Diabetes management system based on big data

Abstract

The invention relates to the technical field of health data management, in particular to a diabetes management system based on big data, which comprises an abnormity monitoring module, a risk analysis module, a health state simulation module and a monitoring management module. According to the invention, through the data processing and analysis technology, the diabetes management is significantly improved, the data quality is ensured, the unqualified data is rejected, the abnormal points are identified through the refined data review and abnormal monitoring strategy, and the development trend of the abnormal points is analyzed through the time sequence, so that the data processing not only improves the data accuracy, but also improves the accuracy of diabetes management. The prediction capability of potential health risks and the evaluation and grading of risk areas are enhanced, so that medical providers can more accurately understand and prevent disease risks, future health transition can be predicted by simulating the health state of a diabetic patient, making of a more prospective recovery plan is helped, and the recovery efficiency of the diabetic patient is improved. The disease management efficiency and effect are improved, and the life quality of the patient is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of health data management, and in particular to a diabetes management system based on big data. Background Art

[0002] The field of health data management technology focuses on effectively collecting, storing, managing and analyzing health-related data, including various technologies from the implementation of electronic health record (EHR) systems to big data analysis, aiming to improve the quality of medical services, reduce costs, promote personalized medicine, and enhance patient management efficiency. By utilizing advanced data processing technologies such as machine learning and artificial intelligence, health data management can reveal statistical trends in etiology and achieve real-time monitoring of patient health status, which plays an important role in preventing various diseases, especially chronic diseases.

[0003] Among them, the diabetes management system refers to a technical solution specifically used to monitor and process the health information of diabetic patients. This system helps medical professionals to control and manage diabetes more accurately by collecting patients' blood sugar data, eating habits, physical activities and other information. Its uses include tracking the progression of the disease, predicting blood sugar changes, and providing patients with customized health advice, which greatly improves the quality of life and recovery of diabetic patients.

[0004] Existing technologies have obvious deficiencies in actual operation, especially in data management and risk prediction. Data processing in existing systems is not in-depth enough, and the identification and analysis of abnormal data are not sophisticated enough, resulting in limited data quality and availability. Existing technologies have limited capabilities in health status prediction and disease progression monitoring, and cannot provide sufficient foresight and recovery advice. This deficiency makes disease management slow to respond, unable to effectively prevent disease risks or adjust recovery strategies to cope with changes in patients' health status. For example, in the long-term management of diabetes, insufficient early warning of blood sugar mutations or complications leads to worsening of patients' conditions and increased recovery costs and complexity. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose a diabetes management system based on big data.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: The diabetes management system based on big data includes:

[0007] The abnormal monitoring module collects blood sugar, blood pressure and weight data of diabetic patients, checks the integrity of the data, removes unqualified data, identifies and analyzes abnormal points in the data, obtains abnormal identification results, and performs time series analysis on the abnormal identification results, analyzes the development trend of abnormal data, and obtains time trend analysis results;

[0008] The risk analysis module analyzes the clustering characteristics of abnormal data based on the time trend analysis results, identifies the clustering distribution, generates an initialization risk area, performs risk assessment on the initialization risk area, identifies the risk level of diabetes symptoms, and obtains a risk assessment index;

[0009] The health status simulation module uses the risk assessment indicators to dynamically simulate the health status of diabetic patients, analyze the differentiated health status transition probabilities, construct a health transition probability matrix, use the health transition probability matrix to predict the health status of patients in the future time period, evaluate the consistency of the prediction, and obtain the health status prediction result;

[0010] The monitoring management module identifies key monitoring indicators and intervention time periods based on the health status prediction results, optimizes health monitoring parameters, generates a health monitoring plan for diabetic patients, adjusts the monitoring frequency, evaluates the effectiveness and impact of the implementation of the health monitoring plan for diabetic patients, and obtains a health management plan for diabetic patients.

[0011] As a further solution of the present invention, the step of obtaining the abnormal identification result is specifically:

[0012] Collect blood sugar, blood pressure and weight data of diabetic patients, check the integrity of the data through data verification, screen the data that meet the quality standards, and generate a quality-screened data set;

[0013] Identifying outliers on the quality-screened data set, by calculating the weighted Euclidean distance between each data point and adjacent data points, and comparing the distance with a dynamically adjusted threshold, identifying data points that deviate from a normal range, and generating an outlier data point set;

[0014] Perform time series analysis on the abnormal data point set to analyze the development trend of the data according to the formula:

[0015]

[0016] Calculate the time-weighted moving average of the abnormal points to obtain the abnormal identification result;

[0017] Among them, E(t) represents the abnormal trend at time point t, a i (t) represents the i-th abnormal data point at time point t, w i represents the time weight of the i-th data point at time point t, and n is the number of abnormal data points.

