A blood glucose management system for pregnancy

Abstract

The present application relates to a blood sugar management system for pregnancy, comprising terminal devices carried by target users and reference users and a cloud server in data connection with the terminal devices, the cloud server being configured to: acquire physiological data and blood sugar data of the target users from the terminal devices, and calculate characteristic indexes of the target users based on the physiological data and the blood sugar data; acquire a plurality of blood sugar data of the reference users from the terminal devices and form an overall reference data set in a manner associated with gestational weeks, and assign the target users to a predetermined characteristic index value interval in the overall reference data set based on the characteristic indexes, so as to provide personalized health management suggestions for the target users based on factors identified to affect the risk of gestational diabetes mellitus of the target users.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical management, and particularly relates to a blood glucose management system for a gestation period. BACKGROUND

[0002] Gestational diabetes mellitus (GDM) refers to the abnormal condition of glucose metabolism that occurs or is diagnosed for the first time during pregnancy, and is one of the most common complications in the perinatal period. According to the statistical data of the International Diabetes Federation (IDF) in 2019, the incidence of GDM worldwide is about 13.2%, and shows a rising trend year by year. As an important public health challenge, GDM poses a significant threat to maternal and infant health, not only increasing the risk of perinatal complications (such as gestational hypertension, preeclampsia, etc.) and adverse pregnancy outcomes (such as macrosomia, increased cesarean section rate, premature birth, etc.), but also associated with an increased likelihood of developing type 2 diabetes mellitus (T2DM) and cardiovascular diseases in the future.

[0003] The mechanism of gestational diabetes mellitus is closely related to the changes in hormone levels in pregnant women. As pregnancy progresses to the middle and late stages, various hormones and cytokines secreted by the placenta (such as tumor necrosis factor alpha, TNF-α) have an antagonistic effect on insulin, leading to gradually increasing insulin resistance. When the pregnant woman's own pancreatic beta cells cannot fully respond to this physiological insulin resistance, or accompanied by beta cell dysfunction, GDM may occur.

[0004] Effective blood glucose management is crucial in mitigating the risks associated with gestational diabetes mellitus (GDM). Traditional interventions emphasize controlling blood glucose levels through diet plans customized by professional medical personnel, which usually include balanced nutrient intake, with special attention to the quality and quantity of carbohydrates.

[0005] In developing a diet plan, a key consideration factor is the impact of carbohydrates in food on blood glucose levels, which can be measured by the Glycemic Index (GI value). GI value is an index used to describe the speed and extent of blood glucose rise caused by carbohydrates in a specific food. It is based on standard tests conducted on a healthy population of non-diabetic adults, requiring subjects to have normal fasting blood glucose levels and no other conditions affecting carbohydrate metabolism. GI value helps identify which foods cause rapid blood glucose rise (high GI) and which foods provide more stable and sustained energy release (low GI).

[0006] However, existing blood glucose management solutions generally lack the ability to adjust and dynamically monitor individual needs, making it difficult to adapt to the specific needs of different pregnant women. In addition, traditional dietary recommendations mainly rely on static GI values, failing to fully consider individual differences and the impact of real-time blood glucose fluctuations.

[0007] CN114743673A discloses a health management method and device for pregnant women with gestational diabetes mellitus (GDM) and an electronic device. The scheme provides personalized dietary recommendations and exercise guidance for pregnant women by interacting with target users, and provides professional dietary and exercise counseling to promote active health management of target users, thereby reducing the probability of complications in GDM patients.

[0008] However, the above technical solution mainly relies on pre-set static data information to develop dietary and exercise programs, and fails to fully consider the impact of individual differences on health management. This fixed parameter-based method is limited in applicability and effectiveness when faced with users of different physical conditions, living habits and health conditions. In addition, since there is no effective feedback mechanism, the scheme cannot dynamically adjust the diet and exercise plan according to the real-time monitoring of user blood glucose data. This results in the inability to update health management measures in a timely manner even if the user's blood glucose level changes, thereby affecting the overall effect of blood glucose management.

[0009] In summary, there is an urgent need to develop a more intelligent and personalized blood glucose management system that can combine real-time blood glucose monitoring data, personal health records and other related factors to provide accurate and flexible treatment recommendations for GDM patients, thereby more effectively controlling blood glucose levels and reducing the risk of maternal and infant complications.

[0010] In addition, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant has studied a large number of literatures and patents when making the invention, but due to space limitations, all details and contents are not listed in detail. However, this does not mean that the present invention does not have these prior art characteristics. On the contrary, the present invention already has all the characteristics of the prior art, and the applicant reserves the right to add relevant prior art in the background art. SUMMARY

[0011] In view of the deficiencies of the prior art, the present invention aims to provide a blood glucose management system for pregnant women, which forms overall data by a large number of users using CGM devices (continuous glucose monitoring devices), and provides personalized dietary recommendations according to the position of user blood glucose data in the overall data, while dynamically adjusting the GI values in the dietary recommendations according to the changes in CGM readings, to achieve more effective blood glucose management.

[0012] The application relates to a blood glucose management system for the gestation period, comprising terminal equipment carried by a target user and reference users and a cloud server in data connection with the terminal equipment, the cloud server being configured to: acquire physiological data and blood glucose data of the target user from the terminal equipment, and calculate a feature index of the target user according to the physiological data and the blood glucose data; acquire a plurality of blood glucose data of the reference users from a plurality of terminal equipment, and form an overall reference data set in a manner associated with gestational weeks, assign the target user to a predetermined feature index value interval in the overall reference data set based on the feature index, and determine factors affecting the risk of gestational diabetes mellitus of the target user, so that the target user manages corresponding physiological parameters according to the factors.

[0013] The personalized distribution mechanism of the system makes the health management suggestions more suitable for the actual situation of the target user, and improves the effectiveness of health management. The system can timely discover potential health risks by identifying factors affecting the risk of gestational diabetes mellitus of the target user. The risk identification mechanism provides a basis for clinical intervention, helps to reduce the incidence of gestational diabetes mellitus, and improves the health level of pregnant women and fetuses. In addition, by integrating the data of the reference users into the overall reference data set, the system can take advantage of big data analysis to improve the accuracy and reliability of analysis. The integration of group data provides a more extensive reference basis for the calculation of the feature index and the generation of health suggestions.

