data set into multiple time scales and extracting time series characteristic values; defining a reinforcement learning environment according to a prediction result, using a deep reinforcement learning algorithm to generate a dynamic prevention and control strategy scheme; and generating a myopia prevention and control scheme by long-term tracking of changes in the eye habits of a user according to the generated dynamic prevention and control strategy scheme.The application decomposes the structured time series data set into multiple time scales by using wavelet transform and extracts time series characteristic values, realizes multi-level description of the eye habits of a user and changes in the environment, and thus improves the accuracy and reliability of myopia risk prediction.

Description

Technical Field

[0001] This invention relates to the field of myopia prevention and control technology, and in particular to a myopia prevention and control system and method based on multi-source data. Background Technology

[0002] In recent years, with the rapid development of information technology, especially the advancements in hardware technologies such as smart wearable devices, cameras, and environmental sensors, data collection and analysis have been widely applied in the healthcare field. Myopia prevention, as an important component of adolescent health, has received increasing attention. Traditional myopia prevention methods mainly rely on regular eye exams and professional advice from doctors. While effective, this method has certain limitations. First, the intervals between regular eye exams are relatively long, making it impossible to monitor users' eye habits and environmental changes in real time. Second, doctors' professional advice is usually based on static data, making it difficult to dynamically adjust to individual differences.

[0003] While existing myopia prevention methods have made some progress, several shortcomings remain. First, current technologies still have limitations in data collection. Traditional myopia prevention methods typically rely on a single data source, such as ophthalmological examination results or user-reported eye habits, lacking comprehensive analysis of multiple data sources. This leads to biased and incomplete data, making it difficult to fully reflect the user's actual eye usage. For example, relying solely on ophthalmological examinations cannot capture subtle changes in daily eye usage duration, frequency, and rest intervals, factors crucial for assessing myopia risk. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a myopia prevention and control method based on multi-source data, which solves the problems of single data source and inability to effectively extract features at multiple time scales in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a myopia prevention and control method based on multi-source data, comprising: collecting multi-source data through smart wearable devices, cameras, and environmental sensors; preprocessing the collected multi-source data to form a structured time-series dataset; decomposing the structured time-series dataset into multiple time scales and extracting time-series feature values; based on the time-series feature values, weightedly fusing the feature values ​​of different time scales to predict the user's myopia risk and generate a prediction result; defining a reinforcement learning environment based on the prediction result, using a deep reinforcement learning algorithm to generate a dynamic prevention and control strategy; and based on the generated dynamic prevention and control strategy, tracking changes in the user's eye habits over a long period to generate a myopia prevention and control plan.

[0008] As a preferred embodiment of the myopia prevention and control method based on multi-source data described in this invention, the multi-source data includes eye-use behavior data, environmental data, image data, and physiological data.

[0009] As a preferred embodiment of the myopia prevention and control method based on multi-source data described in this invention, the preprocessing of the collected multi-source data to form a structured time-series dataset includes the following specific steps.

[0010] Based on the collected multi-source data, Python is used to read the data and remove missing and duplicate values, and the Scikit-learn library is used to standardize the numerical data.

[0011] The standardized multi-source data are organized into a structured format suitable for time series analysis and then fused to form a structured time series dataset.

[0012] As a preferred embodiment of the myopia prevention and control method based on multi-source data described in this invention, the specific steps of decomposing the structured time-series dataset into multiple time scales and extracting time-series feature values ​​are as follows.

[0013] Wavelet transform is used to perform multi-scale decomposition on structured time series datasets, expressed as follows:

[0014]

[0015] Where W(a,b) represents the decomposition of the structured time series dataset under time scale parameter a and translation parameter b, a represents the scale parameter, b represents the translation parameter, D(t) represents the comprehensive feature value of the preprocessed multi-source data at time t, ψ(t) represents the wavelet basis function at time t, dt represents the integration of the variable at time t, and t represents the time point of the time series data;

[0016] By decomposing structured time series datasets, time series features at different time scales are obtained;

[0017] Extract the time series feature values according to the time series characteristics at different time scales, and the expression is:

[0018]

[0019] where, F k (a) represents the time series feature value of the kth type of feature at the time scale a, δ i represents the weight coefficient of the ith time series feature, n represents the number of time series features, σ i represents the non-linear activation function of the ith time series feature, φ i (t) represents the kernel function of the ith time series feature at time t, k represents the index of feature classification, and i represents the index variable of the time series feature.