[0018] As a further solution of the present invention, the step of obtaining the time trend analysis result is specifically:

[0019] Utilizing the anomaly identification result, extracting the timestamp of the anomaly occurrence, recording the time when the anomaly occurred, tracking the time when the anomaly occurred in real time, and obtaining a time stamp list of the anomaly values;

[0020] Performing time window analysis on the time tag list of the outliers, evaluating the frequency of anomalies within the differentiated time window through daily and weekly statistics, calculating the anomaly frequency, and obtaining the time window anomaly frequency result;

[0021] Based on the abnormal frequency results in the time window, the development of the abnormality is evaluated using the formula:

[0022]

[0023] Calculate the outlier trend index R within the differentiated time window b , get the time trend analysis results;

[0024] Where b represents the time variable, β0, β1 and γ are regression coefficients, ∈ b is a random error term, α and δ are adjustment coefficients, and e is a natural constant.

[0025] As a further solution of the present invention, the step of obtaining the initial risk area is specifically:

[0026] According to the time trend analysis results, the time period in which the data fluctuation amplitude exceeds the threshold is screened, potential risk events are identified, and a potential risk time period record is obtained;

[0027] Based on the potential risk time period records, the time periods are grouped, risk patterns of differentiated types are identified, and a risk pattern grouping set is obtained;

[0028] By grouping the risk patterns together, the formula is used:

[0029]

[0030] Calculate the risk value R for each group c , if the risk value exceeds the threshold, it is marked to obtain the initialized risk area;

[0031] Among them, w j is the risk weight, x cj is the risk value of the jth indicator in the cth group, and m is the number of indicators.

[0032] As a further solution of the present invention, the step of obtaining the risk assessment index is specifically as follows:

[0033] Extracting key data indicators of diabetic patients from the initialized risk area, including blood sugar level, blood pressure and weight records of diabetic patients, to obtain a set of key data indicators;

[0034] Performing statistical analysis on the key data indicator set, calculating the mean and standard deviation of the differentiated key data indicators, evaluating the risk contribution of each indicator, and obtaining statistical analysis results;

[0035] Using the statistical analysis results, the criticality of the indicator is measured according to the volatility of the data, using the formula:

[0036]

[0037] Calculate the coefficient of variation V for each indicator k , and obtain the risk assessment index;

[0038] Among them, x k is a single data point, μ and σ are the mean and standard deviation of the indicator, respectively, and p is the number of indicators k.

[0039] As a further solution of the present invention, the steps of obtaining the health transition probability matrix are specifically as follows:

[0040] The risk assessment indicators, including blood sugar control, weight changes and diabetic complications records, are used to standardize the risk assessment indicators, convert them into comparable scoring scales, and obtain standardized risk scores;

[0041] The standardized risk score is combined with the health data of diabetic patients to analyze the association, using the formula:

[0042]

[0043] Estimate the transition probability P between differentiated health states or , construct the health transition probability matrix;

[0044] Among them, S or represents the standardized risk score of transitioning from health state o to health state r, α o , β u and β r is the adjustment coefficient, exp is the natural exponential function, S ou Represents the normalized risk score for transitioning from health state o to health state u.

[0045] As a further solution of the present invention, the steps of obtaining the health status prediction result are specifically as follows:

[0046] Using the health transition probability matrix, by analyzing the data adaptability of the health transition probability matrix, the applicability and consistency of the prediction of the health status of the diabetic patient in the future time period are evaluated to obtain a adaptability score record;

[0047] According to the fitness score record, combined with the real-time health data of diabetic patients, the formula is adopted:

[0048]

[0049] Calculate the probability prediction value of the health status and obtain the health status prediction result;

[0050] Among them, P(t+1) is the predicted health status probability at the next time point, P(t) is the health status probability at the real time point, and x j (t) is the jth index at time point t, β j is the regression coefficient, B is the smoothing parameter, and m is the number of indicators.

[0051] As a further solution of the present invention, the steps for obtaining the health management plan for diabetic patients are specifically as follows:

[0052] Utilizing the health status prediction results, analyzing the risk level and potential health problems of diabetic patients, identifying key monitoring indicators and necessary intervention time periods, and generating a health monitoring protocol for diabetic patients;

[0053] Implement the diabetic patient health monitoring protocol, adjust the monitoring frequency, and use the formula:

[0054] R adj =R base ·(1+α·P risk );

[0055] Evaluate the effect and impact of the implementation of the diabetic patient health monitoring plan and obtain the diabetic patient health management plan;

[0056] Among them, R adj represents the adjusted monitoring frequency, R base is the basic monitoring frequency, α is the adjustment coefficient, P risk is the risk score for patients with diabetes.