[0014] In the blood glucose management system for the gestation period, the physiological parameters refer to a basic index set reflecting the individual health status of the pregnant woman and the dynamic changes of the gestation period, including BMI (body mass index), gestational weeks, body weight growth rate, hormone levels (such as progesterone, hCG, estradiol) and continuous blood glucose monitoring data. These parameters are collected in real time by the terminal equipment or actively input by the user, and constitute the original data basis for the system to evaluate the risk of gestational diabetes mellitus (GDM). For example, high BMI may indicate obesity-related insulin resistance risk, and abnormal hormone levels may be directly related to the influence of placental function on glucose metabolism. The system integrates multi-dimensional physiological parameters into a standardized score value by calculating a feature index (FI), eliminates dimensional differences and realizes cross-individual comparison, and provides a unified framework for subsequent analysis.

[0015] The "key parameters" are the core factors selected from the physiological parameters that have the most significant impact on the GDM risk of a specific user. The system classifies the target user into a homogeneous subgroup with reference users who have similar gestational age, BMI, and weight gain rate through a dynamic grouping strategy, and then compares the deviations of their physiological parameters from the average values within the subgroup. For example, if a user's BMI deviates from the average value of the subgroup by 2 standard deviations and the weight gain rate exceeds the medical recommended threshold, both are identified as key parameters; if the estrogen level is abnormally elevated by 40%, the hormone imbalance may further exacerbate insulin resistance, thus forming a compound risk characteristic. This precise identification mechanism avoids the drawbacks of traditional schemes that intervene all parameters comprehensively, and instead focuses on specific indicators that need to be regulated.

[0016] The technical solution of the present application aims to solve the pain point of "blind intervention leading to metabolic disorders" in traditional blood glucose management. The system guides the user to adjust behavior only for key parameters through data-driven dynamic analysis, rather than general dietary or exercise restrictions. For example, for those with high BMI, low GI foods such as oatmeal instead of white bread are recommended to adjust the dietary structure, rather than simply reducing food intake, thus avoiding the rise of cortisol and blood sugar rebound caused by insufficient energy intake; for those with abnormal weight gain rate, a daily 300kcal calorie gap is designed (achieved through dietary record and low-intensity walking), avoiding stress reactions caused by intense exercise. At the same time, the system dynamically adjusts the intervention intensity by continuously monitoring the changes in key parameters: if the user's weight gain rate meets the standard for two consecutive weeks, the GI limit is automatically relaxed to below 65; if it does not improve, a medical collaboration process is triggered to ensure that the risk is controllable. This strategy optimizes blood glucose control while maintaining hormone homeostasis and metabolic balance.

[0017] According to a preferred embodiment, the cloud server further subdivides the reference users of the feature index interval into multiple gestational age groups according to gestational age, and further analyzes the deviations between the physiological parameters of the target user and the average values of the reference users within the same gestational age group, to determine one or more key parameters that affect the blood glucose level of the target user among the physiological parameters, wherein the key parameters include BMI, weight gain rate, and hormone level. This grouping method takes into account the physiological changes at different stages of pregnancy, making the analysis more relevant and accurate, and helping to identify specific pregnancy risks. By analyzing the deviations between the physiological parameters of the target user and the average values of the reference users within the same gestational age group, the system can identify the abnormal values of the target user. This method effectively helps to identify key parameters that affect blood glucose levels, ensuring the relevance of the analysis. Once the key parameters that affect blood glucose levels are identified, the cloud server can provide personalized health management recommendations for the target user based on these factors. For example, for users with high BMI, the system can recommend adjusting diet and increasing exercise, and for users with abnormal weight gain rate, appropriate intervention measures can be developed. This personalized recommendation helps to improve the effectiveness of user health management.

[0018] Based on the feature index FI value, the cloud server can categorize users with similar FI values into the same group. This way of integrating multiple factors affecting the risk of gestational diabetes into a comprehensive index allows the cloud server to standardize the processing of these multi-dimensional data. This means that even if the target user and each reference user have different performances in each individual factor, they can find common points or similarities between each other through the FI value, thereby achieving effective comparison across individuals. This grouping method ensures high homogeneity among group members, i.e., they show similar trends or characteristics in multiple key health indicators. Therefore, it is more meaningful to compare and analyze blood glucose data within such a group, as they are more likely to reflect the real impact of specific factors on blood glucose control rather than accidental fluctuations.

[0019] According to a preferred embodiment, the cloud server can calculate the percentile of the target user's relative position based on the daily trend blood glucose data of the target user in the gestational age group to which the target user belongs, and compare and analyze the blood glucose data with the reference users to identify the target user's blood glucose fluctuation pattern and potential abnormalities. By calculating the percentile of the target user's blood glucose data in the gestational age group, the system can provide a relative health assessment indicator. This assessment can help users understand their blood glucose level in the same gestational age group, and thus better understand their health status.

[0020] According to a preferred embodiment, the terminal device includes a collection unit for acquiring continuous blood glucose data, and the terminal device uploads the blood glucose data recorded by the collection unit to the cloud server in a manner associated with gestational age through the configured communication unit, and receives personalized dietary recommendations from the cloud server. By using a continuous blood glucose data collection unit, continuous monitoring of blood glucose levels can be achieved, providing more comprehensive and accurate information on blood glucose trends compared to traditional point measurement methods. By uploading blood glucose data associated with gestational age, the system can more accurately analyze the characteristics of blood glucose changes during different gestational periods, thereby providing personalized recommendations that better meet the physiological characteristics of the gestational period. In addition, the design of the terminal device allows users to monitor blood glucose and upload data anytime, anywhere, improving the convenience and compliance of use.

[0021] According to a preferred embodiment, the terminal device is integrated with an interaction unit signal connected with the communication unit, the interaction unit is configured to receive and display dietary recommendations from the cloud server and the storage unit to the user, and provide a platform for the user to input and update personal health information including current gestational weeks, weight changes, and dietary records. The interaction unit can receive and display dietary recommendations from the cloud server and the storage unit, providing a real-time information exchange platform for the user. This instant feedback mechanism helps users adjust their dietary habits according to the latest recommendations. In addition, the interaction unit provides a platform for users to input and update personal health information, which encourages users to actively participate in the health management process. Users can input information such as current gestational weeks, weight changes, and dietary records, allowing the system to better understand their personal health status.

[0022] According to a preferred embodiment, the cloud server groups the overall reference dataset based on similar gestational weeks, wherein the cloud server adjusts the time window range of similar gestational weeks according to the stage of the user's pregnancy: for early and late pregnancy stages, a relatively narrower time window is used; for the middle stage of pregnancy, a relatively wider time window is used. By grouping the overall reference dataset based on similar gestational weeks, the cloud server can provide more accurate and relevant data analysis. This grouping method ensures that the data compared and analyzed has higher relevance and comparability. In addition, adjusting the time window range of similar gestational weeks according to different stages of pregnancy reflects the dynamic adaptability of the system. This flexible adjustment mechanism also better matches the physiological changes characteristics during different stages of pregnancy.