[0020] As a preferred solution of the myopia prevention and control method based on multi-source data described in the present invention, wherein: based on the time series feature values, the feature values of different time scales are weighted and fused to predict the myopia risk of the user and generate a prediction result. The specific steps are as follows.

[0021] Based on the time series feature values, the feature values of different time scales are weighted and fused to generate a comprehensive myopia risk prediction value, and the expression is:

[0022]

[0023] where, P(t) represents the comprehensive myopia risk prediction value at time t, m represents the number of time scales, w ν represents the weight coefficient of the νth time scale, and ν represents the index of the time scale;

[0024] Set the low risk threshold M and the high risk threshold N according to the historical risk division data;

[0025] Based on the myopia risk prediction value, divide the risk level through the low risk threshold M and the high risk threshold N, predict the myopia risk of the user, and generate a prediction result;

[0026] If P(t) < M, it is determined as low myopia risk;

[0027] If M ≤ P(t) < N, it is determined as medium myopia risk;

[0028] If P(t) ≥ N, it is determined as high myopia risk.

[0029] As a preferred solution of the myopia prevention and control method based on multi-source data described in the present invention, wherein: according to the prediction result, define the reinforcement learning environment, use the deep reinforcement learning algorithm, and generate a dynamic prevention and control strategy plan. The specific steps are as follows.

[0030] Based on the prediction results, the key features affecting myopia risk are identified, including the user's current myopia risk prediction value, eye use behavior characteristics, environmental characteristics, and user physiological characteristics. These are combined into a state vector to define the state space.

[0031] The actions to be taken to improve users’ eye habits and environmental conditions include adjusting the interval of reminders for eye use time, optimizing indoor light intensity and color temperature, increasing suggestions for outdoor activities, and adjusting the screen’s blue light ratio. These actions are formalized into a set, defining the action space.

[0032] Using the change in users' myopia risk as the primary optimization objective, the reward value is defined as follows:

[0033]

[0034] Where R represents the reward value, ΔP(t) represents the change in the user's comprehensive myopia risk prediction value within the time interval Δt, ξ(A) represents the execution cost function of action set A, η represents the user's acceptance feedback function of the strategy, λ1 represents the risk change weight, λ2 represents the execution cost weight, λ3 represents the weight of user feedback on the reward value, and A represents the action set.

[0035] Based on the defined state space, action space, and reward value, a deep reinforcement learning algorithm is used to collect experience through environmental interaction, and the optimal action is obtained using an experience replay mechanism to generate a dynamic prevention and control strategy.

[0036] As a preferred embodiment of the myopia prevention and control method based on multi-source data described in this invention, the specific steps for generating a myopia prevention and control plan by long-term tracking of changes in users' eye habits according to the generated dynamic prevention and control strategy are as follows:

[0037] Based on the generated dynamic prevention and control strategy, the action value of each eye-use behavior feature is calculated, expressed as:

[0038]

[0039] Where V(U,t) represents the action value of visual behavior feature U at time t, U represents visual behavior feature, α represents the total number of actions in the action space, and β represents the action value of visual behavior feature U at time t. j This represents the weight coefficient of the action in the j-th spatial action, where j represents the index variable of the action in the spatial action;

[0040] Based on the action value of eye-use behavior characteristics, a myopia prevention and control plan is generated through weighted fusion, expressed as:

[0041]

[0042] Where O(t) represents the myopia prevention and control plan at time t. ω represents the number of visual behavior characteristics. f This represents the weight coefficient of the f-th eye-use behavior feature, where f represents the index variable of the eye-use behavior feature.

