Personalized blood glucose management device and metabolic health assessment system

Abstract

The invention provides a personalized blood glucose management device and a metabolism health assessment system. The personalized blood glucose management device comprises a blood glucose monitoring module, a data transmission module, an intelligent analysis module and a feedback module. According to the device, real-time blood glucose monitoring is achieved through a non-invasive or minimally invasive technology, and a blood glucose fluctuation prediction model and a metabolic health assessment model are built in combination with multi-dimensional data. The system uses a machine learning algorithm to generate personalized health guidance schemes, including diet suggestions, exercise plans and medication adjustment, and pushes prompts through a mobile phone terminal. The device also supports dynamic evaluation of the metabolic health condition of the user, provides a scientific health report through a metabolic load index and a health scoring model, and helps the user to early warn abnormal blood glucose fluctuation and metabolic disease risks. According to the method, the defects of single monitoring mode, data isolation and lack of personalized guidance in the traditional technology are overcome, and efficient and comfortable blood glucose monitoring and comprehensive metabolism health management are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of blood glucose monitoring, and particularly to a personalized blood glucose management device and a metabolic health assessment system. Background Art

[0002] With the development of the global economy and the change of lifestyle, the incidence of diabetes and metabolic diseases has been continuously rising, becoming a major public health problem worldwide. According to the statistics of the World Health Organization (WHO), as of 2023, there are more than 500 million people suffering from diabetes globally, and more than 1 billion people are in the pre-diabetes stage or suffering from other metabolic syndromes (such as hypertension, obesity, hyperlipidemia, etc.). These chronic diseases not only significantly reduce the quality of life of patients, but also bring a huge social and economic burden. Therefore, how to effectively manage blood glucose levels and metabolic health, and prevent or delay the occurrence and development of diseases has become an important topic in medical research.

[0003] The common blood glucose monitoring and health management devices on the current market mainly include the following categories:

[0004] Fingertip blood sampling detection devices: Users extract a small amount of blood through a blood sampling device to measure the instantaneous blood glucose level. The characteristics of this method are mature technology and simple operation, but there are the following problems:

[0005] The measurement method is highly invasive. Users need to frequently prick their skin to draw blood, and the experience is poor;

[0006] The data is not continuous and can only reflect the blood glucose level at the time of a single sampling, making it difficult to comprehensively understand the blood glucose change trend;

[0007] Users rely on subjective judgment to adjust their lifestyles and lack scientific guidance.

[0008] CGM devices can continuously monitor the blood glucose level of users and provide real-time data by subcutaneously implanting sensors. Such devices solve the problems of frequent blood sampling and discontinuous data, but there are still limitations:

[0009] The device is expensive, the wearing period is limited (generally 7 - 14 days), and the replacement cost is high;

[0010] The implanted sensor may cause skin irritation or infection;

[0011] The data analysis ability is limited, usually only providing a blood glucose curve without personalized suggestions.

[0012] Health management mobile applications help users analyze their health status by recording users' diet, exercise, and blood glucose levels. Although these applications can provide certain health guidance, there are the following deficiencies: relying on users to manually input data, which is likely to lead to inaccurate or incomplete information;

[0013] It is impossible to monitor blood glucose changes in real time, and it is also difficult to generate a scientific metabolic health assessment;

[0014] Lack of integration with hardware devices, unable to achieve closed-loop management.

[0015] Main deficiencies of the existing technology

[0016] Although there are currently various blood glucose monitoring and health management technologies, in practical applications, there are still the following significant deficiencies:

[0017] The fingertip blood sampling method is highly invasive, with a poor user experience and is difficult to adhere to in the long term;

[0018] Most continuous blood glucose monitoring devices are implantable, causing discomfort during wearing, and some users may experience adverse reactions such as allergies or infections.