[0057] Compared with the prior art, the advantages and positive effects of the present invention are:

[0058] In the present invention, through data processing and analysis technology, significant improvements are brought to diabetes management. Refined data review and abnormal monitoring strategies ensure data quality, eliminate unqualified data, identify abnormal points, and analyze the development trend of abnormal points through time series analysis. Such data processing not only improves the accuracy of data, but also enhances the ability to predict potential health risks. The assessment and classification of risk areas allow medical providers to understand and prevent disease risks more accurately. By simulating the health status of diabetic patients, future health changes can be predicted, helping to formulate more forward-looking recovery plans, improving the efficiency and effectiveness of disease management, and promoting the improvement of patients' quality of life. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a system flow chart of the present invention;

[0060] Figure 2 It is a flow chart of the abnormal identification result in the present invention;

[0061] Figure 3 It is a flow chart of the time trend analysis results in the present invention;

[0062] Figure 4 A flow chart of initializing the risk area in the present invention;

[0063] Figure 5 It is a flow chart of the risk assessment indicators in the present invention;

[0064] Figure 6 It is a flow chart of the health transition probability matrix in the present invention;

[0065] Figure 7 A flow chart of the health status prediction results in the present invention;

[0066] Figure 8 The figure is a flow chart of the health management program for diabetic patients in the present invention. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0068] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0069] See also Figure 1 , the diabetes management system based on big data includes:

[0070] The abnormal monitoring module collects blood sugar, blood pressure and weight data of diabetic patients, checks the integrity of the data, removes unqualified data, identifies and analyzes abnormal points in the data, obtains abnormal identification results, and performs time series analysis on the abnormal identification results, analyzes the development trend of abnormal data, and obtains time trend analysis results;

[0071] The risk analysis module analyzes the clustering characteristics of abnormal data based on the time trend analysis results, identifies the clustering distribution, generates an initial risk area, performs risk assessment on the initial risk area, identifies the risk level of diabetes symptoms, and obtains risk assessment indicators;

[0072] The health status simulation module uses risk assessment indicators to dynamically simulate the health status of diabetic patients, analyze the differentiated health status transition probabilities, construct a health transition probability matrix, use the health transition probability matrix to predict the health status of patients in the future time period, evaluate the consistency of the prediction, and obtain the health status prediction results;

[0073] The monitoring management module identifies key monitoring indicators and intervention time periods based on health status prediction results, optimizes health monitoring parameters, generates health monitoring protocols for diabetic patients, adjusts monitoring frequency, evaluates the effectiveness and impact of the implementation of health monitoring protocols for diabetic patients, and obtains a health management plan for diabetic patients.

[0074] The results of anomaly identification include anomaly point identification, data integrity check, and elimination of unqualified data. The results of time trend analysis include data development trends, time series characteristics, and anomaly point analysis. Initialization of risk areas includes risk clustering characteristics, clustering distribution, and risk area assessment. Risk assessment indicators include risk level, risk assessment results, and cluster analysis results. The health transition probability matrix includes health state transition probability, model parameters, and health state dynamic simulation. The health status prediction results include health status prediction in the future time period, prediction model effect, and health state transfer analysis. The draft of health monitoring for diabetic patients includes monitoring indicators, intervention time period, and health monitoring parameter optimization. The health management plan for diabetic patients includes monitoring frequency adjustment, health management effect evaluation, and the impact of monitoring plan implementation.

[0075] See also Figure 2 , the specific steps for obtaining abnormal recognition results are:

[0076] Collect blood glucose, blood pressure and weight data of diabetic patients, verify the data integrity, screen the data that meet the quality standards, and generate a quality-screened data set;

[0077] During the data collection phase of clinical research, collecting blood sugar, blood pressure and weight data from diabetic patients is a critical first step. This process involves regular monitoring using a variety of medical devices, including blood glucose meters, blood pressure monitors and electronic scales. Data verification and screening are performed using professional data verification algorithms that detect the integrity and accuracy of the data and automatically mark any abnormal or significantly deviated readings from the normal range. For example, blood sugar values ​​should not be lower than 3.9 or higher than 7.8 in the non-meal state, and the normal range of blood pressure should be between 90 / 60 and 120 / 80. Data cleaning and processing techniques are also used to screen out data that meets medical quality standards, including removing any erroneous data points caused by equipment failure or operational errors to generate quality-screened data sets.

[0078] Identify outliers on the quality-screened data set by calculating the weighted Euclidean distance between each data point and its neighboring data points and comparing it with the dynamically adjusted threshold to identify data points that deviate from the normal range and generate an abnormal data point set;

[0079] Detailed anomaly detection of the data set is directly related to the accuracy and credibility of subsequent analysis. Anomaly detection is mainly performed by calculating the weighted Euclidean distance between each data point and the neighboring data. This not only involves direct data point comparison, but also introduces the concept of dynamic thresholds to better adapt to the physiological differences between individuals. In specific operations, the neighborhood of each data point is defined, and the weighted distance between the target data point and the data points in the neighborhood in terms of blood sugar, blood pressure, and weight is calculated. The setting of weights depends on the medical importance of each dimension and the degree of contribution to diabetes control. For example, for blood sugar data, the weight is higher than that of weight data, because blood sugar directly affects the daily management of diabetes. It can effectively identify abnormal data points that deviate from the normal physiological range, lay a solid foundation for subsequent data analysis, and generate a set of abnormal data points.