[0043] Secondly, this invention provides a myopia prevention and control system based on multi-source data, comprising: a multi-source data acquisition module, a time-series dataset formation module, a time-series feature value extraction module, a prediction result generation module, a dynamic prevention and control strategy generation module, and a myopia prevention and control scheme generation module; the multi-source data acquisition module is used to collect multi-source data through smart wearable devices, cameras, and environmental sensors; the time-series dataset formation module is used to preprocess the collected multi-source data to form a structured time-series dataset; the time-series feature value extraction module is used to decompose the structured time-series dataset into multiple time scales and extract time-series feature values; the prediction result generation module is used to weightedly fuse feature values ​​from different time scales based on the time-series feature values ​​to predict the user's myopia risk and generate a prediction result; the dynamic prevention and control strategy generation module is used to define a reinforcement learning environment based on the prediction result and use a deep reinforcement learning algorithm to generate a dynamic prevention and control strategy; and the myopia prevention and control scheme generation module is used to track changes in the user's eye habits over a long period based on the generated dynamic prevention and control strategy to generate a myopia prevention and control scheme.

[0044] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the myopia prevention and control method based on multi-source data as described in the first aspect of the present invention.

[0045] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the myopia prevention and control method based on multi-source data as described in the first aspect of the present invention.

[0046] The beneficial effects of this invention are as follows: By using wavelet transform to decompose structured time-series datasets into multiple time scales and extracting time-series feature values, a multi-level description of user eye-use behavior and environmental changes is achieved, thereby improving the accuracy and reliability of myopia risk prediction. Secondly, based on the prediction results, a reinforcement learning environment is defined, and a deep reinforcement learning algorithm is used to generate dynamic prevention and control strategy schemes. This enables the system to dynamically adjust prevention and control measures according to the user's real-time data, providing personalized myopia prevention and control suggestions. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart of the myopia prevention and control method based on multi-source data in Example 1.

[0049] Figure 2 This is a schematic diagram of the myopia prevention and control system based on multi-source data in Example 1. Detailed Implementation

[0050] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0051] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0052] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0053] Example 1, referring to Figure 1 and Figure 2 This is the first embodiment of the present invention, which provides a myopia prevention and control method based on multi-source data, including the following steps:

[0054] S1. Collect multi-source data through smart wearable devices, cameras, and environmental sensors;

[0055] The multi-source data includes eye-use behavior data, environmental data, image data, and physiological data.

[0056] S2. Preprocess the collected multi-source data to form a structured time series dataset;

[0057] Based on the collected multi-source data, Python is used to read the data and remove missing and duplicate values, and the Scikit-learn library is used to standardize the numerical data.

[0058] It should be noted that missing and duplicate values ​​were removed to ensure data integrity and uniqueness. Numerical data was standardized using the Scikit-learn library to ensure that data from different sources have the same units of measurement, facilitating subsequent time series analysis.

[0059] The standardized multi-source data is organized into a structured format suitable for time series analysis and then fused to form a structured time series dataset.

[0060] It should be noted that data from different sources are aligned according to timestamps to ensure consistency across all data at the same point in time. Multi-source data is then merged into a unified DataFrame to form a structured time-series dataset, facilitating subsequent time-series analysis.

[0061] S3. Decompose the structured time series dataset into multiple time scales and extract time series feature values;

[0062] Wavelet transform is used to perform multi-scale decomposition on structured time series datasets, expressed as follows:

[0063]

[0064] Where W(a,b) represents the decomposition of the structured time series dataset under time scale parameter a and translation parameter b, a represents the scale parameter, b represents the translation parameter, D(t) represents the comprehensive feature value of the preprocessed multi-source data at time t, ψ(t) represents the wavelet basis function at time t, dt represents the integration of the variable at time t, and t represents the time point of the time series data;

[0065] It should be noted that 'a' controls the scaling of the wavelet basis function, with a larger 'a' value corresponding to a lower frequency and a smaller 'a' value corresponding to a higher frequency; 'b' controls the position of the wavelet basis function on the time axis; 'ψ(t)' is usually an oscillating function with finite energy; 'D(t)' is usually standardized numerical data, representing the user's eye behavior, environmental conditions, or physiological state at a certain moment. The integration operation multiplies the wavelet basis function with the original signal and sums the results over the entire time span.

[0066] By decomposing structured time series datasets, time series features at different time scales are obtained;

[0067] It should be noted that the number of decomposition levels indicates how many components of different time scales the signal is decomposed into. A higher number of decomposition levels means more scale components, but it also increases computational complexity. By capturing feature changes at different time scales, it is possible to more accurately predict a user's myopia risk.