[0023] According to a preferred embodiment, the cloud server can comprehensively group the overall reference dataset according to the physiological data input by the user through the interaction unit of the terminal device, combining multi-dimensional grouping methods to improve the homogeneity of members within each group. Physiological data includes the user's own BMI and weight gain rate input or updated regularly. By comprehensively grouping the reference dataset according to the physiological data input by the user (such as BMI and weight gain rate), the homogeneity of members within each group can be significantly improved. This grouping method ensures the comparability and consistency of data within the group, especially improving the accuracy of the user's corrected characteristic index.

[0024] According to a preferred embodiment, the cloud server calculates basic statistical indicators for each group in the overall reference dataset, combines the specific percentile position of the user's blood glucose data in its group, judges the current blood glucose state of the user by comparing the user's blood glucose data with the preset healthy blood glucose range, and generates a graded diet suggestion accordingly, which optimizes the diet structure of the target user by selecting food categories with different GI values. By calculating the basic statistical indicators of each group, the cloud server can accurately locate the specific percentile of the user's blood glucose data in its group. This method makes the blood glucose evaluation more accurate and can identify the current blood glucose state of the user. Comparing the user's blood glucose data with the preset healthy blood glucose range helps to evaluate the user's health status individually. This allows customized health recommendations for users with different blood glucose states. In addition, by selecting food categories with different GI values (glycemic index), the system can help users optimize their diet structure. This method helps to smooth blood glucose fluctuations and improve overall health.

[0025] According to a preferred embodiment, the cloud server is configured to divide users into different groups according to gestational weeks, and in each gestational week group, set multiple BMI ranges so that users can be subdivided into multiple BMI subgroups. In the same BMI subgroup, the cloud server can further subdivide users according to weight gain speed to identify users with excessively fast or slow weight gain and assign them to a special attention group. The division of different groups and subgroups allows for personalized health guidance and recommendations. This guidance can be adjusted according to the specific gestational weeks and weight changes of the user, improving the effectiveness of health management.

[0026] According to a preferred embodiment, the cloud server can periodically reevaluate the gestational week information of the target user and adjust its group accordingly; the cloud server allows users to object to the automatically assigned group and receives the manually adjusted gestational week information from the user through the interaction unit. Allowing users to object to the automatically assigned group and manually adjusting the gestational week information helps the system correct possible errors or inaccuracies, improving the overall accuracy of the data; on the other hand, it enhances the user's sense of participation and control, improving the user's acceptance and compliance of health management recommendations. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 is a preferred hardware topology of a pregnancy blood glucose management system provided by the present application;

[0028] Figure 2 is a preferred flowchart of the cloud server obtaining a feature index representing the risk level of gestational diabetes mellitus blood glucose of a user provided by the present application;

[0029] Figure 3is a preferred cloud server process flow diagram provided by the present application for dividing gestational weeks;

[0030] Figure 4 is a preferred cloud server process flow diagram provided by the present application for grouping the overall reference dataset by gestational weeks, BMI and weight gain rate;

[0031] Figure 5 is a preferred hardware scenario diagram of the pregnancy blood glucose management system provided by the present application;

[0032] Figure 6a is a preferred example diagram of the characteristic index drawn by the cloud server for a target user changing with gestational weeks in the early pregnancy stage provided by the present application;

[0033] Figure 6b is a preferred example diagram of the characteristic index drawn by the cloud server for a target user changing with gestational weeks in the middle pregnancy stage provided by the present application;

[0034] Figure 6c is a preferred example diagram of the characteristic index drawn by the cloud server for a target user changing with gestational weeks in the late pregnancy stage provided by the present application;

[0035] Figure 7 is a preferred comparison diagram of the blood glucose data of a target user and the blood glucose data of other reference users in the same FI interval drawn by the cloud server provided by the present application;

[0036] Figure 8 is a preferred target user daily trend blood glucose comparison diagram provided by the present application.

[0037] List of reference signs

[0038] 100: terminal device; 110: acquisition unit; 120: communication unit; 130: storage unit; 140: interaction unit; 200: cloud server. DETAILED DESCRIPTION

[0039] The following will be described in detail with reference to the accompanying drawings.

[0040] In the present application, target users and reference users refer to two types of pregnant women participating in the pregnancy blood glucose management system. These two terms are used to distinguish individuals who play different roles in the same system but are related to each other.

[0041] Target users refer to pregnant women who are currently using the system and seeking personalized glycemic management and dietary recommendations. These women usually have higher blood glucose levels or have the need for glycemic management, and through the help of the system, they optimize their own health management, especially effective control of blood glucose levels. Their data are used to generate personalized health guidance programs.

[0042] Reference users refer to other pregnant women who are also in the same period of pregnancy and can share their own blood glucose data with the system. The blood glucose levels of these pregnant women are usually within the normal range, and the anonymized data they provide constitute a large reference database, providing a comparative benchmark and group analysis support for target users. The data of reference users help to establish typical blood glucose patterns at each stage of pregnancy.

[0043] The identities of target users and reference users are not fixed. If a reference user who originally contributes data starts to need personalized glycemic management services, she can activate the full functions of the intelligent monitoring terminal device 100 and become a target user who enjoys comprehensive services. This situation can occur when a reference user finds that her blood glucose level is rising. Conversely, if a target user has achieved effective glycemic control or only wants to continue to contribute data to help others, she can choose to keep the basic monitoring function of the device open and continue to provide reference data without receiving further personalized services.

[0044] Embodiment 1

[0045] The present application relates to an intelligent system for glycemic management during pregnancy, as shown in Figure 1 , Figure 4 The system includes a plurality of terminal devices 100 and a cloud server 200. The terminal device 100 can be a smart bracelet, a smart watch or other wearable device, which is worn by target users and reference users to collect and view blood glucose data in real time. At the same time, the terminal device 100 can visually display personalized dietary recommendations that match the blood glucose status of the target user. The cloud server 200, with its powerful computing power and storage capacity, can quickly process and deeply analyze the massive data collected by the terminal, and generate optimized dietary recommendations for the target user.

[0046] Preferably, as shown in Figure 1 , the system can support the access of multiple terminal devices 100 to cover more target users and reference users. Each terminal device 100 interacts with the cloud server 200 through data connection to realize data uploading and feedback. The terminal device 100 includes a blood glucose collection unit 110, which can take the form of a continuous glucose monitor (CGM) to dynamically monitor the change of glucose concentration in the interstitial fluid of the user's subcutaneous tissue by using a glucose sensor, thereby realizing continuous and dynamic blood glucose data collection.

