Diabetes blood glucose dynamic management system and method based on physiological rhythm segmentation

Abstract

The invention discloses a diabetes mellitus blood glucose dynamic management system and method based on physiological rhythm segmentation, and the system comprises a threshold storage and inheritance module which is used for storing a personalized safety threshold library of five key points established in a re-intervention mode; the five-point monitoring and scheduling engine module is used for monitoring and collecting blood glucose data of five key points based on the physiological rhythm in a light intervention mode; the grading abnormity evaluation module is used for carrying out grading evaluation; the cross-mode response linkage module is used for the system to take corresponding measures according to the grading evaluation result; the central data management module is used for uniformly storing blood glucose data and the like; the abnormity deviating from the inheritance threshold value is found in time; a precise personalized reference is provided for light intervention mode monitoring, a regular response trigger mechanism is realized according to a graded response strategy of point location anomaly types and severity, seamless inheritance of personalized safety thresholds from a heavy intervention mode to the light intervention mode is realized, and management closed loop and effect evaluation throughout the whole process are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to a diabetes blood glucose dynamic management system and method, in particular to a diabetes blood glucose dynamic management system and method based on physiological rhythm segmentation. BACKGROUND

[0002] The existing diabetes blood glucose dynamic management system lacks a clear and easy-to-operate target monitoring mechanism for key points when the user switches from heavy intervention (intensive monitoring) to light intervention (self-management), resulting in the inability to timely detect potential risk points such as nocturnal hypoglycemia, dawn phenomenon, postprandial hyperglycemia, and post-exercise hyperglycemia, causing a silent hyperglycemic state, and there is a monitoring blind area. The light intervention stage fails to effectively inherit and utilize the individualized safety threshold established in the heavy intervention stage, which fails to reflect the individualized control target and vulnerable point, resulting in a lack of individualized benchmark for monitoring, and inaccurate early warning results. The abnormality found in the light intervention stage lacks a response triggering mechanism that is connected and regularized with the heavy intervention stage, such as automatically upgrading the intervention level, resulting in problems such as delayed problem handling or low efficiency. The data of the light intervention stage and the heavy intervention stage, especially the key threshold and monitoring results, lack systematic integration and correlation analysis, making it difficult to form a management closed loop and effect evaluation throughout the whole process. SUMMARY

[0003] In order to solve the above technical problems, the present application provides a diabetes blood glucose dynamic management system and method based on physiological rhythm segmentation.

[0004] In order to solve the above technical problems, the present application adopts the technical scheme: a diabetes blood glucose dynamic management system based on physiological rhythm segmentation, comprising a threshold storage and inheritance module for storing five key point individualized safety threshold libraries established in the heavy intervention mode, and automatically loading and updating each individualized safety threshold library when switching to the light intervention mode; A five-point monitoring scheduling engine module is used to schedule and execute blood glucose data monitoring and collection of five key points of night, morning fasting, any point, postprandial and post-exercise in the light intervention mode based on physiological rhythm according to a preset rule; A hierarchical abnormality evaluation module is used to compare the collected five-point monitoring data with the corresponding individualized safety threshold, and perform hierarchical evaluation according to the point type, deviation and frequency; A cross-mode response linkage module is used to take corresponding measures according to the hierarchical evaluation results; A central data management module is configured to store blood glucose data, life events, a personalized safety threshold library, monitoring scheduling records, abnormality evaluation results, and response logs, thereby ensuring data continuity and accessibility between heavy intervention mode and light intervention mode.

[0005] The diabetes blood glucose dynamic management method based on physiological rhythm segmentation comprises the following steps: In step S1, the threshold storage and inheritance module is used to divide the specific period in the heavy intervention mode into five key points based on physiological rhythm, and monitor the blood glucose at the five key points. In step S2, a personalized safety threshold library is established, and the monitoring data is stored in the personalized safety threshold library. When used, the monitoring data is automatically loaded and set as the blood glucose monitoring evaluation benchmark in the light intervention mode. In step S3, after the user enters the light intervention mode, the five-point monitoring scheduling engine module is used to schedule and perform blood glucose monitoring at the five key points, execute the five-in-one targeted monitoring, and make monitoring scheduling strategy decisions. In step S4, if the monitoring data is normal, the current scheduling strategy is maintained; if the monitoring data is abnormal, the hierarchical abnormality evaluation module is used to respond to the monitoring data deviating from the corresponding personalized safety threshold at any key point, and perform hierarchical evaluation according to the type, degree, and frequency of deviation. In step S5, if the hierarchical evaluation result is mild abnormality or moderate abnormality, the cross-mode response linkage module generates and pushes corresponding guidance information in the light intervention mode; if the hierarchical evaluation result is severe abnormality, the cross-mode response linkage module automatically triggers the warning disposal protocol associated with the heavy intervention mode.