[0080] Perform time series analysis on the abnormal data point set and analyze the development trend of the data according to the formula:

[0081]

[0082] Calculate the time-weighted moving average of the abnormal points to obtain the abnormal identification result;

[0083] Among them, E(t) represents the abnormal trend at time point t, a i (t) represents the i-th abnormal data point at time point t, w i represents the time weight of the i-th data point at time point t, and n is the number of abnormal data points;

[0084] The time weight is set based on the location and importance of the data point during the entire monitoring period. For example, in the intraday blood sugar fluctuation monitoring, the reading after the meal will be given a higher weight, which is particularly critical for understanding the patient's blood sugar control status;

[0085] Through actual data monitoring, set at a specific time point, a total of 5 data points were monitored, among which the abnormal data point was identified as a(t) = [1, 0, 1, 1, 0];

[0086] The weights are set to w = [0.5, 0.2, 0.5, 0.7, 0.1] according to the importance of the data;

[0087] The calculation process is as follows:

[0088] Calculate the sum of weights:

[0089]

[0090] Calculate the sum of weighted outliers:

[0091]

[0092] Apply the formula to calculate E(t):

[0093]

[0094] The results show that the weighted average of the abnormal trend at time point t is 0.85. The closer this value is to 1, the higher the degree of abnormality at that time point, which reflects the degree of abnormality and distribution of abnormal data points, helps to further analyze the development trend of the data and provide data support for medical intervention.

[0095] See also Figure 3 , the specific steps for obtaining the time trend analysis results are:

[0096] Using the anomaly recognition results, extract the timestamp of the anomaly occurrence, record the time when the anomaly occurred, track the time when the anomaly occurred in real time, and obtain a time-stamped list of anomaly values;

[0097] Determine the existence of anomalies at each time point. This operation involves recording the time when the anomaly occurred, accurate to a specific timestamp, such as the year, month, day, hour, minute, and second when the anomaly occurred. This operation ensures that each abnormal event can be accurately tracked. The time stamp list includes the specific time of occurrence of each anomaly, providing the necessary basic data for subsequent analysis. Through efficient data indexing technology, data at any time point can be quickly accessed, which is crucial for processing batch time series data. Each timestamp is also associated with a measure of anomaly intensity, which reflects the severity of the anomaly and provides a basis for the abnormal frequency and trend analysis in subsequent steps, resulting in a time stamp list of anomaly values.

[0098] Perform time window analysis on the time-stamped list of outliers, evaluate the frequency of anomalies in differentiated time windows through daily and weekly statistics, calculate the anomaly frequency, and obtain the time window anomaly frequency result;

[0099] Perform a detailed time window analysis on the time-stamped list of outliers. By selecting an appropriate window length (such as one day or one week), count the number of anomalies in each window. This statistical analysis is done through advanced data aggregation technology, allowing the anomaly frequency to be quickly calculated from batches of time tags and the time distribution trend of anomalies to be evaluated. This time window analysis also includes seasonal or periodic evaluation of anomaly occurrence patterns, such as analyzing whether there is an anomaly pattern that occurs at a specific time each week. This information is extremely important for understanding the root cause of the anomaly, and not only provides data support for the next trend analysis, but also provides a scientific basis for decision-making to obtain the time window anomaly frequency results.

[0100] According to the abnormal frequency results of the time window, the development and change of the abnormality are evaluated using the formula:

[0101]

[0102] Calculate the outlier trend index R within the differentiated time window b , get the time trend analysis results;

[0103] Where b represents the time variable, β0, β1 and γ are regression coefficients, ∈ b is the random error term, α and δ are adjustment coefficients, and e is a natural constant;

[0104] The formula combines exponential and square root transformations to improve sensitivity to fast-changing trends in data and resistance to small random fluctuations, effectively balancing the response to abnormal data and the stability of the overall data trend. It is particularly suitable for abnormal trend analysis in a dynamically changing environment.

[0105] Given data example: b = 5 (days);

[0106] α = 1.2 (dynamically adjusted according to the change range of historical data);

[0107] β0 = 0.5 (baseline value obtained through stability analysis of historical data);

[0108] β1 = 0.3 (based on regression analysis results of long-term data trends);

[0109] γ = 0.05 (trend acceleration factor identified from historical data);

[0110] δ = 1.5 (calculated based on the standard deviation of historical errors);

[0111] ∈ b =0.2 (mean error calculated from previous anomaly detection data);

[0112] Calculation process:

[0113] e γb =e 0.05×5 ≈1.284; Display time growth factor;

[0114] Displays adjustments for random errors;

[0115] R b =1.2×0.5+0.3×1.284+1.5×0.447≈2.095;

[0116] The results show that the outlier trend index predicted for 5 days later is 2.095, which means that the development trend of anomalies will significantly increase during this period, and special monitoring and response measures need to be taken at this time point to ensure the stability and security of the system.