[0019] Existing devices often only focus on blood glucose indicators and ignore other metabolic health-related factors such as diet, exercise, and sleep;

[0020] Data is isolated and lacks integration, making it difficult to reflect the overall metabolic health status of users;

[0021] It is impossible to combine real-time monitoring data with the individual characteristics of users (such as age, gender, lifestyle habits, genetic characteristics, etc.) for comprehensive evaluation.

[0022] Current devices or applications only provide a single data display function, unable to dynamically predict blood glucose fluctuations; unable to generate personalized health management plans based on users' real-time data, such as adjusting diet, exercise, or medication strategies according to the predicted blood glucose trend; data analysis is mostly based on static rules and is difficult to adapt to the dynamic changes in users' health conditions.

[0023] Traditional blood glucose monitoring devices or management systems mostly focus on short-term data and ignore the long-term change trends of metabolic health;

[0024] Metabolic diseases often involve dynamic changes in multiple indicators (such as blood glucose, blood lipids, body fat, insulin sensitivity, etc.), and existing technologies are difficult to track and jointly analyze these indicators in the long term.

[0025] Security issues during data transmission and storage may lead to the leakage of users' privacy;

[0026] Some devices or applications lack a clear data security guarantee mechanism, and the opacity of data use by users affects their willingness to use.

[0027] Therefore, we urgently need to design a personalized blood glucose management device and a metabolic health assessment system to solve the above problems. Summary of the Invention

[0028] On the one hand, the present invention provides a personalized blood glucose management device, comprising:

[0029] A blood glucose monitoring module: for real-time monitoring of the user's blood glucose level;

[0030] A data transmission module: for sending blood glucose data to an external device or a cloud platform;

[0031] An intelligent analysis module: for generating a personalized blood glucose fluctuation prediction model based on the user's blood glucose data and historical data, and the model is based on the formula:

[0032]

[0033] Wherein, is the predicted blood glucose level, G 0 is the reference blood glucose value, α i is the weight of each factor, F i (t) is the influence factor function, and n is the number of influence factors.

[0034] As a preferred technical solution of the present invention, the influence factor function F i (t) includes the user's dietary intake D(t), exercise consumption E(t), and the user's insulin sensitivity I(t), and its specific form is:

[0035] F i (t) = β 1 ·D(t) - β 2 ·E(t) + β 3 ·I(t)

[0036] Wherein, β 1 , β 2 , β 3 are the influence factor adjustment coefficients.

[0037] As a preferred technical solution of the present invention, the dietary intake D(t) is calculated by the following formula:

[0038]

[0039] Wherein, C j is the intake of the jth type of food, N j is the glycemic index of this type of food, and m is the number of food types.

[0040] As a preferred technical solution of the present invention, the exercise consumption E(t) is based on the user's exercise time and intensity, and the calculation formula is:

[0041]

[0042] Among them, A(v) is the exercise intensity function, T(v) is the exercise time function, k is the user's personalized metabolic efficiency coefficient, and [[t 1 , t 2 is the exercise time interval.

[0043] As a preferred technical solution of the present invention, the user's insulin sensitivity I(t) is represented by the following recursive model:

[0044] I(t + 1) = γ·I(t) + δ·ΔG(t)

[0045] Among them, ΔG(t) is the blood glucose change amount, and γ, δ are weight parameters used to describe the dynamic change of insulin sensitivity.

[0046] The present invention also provides a metabolic health assessment system, including:

[0047] Data collection module: used to collect the user's blood glucose data, diet data, exercise data and other metabolism-related data;

[0048] Comprehensive evaluation module: used to calculate the metabolic health score based on multi-dimensional data, and the scoring formula is:

[0049]

[0050] Among them, S is the metabolic health score, is the predicted blood glucose value, G(t) is the actual blood glucose value, G max is the upper limit of the user's blood glucose reference, and [[t 0 , t n is the evaluation time interval.