[0068] Based on the time series characteristics at different time scales, time series feature values ​​are extracted, expressed as follows:

[0069]

[0070] Among them, F k (a) represents the time series feature value of the k-th type of feature at time scale a, δ i σ represents the weight coefficient of the i-th time series feature, n represents the number of time series features, and σ represents the weight coefficient of the ith time series feature. i φ represents the nonlinear activation function for the i-th time series feature. i (t) represents the kernel function of the i-th time series feature at time t, k represents the index of feature classification, and i represents the index variable of the time series feature;

[0071] It should be noted that φ i (t) is typically a set of orthogonal basis functions used to extract useful features from the decomposed coefficients; σ i Used to perform nonlinear transformations on inner product results.

[0072] S4. Based on time series feature values, the feature values ​​of different time scales are weighted and fused to predict the user's myopia risk and generate prediction results;

[0073] Based on time series feature values, feature values ​​from different time scales are weighted and fused to generate a comprehensive myopia risk prediction value, expressed as follows:

[0074]

[0075] Where P(t) represents the comprehensive myopia risk prediction value at time t, m represents the number of time scales, and w ν This represents the weighting coefficient for the ν-th time scale, where ν represents the index of the time scale;

[0076] It should be noted that F k (a) The eigenvalues ​​reflect the user's eye-use behavior, environmental conditions, and physiological state at different time scales; w ν This indicates the importance of each time scale. Different time scales may have varying degrees of impact on myopia risk, therefore, it is necessary to assign appropriate weights to each time scale. Weighted fusion is performed by multiplying the feature values ​​of different time scales by their corresponding weight coefficients and then normalizing them to obtain a comprehensive myopia risk prediction value.

[0077] Based on historical risk classification data, a low-risk threshold M and a high-risk threshold N are set.

[0078] It should be noted that historical risk classification data refers to the results of users' myopia risk assessments collected over a past period of time. The data usually includes users' eye-using behaviors, environmental conditions, physiological states, and corresponding myopia risk levels.

[0079] Based on the myopia risk prediction value, the risk level is classified through a low-risk threshold M and a high-risk threshold N to predict the myopia risk of the user and generate a prediction result.

[0080] If P(t) < M, it is determined as a low myopia risk.

[0081] If M ≤ P(t) < N, it is determined as a medium myopia risk.

[0082] If P(t) ≥ N, it is determined as a high myopia risk.

[0083] It should be noted that when the myopia risk is low, the user's myopia risk is relatively low, indicating that their current eye-using behaviors, environmental conditions, and physiological states are relatively healthy; when the myopia risk is medium, the user's myopia risk is at a medium level, indicating that some of their eye-using behaviors or environmental conditions may need to be adjusted; when the myopia risk is high, the user's myopia risk is relatively high, indicating that there may be relatively large problems with their current eye-using behaviors, environmental conditions, or physiological states.

[0084] S5. According to the prediction result, define the reinforcement learning environment and use the deep reinforcement learning algorithm to generate a dynamic prevention and control strategy plan.

[0085] According to the prediction result, determine that the key features affecting the myopia risk include the user's current myopia risk prediction value, eye-using behavior features, environmental features, and user physiological features, combine them into a state vector, and define the state space.

[0086] It should be noted that the eye-using behavior features describe the user's eye-using habits over a period of time, such as: the time of using electronic devices every day, the time of reading or doing close work, the frequency and duration of eye rest; the environmental features describe the impact of the environment where the user is located on the eyes, such as: indoor light intensity, screen brightness and contrast, environmental noise level; the user physiological features describe the user's physical condition and physiological parameters, such as: changes in pupil size, trend of vision change, sleep quality and duration. Combining the above key features into a state vector can comprehensively describe the user's current state.

[0087] Determine the actions to be taken to improve the user's eye-using habits and environmental conditions, including adjusting the reminder interval of eye-using time, optimizing the indoor light intensity and color temperature, increasing suggestions for outdoor activities, and adjusting the screen blue light ratio. The actions are formalized into a set and the action space is defined.