[0006] Further, in step S1, the night blood glucose monitoring adopts two monitoring methods. The first method is to use a CGM device to perform continuous glucose monitoring to obtain whole-night blood glucose trend data. The second method is to perform two fingertip blood point glucose monitoring before going to bed and before getting up in the morning, and take the average of the two monitoring results as the representative value.

[0007] Further, in step S1, the morning fasting blood glucose monitoring is fingertip blood point glucose monitoring after getting up in the morning and before having breakfast. The arbitrary point blood glucose monitoring is fingertip blood point glucose monitoring at an arbitrary time point after not having a meal and not exercising. The postprandial blood glucose monitoring is fingertip blood point glucose monitoring at a specified time point after a meal. The post-exercise blood glucose monitoring is fingertip blood point glucose monitoring at a specified time point after planned exercise.

[0008] Further, in step S3, the five-point monitoring scheduling engine module is used to make monitoring scheduling strategy decisions, which specifically comprises the following steps: Step A1: the user enters the light intervention mode; Step A2: risk assessment is performed, and based on the risk assessment result, the system specifies a monitoring mode to monitor the user's blood glucose; Step A3: the monitoring result data is fed back to the evaluator, and it is judged whether the data is abnormal; if the data is not abnormal, the current monitoring scheduling strategy is maintained; if the data is abnormal, a dynamic adjustment strategy is triggered to perform reverse optimization scheduling of abnormal data.

[0009] Further, in step A2, based on the risk assessment result, the system specifies a monitoring mode to monitor the user's blood glucose, which includes the following cases: If the risk assessment result is high risk, or the user is a new transferred user, enter the high-frequency FKPP monitoring mode, and monitor the blood glucose of five key points not less than 3 times a week; If the risk assessment result is stable state, enter the low-frequency KFPP monitoring mode, and monitor the blood glucose of any point not less than 2 times a week; monitor the night blood glucose and postprandial blood glucose each once a week; If the risk assessment result triggers a specific event, enter the event-driven monitoring mode, and take corresponding measures according to the event type.

[0010] Further, in step A3, the dynamic adjustment strategy includes increasing the monitoring frequency of abnormal points by 50%, expanding the blood glucose monitoring of associated points, and triggering a hierarchical response protocol for hierarchical evaluation.

[0011] Further, in step S5, it is judged that the blood glucose monitoring data meets one of the following conditions: One, the night blood glucose monitoring data is less than 4.4 mol / L for 2 consecutive hours; Two, the frequency of night blood glucose monitoring data less than 4.4 mol / L exceeds the normal frequency; Three, any one of the monitoring data of morning fasting blood glucose monitoring data, arbitrary point blood glucose monitoring data, postprandial blood glucose monitoring data or post-exercise blood glucose monitoring data exceeds 20% of the upper limit of the corresponding personalized safety threshold of each key point for N consecutive times, wherein N is an integer not less than 2; Four, within a preset statistical period, the total time percentage of the comprehensive blood glucose monitoring data of a specific key point or all key points exceeding the corresponding personalized safety threshold of each key point exceeds the preset safety threshold.

[0012] Further, in step S5, in the light intervention mode, the blood glucose control index of the user is continuously stable and meets the preset condition, then the personalized safety threshold library is fine-tuned and optimized based on the new blood glucose monitoring data and is updated and stored; when the heavy intervention mode is switched back due to triggering the early warning treatment protocol, the personalized safety threshold library is updated in the heavy intervention mode based on the threshold storage and inheritance module.

[0013] Further, the early warning treatment protocol includes sending an advanced alarm to the user, automatically locking the light intervention mode function, automatically downgrading the light intervention mode function, triggering system switching or instructing to switch back to the heavy intervention mode, and providing preliminary emergency guidance.

[0014] The application discloses a diabetes blood glucose dynamic management system and method based on physiological rhythm segmentation, in the light intervention mode stage, based on a five-point monitoring scheduling engine module, the blood glucose data of the five key points are targetedly monitored and collected based on physiological rhythm, and abnormalities deviating from the inherited threshold are timely found; Based on the threshold storage and inheritance module, the personalized safety threshold library is automatically updated after the heavy intervention mode is switched to the light intervention mode, so as to provide accurate personalized benchmarks for light intervention mode monitoring; Based on the hierarchical abnormality evaluation module and the cross-mode response linkage module, the hierarchical response strategy according to the point abnormality type and the severity is used to realize a regular response triggering mechanism; Based on the central data management module, the personalized safety threshold, the blood glucose monitoring result and the early warning record are seamlessly transferred and shared between the light intervention mode and the heavy intervention mode, the personalized safety threshold is seamlessly inherited from the heavy intervention mode to the light intervention mode, and the management closed loop and the effect evaluation throughout the whole process are realized. BRIEF DESCRIPTION OF DRAWINGS

[0015] Fig. 1 The system architecture of the application is shown.

[0016] Fig. 2 The flowchart of the scheduling strategy decision of example one in the application is shown.

[0017] Fig. 3 The flowchart of the hierarchical evaluation of example one in the application is shown. DETAILED DESCRIPTION

[0018] The application will be further described in detail in combination with the drawings and specific embodiments.