[0117] See also Figure 4 , the specific steps for obtaining the initial risk area are:

[0118] According to the time trend analysis results, filter the time periods where the data fluctuations exceed the threshold, identify potential risk events, and obtain the potential risk time period records;

[0119] By carefully analyzing the patients' diabetes-related indicator data over a period of time, we identified the fluctuation characteristics in the data, especially the abnormal data points that exceeded the predetermined threshold. This threshold is set based on the needs of disease management, and mainly refers to the daily measurement changes of blood sugar, including statistical parameters such as the mean value, standard deviation and coefficient of variation of blood sugar measurements. The statistical parameters are obtained through actual data monitoring. When determining the fluctuation range, not only the degree of deviation of a single measurement value is considered, but also the fluctuation trend of data for multiple consecutive days is evaluated, so as to more accurately capture the estimated risk points. After preliminary screening, the data will be used for cluster analysis to more carefully identify and classify different risk patterns. Each risk time period is confirmed by the actual situation of a specific blood sugar measurement value exceeding the threshold, ensuring the practical application value and accuracy of the analysis results and obtaining a record of potential risk time periods.

[0120] Based on the records of potential risk time periods, the time periods are grouped, and risk patterns of different types are identified to obtain a risk pattern grouping set;

[0121] The time periods were analyzed and grouped in detail using a clustering algorithm. The process involved multiple steps, the first of which was to select a suitable clustering algorithm. The K-means clustering algorithm was used, which is a widely used clustering technology suitable for processing batch data sets. Before applying the clustering algorithm, the data was standardized to eliminate the dimensional impact between different indicators and ensure the accuracy of the clustering results. The main purpose of clustering is to group data with similar risk characteristics. The distance between each data point and the remaining points is calculated using the Euclidean distance, because the Euclidean distance has an intuitive advantage in expressing the absolute difference between points. The clustering center is iteratively optimized to ensure the pertinence of the analysis and the feasibility of the operation, and a risk pattern grouping set is obtained.

[0122] By risk pattern grouping set, the formula is used:

[0123]

[0124] Calculate the risk value R for each group c , if the risk value exceeds the threshold, it is marked to obtain the initialized risk area;

[0125] Among them, w j is the risk weight, x cjis the risk value of the jth indicator in group c, and m is the number of indicators;

[0126] Multiple risk indicators are combined by weighted summation to calculate the comprehensive risk value of each risk pattern grouping. The weighting factor w j It reflects the importance of different indicators in risk assessment, making the results closer to the actual risk status;

[0127] There is a grouping that includes two indicators, blood sugar level and body mass index, with corresponding risk weights of 0.6 and 0.4. The average blood sugar level and body mass index in this group are 180 and 25 respectively.

[0128] Then the calculation process is:

[0129] R c =0.6·180+0.4·25=108+10=118;

[0130] The results show that for this specific risk pattern grouping, the comprehensive risk value is 118. The size of this value can be used directly to assess the risk level of the group. After exceeding the set risk threshold, the grouping is marked as a high-risk area, which will directly affect subsequent disease management and intervention plans. It can effectively determine that some risk pattern groups have higher risks and need to be managed.

[0131] See also Figure 5 , the specific steps for obtaining risk assessment indicators are:

[0132] Extracting key data indicators of diabetic patients from the initial risk area, including blood sugar level, blood pressure and weight records of diabetic patients, to obtain a set of key data indicators;

[0133] The division of the initial risk area is carried out through a detailed analysis of key data indicators, including key information such as blood sugar levels, blood pressure and weight records of diabetic patients. Through comprehensive processing and analysis of data, the estimated health risk areas can be identified more accurately. This process involves data collection and processing, including data cleaning, in which the removal of outliers is very critical, such as blood pressure values ​​that are too high or too low will be removed to ensure the quality and reliability of the data set. Through statistical analysis methods, the mean and standard deviation of each data indicator are calculated. Statistics not only help understand the overall distribution of the data, but also evaluate the contribution of each indicator to the overall risk assessment. A higher standard deviation in blood sugar data indicates the instability of the patient's blood sugar control, which is an important factor in assessing the risk of diabetic patients, and a set of key data indicators is obtained.

[0134] Conduct statistical analysis on the key data indicator set, calculate the mean and standard deviation of differentiated key data indicators, evaluate the risk contribution of each indicator, and obtain statistical analysis results;

[0135] When grouping risk patterns, the clustering algorithm used groups risk areas based on the similarity of data in different time periods. This process involves several key steps. The selection and parameter setting of the clustering algorithm are based on the specific characteristics of the data, such as the number and type of clusters. The correct selection of parameters is crucial to correctly identifying risk patterns. Each grouped data reflects a specific risk pattern, which is achieved by analyzing the similarities and differences of data points within each group. By determining that some time periods show similar patterns of elevated blood sugar, potential risks of diabetes complications can be identified. Further analysis also includes evaluating the prevalence of risk patterns in different patient groups and their potential impact on patient health, which helps medical providers develop targeted interventions to mitigate the impact of potential risks and obtain statistical analysis results.