[0088] It should be noted that the interval between reminders for eye use time will be adjusted dynamically based on the user's eye-use behavior characteristics. Indoor light intensity and color temperature will be optimized to provide a more suitable eye-use environment. Outdoor activity suggestions will be added to encourage users to increase outdoor activity time and reduce close-range eye use, which helps reduce the risk of myopia. The proportion of blue light emitted from the screen will be adjusted to reduce eye strain, especially when using electronic devices at night. These actions will be formalized into a set, defining an action space.

[0089] Using the change in users' myopia risk as the primary optimization objective, the reward value is defined as follows:

[0090]

[0091] Where R represents the reward value, ΔP(t) represents the change in the user's comprehensive myopia risk prediction value within the time interval Δt, ξ(A) represents the execution cost function of action set A, η represents the user's acceptance feedback function of the strategy, λ1 represents the risk change weight, λ2 represents the execution cost weight, λ3 represents the weight of user feedback on the reward value, and A represents the action set.

[0092] It should be noted that a negative ΔP(t) indicates that the user's myopia risk has decreased, while a positive ΔP(t) indicates that the user's myopia risk has increased; λ1 is used to adjust the degree of impact of risk changes on the reward value, and a larger λ1 indicates that the system pays more attention to changes in myopia risk; ξ(A) is a cost function that includes factors such as time, effort, and economy; a higher η indicates that the user is more satisfied with the current strategy, while a lower η indicates that the user is less satisfied.

[0093] Based on the defined state space, action space and reward value, a deep reinforcement learning algorithm is used to collect experience through environmental interaction, and the best action is obtained by using the experience replay mechanism to generate a dynamic prevention and control strategy.

[0094] It should be noted that the experience replay mechanism is used to improve learning efficiency and stability. The agent stores the experience (state, action, reward, next state) of each interaction with the environment in an experience pool, and randomly draws a batch of experiences for learning during training. This helps to break the correlation between data and improve the model's generalization ability.

[0095] S6. Based on the generated dynamic prevention and control strategy, track changes in users' eye habits over a long period of time and generate a myopia prevention and control plan;

[0096] Based on the generated dynamic prevention and control strategy, the action value of each eye-use behavior feature is calculated, expressed as:

[0097]

[0098] Where V(U,t) represents the action value of visual behavior feature U at time t, U represents visual behavior feature, α represents the total number of actions in the action space, and β represents the action value of visual behavior feature U at time t. j This represents the weight coefficient of the action in the j-th spatial action, where j represents the index variable of the action in the spatial action;

[0099] It should be noted that the action value V(U,t) is a comprehensive index used to measure the overall effect of a series of actions taken to improve a specific eye-use behavior characteristic U at a specific time point t. A higher value indicates a better improvement effect of these actions on that eye-use behavior characteristic. β j This indicates the importance or priority of each action. Different actions may have different effects on the user's eye behavior, so it is necessary to assign appropriate weight coefficients to each action.

[0100] Based on the action value of eye-use behavior characteristics, a myopia prevention and control plan is generated through weighted fusion, expressed as:

[0101]

[0102] Where O(t) represents the myopia prevention and control plan at time t. ω represents the number of visual behavior characteristics. f This represents the weight coefficient of the f-th eye-use behavior feature, where f represents the index variable of the eye-use behavior feature;

[0103] It should be noted that O(t) is used to measure the overall effect of taking a series of actions on all eye-use behavior characteristics at a specific time point t. The higher this value, the better these actions are in improving users' eye-use habits and reducing the risk of myopia. Eye-use behavior characteristics include daily time spent using electronic devices, time spent reading or working at close range, and the frequency and duration of eye rest; ω f This indicates the importance or priority of each eye-use behavior feature. Different eye-use behavior features may have different impacts on a user's myopia risk, so it is necessary to assign appropriate weight coefficients to each feature.