[0051] As a preferred technical solution of the present invention, the comprehensive evaluation module combines the user's metabolic load index M(t), and the formula is:

[0052]

[0053] Among them, M(t) reflects the cumulative impact of blood glucose deviation from the reference value and is used to evaluate the user's metabolic health risk.

[0054] As a preferred technical solution of the present invention, the comprehensive evaluation module comprehensively analyzes the metabolic load index M(t) and the health score S to generate a health risk prediction model, and the model formula is:

[0055]

[0056] Among them, R is the health risk index, ω 1 , ω 2 are weighting coefficients.

[0057] As a preferred technical solution of the present invention, the health risk prediction model combines machine learning algorithms to optimize the user's individual characteristics, and the algorithm is based on the following objective function:

[0058]

[0059] Wherein, is the actual health risk index, and p is the number of samples.

[0060] Beneficial effects

[0061] The present invention adopts non-invasive or minimally invasive blood glucose monitoring technology, significantly improving the monitoring comfort of users and avoiding the pain and inconvenience of traditional blood sampling methods. The device supports continuous real-time monitoring, can accurately record the trend of blood glucose changes, and provides comprehensive and dynamic blood glucose data support for users.

[0062] The present invention combines an intelligent analysis module with machine learning algorithms to integrate the user's blood glucose data, diet records, exercise habits, and other metabolic health data to generate a personalized health guidance plan. The blood glucose fluctuation prediction model can give early warnings of abnormal fluctuations and dynamically adjust diet, exercise, and medication recommendations, thereby optimizing the blood glucose management effect and reducing the risk of metabolic diseases.

[0063] The present invention integrates multi-dimensional health data, constructs a metabolic health score and a metabolic load index model, and provides a scientific and comprehensive health assessment for users. The system dynamically optimizes the management plan by long-term tracking of the user's health status, realizes personalized and sustainable metabolic health management, and at the same time ensures user privacy through a strict data security protection mechanism, enhancing user trust. Brief description of the drawings

[0064] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0065] Figure 1 It is a framework diagram of a personalized blood glucose management device. Detailed implementation manners

[0066] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0067] The following combines multiple embodiments to describe in detail the specific implementation manners of the present invention.

[0068] Embodiment 1: Real-time monitoring and personalized feedback;

[0069] A diabetic patient wears the personalized blood glucose management device of the present invention, and the blood glucose level G(t) is collected in real time through the blood glucose monitoring module. The device calculates the next moment's blood glucose level using the built-in prediction model according to factors such as dietary intake, exercise volume, and drug use records.

[0070] The prediction result shows that the user's blood glucose level may reach 9.5 mmol / L (exceeding the normal range) 1 hour after a meal.

[0071] The system pushes the following suggestions to the user's mobile terminal through the feedback module: Dietary adjustment suggestion: Reduce carbohydrate intake and prefer foods with a low glycemic index.

[0072] Exercise plan suggestion: Perform 15 minutes of light exercise (such as walking) 20 minutes after a meal to promote blood glucose consumption.

[0073] Medication suggestion: Prompt the user to inject insulin in a timely manner according to the doctor's advice and adjust the dose to 6 units. The user makes adjustments according to the suggestions, and subsequent blood glucose data collection shows that the blood glucose level drops to 7.5 mmol / L 2 hours later, entering the healthy range.

[0074] Embodiment 2: Automatic identification and data isolation in a multi-user family;

[0075] A family of three (the father is a diabetic patient, the mother is a healthy person, and the son is a pre-diabetic population) jointly uses the personalized blood glucose management device of the present invention. The device records information such as the blood glucose data, eating habits, and exercise records of each user through the built-in user identity recognition module.

[0076] After the father gets up in the morning and wears the device, after the device confirms the identity through fingerprint recognition, it automatically loads his historical data and performs real-time monitoring and prediction. The system recognizes that the father's blood glucose fluctuates greatly at night (from 22:00 to 6:00, the lowest blood glucose value is 4.2 mmol / L), and prompts that there may be a risk of hypoglycemia, and suggests increasing the staple food intake appropriately after dinner and monitoring the blood glucose before going to bed.