[0136] Using the statistical analysis results, the criticality of the indicator is measured according to the volatility of the data, using the formula:

[0137]

[0138] Calculate the coefficient of variation V for each indicator k , and obtain the risk assessment index;

[0139] Among them, x k is a single data point, μ and σ are the mean and standard deviation of the indicator, respectively, and p is the number of indicators k;

[0140] The risk is measured by calculating the coefficient of variation of each indicator. Indicators with a high coefficient of variation mean that the data points fluctuate more relative to the mean, indicating a higher risk. Therefore, the formula can help effectively identify and prioritize data indicators with high risks.

[0141] Set the mean to 100, the standard deviation to 15, the sample size to 50, and the data point distribution to (90, 110, 95, 105, 115);

[0142] The coefficient of variation is calculated as follows:

[0143]

[0144] The results showed that greater volatility in a given data indicator indicates a higher risk and is critical for the management of diabetic patients, as highly volatile blood sugar levels require more intensive monitoring and intervention.

[0145] See also Figure 6 , the specific steps for obtaining the health transition probability matrix are:

[0146] Risk assessment indicators, including blood sugar control, weight changes, and diabetic complications records, were standardized and converted into comparable scoring scales to obtain standardized risk scores;

[0147] By applying risk assessment indicators including blood sugar control, weight change, and diabetic complications records, data standardization is performed to convert indicators into comparable scoring scales. This process involves multiple data processing steps, quantifying each indicator, quantifying blood sugar levels into blood sugar control scores, and converting weight changes into weight change scores based on percentage changes. Diabetic complications are classified and scored according to severity, and the scores are normalized to eliminate the impact of different scales and ensure data consistency and comparability. Through this series of operations, not only the quality of the data is ensured, but also the accuracy of subsequent model predictions is improved. The key to this process is to ensure that each risk indicator can fairly reflect the influence in the assessment model and obtain a standardized risk score.

[0148] By combining the standardized risk score with the health data of diabetic patients, the association was analyzed using the formula:

[0149]

[0150] Estimate the transition probability P between differentiated health states or , construct the health transition probability matrix;

[0151] Among them, S or represents the standardized risk score of transitioning from health state o to health state r, α o , β u and β r is the adjustment coefficient, exp is the natural exponential function, S ou represents the standardized risk score for moving from health state o to health state u;

[0152] By introducing a standardized risk score S from state o to r or and the adaptability parameter α o and β r , which enhances the model's sensitivity to health status transitions and its ability to adapt to individual differences, helps to accurately predict the health transition probability of diabetic patients and provides a basis for formulating personalized recovery recommendations;

[0153] There are three health states A, B and C, and the corresponding standardized risk scores are S oA =0.5, S oB =1.0,S oC =1.5; Adaptive parameter α o =0.2,βA =0.3,β B =0.5,β C =0.7;

[0154] Calculate the transition probability from state o to A according to the formula:

[0155]

[0156] The result shows that the transition probability from state o to state A is 25.5%, reflecting the probability of the patient transitioning from state o to state A under the current health score and parameter configuration.

[0157] See also Figure 7 , the specific steps for obtaining the health status prediction results are:

[0158] By using the health transition probability matrix and analyzing the data adaptability of the health transition probability matrix, the applicability and consistency of the prediction of the health status of diabetic patients in the future time period are evaluated to obtain the adaptability score record;

[0159] Using the constructed health transition probability matrix, the historical data adaptability analysis of the model is performed to evaluate the applicability and accuracy of the model for predicting future health status. By comparing the differences between the model's past predictions and actual health outcomes, and using statistical error analysis methods such as mean square error and determination coefficient to evaluate the accuracy of the prediction, the adaptability of the model can be quantified, reflecting the reliability of the model in processing future data, indicating that the more accurate the model is in predicting future patient health status, this scoring mechanism not only helps medical professionals understand the performance of the model, but also ensures that the health prediction model can be effectively adjusted and optimized in actual applications to obtain a fitness score record.

[0160] According to the fitness score records, combined with the real-time health data of diabetic patients, the formula is used:

[0161]

[0162] Calculate the probability prediction value of the health status and obtain the health status prediction result;

[0163] Among them, P(t+1) is the predicted health status probability at the next time point, P(t) is the health status probability at the real time point, and x j (t) is the jth index at time point t, β j is the regression coefficient, B is the smoothing parameter, and m is the number of indicators;

[0164] Through the smoothing parameter B and the weighting coefficient β of the health index jWith the introduction of the model, the model can flexibly adjust its response to historical and new input data, improving the accuracy and adaptability of predictions. This approach enables the model to quickly adapt to recent health data changes while maintaining the influence of historical data;

[0165] The health status probability P(t) at the current time point t is set to 0.7, and the smoothing parameter B is set to 0.3. This parameter reflects the weight given to new data while retaining the influence of historical data. There are three health indicators, corresponding to the coefficients β j are 0.5, 0.3, and 0.2 respectively, quantifying the contribution of each health indicator to the prediction model. The current value of each health indicator x j (t) are 1, 0, and 1 respectively, representing specific health status data such as blood sugar control and weight changes:

[0166] P(t+1)=(1-0.3)×0.7+0.3×(0.5×1+0.3×0+0.2×1)=0.49+0.3×(0.5+0+0.2)=0.49+0.3×0.7=0.49+0.2=0.70;

[0167] The result shows that at the next time point t+1, the probability prediction value of the health status is 0.70, which indicates the probability that the patient will maintain a stable health status. This numerical result directly supports the output of the health status prediction, showing that the model can effectively maintain the prediction accuracy.