[0019] Example one:

[0020] As Figs. 1 to 3The physiological rhythm-based segmented diabetes blood glucose dynamic management system shown includes a threshold storage and inheritance module for storing five key point personalized safety threshold libraries established in the heavy intervention mode and automatically loading each personalized safety threshold library when switching to the light intervention mode. It should be noted that the personalized safety threshold library, as a core asset, is updated in real time and synchronized to the database during the heavy intervention phase. When the light intervention is started, the latest valid threshold library of the user is directly loaded from the database. If there is a fine-tuning in the light intervention, the fine-tuned data is updated to the database. The system has the ability to store and inherit five personalized safety thresholds from the heavy intervention mode, and uses them as the core evaluation benchmark for the light intervention. The five-point monitoring scheduling engine module, i.e., the FKPP targeted monitoring engine module, is used to schedule and execute blood glucose data monitoring and collection at five key points, i.e., night, morning fasting, arbitrary point, postprandial, and post-exercise, based on physiological rhythms according to preset rules in the light intervention mode. The hierarchical abnormality evaluation module is used to compare the collected five-point monitoring data with the respective corresponding personalized safety threshold, i.e., the inherited threshold, and perform hierarchical evaluation according to the point type, deviation, and frequency. That is, the FKPP monitoring data is compared with the inherited threshold in real time, and hierarchical evaluation is performed according to the point, deviation, and frequency. The cross-mode response linkage module is used to generate evaluation results and push light intervention guidance information on the system interface according to the mild abnormality evaluation results or moderate abnormality evaluation results. According to the severe abnormality evaluation results, the warning disposal protocol linked with the heavy intervention mode is automatically activated and executed. That is, the cross-mode hierarchical responder is internally set to generate reminders within the light intervention, i.e., the first-level response and the second-level response. When the preset conditions of the heavy intervention mode, i.e., the third-level response, are met, the warning disposal protocol of the heavy intervention mode is automatically triggered to realize mode switching or forced upgrade.

[0021] The central data management module, i.e., the cross-light and heavy data continuity module, is used to uniformly store blood glucose data, life events, personalized safety threshold libraries, monitoring scheduling records, abnormality evaluation results, and response logs to ensure the continuity and accessibility of data between the heavy intervention mode and the light intervention mode. It is ensured that all blood glucose data and user profiles under the unified platform are seamlessly transferred and shared before and after mode switching. It should be noted that the unified data platform is set in the central data management module to store all blood glucose data, life event records, KFPP monitoring records, personalized safety threshold libraries, and response event logs in the light intervention mode and the heavy intervention mode. The blood glucose data includes blood glucose data from BGM fingertip blood point blood glucose monitoring and blood glucose data from CGM subcutaneous interstitial fluid continuous blood glucose monitoring. The life event records include diet, exercise, and medication, etc.

[0022] The physiological rhythm-based segmented diabetes blood glucose dynamic management method specifically includes the following steps: Step S1: Based on the threshold storage and inheritance module, the specific period in the re-intervention mode is divided into five key points, namely the five points, which are five intervals based on physiological rhythms. These five points are crucial for evaluating the blood glucose fluctuation mechanism and intervention effect, including nighttime, fasting morning blood glucose, random blood glucose, postprandial blood glucose, and post-exercise blood glucose. That is, monitor nighttime blood glucose, fasting morning blood glucose, random blood glucose, postprandial blood glucose, and post-exercise blood glucose. The intensive intervention mode refers to a management model that involves high-frequency monitoring and improvement recommendations during specific periods, enhanced medical nutrition therapy or exercise prescription supervision, and frequent medical interventions. For example, it may be implemented after hospitalization or during periods of uncontrolled blood glucose, with more than 15 daily BGM / CGM measurements covering all or intensively, MNT medical nutrition therapy, and corresponding medication adjustments. BGM refers to fingertip blood glucose monitoring, and CGM refers to continuous subcutaneous interstitial fluid blood glucose monitoring. The intensive intervention mode, transitioning after the intensive intervention has achieved results, is a user-self-management-based model where the system provides on-demand monitoring and precise guidance. Its core is risk warning and mild intervention based on key point monitoring and inherited thresholds. Personalized safety thresholds refer to the range of blood glucose values ​​set for different physiological periods during the intensive intervention phase based on individual blood glucose control goals, risk factors, and actual control status. In this embodiment, the upper, lower, or range of blood glucose values ​​are set according to five points based on an individual's history of hypoglycemia, body fat percentage, and actual control status during the intensive intervention phase.