[0077] When the mother uses the device, the system recognizes her health status and switches to the normal monitoring mode, only recording the blood glucose change trend without performing personalized prediction and feedback.

[0078] When my son wore the device in the afternoon, it reminded him that his current blood sugar level was 6.8mmol / L and recommended that he continue to maintain a healthy diet and 30 minutes of moderate exercise every day.

[0079] Example 3: Prediction of blood sugar fluctuations assisted by dietary intake data;

[0080] A user recorded his or her breakfast dietary data for the day through the device's mobile phone application: consuming 100 grams of white rice (glycemic index 75) and 30 grams of eggs (glycemic index 0).

[0081] The device calculates the glycemic impact D(t) of its total dietary intake using the following formula:

[0082] D(t)=C 1 ·N 1 +C 2 ·N 2

[0083] Where: C 1 =100 g, N 1 =75; C 2 =30 g, N 2 =0.

[0084] The calculation results are:

[0085] D(t)=100·75+30·0=7500

[0086] Based on the user's historical data, the device predicts that his blood sugar level will rise to 8.3mmollL one hour after the meal.

[0087] The system recommends that users: Give priority to whole grains with a low glycemic index (such as oats, quinoa, etc.) for lunch; avoid consuming high-sugar beverages.

[0088] Example 4: Dynamic regulation of blood sugar by exercise consumption;

[0089] A user starts jogging for 45 minutes at 9:00 am. The device synchronizes the exercise data with the smart bracelet through the data transmission module to record the exercise intensity and duration. According to the exercise consumption formula:

[0090]

[0091] Assumptions: k = 0.85, representing the user's metabolic efficiency coefficient; A(v) = 6 MET (exercise metabolic equivalent, jogging intensity); T(v) = 45 minutes.

[0092] The results of exercise consumption calculation are:

[0093] E(t)=0.85·6·45=229.5

[0094] The device uses this data to correct the user's blood glucose fluctuation prediction model and concludes that the user's blood glucose level will drop from 7.5 mmol / L to 6.2 mmol / L after exercise and then enter a stable state. The user adds a serving of fruit as an energy supplement according to the system's recommendation to avoid hypoglycemia.

[0095] Example 5: Metabolic health assessment and long-term health management;

[0096] A pre-diabetic user conducts 3-month metabolic health management through the device and the supporting application. The device monitors their blood glucose level daily and generates a health score S through the metabolic health assessment module:

[0097]

[0098] After 3 months, the health report generated by the system shows that the user's metabolic health score has increased from 62 at the beginning to 85;

[0099] The blood glucose fluctuation range has narrowed, from 5.8 - 9.2 mmol / L to 5.3 - 8.0 mmol / L; the diet and exercise records show that the intake of high glycemic index foods has decreased by 30%, and the average daily exercise time has increased by 20 minutes.

[0100] Based on the assessment results, the system recommends that the user further increase the frequency of low-intensity aerobic exercise and perform strength training at least once a week to further improve insulin sensitivity.

[0101] Example 6: Correlation analysis between sleep and blood glucose fluctuations;

[0102] A user records the blood glucose changes and sleep duration data for 7 consecutive days through the device. The intelligent analysis module finds that the amplitude of their blood glucose fluctuations is positively correlated with sleep deprivation (less than 6 hours).

[0103] Specifically, during the nights of sleep deprivation, the lowest blood glucose point increases and the morning blood glucose is on the high side. The system analyzes the following trend relationship:

[0104] G min = G 0 + α·(T sleep - 6)

[0105] Where:

[0106] G min is the lowest blood glucose value at night;

[0107] G 0 is the reference value;

[0108] T sleep is the sleep duration; α is the correlation coefficient (the user's personalized characteristic).