[0168] See also Figure 8 , the specific steps for obtaining the health management plan for diabetic patients are:

[0169] Analyze the risk level and potential health problems of diabetic patients using the health status prediction results, identify key monitoring indicators and necessary intervention time periods, and generate health monitoring protocols for diabetic patients;

[0170] The initial analysis is conducted through the health status prediction results to assess the current health status of diabetic patients, including physiological indicators such as blood sugar level, blood pressure, heart rate, and living habits such as diet, activity level, and sleep patterns. The data comes from the patient's daily monitoring records and is used to compare the patient's historical health data to determine new or aggravated health risks faced by the patient. Based on the prediction results, the patient's acute or chronic health problems are identified. The judgment is based on big data health analysis, such as sudden fluctuations in blood sugar or long-term high blood pressure, which are important bases for determining monitoring indicators. According to the risk factors, the necessary intervention time period is determined. The time period is set based on the patient's past medical records and pattern recognition results, so that intervention can be carried out in advance when health problems occur. Such intervention is not only timely and effective, but also customized based on the individual needs of patients, with the goal of preventing potential problems before they break out. The adjustment of monitoring frequency is equally important. The monitoring frequency and intervention measures will be adjusted according to the patient's real-time data and long-term health trends to maximize the monitoring effect and reduce patient discomfort, accurately reflect the patient's actual needs, and improve the timeliness and accuracy of monitoring, and generate a health monitoring draft for diabetic patients.

[0171] Implement the health monitoring protocol for diabetic patients, adjust the monitoring frequency, and use the formula:

[0172] R adj =R base ·(1+α·P risk );

[0173] Evaluate the effect and impact of the implementation of the diabetic patient health monitoring plan and obtain the diabetic patient health management plan;

[0174] Among them, R adj represents the adjusted monitoring frequency, R base is the basic monitoring frequency, α is the adjustment coefficient, P risk is the risk score for patients with diabetes;

[0175] Adjusting the frequency of health monitoring based on the patient's risk score makes the monitoring plan for diabetes patients more personalized and targeted. The monitoring frequency can be adjusted according to the patient's specific health status instead of a one-size-fits-all approach, which helps to improve the effectiveness of monitoring and reduce unnecessary interventions;

[0176] Set the basic monitoring frequency R base Once a week, the patient's risk score P risk is 0.3, and the adjustment coefficient α is set to 0.5, which means that for every additional risk point, the monitoring frequency increases by 50%;

[0177] Calculate the adjusted monitoring frequency R adj for:

[0178] R adj =1·(1+0.5·0.3)=1·1.15=1.15;

[0179] This means that the monitoring frequency should be adjusted from once a week to 1.15 times a week. The results show that for patients with moderate health risks, increasing the monitoring frequency can better capture potential health problems and intervene in time to improve the overall effect of health management. This adjustment allows the monitoring plan for diabetic patients to be flexibly adjusted according to the actual health status of the patient, ensuring that the patient can receive timely attention and treatment in the face of any sudden or predicted health risks, achieving the goal of reducing the incidence of acute events and optimizing chronic disease management.

[0180] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A diabetes management system based on big data, characterized in that: The system comprises: The abnormal monitoring module collects blood sugar, blood pressure and weight data of diabetic patients, checks the integrity of the data, removes unqualified data, identifies and analyzes abnormal points in the data, obtains abnormal identification results, and performs time series analysis on the abnormal identification results, analyzes the development trend of abnormal data, and obtains time trend analysis results; The risk analysis module analyzes the clustering characteristics of abnormal data based on the time trend analysis results, identifies the clustering distribution, generates an initialization risk area, performs risk assessment on the initialization risk area, identifies the risk level of diabetes symptoms, and obtains a risk assessment index; The health status simulation module uses the risk assessment indicators to dynamically simulate the health status of diabetic patients, analyze the differentiated health status transition probabilities, construct a health transition probability matrix, use the health transition probability matrix to predict the health status of patients in the future time period, evaluate the consistency of the prediction, and obtain the health status prediction result; The monitoring management module identifies key monitoring indicators and intervention time periods based on the health status prediction results, optimizes health monitoring parameters, generates a health monitoring plan for diabetic patients, adjusts the monitoring frequency, evaluates the effectiveness and impact of the implementation of the health monitoring plan for diabetic patients, and obtains a health management plan for diabetic patients.