[0047] Preferably, as shown in Figure 1 The terminal device 100 is also equipped with a communication unit 120 for uploading and receiving data. On the one hand, the communication unit 120 can upload the blood glucose data recorded by the acquisition unit 110 to the cloud server 200 in a manner associated with gestational age; on the other hand, it can also receive the personalized dietary recommendations generated by the cloud server 200 for the target user. The communication unit 120 can be composed of various wireless communication modules (such as Bluetooth Low Energy BLE, Wi-Fi, Zigbee, etc.) to ensure the security, reliability and real-time performance of data transmission.

[0048] Preferably, as shown in Figure 1 The terminal device 100 also includes a storage unit 130 equipped with large storage space, which can save the personalized dietary recommendations fed back by the cloud server 200 on the one hand, and can store the blood glucose change data of the target user since wearing the terminal device 100 on the other hand, to meet the needs of the user to view at any time.

[0049] Preferably, as shown in Figure 1 The terminal device 100 also integrates an interactive unit 140, which can exist in the form of a mobile application interface. The interactive unit 140 is connected with the communication unit 120 and can directly interact with the target user. It is not only responsible for providing personalized health recommendations, reminders and historical trend charts and other information to the user, but also supports the user to input and update key personal information such as current gestational age, weight change, dietary records, etc.

[0050] Preferably, the cloud server 200, with its stronger computing power and storage capacity than the terminal device 100, undertakes the core data processing task in the system. The cloud server 200 not only saves the blood glucose data and related information of all target users and reference users, but also deploys an analysis model for in-depth analysis of complex blood glucose data and generates scientific and precise personalized dietary recommendations, thereby effectively assisting the target user in managing blood glucose during pregnancy.

[0051] Preferably, as shown in Figure 1As shown, the cloud server 200 can be configured to obtain the four physiological data of the target user's gestational age, BMI, weight gain rate and hormone level from the terminal device 100 of the target user, and preliminarily calculate the feature index FI by substituting the four data into the built-in calculation model. The feature index FI is a comprehensive quantitative index calculated by the cloud server 200 according to the physiological parameters of the target user. It aims to unify pregnant women of different gestational ages, BMIs, weight gain rates and hormone levels into a standardized comparison framework, so that the target user can be assigned to a small group with similar or close FI values for research. By comparing the differences in blood glucose data of the target users in these groups and the reference users, the specific factors that have the greatest impact on the risk of gestational diabetes can be identified, and personalized health management recommendations can be provided to the target user.

[0052] Preferably, the cloud server 200 can calculate the feature index FI by the following formula:

[0053]

[0054] Since there are differences in the physiological conditions of each pregnant woman, directly comparing their blood glucose data may not accurately reflect individual health conditions. Therefore, the system calculates multiple key health indicators of pregnant women (such as body mass index, gestational age, weight gain rate and hormone level) through the feature index (FI), and adjusts the importance weight of different indicators, finally generating a unified score value. This process eliminates the unit difference and magnitude influence of different indicators, so that the complex health data of all pregnant women can be compared fairly under the same standard. For example, the system will assign pregnant women with excessive weight gain or abnormal hormone levels to a specific score interval, and match their blood glucose data with other pregnant women in the same interval for analysis. In this way, each pregnant woman can intuitively understand her health status in the similar population and clearly identify the health direction that needs to be focused on (such as controlling weight gain or monitoring hormone changes). This standardized scoring and grouping mechanism not only simplifies the complexity of data processing, but also helps users to manage health risks more targetedly.

[0055] wherein,

[0056] B represents the body mass index (BMI), which is obtained by dividing the weight (P) by the square of the height (H s ) of the pregnant woman:

[0057]

[0058] Body Mass Index (BMI) has a significant impact on the risk of developing gestational diabetes. Higher BMI is often associated with insulin resistance, where obesity leads to decreased sensitivity to insulin in the body, resulting in elevated blood glucose levels and an increased risk of gestational diabetes. Additionally, obesity can trigger chronic low-grade inflammation, which further affects insulin function and glucose metabolism. During pregnancy, the body undergoes various metabolic changes, and a high BMI can complicate this adaptation process, increasing the risk of abnormal glucose metabolism.

[0059] W represents the gestational age, which is the number of weeks of pregnancy calculated from the first day of the last menstrual cycle:

[0060] W = D c -LMP

[0061] D c D is the current date, and LMP is the last menstrual date. An increase in gestational age may be associated with an increased risk of gestational diabetes, especially in the later stages of pregnancy.

[0062] Current date D c This can be automatically obtained through the system time of the target user's terminal device 100, ensuring the real-time and accuracy of the time. The last menstrual date LMP can be input by the user independently, and specific format requirements (such as YYYY-MM-DD or other standard date formats) are provided for user input.

[0063] G represents the weight gain rate, indicating the value of weight gain per week (G) during pregnancy, which can be expressed in kilograms per week.

[0064]

[0065] T c represents the current weight (kg), and T pwherein W represents the current weight (kg), W represents the previous measured weight (kg), and Δt represents the time interval (weeks). Both the current weight and the previous measured weight can be uploaded by the user periodically (manually or automatically synchronized to the terminal device 100 through a Bluetooth scale). If the user uploads data at an irregular time, the timestamp information of the uploaded data can be combined to dynamically calculate Δt, ensuring the accuracy of the data. For users with low upload frequency, the cloud server 200 can set a reminder mechanism to prompt the user through the terminal device 100 to upload physiological data regularly to ensure the accuracy of the evaluation results. The weight gain rate has an important influence on the risk of gestational diabetes. Rapid weight gain can lead to increased insulin resistance, making the body face greater challenges in adjusting blood sugar levels. In addition, rapid weight gain during pregnancy increases the metabolic burden, especially in the later stages of pregnancy, which can make blood sugar control more difficult. At the same time, rapid weight gain is often associated with unhealthy diet and lack of exercise, which can further increase the risk of gestational diabetes.

[0066] H represents the hormone level, and a comprehensive hormone level index is used to reflect the combined effect of multiple hormones.

[0067] H = h1 · H1 + h2 · H2 + h3 · H3

[0068] wherein:

[0069] H1, H2, H3 represent the concentration values of different hormones (such as progesterone, human chorionic gonadotropin (hCG), estradiol); h1, h2, h3 represent the corresponding weight coefficients.