[0109] The system recommends that users ensure more than 7 hours of sleep per day and monitor their blood glucose before going to bed to prevent nocturnal hypoglycemia.

[0110] After 3 weeks, the user's morning blood glucose level decreased from 7.1 mmol / L to 6.5 mmol / L.

[0111] Through the above embodiments, the present invention demonstrates its various application scenarios in real-time monitoring, personalized analysis, health management, and long-term intervention, and has good practicality and promotion value.

[0112] Example 7: Comprehensive evaluation of the metabolic load index and health score;

[0113] A user records the continuous 7-day blood glucose fluctuation data through the device every day. The metabolic health assessment module uses a formula to calculate the user's metabolic load index M(t):

[0114]

[0115] Where: G(t) is the real-time blood glucose level; G 0 is the user's personalized baseline blood glucose value; [t 0 , t n is the evaluation time interval (such as one day or one week).

[0116] The calculation results of the user's metabolic load index from the 1st day to the 7th day are as follows: M(1)=15.6 on the 1st day, M(2)=12.3 on the 2nd day... M(7)=8.5 on the 7th day. Through the analysis of the change trend of M(t), the system finds that the user's metabolic load has decreased significantly, indicating that the blood glucose management effect has been gradually improved.

[0117] Combined with the metabolic health score formula:

[0118]

[0119] The system calculates that the user's metabolic health score has increased from 65 on the 1st day to 82 on the 7th day.

[0120] Results and suggestions: The system gives the following feedback in the health report:

[0121] The user's blood glucose control ability has been significantly enhanced. It is recommended to continue to maintain the current diet and exercise habits.

[0122] Further optimize the sleep duration (increase to 7 - 8 hours) to improve the score.

[0123] Example 8: Application of the health risk prediction model;

[0124] A user records their metabolic health score S and metabolic load index M(t) through a device, and uses a health risk prediction model to evaluate long-term health risks:

[0125]

[0126] Where: R is the health risk index;

[0127] ω 1 = 0.6, ω 2 = 0.4 are the weighted parameters of the model.

[0128] Suppose a user's metabolic health score is 75 and the metabolic load index M(t) is 10.5. The calculation shows:

[0129]

[0130] The system evaluates the health risk according to the magnitude of R:

[0131] If R < 3, it indicates a good metabolic health status;

[0132] If R ∈ [3, 5), it indicates a moderate risk;

[0133] If R ≥ 5, it indicates a high risk and immediate intervention is required.

[0134] Results and suggestions: The user is currently at moderate risk. The system recommends that the user increase moderate-intensity exercise by 30 minutes per day and reduce the proportion of high-glycemic-index foods in dinner.

[0135] According to the model update, if the metabolic load index drops to 8, the calculation result R = 3.2 of the health risk prediction model, and the user's health status is significantly improved.

[0136] Example 9: Optimize the dynamic adjustment of weight parameters;

[0137] The weighted parameters ω 1 , ω 2 of the health risk prediction model are dynamically adjusted according to the user's historical health data and behavior habits. The optimization process is based on the following objective function:

[0138]

[0139] Where: R i is the predicted health risk index; is the actual health risk index;

[0140] p is the number of samples.

[0141] The specific optimization steps are as follows:

[0142] Data collection: Obtain historical risk index records of different users from the cloud platform of the system to form a data set

[0143] Optimization iteration: Use the gradient descent algorithm to perform iterative updates on ω 1 , ω 2 until the objective function converges.

[0144] Update the model: Apply the optimized parameters to the user's health risk prediction model.

[0145] Example 10: Long-term monitoring combined with an insulin sensitivity recursive model;

[0146] A user has been using a device for long-term blood glucose management, and the system dynamically evaluates their insulin sensitivity I(t) through a recursive model:

[0147] I(t + 1) = γ·I(t) + δ·ΔG(t)

[0148] where: I(t) is the current insulin sensitivity; γ = 0.8, δ = 0.2 are model parameters; ΔG(t) = G(t + 1) - G(t) is the blood glucose change.