[0023] Among the five key blood glucose monitoring points, nighttime blood glucose monitoring uses two monitoring methods. The first method is to use a continuous glucose monitoring (CGM) device to obtain the blood glucose trend data throughout the night. The second method is to perform two finger-prick blood glucose monitoring (BGM) tests before going to bed and before waking up in the morning and calculate the average value as a representative value. Monitoring fasting blood glucose in the morning involves a finger-prick blood glucose test performed after waking up and before breakfast. Random blood glucose monitoring refers to finger-prick blood glucose testing at any time point other than after a meal and after exercise. Postprandial blood glucose monitoring involves finger-prick blood glucose testing at designated postprandial time points. Post-exercise blood glucose monitoring involves finger-prick blood glucose testing at a designated time point after the planned exercise session.

[0024] Step S2: Establish a personalized safety threshold library, store the monitoring data in the personalized safety threshold library, and automatically load and set it as the blood glucose monitoring assessment benchmark in the light intervention mode when used; During the heavy intervention phase, the system dynamically calculates and stores the personalized safety thresholds for each FKPP based on individual blood glucose control targets (HbA1c), hypoglycemia risk, and real-time blood glucose data from five points. In this embodiment, the nighttime blood glucose monitoring range is 5.0-6.0 mmol / L; the morning fasting blood glucose monitoring range is 5-6.5 mmol / L; the 2-hour postprandial blood glucose monitoring range is less than 10 mmol / L; the 30-minute post-exercise blood glucose monitoring range is less than 9 mmol / L, and there is no rapid decline; the monitoring range for blood glucose at any point is 3.9 to 10.0 mmol / L. These threshold ranges constitute the personalized safety threshold library.

[0025] When a user meets the conditions to switch to the light intervention mode, the system will automatically inherit the personalized safety threshold library established in the heavy intervention phase as the monitoring benchmark for the light intervention mode. All monitoring, evaluation and early warning in the light intervention mode are based on the inherited personalized safety thresholds.

[0026] Step S3: After the user enters the light intervention mode, based on preset strategies or trigger conditions, the five-point monitoring scheduling engine module schedules and executes blood glucose monitoring at five key points, performing five-in-one targeted monitoring, i.e., five-in-one FKPP, and making monitoring scheduling strategy decisions. In light intervention mode, continuous intensive monitoring is not required, such as CGM throughout the day or multiple BGMs daily. It is only necessary to strategically, on demand, and according to plan monitor blood glucose at the five key physiological points as the main means of assessing the overall blood glucose control stability and detecting persistent risks. Among them, the monitoring methods can flexibly combine BGM and CGM segments. The purpose of monitoring blood glucose at night is to prevent nocturnal hypoglycemia and identify the Somogyi effect. In this embodiment, two BGM (Body Glucose Measurement) tests are performed between 5:00 AM and 6:00 AM before bedtime and before waking up. The system calculates the average of these two test results or evaluates them separately as representative values ​​for nighttime blood glucose. The system focuses on identifying test values ​​less than 4.4 mmol / L or a significant downward trend. At the same time, the nighttime safety threshold of the heavy intervention mode is inherited, and the timestamp, test value, and whether CGM or BGM is used are recorded. It should be noted that for nighttime blood glucose monitoring, after an abnormality is detected in the first monitoring mode, the overnight blood glucose trend can be automatically monitored using only the CGM device without activating the heavy intervention mode.

[0027] The purpose of monitoring fasting blood glucose upon waking is to identify the dawn phenomenon and assess baseline blood glucose control. BGM measurements are performed at multiple points after waking and before breakfast, with the fasting threshold of the morning inherited from the heavy intervention mode and correlated with the previous night's data.

[0028] The purpose of monitoring the blood glucose at any point is to check the blood glucose stability throughout the day and infer the previous management behaviors, such as whether there is urine leakage, snack intake and two-hour blood glucose emptying condition; the BGM point measurement is performed at any time point in the day after meals and after exercise, and the system can randomly prompt or the user actively performs the measurement; wherein, the any-point threshold value of the heavy intervention mode is inherited, and the measurement situation is recorded, that is, in a working condition or in a resting condition.

[0029] The purpose of monitoring the postprandial blood glucose is to control the postprandial blood glucose peak and evaluate the effects of diet or medication; in the specific postprandial time point, the BGM point measurement is performed 2 hours after starting the meal in the embodiment; it needs to be noted that the diet structure influence can be evaluated in combination with the shared protein ratio algorithm; wherein, the postprandial threshold value of the heavy intervention mode is inherited, and the user's meal record is associated, such as the type, quality and time of food and the like.

[0030] The purpose of monitoring the post-exercise blood glucose is to warn the post-exercise rebound hyperglycemia, evaluate the effect of exercise prescription, and impose guidance and whether a meal is needed after exercise; in the specific time point after the planned exercise, the BGM point measurement is performed 30 minutes, 60 minutes and 2 hours after the exercise end according to the exercise intensity preset in the embodiment, and the rise and fall of blood glucose is paid attention to, whether it has the change of first falling and then rising; wherein, the post-exercise threshold value of the heavy intervention mode is inherited, and the type, intensity and duration of the user's exercise are associated.