[0070] The measurement of hormone levels (H1, H2, H3) can be obtained by the laboratory of the medical institution when the user regularly checks the pregnancy, such as the detection of progesterone, human chorionic gonadotropin (hCG), estradiol, etc. The user can upload the detection report to the terminal device 100, and the terminal device 100 collects the data through OCR (image recognition) or user manual input. In this case, h1, h2 and h3 can be set based on the initial value of the weight of medical research. By consulting authoritative medical literature and clinical research, the strength and relevance of progesterone (H1), hCG (H2), estradiol (H3) on the risk of gestational diabetes mellitus (GDM) are analyzed. According to the effect size reported in the study (such as regression coefficient, correlation coefficient, etc.), an initial weight is assigned to each hormone. When the hormone detection value received by the communication unit 120 exceeds the pre-stored safety threshold value of the storage unit 130, the corresponding sub-weight is automatically increased to 1.5-2.0 times of the original value. In addition, the weight coefficient can also be modified based on statistical analysis. For example, a large number of real pregnant women's data can be collected, including progesterone (H1), hCG (H2), estradiol (H3) concentration values and the diagnosis results of gestational diabetes mellitus. Then, the independent contribution or correlation of each hormone to the risk of GDM is evaluated by means of Pearson correlation coefficient, multiple linear regression analysis or generalized linear model.

[0071] The change of hormone levels during pregnancy has an important influence on the risk of gestational diabetes mellitus. With the progress of pregnancy, the increase of hormone levels such as progesterone, human chorionic gonadotropin (hCG) and estradiol will gradually affect the sensitivity of insulin and glucose metabolism, leading to the increase of insulin resistance, and thus the risk of gestational diabetes mellitus shows differences in different gestational weeks.

[0072] w1, w2, w3, w4 are weight coefficients, each factor's relative importance weight in risk assessment. By setting the weight, the different influence degree of each factor on the risk of gestational diabetes can be reflected. The setting of the weight coefficient can be based on large-scale medical statistical data, and the weight value can be optimized by data analysis and other methods. For example, the initial value of the weight coefficient w1-w4 is set in association with the pre-stored clinical standard database in the storage unit 130 of the terminal device 100, which includes the "Pregnancy Health Management Guidelines" module regularly updated by cooperative medical institutions, wherein the BMI and gestational age adopt the weight recommended by the guidelines. The "Pregnancy Weight Management Clinical Pathway" as a submodule of the database is formulated by endocrinology experts consensus, which includes differentiated weight gain threshold in early, middle and late pregnancy, and the system automatically calls the w3 coefficient calculation logic of the corresponding stage according to the current gestational age value, while the dynamic adjustment range of the hormone level weight w4 can be constrained by the gestational age grouping rule. In addition, linear regression or logistic regression models are trained using historical data to optimize the weight of each parameter. For different populations, an adaptive weight adjustment mechanism can be provided. For example, by inputting the basic information of the user, the corresponding weight set is loaded, and the weight is dynamically adjusted to reflect the actual risk of different populations.

[0073] K represents an exponential constant of weight gain rate, which is used to adjust the strength of the impact of weight gain rate on risk. By changing the value of k, the non-linear influence of weight gain rate in the risk index can be controlled. The value of the weight gain rate exponential constant K can be based on historical medical data, and the initial value range suitable for most pregnant women can be obtained by fitting a non-linear regression model. At the same time, dynamic adjustment technology can also be used, that is, the value of K is adjusted according to the actual weight gain of each user. For example, if the weight gain rate is fast, the value of K can be increased to better reflect the risk.

[0074] Z is a standardization factor, which is used to normalize the results of each item. It can avoid the result deviation caused by variables with different dimensions and ranges, making the FI more comparable. Its calculation formula is as follows:

[0075] Z = σ(B) + σ(W) + σ(G) + σ(H)

[0076] The calculation of the standardization factor Z can be based on a large-scale sample data set. This data set can include, but is not limited to, the terminal device 100 of the cloud server 200, which can obtain a large number of reference user data, and can contain historical statistical data of different populations (such as pregnant women). Regularly update the calculation parameters of the standardization factor using the latest user data to maintain the accuracy and timeliness of the model. If the user data is insufficient, the cloud server 200 can combine public medical big data or estimate based on statistical models.

[0077] σ(B), σ(W), σ(G) and σ(H) represent the standard deviation of BMI, the standard deviation of gestational weeks, the standard deviation of weight gain rate and the standard deviation of comprehensive hormone level respectively, wherein,

[0078]

[0079] wherein, B i is the BMI of the i-th user (target user);

[0080] is the average BMI of all users (reference users);

[0081] N is the total number of samples.

[0082] Similarly, the calculation formula of other standard deviations is as follows:

[0083]

[0084] wherein, W i , G i and H i are the gestational weeks, weight gain rate and comprehensive hormone level of the i-th user (target user) respectively, are the average gestational weeks, weight gain rate and comprehensive hormone level of all users (reference users) respectively.

[0085] In summary, the cloud server 200 can obtain the feature index FI for evaluating the risk level of gestational diabetes mellitus of the target user based on the built-in calculation model. Preferably, as shown in Figure 1 , the cloud server 200 can be configured to form a comprehensive overall reference data set by collecting various blood glucose data of each gestational period from the terminal device 100 of each reference user. On this basis, after the FI value of the target user is calculated, the cloud server 200 can include it in the FI value interval corresponding to the FI value which has been determined in advance. These FI value intervals are optimized by the cloud server 200 according to large-scale reference user data through statistical analysis algorithm for homogenization grouping, aiming to ensure that the users in each interval have highly similar risk characteristics. The purpose of this step is to narrow the comparison range from a large number of reference users, so that the target user can be compared and analyzed more accurately with other reference users with similar FI values.

[0086] Within the same FI value interval, the cloud server 200 first further subdivides the reference users into multiple gestational period groups by gestational weeks. This hierarchical strategy takes into account the physiological changes characteristic of different gestational weeks, ensuring comparison under the same or similar gestational weeks, improving the relevance and reliability of the analysis results.

[0087] In summary, the FI is not only an important bridge connecting individual and group health information but also an indispensable technical tool for intelligent, personalized blood glucose management during pregnancy. Through mathematical models, it transforms complex physiological variables into easily understood and applicable numerical values, contributing to improving users' blood glucose status.

[0088] Preferably, to improve the homogeneity of the grouping, Figure 3 As shown, the cloud server 200 can first group the data based on gestational age to improve the accuracy of data analysis. Specifically, the cloud server 200 automatically assigns the target user to the corresponding gestational age group based on the latest gestational age information reported by the target user. In order to maintain the timeliness and accuracy of the data, the cloud server 200 can re-evaluate the gestational age of the target user at a certain period and adjust the group to which it belongs accordingly. Taking into account the continuity of gestational age, the cloud server 200 can set a reasonable range to define "similar" gestational age, such as ±2 weeks. For example, if a target pregnant woman is in her 20th week, her blood glucose data will be compared with other reference pregnant women between 18 and 22 weeks.