2. The diabetes management system based on big data according to claim 1, characterized in that: The steps for obtaining the abnormal identification result are specifically as follows: Collect blood glucose, blood pressure and weight data of diabetic patients, verify the data integrity, screen the data that meet the quality standards, and generate a quality-screened data set; Identifying outliers on the quality-screened data set, by calculating the weighted Euclidean distance between each data point and adjacent data points, and comparing the distance with a dynamically adjusted threshold, identifying data points that deviate from a normal range, and generating an outlier data point set; Perform time series analysis on the abnormal data point set to analyze the development trend of the data according to the formula: Calculate the time-weighted moving average of the abnormal points to obtain the abnormal identification result; Among them, E(t) represents the abnormal trend at time point t, a i (t) represents the i-th abnormal data point at time point t, w i represents the time weight of the i-th data point at time point t, and n is the number of abnormal data points.

3. The diabetes management system based on big data according to claim 2, characterized in that: The steps for obtaining the time trend analysis results are specifically as follows: Utilizing the anomaly identification result, extracting the timestamp of the anomaly occurrence, recording the time when the anomaly occurred, tracking the time when the anomaly occurred in real time, and obtaining a time stamp list of the anomaly values; Performing time window analysis on the time tag list of the outliers, evaluating the frequency of anomalies within the differentiated time window through daily and weekly statistics, calculating the anomaly frequency, and obtaining the time window anomaly frequency result; Based on the abnormal frequency results in the time window, the development of the abnormality is evaluated using the formula: Calculate the outlier trend index R within the differentiated time window b , get the time trend analysis results; Where b represents the time variable, β0, β1 and γ are regression coefficients, ∈ b is a random error term, α and δ are adjustment coefficients, and e is a natural constant.

4. The diabetes management system based on big data according to claim 3, characterized in that: The steps of obtaining the initial risk area are specifically as follows: According to the time trend analysis results, the time period in which the data fluctuation amplitude exceeds the threshold is screened, potential risk events are identified, and a potential risk time period record is obtained; Based on the potential risk time period records, the time periods are grouped, risk patterns of differentiated types are identified, and a risk pattern grouping set is obtained; By grouping the risk patterns together, the formula is used: Calculate the risk value R for each group c , if the risk value exceeds the threshold, it is marked to obtain the initialized risk area; Among them, w j is the risk weight, x cj is the risk value of the jth indicator in the cth group, and m is the number of indicators.

5. The diabetes management system based on big data according to claim 4, characterized in that: The steps for obtaining the risk assessment indicators are specifically as follows: Extracting key data indicators of diabetic patients from the initialized risk area, including blood sugar level, blood pressure and weight records of diabetic patients, to obtain a set of key data indicators; Performing statistical analysis on the key data indicator set, calculating the mean and standard deviation of the differentiated key data indicators, evaluating the risk contribution of each indicator, and obtaining statistical analysis results; Using the statistical analysis results, the criticality of the indicator is measured according to the volatility of the data, using the formula: Calculate the coefficient of variation V for each indicator k , and obtain the risk assessment index; Among them, x k is a single data point, μ and σ are the mean and standard deviation of the indicator, respectively, and p is the number of indicators k.

6. The diabetes management system based on big data according to claim 5, characterized in that: The steps for obtaining the health transition probability matrix are specifically as follows: The risk assessment indicators, including blood sugar control, weight changes and diabetic complications records, are used to standardize the risk assessment indicators, convert them into comparable scoring scales, and obtain standardized risk scores; The standardized risk score is combined with the health data of diabetic patients to analyze the association, using the formula: Estimate the transition probability P between differentiated health states or , construct the health transition probability matrix; Among them, S or represents the standardized risk score of transitioning from health state o to health state r, α o , β u and β r is the adjustment coefficient, exp is the natural exponential function, S ou Represents the normalized risk score for transitioning from health state o to health state u.

7. The diabetes management system based on big data according to claim 6, characterized in that: The steps for obtaining the health status prediction result are specifically as follows: Using the health transition probability matrix, by analyzing the data adaptability of the health transition probability matrix, the applicability and consistency of the prediction of the health status of the diabetic patient in the future time period are evaluated to obtain a adaptability score record; According to the fitness score record, combined with the real-time health data of diabetic patients, the formula is adopted: Calculate the probability prediction value of the health status and obtain the health status prediction result; Among them, P(t+1) is the predicted health status probability at the next time point, P(t) is the health status probability at the real time point, and x j (t) is the jth index at time point t, β j is the regression coefficient, B is the smoothing parameter, and m is the number of indicators.

8. The diabetes management system based on big data according to claim 7, characterized in that: The steps for obtaining the health management plan for diabetic patients are specifically as follows: Utilizing the health status prediction results, analyzing the risk level and potential health problems of diabetic patients, identifying key monitoring indicators and necessary intervention time periods, and generating a health monitoring protocol for diabetic patients; Implement the diabetic patient health monitoring protocol, adjust the monitoring frequency, and use the formula: R adj =R base ·(1+α·P risk ); Evaluate the effect and impact of the implementation of the diabetic patient health monitoring plan and obtain the diabetic patient health management plan; Among them, R adj represents the adjusted monitoring frequency, R base is the basic monitoring frequency, α is the adjustment coefficient, P risk is the risk score for patients with diabetes.

Images