[0031] The five-point monitoring scheduling engine module performs monitoring scheduling strategy decision, specifically including the following steps: Step A1: the user enters the light intervention mode; Step A2: risk assessment judgment is performed, based on the risk assessment result, the system specifies the monitoring mode to monitor the blood glucose of the user; if the risk assessment result is high risk or the user is a newly transferred user, the high-frequency FKPP monitoring mode is entered; if the risk assessment result is stable state, the low-frequency FKPP monitoring mode is entered; if a specific event is triggered, the event-driven monitoring mode is entered; it needs to be noted that the risk level is divided according to the HbA1c target, TIR and hypoglycemia history, at the same time, the newly transferred user is automatically marked as high risk; I. If the high-frequency FKPP monitoring mode is entered, the blood glucose of all key points is monitored, and the blood glucose of each key point is monitored not less than 3 times per week; to ensure that the fragile period is fully covered.

[0032] II. If the low-frequency FKPP monitoring mode is entered, the blood glucose of specific key points is monitored, the blood glucose at any point is monitored not less than 2 times per week, and the blood glucose at the key points is strengthened, the night blood glucose monitoring and the postprandial blood glucose monitoring are each once per week; the embodiment performs the postprandial blood glucose monitoring on Monday and Thursday every week, and the night blood glucose monitoring on Tuesday and Friday every week, to ensure that the resources are tilted to the high-impact point.

[0033] III. If the event-driven monitoring mode is entered, first determine the event type. If high-intensity exercise is planned in the near future, the system will forcibly schedule post-exercise monitoring. The system will push a message "Please measure blood glucose 30 minutes after exercise"; if TIR has decreased by more than 10% in the near future, temporary increase in morning blood glucose monitoring and nighttime blood glucose monitoring is required; if there is a holiday or a warning of a banquet, postprandial blood glucose monitoring needs to be added; wherein TIR is the percentage of blood glucose compliance time, usually 3.9-10.0 mol / L or 3.9-7.8 mol / L; Step A3: feedback the monitoring result data to the evaluator, and determine whether the data is abnormal; Step A31: if the data is not abnormal, maintain the current monitoring scheduling strategy; Step A32: if the data is abnormal, trigger a dynamic adjustment strategy for reverse optimization scheduling of abnormal data. If a postprandial blood glucose is out of range, the frequency of blood glucose monitoring for that meal in the next week is increased by one. The dynamic adjustment strategy includes increasing the monitoring frequency of the abnormal point by 50%, expanding the blood glucose monitoring of the associated point, and triggering a hierarchical response protocol for hierarchical evaluation; wherein the expanded monitoring of the associated point is used for the abnormal point linkage related physiological period. If the morning fasting blood glucose is abnormal, the nighttime 3 AM monitoring, i.e., blood glucose monitoring at 3 AM, is added.

[0034] Step S4: the hierarchical response protocol is triggered in response to the monitoring data of any one key point deviating from its corresponding individualized safety threshold. The hierarchical evaluation is performed according to the point type, deviation degree, and deviation frequency. That is, the hierarchical response strategy is set as a link between the light intervention mode and the heavy intervention mode. When the KFPP monitoring data deviates from the inherited individualized safety threshold in the light intervention mode, the system triggers the hierarchical response according to the point type, deviation degree, and deviation frequency. Step S5: if the evaluation result is mild abnormality or moderate abnormality, the corresponding guidance information is generated and pushed in the light intervention mode based on the cross-mode response linkage module; if the evaluation result is severe abnormality, the pre-warning disposal protocol associated with the heavy intervention mode is automatically triggered based on the cross-mode response linkage module; mild abnormality is a first-level response, moderate abnormality is a second-level response, and severe abnormality is a third-level response. Abnormality-driven, specifically including the following steps: Step B1: monitoring blood glucose data out of range is an abnormal state; Step B2: combined with the abnormal blood glucose data, hierarchical evaluation is performed according to the point type, deviation degree, and deviation frequency; Step B21: if the hierarchical evaluation result is a first-level response, the system will give a reminder and maintain the original monitoring schedule; Step B22: If the grading evaluation result is a secondary response, the monitoring frequency of the point where the abnormal data is located is increased, and the grading evaluation is repeated for 3 consecutive days to determine whether the grading evaluation result meets the standard. If it meets the standard, the baseline monitoring schedule is restored; if it is determined that the grading evaluation result is abnormal, the heavy intervention mode is switched back; Step B23: If the grading evaluation result is a tertiary response, the heavy intervention mode is switched back.

[0035] Specifically, mild abnormalities are a primary response, and the triggering condition is that single-point data deviates from the threshold value by a small amount, and it occurs non-continuously and frequently. In this embodiment, the mild deviation threshold value is that the single-point data exceeds the upper limit of the threshold value by less than 10%, or is lower than the lower limit range of 0.4 mmol / L to 0.6 mmol / L. After triggering the primary response, the system will push mild reminders and education information, and record abnormal events. For example, in this embodiment, after triggering the primary response, the system will push “Your fasting blood glucose in the morning is slightly high, please pay attention to the blood glucose after breakfast” or “There is a downward trend in blood glucose last night, consider a small amount of snacks before sleep”.