[0089] Preferably, the cloud server 200 can adjust the definition of "similar" gestational age based on key events in different trimesters. The key events may include fetal development milestones and maternal physiological changes.

[0090] In the early stages of pregnancy, hormone levels in the body rise rapidly, especially changes in hormones such as human chorionic gonadotropin (hCG), estrogen and progesterone significantly affect metabolic function. This period is the most critical period for the formation of embryonic organs. As the placenta begins to form and secrete a variety of hormones, the mother's insulin sensitivity begins to change, which may lead to unstable blood sugar levels. Any external factors (such as blood sugar levels) may have long-term effects on the health of the fetus. In the late stages of pregnancy, the fetus grows rapidly, and the metabolic burden on the mother increases, especially the demand for glucose increases significantly. As the placenta secretes more hormones that antagonize the action of insulin, the mother's insulin resistance gradually increases, and the physiological changes are more significant, which makes blood sugar control more difficult. Therefore, in the early stages of pregnancy (0 to 12 weeks) and the late stages of pregnancy (after 28 weeks), the cloud server 200 adopts a more fine-grained time window (such as ±1 week or a specific number of days).

[0091] In contrast to early pregnancy, the hormone fluctuations during the mid-pregnancy stage (12-28 weeks) tend to be more stable, and fetal growth accelerates but is not yet as highly active as in late pregnancy. During this period, the pregnant woman's weight gain gradually accelerates, but the physiological changes during this period are relatively stable, and it is suitable for a large-scale comparison of groups. The cloud server 200 uses a time window of ±2 weeks to cover a sufficient number of comparison objects, and does not lose comparability due to a too large time span. This setting helps to improve the effectiveness and reliability of statistical analysis.

[0092] After assigning the pregnancy groups, the cloud server 200 will conduct a detailed comparative analysis of the target user's blood glucose data with the reference user's blood glucose data in the same pregnancy group. The specific steps are as follows: First, calculate the target user's blood glucose data in the pregnancy group to which it belongs to the percentile, in order to quantify its relative position relative to other reference users. The blood glucose data referred to here is preferably the daily trend blood glucose, i.e. the blood glucose changes of the user at different time points in a day obtained by the continuous blood glucose monitor, which shows the trend of blood glucose level changes over time and daily activities in a day, which can be represented by a continuously changing blood glucose line. For example, Figure 7 The daily trend blood glucose of the reference user and the target user is shown. Through this method, the target user's blood glucose fluctuation pattern can be more comprehensively understood, rather than just a snapshot at a single time point.

[0093] If the target user's daily trend blood glucose data is higher than the set threshold (e.g. more than 75% of the reference users), it is considered to be abnormal. This abnormality may indicate an increased risk of gestational diabetes or the presence of other health problems. For target users marked as abnormal, the cloud server 200 will further analyze which key parameter in the physiological parameters (including hormone levels, BMI, weight gain speed) shows a significant difference compared to other reference users. This step aims to identify the specific factors that may cause abnormal blood glucose and provide a basis for subsequent health management. The setting of the 75% percentile threshold is dynamically linked to the blood glucose control benchmark library stored in the cloud server 200, and when the 75th percentile blood glucose value of the target user's group exceeds the safety threshold of the corresponding gestational age in the benchmark library, the system automatically performs a blood glucose range warning judgment. The benchmark library data can be updated regularly from the cooperative medical institution data interface through the communication unit 120, ensuring compliance with clinical practice requirements.

[0094] To achieve this purpose, the cloud server 200 will compare each physiological parameter of the target user with the average or median of the reference users in the same pregnancy group, and find out the parameters that deviate greatly. This way can quickly lock in those factors that are significantly different from the average level of the group, so as to preliminarily determine the possible influencing factors.

[0095] Preferably, asFigure 1 、 Figure 4 As shown in FIG. 2B, in addition to gestational weeks, the cloud server 200 can also group the users based on other physiological data inputted by the users through the interactive unit 140 of the terminal device 100. In this embodiment, the physiological data can be the user's own BMI and weight gain rate which are inputted or updated regularly (e.g. weekly) by the user.

[0096] Preferably, the BMI and weight gain rate are manually inputted by the user through the interactive unit 140 of the terminal device 100, and these information can also be automatically synchronized by using the communication unit 120 of the terminal device 100 to obtain data from a smart scale or other devices. The cloud server 200 can send a prompt message to the user who uses the terminal device 100 for the first time so that the user can provide complete height and weight information through the interactive unit 140, and the cloud server 200 can calculate the initial BMI accordingly and record the pre-pregnancy weight as a reference.

[0097] Preferably, the cloud server 200 can convert the BMI data collected by each terminal device 100 into a standard range (e.g. 18.5-24.9 kg / m 2 ), and represent the weight gain rate as an average value per week. For BMI, it can be classified according to the "Recommended Values for Weight Gain during Pregnancy" (WS / T 801-2022) issued by the National Health Commission of the People's Republic of China, for example: the BMI range for underweight is ≤18.5 kg / m 2 ; the BMI range for normal weight is 18.5-24.9 kg / m 2 ; the BMI range for obese weight is ≥25 kg / m 2 . For the weight gain rate threshold, the cloud server 200 can be reasonably set according to the latest pregnancy health guidelines, and divided into at least three groups accordingly. For example, the first weight gain rate threshold can be ≤0.2 kg / week, the second weight gain rate threshold can be 0.2-0.5 kg / week, and the third weight gain rate threshold can be ≥0.5 kg / week. In addition, the cloud server 200 can also apply statistical methods such as box plot method and Z-score method to identify potential outliers, so as to avoid the influence of these extreme values on the accuracy of the overall statistical data. Preferably, the cloud server 200 can automatically optimize the grouping strategy according to the actual feedback of the users and the learning process within the system. For example, if it is found that the difference between the users in a certain group is large, the system will automatically adjust the threshold range of BMI or weight gain rate to ensure the rationality of the grouping.