[0149] Assume that the user's initial insulin sensitivity on the first day is I(0) = 1.5, and their daily blood glucose changes are as follows: on the first day, ΔG(1) = -0.2; on the second day, ΔG(2) = -0.1; on the third day, ΔG(3) = +0.3;

[0150] Calculating recursively:

[0151] I(1) = 0.8·1.5 + 0.2·(-0.2) = 1.16

[0152] I(2) = 0.8·1.16 + 0.2·(-0.1) = 0.908

[0153] I(3) = 0.8·0.908 + 0.2·0.3 = 0.926

[0154] Based on the dynamic change trend of insulin sensitivity, the system advises the user to: reduce the intake of high-fat foods to improve insulin sensitivity; increase daily strength training to promote the glucose metabolism ability of muscle tissue.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A personalized blood sugar management device, characterized in that: include: Blood glucose monitoring module: used to monitor the user's blood glucose level in real time; Data transmission module: used to send blood glucose data to external devices or cloud platforms; Intelligent analysis module: used to generate a personalized blood sugar fluctuation prediction model based on the user's blood sugar data and historical data. The model is based on the formula: in, is the predicted blood glucose level, G0 is the baseline blood glucose value, α i is the weight of each factor, F i (t) is the impact factor function, and n is the number of impact factors.

2. The personalized blood sugar management device according to claim 1, characterized in that: The impact factor function F i (t) includes the user's dietary intake D(t), exercise consumption E(t), and the user's insulin sensitivity I(t), and its specific form is: F i (t)=β1·D(t)-β2·E(t)+β3·I(t) Among them, β1, β2, and β3 are the impact factor adjustment coefficients.

3. The personalized blood sugar management device according to claim 2, characterized in that: The dietary intake D(t) was calculated by the following formula: Among them, C j is the intake of food of type j, N j is the glycemic index of this type of food, and m is the number of food types.

4. The personalized blood sugar management device according to claim 2, characterized in that: The exercise consumption E(t) is based on the user's exercise time and intensity, and the calculation formula is: Among them, A(v) is the exercise intensity function, T(v) is the exercise time function, k is the user's personalized metabolic efficiency coefficient, and [t1, t2] is the exercise time interval.

5. The personalized blood sugar management device according to claim 2, characterized in that: The user's insulin sensitivity I(t) is represented by the following recursive model: I(t+1)=γ·I(t)+δ·ΔG(t) Among them, ΔG(t) is the change in blood glucose, and γ and δ are weight parameters used to describe the dynamic changes of insulin sensitivity.

6. A metabolic health assessment system, based on a personalized blood glucose management device according to any one of claims 1 to 5, characterized in that: include: Data collection module: used to collect user blood sugar data, diet data, exercise data and other metabolism-related data; Comprehensive evaluation module: used to calculate metabolic health scores based on multidimensional data. The scoring formula is: Where S is the metabolic health score, is the predicted blood sugar value, G(t) is the actual blood sugar value, G max is the user's blood sugar reference upper limit, [t0,t n ] is the evaluation time interval.

7. The metabolic health assessment system according to claim 6, characterized in that: The comprehensive evaluation module combines the user's metabolic load index M(t) with the formula: Among them, M(t) reflects the cumulative impact of blood sugar deviation from the baseline value and is used to assess the user's metabolic health risk.

8. The metabolic health assessment system according to claim 7, characterized in that: The comprehensive evaluation module conducts a comprehensive analysis of the metabolic load index M(t) and the health score S to generate a health risk prediction model. The model formula is: Among them, R is the health risk index, ω1, ω2 are weighting coefficients.

9. The metabolic health assessment system according to claim 8, characterized in that: The health risk prediction model combines machine learning algorithms to optimize user individual characteristics. The algorithm is based on the following objective function: in, is the actual health risk index, and p is the sample size.

Images