[0036] Moderate abnormalities are a secondary response, which has a trend, and the same point is continuously monitored for 2-3 times to deviate from the threshold value, or different points have mild abnormalities for multiple times, or a single point deviates relatively obviously, for example, a single point exceeds the upper limit of the threshold value by 10% to 20%. After triggering the secondary response, the system will push more targeted guidance suggestions, generate a brief report for the user or user family members to view, and mark it as a need-to-attention event. In this embodiment, the more targeted guidance suggestions are adjusting the next meal carbohydrate, suggesting specific period activities, reminding to review blood glucose, and the like.

[0037] Severe abnormalities are a tertiary response, which is judged as a severe abnormality, and the blood glucose monitoring data needs to meet one of the following conditions: One, the night blood glucose monitoring data is less than 4.4 mol / L for 2 consecutive hours Two, the number of times of the night blood glucose monitoring data being less than 4.4 mol / L exceeds the normal frequency, which immediately triggers a hypoglycemia protocol; it should be noted that the normal frequency of the night blood glucose monitoring data being less than 4.4 mol / L is well known to those skilled in the art; Three, any one of the monitoring data of the fasting blood glucose in the morning, the blood glucose at any point, the postprandial blood glucose, or the post-exercise blood glucose continuously exceeds the upper limit value of the corresponding individualized safety threshold value by 20% for N times, where N is an integer not less than 2; Four, within a preset statistical period, the comprehensive blood glucose monitoring data of a specific key point or all key points exceeds the total time percentage of the corresponding individualized safety threshold value of each key point by more than a preset safety line threshold value; Specific risk patterns are explicitly identified through KFPP data analysis, such as night hypoglycemia leading to the phenomenon of Su Mu Jie, persistent dawn phenomenon, frequent post-exercise hyperglycemic rebound, and persistent high basal glucose levels above the baseline; the dawn phenomenon is the increase in blood glucose in the morning due to hormone secretion; the Su Mu Jie phenomenon is the rebound hyperglycemia after night hypoglycemia; after triggering a level 3 response, the system will automatically activate the early warning treatment protocol. The early warning treatment protocol in this embodiment includes sending a high-level alarm, locking or downgrading the light intervention mode function, triggering a switch back to the heavy intervention mode or a strong prompt to switch; specifically, a high-level alarm is immediately sent to the user, the user's family members, i.e., the medical personnel authorized by the user; the functions of the light intervention module are automatically locked or downgraded; the system is triggered to switch or strongly prompted to switch back to the heavy intervention mode, and the historical heavy intervention configuration of the user is loaded, and the previous personalized safety threshold library is used as a reference benchmark; preliminary emergency guidance is provided; the relevant FKPP abnormal data and triggering events are completely transferred to the heavy intervention module in the heavy intervention mode as the basis for restarting intensive management.

[0038] In the light intervention mode, if the user's blood glucose control indicators continue to meet the preset conditions, the system will fine-tune and optimize the personalized safety threshold library based on new monitoring data according to preset rules or professional instructions; it should be noted that the preset conditions are common technical knowledge for those skilled in the art, such as lowering the upper limit, and updating the storage; if the heavy intervention mode is triggered, when the heavy intervention mode is switched back due to the triggering of the early warning treatment protocol, the personalized safety threshold library is updated in the new round of heavy intervention mode, and the light intervention is inherited again.

[0039] It should be noted that all KFPP monitoring events, abnormal evaluation results and response actions triggered in the light intervention mode are recorded as continuous event streams in the database; when switching back to the heavy intervention mode, the events in the continuous event stream will be seamlessly loaded, providing a complete transition period history view for the AI system, intervention service personnel or medical personnel; and the user's individual characteristics, historical control targets, risk factors, past intervention records and other core archival information are completely shared between the light intervention mode and the heavy intervention mode, maintaining the continuity of the user's archives.

[0040] In the light intervention mode, based on the five-point monitoring scheduling engine module, the blood glucose data of the five key points are targetedly monitored and collected based on the physiological rhythm, and abnormalities deviating from the inheritance threshold are discovered in a timely manner; Based on the threshold storage and inheritance module, the personalized safety threshold library is automatically updated after the heavy intervention mode is switched to the light intervention mode, providing accurate personalized benchmarks for light intervention mode monitoring; Based on the hierarchical abnormality evaluation module and the cross-mode response linkage module, the hierarchical response strategy according to the point abnormality type and severity, realizes the regular response triggering mechanism, from the mild intervention mode to the linear penetration of the warning disposal protocol of the heavy intervention mode; Based on the central data management module, the personalized safety threshold, the blood glucose monitoring result and the warning record are ensured to be seamlessly shared between the mild intervention mode and the heavy intervention mode, the personalized safety threshold is seamlessly inherited from the heavy intervention mode to the mild intervention mode, and the management closed loop and the effect evaluation throughout the whole process are realized.