[0098] Preferably, the cloud server 200 can employ multi-dimensional grouping logic, such that the reference users within each group have higher homogeneity, thereby improving the relevance and guidance of the target user comparison results. Specifically, the cloud server 200 can combine the gestational age and BMI, i.e. first divide the users into different groups according to the gestational age, and then further subdivide within each gestational age group according to the BMI. Within the same BMI subgroup, the cloud server 200 can further subdivide the user groups according to the user's weight gain rate. For example, for users with excessively fast or slow weight gain, the system will assign them to a special attention group, providing more personalized dietary recommendations and support, including but not limited to regularly reminding the user's terminal device 100 to pay attention to dietary habits, suggesting contacting a nutritionist or a doctor.

[0099] Preferably, the cloud server 200 can set a reasonable gestational age information update period, such as once a week or once every two weeks, and remind the user to update personal information in a timely manner through the interactive interface. At the same time, the user is allowed to manually trigger an immediate update at any time, especially after experiencing a major life event or medical examination. In addition, the cloud server 200 also allows the user to object to the automatically assigned group, and provides a simplified manual adjustment option. In this way, not only can the user experience be enhanced, but also valuable user feedback can be collected for further optimization of the grouping algorithm.

[0100] According to an example embodiment, the target user is a pregnant woman at 24 weeks of gestation, with a BMI of 28.5 (obesity classification) and a weight gain rate of 0.6 kg / week (exceeding the upper limit of 0.5 kg / week recommended by medical guidelines). The terminal device 100 worn by the target user integrates the acquisition unit 110, which synchronizes the physiological parameters including gestational age, BMI, and blood glucose data to the cloud server 200 in real time through the communication unit 120. Based on the target user's gestational age, BMI, and weight gain rate, the cloud server 200 combines the aforementioned preset homogeneous subgroup division rules (such as Figure 4 indicated), and dynamically assigns it to the "gestational age 22-26 weeks + obesity BMI + rapid weight gain" subgroup, where the homogeneous subgroup division is based on the multi-dimensional parameter correlation group homogeneity optimization algorithm.

[0101] The cloud server 200 extracts the target user's daily trend blood glucose curve (as shown by the middle red dashed line in Figure 8 ), and compares it with the blood glucose data of the reference users in the homogeneous subgroup at each time point. Specifically, the cloud server 200 calculates the 25% to 75% percentile interval (as shown by the light blue filled area in Figure 8 ) and the median (as shown by the dark blue dashed line in Figure 8The analysis result shows that the target user's blood glucose peak value in the breakfast period (7:00-9:00) is 7.2 mmol / L (8:00), which exceeds the 75% percentile threshold (6.8 mmol / L) of the homogeneous subgroup; the blood glucose peak value in the lunch period (12:00-14:00) is 7.5 mmol / L (12:30), which exceeds the 75% percentile threshold (7.0 mmol / L). The system determines that the target user's total abnormal period is 4.5 hours, and generates a "pregnancy-induced diabetes medium-high risk" warning signal in combination with the risk characteristics (obesity BMI and rapid weight gain) of the subgroup to which the target user belongs.

[0102] The cloud server 200 further analyzes the physiological parameter deviation of the target user from the reference users of the homogeneous subgroup: the BMI (28.5) of the target user deviates from the mean value (26.2) of the subgroup by +2.3 standard deviations (Z value), and is located at the 95% percentile of the subgroup; the weight gain rate (0.6 kg / week) exceeds the median value (0.35 kg / week) of the subgroup by 71%, and deviates from the medically recommended value (0.28-0.50 kg / week) by 20%. In addition, the estradiol concentration (450 pg / mL) of the target user is increased by 40% compared with the mean value (320 pg / mL) of the subgroup, and the progesterone level (18 ng / mL) is lower than the mean value (22 ng / mL). Based on the correlation model of hormone level and insulin resistance, the system determines that the hormone imbalance may lead to the aggravation of insulin resistance risk.

[0103] Based on the above analysis result, the cloud server 200 can generate the following dynamic hierarchical health management scheme for the target user:

[0104] 1. Dietary structure adjustment: replace high GI food (such as white bread, GI=75) with low GI substitutes (such as oatmeal, GI=55), which is expected to reduce the 1-hour postprandial glucose peak value by 15%-20%; increase the proportion of non-starch vegetables to 60% and match with low GI staple food (such as quinoa, GI=53) to slow down the glucose absorption rate.

[0105] 2. Weight control strategy: record daily dietary calories through the interaction unit 140, and form a daily calorie gap of about 300 kcal through low-intensity walking.

[0106] 3. Dynamic feedback mechanism: if the target user's weight gain rate decreases to below 0.4 kg / week for two consecutive weeks, the system automatically relaxes the GI limit to below 65; if the target user does not meet the standard, the doctor collaborative intervention process is triggered, and the warning notification is sent to the designated medical terminal through the communication unit 120.

[0107] Preferably, the cloud server 200 of the present application does not simply rely on the preset blood glucose level fixed reference threshold in the existing medical guidelines (such as the “China Gestational Diabetes Mellitus Maternal and Infant Joint Management Guidelines”) to set a unified control recommendation level for the target user. On the contrary, the system dynamically adjusts the recommendation parameters based on the in-depth analysis of the distribution characteristics of the target user's blood glucose data in the overall reference data set in a larger range, and combines the reference user data highly similar to the physiological state of the target user. Through this dynamic analysis and adjustment mechanism, individual differences can be more accurately reflected, thereby providing personalized dietary management recommendations for users. Compared with the scheme generated based on the fixed reference value, this method is more in line with the actual needs of users, and helps to improve the trust and compliance of users to the system recommendations.

[0108] In addition, the cloud server 200 continuously optimizes the overall reference data set through real-time analysis and dynamic updating of massive user data, so that the generated control recommendations can maintain high timeliness and scientificity. This data-driven personalized recommendation generation mode has higher flexibility and adaptability compared with the traditional method based on static medical guidelines, not only can reflect the latest medical research results and clinical experience, but also can effectively adapt to the dynamic health status of users, provide more scientific and practical management schemes for users, and thus improve the overall effect of health management.

[0109] Embodiment 2

[0110] This embodiment is a supplement to the foregoing embodiments, and the repeated contents will not be described again.

[0111] Preferably, the cloud server 200 can also use a segmented function model to describe the feature index FI based on the different stages of pregnancy:

[0112]

[0113] Among them, the early pregnancy stage f1(W) uses a nonlinear regression model, focusing on the influence of hormone levels on insulin sensitivity; the middle pregnancy stage f2(W): uses a linear regression model, suitable for relatively stable change trend; the late pregnancy stage f3(W): uses a nonlinear regression or logistic regression model, focusing on the influence of insulin resistance.

[0114] Preferably, in this embodiment, the calculation of the feature index FI weakens the influence of gestational weeks, and only sets different calculation models for the early, middle and late stages of pregnancy, respectively, and more considers the changes of BMI, body weight growth rate and / or hormone levels, which have obvious influence on gestational diabetes blood glucose.