[0104] This embodiment also provides a myopia prevention and control system based on multi-source data, including: a multi-source data acquisition module, a time-series dataset formation module, a time-series feature value extraction module, a prediction result generation module, a dynamic prevention and control strategy generation module, and a myopia prevention and control scheme generation module. The multi-source data acquisition module is used to collect multi-source data through smart wearable devices, cameras, and environmental sensors. The time-series dataset formation module is used to preprocess the collected multi-source data to form a structured time-series dataset. The time-series feature value extraction module is used to decompose the structured time-series dataset into multiple time scales and extract time-series feature values. The prediction result generation module is used to weightedly fuse feature values ​​from different time scales based on the time-series feature values ​​to predict the user's myopia risk and generate a prediction result. The dynamic prevention and control strategy generation module is used to define a reinforcement learning environment based on the prediction result and use a deep reinforcement learning algorithm to generate a dynamic prevention and control strategy. The myopia prevention and control scheme generation module is used to track changes in the user's eye habits over a long period based on the generated dynamic prevention and control strategy to generate a myopia prevention and control scheme.

[0105] This embodiment also provides a computer device applicable to the myopia prevention and control method based on multi-source data, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the myopia prevention and control method based on multi-source data as proposed in the above embodiment.

[0106] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0107] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the myopia prevention and control method based on multi-source data as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0108] In summary, this invention improves the accuracy and reliability of myopia risk prediction by: using wavelet transform to decompose structured time-series datasets into multiple time scales and extracting time-series feature values; and achieving a multi-level description of user eye-use behavior and environmental changes. Secondly, it defines a reinforcement learning environment based on the prediction results and uses deep reinforcement learning algorithms to generate dynamic prevention and control strategies, enabling the system to dynamically adjust prevention and control measures based on real-time user data and provide personalized myopia prevention and control recommendations.

[0109] It should be noted that the above embodiments are only used to illustrate the technical solutions 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 solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A myopia prevention and control method based on multi-source data, characterized in that: The application relates to a myopia risk prediction and prevention method based on multi-source data fusion and deep reinforcement learning. Multi-source data is collected through smart wearable devices, cameras and environmental sensors; The collected multi-source data is preprocessed to form a structured time series data set; The structured time series data set is decomposed into multiple time scales, and time series characteristic values are extracted; Based on the time series characteristic values, the characteristic values of different time scales are weighted and fused to predict the myopia risk of the user and generate a prediction result; According to the prediction result, a reinforcement learning environment is defined, and a dynamic prevention and control strategy scheme is generated using a deep reinforcement learning algorithm; According to the generated dynamic prevention and control strategy scheme, the user's eye habit changes are tracked in the long term, and a myopia prevention and control scheme is generated. The multi-source data includes eye behavior data, environmental data, image data and physiological data. The specific steps of preprocessing the collected multi-source data to form a structured time series data set are as follows, Based on the collected multi-source data, Python is used to read the data and delete missing values and repeated values, and the Scikit-learn library is used to standardize the numerical value type data; The standardized multi-source data is organized into a structured format suitable for time series analysis and fused to form a structured time series data set; The specific steps of decomposing the structured time series data set into multiple time scales and extracting time series characteristic values are as follows, The structured time series data set is decomposed into multiple time scales using wavelet transform, and the expression is as follows: ； in, Representing time scale parameters Translation parameters Decomposition of structured time series datasets Indicates the scale parameter. Indicates the translation parameter. This indicates the time sequence of the preprocessed multi-source data. The comprehensive eigenvalues, Indicates time wavelet basis functions, Indicates time Integrate the variables, This represents a point in time in a time series data set. Through the decomposition of the structured time series data set, the time series characteristics under different time scales are obtained; According to the time series characteristics under different time scales, the time series characteristic values are extracted, and the expression is as follows: ； wherein, denotes a time scale the lower time series characteristic value of the class feature, denotes a weight coefficient of the time series characteristic, denotes the number of time series characteristics, denotes a non-linear activation function of the time series characteristic, denotes a time series characteristic, kernel function of the characteristic classification, denotes an index variable of the time series characteristic.