[0041] Example two: User A has type 2 diabetes, and has been transferred to the mild intervention mode after four weeks of the heavy intervention mode; in the heavy intervention mode stage, the safety threshold of the personalized safety threshold established in the night is 5.0-6.5 mmol / L, on the tenth day of the mild intervention, the system automatically schedules or user A actively requires to measure the night blood glucose by BGM at ten o'clock at night, i.e. before sleep, and at five thirty in the morning, i.e. before getting up, and the blood glucose of user A before sleep is 6.2 mmol / L, and the blood glucose before getting up is 3.8 mmol / L; The system evaluates user A, and the night blood glucose monitoring result of user A is 3.8 mmol / L, which meets the third level response condition; the system will immediately respond, i.e. activate the warning disposal protocol; Firstly, a strong vibration or ring alarm is sent to the mobile phone of Li, which is “warning, night hypoglycemia risk is detected! Please immediately supplement food equivalent to 15g fast sugar and retest blood glucose!” At the same time, the alarm and detailed data, i.e. time, value and threshold, are pushed to the family APP of user A and the care physician workstation; the “exercise plan” module in the mild intervention mode is automatically locked, and the system interface is obviously prompted to “detect serious abnormality, suggest to immediately contact the doctor or switch back to the intensive management mode”, finally, the complete record of this hypoglycemia event is recorded and synchronized to the central data management module for subsequent reference of the heavy intervention mode; user A timely handles it, avoids serious consequences, and according to the alarm information, it is suggested to temporarily switch back to the heavy intervention mode, and reevaluate the insulin dose and the pre-sleep meal plan.

[0042] Example three: user B has been transferred to the mild intervention mode, and in the heavy intervention mode stage, the safety threshold after meal in the personalized safety threshold library established by user B is less than 7.8 mmol / L; in one week, the three postprandial 2-hour blood glucose monitoring values recorded by the system are 8.5 mmol / L, 8.9 mmol / L and 9.1 mmol / L, all of which are greater than 7.8 mmol / L, and greater than 20% reach 8.8 mmol / L; the system evaluates that user B meets the third level response condition, and will automatically activate the warning disposal protocol; Firstly, the system sends an alarm to user B and the doctor "reminder: continuous postprandial blood glucose seriously exceeds the standard! Please review the recent diet as soon as possible and consult the management teacher." The system interface will prompt to switch back to the heavy intervention mode to carry out the "diet reconstruction module" for evaluation, and automatically generate a brief report containing the over-standard record. If user B confirms the switching or the doctor performs remote operation, the state will be switched back to the heavy intervention mode, and the historical configuration and this time trigger data will be loaded. User B will switch back to the heavy intervention mode under the guidance, find that the recent carbohydrate intake has increased significantly through the diet diary and intensive monitoring, and will adjust the diet plan and restore blood glucose.

[0043] The above embodiments are not a limitation of the present application, and the present application is not limited to the above examples. Changes, modifications, additions or substitutions made by those skilled in the art within the scope of the technical solutions of the present application also belong to the protection scope of the present application.

Claims

1. A diabetes blood glucose dynamic management system based on circadian rhythm segmentation, characterized in that, The threshold storage and inheritance module is used for storing the personalized safety threshold library of five key points of the heavy intervention mode and automatically loading and updating the personalized safety threshold library when switching to the light intervention mode; The five-point monitoring scheduling engine module is used for scheduling and executing the blood glucose data monitoring and collection of the five key points of night, morning fasting, any point, postprandial and post-exercise in the light intervention mode based on the physiological rhythm according to the preset rules; The hierarchical abnormality evaluation module is used for comparing the collected five-point monitoring data with the corresponding personalized safety threshold and performing hierarchical evaluation according to the point type, deviation degree and frequency; The cross-mode response linkage module is used for taking corresponding measures according to the hierarchical evaluation result; The central data management module is used for uniformly storing the blood glucose data, life events, personalized safety threshold library, monitoring scheduling record, abnormality evaluation result and response log, so as to ensure the coherence and accessibility of the data between the heavy intervention mode and the light intervention mode.

2. The method for dynamic blood glucose management of diabetes based on circadian segmentation, applied to the system for dynamic blood glucose management of diabetes based on circadian segmentation as claimed in claim 1, characterized in that, The method comprises the following steps: Step S1: based on the threshold storage and inheritance module, the blood glucose of the five key points is monitored in the heavy intervention mode based on the physiological rhythm; Step S2: the personalized safety threshold library is established, the monitoring data is stored in the personalized safety threshold library, and the personalized safety threshold library is automatically loaded and set as the blood glucose monitoring and evaluation reference of the light intervention mode; Step S3: after the user enters the light intervention mode, the blood glucose monitoring of the five key points is scheduled and executed based on the five-point monitoring scheduling engine module, the five-in-one targeted monitoring is performed, and the monitoring scheduling strategy decision is made; Step S4: if the monitoring data is normal, the current scheduling strategy is maintained; if the monitoring data is abnormal, the hierarchical abnormality evaluation module is used to respond to the monitoring data deviation of any key point from the corresponding personalized safety threshold, and the hierarchical evaluation is performed according to the point type, deviation degree and deviation frequency; Step S5: if the hierarchical evaluation result is mild abnormality or moderate abnormality, the cross-mode response linkage module is used to generate and push the corresponding guidance information in the light intervention mode; if the hierarchical evaluation result is severe abnormality, the cross-mode response linkage module is used to automatically trigger the early warning disposal protocol associated with the heavy intervention mode.