[0115] Specifically,

[0116] f1(W) = w1·(ek·G(W) -1)+w2·ln(B+1)

[0117] exponential term e k·G(W) represents the risk increase due to hormonal changes, the logarithmic term ln(B+1) adjusts the impact of BMI on early risk, avoiding instability of the model due to too large values. G(W) is a function of the weight gain rate; k is a parameter that adjusts the rate of hormonal changes; B is the BMI value; w1 and w2 are weight coefficients. Preferably, Figure 6a An example graph showing the evolution of the feature index plotted for a target user as a function of gestational weeks in the early pregnancy stage is shown.

[0118] f2(W) = w3·G(W) + w4·B + w5·H(W)

[0119] The linear terms G(W) and B mainly reflect the impact of weight gain and BMI of the pregnant woman on the risk of GDM. The hormonal level term H(W) is kept at a lower weight, as hormonal fluctuations are relatively stable. H(W) is a function of the hormonal level (which is relatively small at this time). w3, w4 and w5 are weight coefficients. Preferably, Figure 6b An example graph showing the evolution of the feature index plotted for a target user as a function of gestational weeks in the mid-pregnancy stage is shown.

[0120] f3(W) = w6·(W-28) 2 + w7·G(W) + w8·H(W)

[0121] the quadratic term (W-28) 2 emphasizes the non-linear risk increase in the late pregnancy stage. The weight gain rate term G(W) and the hormonal level term H(W) superimpose the impact on the risk of GDM. w6, w7, w8 are weight coefficients. Preferably, Figure 6c An example graph showing the evolution of the feature index plotted for a target user as a function of gestational weeks in the late pregnancy stage is shown.

[0122] It should be noted that the above-mentioned specific embodiments are exemplary, and those skilled in the art can think of various solutions under the inspiration of the disclosure of the present application, and these solutions also belong to the disclosed range of the present application and fall within the protection scope of the present application. Those skilled in the art should understand that the specification and drawings of the present application are illustrative and not constitute a limitation on the claims. The protection scope of the present application is defined by the claims and their equivalents. The specification of the present application contains multiple inventive concepts, such as "preferably" or "according to a preferred embodiment", which means that the corresponding paragraph discloses an independent concept, and the applicant reserves the right to file a divisional application according to each inventive concept. Throughout the text, the features introduced by "preferably" are only optional ways, and should not be understood as necessarily set, therefore the applicant reserves the right to abandon or delete the relevant preferred features at any time.

Claims

1. A blood glucose management system for a gestational period, comprising a terminal device (100) carried by a target user and a reference user, and a cloud server (200) in data connection with the terminal device (100), characterized in that, The cloud server (200) is configured to: obtain physiological data and blood glucose data of a target user from the terminal device (100), and calculate a feature index of the target user according to the physiological data and the blood glucose data; obtain a plurality of blood glucose data of reference users from a plurality of terminal devices (100), and form an overall reference data set in a manner associated with gestational weeks, assign the target user to a predetermined feature index value interval in the overall reference data set based on the feature index, determine factors affecting the risk of gestational diabetes mellitus of the target user, so that the target user manages corresponding physiological parameters according to the factors, the cloud server (200) further subdivides the reference users in the feature index value interval into a plurality of gestational week groups according to gestational weeks, and the cloud server (200) further analyzes the deviation between the physiological parameters of the target user and the average values of the reference users in the same gestational week group, so as to determine one or more key parameters affecting the blood glucose level of the target user in the physiological parameters, wherein the physiological parameters include body mass index BMI, body weight gain rate and hormone level, the cloud server (200) can calculate the percentile of the relative position of the target user based on the daily trend blood glucose data of the target user in the gestational week group to which the target user belongs, and compare and analyze the blood glucose data of the target user with the blood glucose data of the reference users to identify the blood glucose fluctuation pattern and potential abnormalities of the target user, The cloud server (200) calculates the feature index FI by the following formula: In the formula, B is the body mass index BMI, W is the gestational week, G is the body weight gain rate, H is the hormone level, w1, w2, w3, w4 are weight coefficients, K is the exponential constant of the body weight gain rate, and Z is the standardization factor, The cloud server (200) is configured to divide users into different groups according to gestational weeks, in each gestational week group, a plurality of BMI ranges are set so that users can be subdivided into a plurality of BMI subgroups, and in the same BMI subgroup, the cloud server (200) can further subdivide users according to the body weight gain rate to identify users with excessively fast or slow body weight gain and assign them to a special attention group.

2. The system of claim 1, wherein, The terminal device (100) includes an acquisition unit (110) for obtaining continuous blood glucose data, and the terminal device (100) uploads the blood glucose data recorded by the acquisition unit (110) to the cloud server (200) in a manner associated with gestational weeks through the configured communication unit (120), and receives personalized dietary recommendations from the cloud server (200).

3. The system of claim 2, wherein, The terminal device (100) is integrated with an interaction unit (140) signal connected with the communication unit (120), the interaction unit (140) is configured to receive and display dietary recommendations from the cloud server (200) and the storage unit (130) to the user, and provide a platform for the user to input and update personal health information including the current gestational week, weight change, and dietary record.

4. The system of claim 3, wherein, The cloud server (200) groups the acquired overall reference dataset based on similar gestational weeks, wherein the cloud server (200) adjusts the time window range of similar gestational weeks according to the stage of the user's pregnancy: for the middle stage of pregnancy, the time window is wider than that of the early and late stages of pregnancy.

5. The system of claim 4, wherein, The cloud server (200) can comprehensively group the overall reference dataset according to the physiological data input by the user through the interactive unit (140) of the terminal device (100), and combine the multi-dimensional grouping method to improve the homogeneity of the members in each group, wherein, The physiological data includes the user's own BMI and weight gain rate which are regularly input or updated.

6. The system of claim 1, wherein, The cloud server (200) calculates the basic statistical indicators for each group in the overall reference dataset, combines the specific percentile position of the user's blood glucose data in its own group, compares the user's blood glucose data with the preset healthy blood glucose range to judge the user's current blood glucose state, and accordingly generates a graded diet suggestion, which optimizes the target user's diet structure by selecting food categories with different GI values.

7. The system according to any one of claims 1 to 6, characterized in that The cloud server (200) can periodically reevaluate the gestational week information of the target user and adjust its group accordingly. The cloud server (200) allows the user to object to the automatically assigned group and receives the gestational week information manually adjusted by the user through the interactive unit (140).

Images