2. The myopia prevention and control method based on multi-source data according to claim 1, characterized in that: The specific steps of weighting and fusing the characteristic values of different time scales based on the time series characteristic values to predict the myopia risk of the user and generate a prediction result are as follows, Based on the time series characteristic values, the characteristic values of different time scales are weighted and fused to generate a comprehensive myopia risk prediction value, and the expression is as follows: ； wherein, denotes the time of the comprehensive myopia risk prediction value, denotes the number of time scales, denotes the weight coefficient of the time scale, denotes the index of the time scale; Setting a low risk threshold and a high risk threshold based on historical risk partitioning data and high risk threshold values ; based on the myopia risk prediction value, dividing risk levels by a low risk threshold and a high risk threshold to predict the myopia risk of the user and generate a prediction result;​ If myopia is low; If the time is between 18 and 22 hours, the myopia risk is determined to be medium; If the time, the person is determined to be at high risk for myopia.

3. The myopia prevention and control method based on multi-source data of claim 2, wherein: The specific steps of defining a reinforcement learning environment according to the prediction result and generating a dynamic prevention and control strategy scheme using a deep reinforcement learning algorithm are as follows, According to the prediction result, the key features affecting the myopia risk are determined, including the current myopia risk prediction value of the user, the eye behavior feature, the environmental feature and the physiological feature of the user, which are combined into a state vector to define a state space; The actions taken to improve the eye habits and environmental conditions of the user include adjusting the eye use time reminder interval, optimizing the indoor light intensity and color temperature, increasing the outdoor activity suggestion and adjusting the screen blue light proportion, and the actions are formalized into a set to define an action space; The change amount of the myopia risk of the user is taken as the main optimization target, and the reward value is defined, and the expression is as follows: ； wherein, represents a reward value, represents a time interval a change in the integrated myopia risk prediction value of the user within, represents a set of actions an execution cost function, represents a user acceptance of the policy feedback function, represents a risk change weight, represents an execution cost weight, represents a weight of the user feedback on the reward value, represents a set of actions; Based on the defined state space, action space and reward value, a deep reinforcement learning algorithm is used to collect experience through environment interaction, and the best action is obtained using an experience replay mechanism to generate a dynamic prevention and control strategy scheme.

4. The myopia prevention and control method based on multi-source data of claim 3, wherein: The specific steps of tracking the eye habit changes of the user in the long term according to the generated dynamic prevention and control strategy scheme and generating a myopia prevention and control scheme are as follows, According to the generated dynamic prevention and control strategy scheme, the action value of each eye use behavior characteristic is calculated, and the expression is: ； wherein, represents time the eye behavior feature of the action value, represents the eye behavior feature, represents the total number of actions in the action space, represents the weight coefficient of the action in the th space action, represents the index variable of the action in the space action; Based on the action value of the eye use behavior characteristic, a myopia prevention and control scheme is generated through weighted fusion, and the expression is: ； wherein, denotes time myopia prevention and control scheme, denotes a number of eye behavior characteristics, denotes a weight coefficient of the eye behavior characteristic, denotes an index variable of the eye behavior characteristic.

5. A myopia prevention and control system based on multi-source data, based on the myopia prevention and control method based on multi-source data according to any one of claims 1-4, characterized in that: It comprises a multi-source data acquisition module, a time series data set forming module, a time series characteristic value extraction module, a prediction result generation module, a dynamic prevention and control strategy scheme generation module and a myopia prevention and control scheme generation module. The multi-source data acquisition module is used for acquiring multi-source data through intelligent wearable devices, cameras and environmental sensors. The time series data set forming module is used for preprocessing the acquired multi-source data to form a structured time series data set. The time series characteristic value extraction module is used for decomposing the structured time series data set into multiple time scales and extracting time series characteristic values. The prediction result generation module is used for weighting and fusing the characteristic values of different time scales based on the time series characteristic values to predict the myopia risk of the user and generate a prediction result. The dynamic prevention and control strategy scheme generation module is used for defining a reinforcement learning environment according to the prediction result, using a deep reinforcement learning algorithm to generate a dynamic prevention and control strategy scheme. The myopia prevention and control scheme generation module is used for long-term tracking of the change of the eye use habits of the user according to the generated dynamic prevention and control strategy scheme to generate a myopia prevention and control scheme.

6. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that: The processor executes the computer program to realize the steps of the myopia prevention and control method based on multi-source data according to any one of claims 1-4.

7. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the myopia prevention and control method based on multi-source data according to any one of claims 1-4.

Images