3. The circadian rhythm based segmented diabetes blood glucose dynamic management method of claim 2, wherein: In the step S1, the night blood glucose monitoring adopts two monitoring modes, the first mode is to use the CGM device to continuously monitor the glucose to obtain the whole night blood glucose trend data; the second mode is to perform two fingertip blood point glucose monitoring before going to bed and before getting up in the morning, and the average value of the two monitoring results is taken as the representative value.

4. The circadian rhythm based segmented diabetes blood glucose dynamic management method of claim 3, wherein: In the step S1, the morning fasting blood glucose monitoring is fingertip blood point glucose monitoring after getting up and before having breakfast; The arbitrary point blood glucose monitoring is fingertip blood point glucose monitoring at any time point after meal and after exercise; The postprandial blood glucose monitoring is fingertip blood point glucose monitoring at a specified time point after meal; The post-exercise blood glucose monitoring is fingertip blood point glucose monitoring at a specified time point after planned exercise.

5. The circadian rhythm based segmented diabetes blood glucose dynamic management method of claim 4, wherein: In the step S3, the five-point monitoring scheduling engine module is used to make the monitoring scheduling strategy decision, which specifically comprises the following steps: Step A1: the user enters the light intervention mode; Step A2: risk assessment is performed, and based on the risk assessment result, the system specifies a monitoring mode to monitor the user's blood glucose; Step A3: the monitoring result data is fed back to the evaluator, and it is judged whether the data is abnormal; if the data is not abnormal, the current monitoring scheduling strategy is maintained; if the data is abnormal, the dynamic adjustment strategy is triggered to perform reverse optimization scheduling of abnormal data. 6.The circadian rhythm segmented-based diabetes blood glucose dynamic management method according to claim 5, characterized in that: In step A2, based on the risk assessment result, the system specifies a monitoring mode to monitor the user's blood glucose, specifically including the following cases: If the risk assessment result is high risk, or the user is a new user, enter high-frequency FKPP monitoring mode, monitor blood glucose at five key points not less than 3 times a week; If the risk assessment result is stable state, enter low-frequency KFPP monitoring mode, monitor blood glucose at any point not less than 2 times a week; Monitor blood glucose at night and after meals each week; If the risk assessment result triggers a specific event, enter event-driven monitoring mode, and take corresponding measures according to the event type.

7. The circadian rhythm based segmented diabetes blood glucose dynamic management method of claim 6, wherein: The dynamic adjustment strategy in step A3 includes increasing the monitoring frequency of abnormal point by 50%, expanding the blood glucose monitoring of associated point, and triggering the hierarchical response protocol for hierarchical evaluation.

8. The circadian rhythm based segmented diabetes blood glucose dynamic management method of claim 7, wherein: In step S5, if it is judged as severe abnormality, the blood glucose monitoring data meets one of the following conditions: One, the night blood glucose monitoring data is less than 4.4 mol / L for 2 hours in a row; Two, the frequency of night blood glucose monitoring data less than 4.4 mol / L exceeds the normal frequency; Three, any one of the monitoring data of morning fasting blood glucose monitoring data, arbitrary point blood glucose monitoring data, postprandial blood glucose monitoring data or post-exercise blood glucose monitoring data exceeds 20% of the upper limit of the corresponding personalized safety threshold of each key point for N times in a row, where N is an integer not less than 2; Four, within a preset statistical period, the comprehensive blood glucose monitoring data of a specific key point or all key points exceeds the total time percentage of the corresponding personalized safety threshold of each key point, which exceeds the preset safety threshold.

9. The circadian rhythm segmented based diabetes blood glucose dynamic management method of claim 8, wherein: In step S5, in the light intervention mode, if the user's blood glucose control index is continuously stable and meets the preset condition, the personalized safety threshold library is fine-tuned and optimized based on the new blood glucose monitoring data and is updated and stored; When switching back to the heavy intervention mode due to triggering the early warning disposal protocol, the personalized safety threshold library is updated in the heavy intervention mode based on the threshold storage and inheritance module.

10. The circadian rhythm based segmented diabetes blood glucose dynamic management method of claim 9, wherein: The early warning disposal protocol includes sending a high-level alarm to the user, automatically locking the light intervention mode function, automatically degrading the light intervention mode function, triggering system switching or strongly prompting switching back to the heavy intervention mode instruction, and providing preliminary emergency guidance.

Images