Diabetic health data management method and system

Abstract

The invention discloses a diabetic health data management method and system, and relates to the technical field of data management. A diabetic health data management system comprises a health data monitoring module and a health data management module. According to the method, blood glucose fluctuation is considered, multi-dimensional signals such as a molecular marker, a sleep cycle and bioelectricity feedback are introduced, and a patient molecular physiological interaction index is established through mapping of a unified time axis and correction of a causal inference model; misjudgment caused by accidental fluctuation is effectively reduced by capturing cross-channel collaborative abnormity; according to the method, the patient individual metabolism adaptation domain and the similar patient metabolism adaptation domain are constructed, a double-gating judgment mode is adopted, excessive early warning of health data management caused by a single threshold standard is avoided, and personalized applicability under patient individual difference is guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data management, in particular to a diabetes patient health data management method and system. BACKGROUND

[0002] Existing diabetes patient health management methods mainly rely on single means such as blood glucose monitoring, drug management and lifestyle intervention. Although these methods can monitor the health status of patients to some extent, they usually have the following shortcomings: first, excessive reliance on blood glucose value as the only indicator, without fully considering the potential impact of other physiological signals on blood glucose fluctuations; second, most existing methods do not consider the lag effect of physiological signals, resulting in delayed disease judgment and early warning.

[0003] Therefore, there is a need for a diabetes health data management method based on multi-dimensional signal fusion and causal inference, which significantly improves the accuracy of diabetes risk assessment and health management. SUMMARY

[0004] The present application aims to provide a diabetes patient health data management method and system that not only improves the accuracy of health risk assessment, but also realizes personalized intervention, significantly improving the long-term management effect and quality of life of diabetes patients.

[0005] A diabetes patient health data management method, comprising the following steps: Correlate the blood glucose fluctuations and molecular marker concentration fluctuations of the patient obtained according to the set monitoring period to obtain the molecular regulation ratio parameter in the patient's body; obtain the sleep cycle fluctuation parameter and the bioelectric feedback fluctuation parameter of the patient in real time; construct the patient's molecular physiological interaction index according to the molecular regulation ratio parameter, the sleep cycle fluctuation parameter and the bioelectric feedback fluctuation parameter; Obtain the original parameter monitoring data of the patient; judge the original parameter monitoring data and the patient's individual metabolic adaptation domain and the metabolic adaptation domain of the same type of patients to obtain the patient's health data abnormal result; if the patient's health data abnormal result is abnormal, perform diabetes health abnormality judgment on the patient's molecular physiological interaction index, and output the patient's diabetes health risk index; continue to execute the corresponding diabetes monitoring and management strategy based on the patient's diabetes health risk index; if the patient's health data abnormal result is normal, continue to manage the health data of the diabetes patient.

[0006] As a preferred technical solution of the present application, the specific steps of constructing the patient's molecular physiological interaction index according to the molecular regulation ratio parameter, the sleep cycle stability parameter and the bioelectric feedback fluctuation parameter include: Correlate the blood glucose fluctuation and the molecular marker concentration fluctuation in time sequence to obtain a plurality of phase segments; wherein, the phase segment contains a before-and-after meal segment, a sleep cycle segment, a before-and-after meal segment and a before-and-after medication segment; the change amplitude of the blood glucose fluctuation and the molecular marker concentration fluctuation in the phase segment is extracted respectively to obtain a corresponding molecular regulation ratio parameter; Calculate the rhythm consistency of the sleep cycle fluctuation parameter to obtain a sleep time adjustment factor; perform mode recognition on the bioelectric feedback fluctuation parameter to extract bioelectric abnormal fluctuation characteristics in different nerve states; synchronize map the molecular regulation ratio parameter, the sleep time adjustment factor and the bioelectric abnormal fluctuation characteristics on a unified time axis to obtain a multi-dimensional blood glucose-molecule interaction matrix; Generate a patient molecular physiological interaction index based on the synchronization deviation degree of different dimension characteristics in the multi-dimensional blood glucose-molecule interaction matrix.

[0007] As a preferred technical solution of the present application, the specific steps of judging the original parameter monitoring data and the patient individual metabolic adaptation domain and the same type patient metabolic adaptation domain include: Based on the parameter monitoring data of the patient in a period of time, stratify according to the diurnal phase and physiological state to obtain a plurality of phase-state segments; obtain the key event anchor point and the patient work-rest medication rhythm based on the original parameter monitoring data; calculate the parameter stable interval and the parameter fluctuation profile for each phase-state segment as a key segment adaptation domain with the key event anchor point as the boundary; calculate the data quality weight of each key segment adaptation domain; sort each key segment adaptation domain according to the patient work-rest medication rhythm and the data quality weight to obtain a patient individual metabolic adaptation domain based on time sequence; Based on the publicly available diabetes patient parameter database; match the corresponding fingerprint characteristics of the patient to output a behavior-physiology twin patient group; construct a same type patient metabolic adaptation domain based on the behavior-physiology twin patient group; Judge the original parameter monitoring data in the patient individual metabolic adaptation domain and the same type patient metabolic adaptation domain to obtain a patient health data abnormal result.

[0008] As a preferred technical solution of the present application, the specific steps of judging the patient molecular physiological interaction index for diabetes health abnormality include: Identify the fluctuation lag intensity based on the patient molecular physiological interaction index and the causal inference model; correct the patient molecular physiological interaction index according to the fluctuation lag intensity to obtain a corrected patient molecular physiological interaction index; use the abnormal coupling effect judgment mechanism to judge the diabetes health abnormality of the corrected patient molecular physiological interaction index to obtain a patient diabetes health risk index; Specifically, the revised patient molecular physiological interaction index is divided into N types of revised diabetes abnormal indexes; the contribution degree of each type of revised diabetes abnormal index is calculated; if the N types of revised diabetes abnormal indexes exist in the same time window and the contribution degrees exceed the preset threshold, it is determined that the patient's diabetes health risk index is a diabetes compound risk; if only part of the revised diabetes abnormal indexes exist in the same time window or any contribution degree is lower than the preset threshold, it is determined that the patient's diabetes health risk index is a diabetes observation risk.

[0009] As a preferred technical solution of the present application, the specific steps of continuing to execute the corresponding diabetes monitoring and management strategy based on the patient's diabetes health risk index include: When the patient is in the diabetes compound risk, a high-frequency data acquisition mode is executed on the patient, and the length and density of the monitoring cycle are shortened; When the patient is in the diabetes observation risk, at least two set monitoring cycles are shortened; if the patient is not in the diabetes compound risk or the diabetes observation risk in two consecutive monitoring cycles, the set monitoring cycle is restored; otherwise, the high-frequency data acquisition mode is executed on the patient.

[0010] As a preferred technical solution of the present application, the abnormal coupling effect determination mechanism is based on a causal inference algorithm.

[0011] A diabetes patient health data management system, comprising: A health data monitoring module comprising a data monitoring unit; the data monitoring unit is used for correlatively calculating the blood glucose fluctuation and the molecular marker concentration fluctuation of the patient according to a set monitoring cycle to obtain a molecular regulation ratio parameter in the patient; real-time acquisition of the sleep cycle fluctuation parameter and the bioelectric feedback fluctuation parameter of the patient; construction of a patient molecular physiological interaction index according to the molecular regulation ratio parameter, the sleep cycle fluctuation parameter and the bioelectric feedback fluctuation parameter; A health data management module comprising a diabetes abnormality judgment unit and a diabetes health judgment unit; the diabetes abnormality judgment unit is used for acquiring the original parameter monitoring data of the patient; the original parameter monitoring data and the individual metabolic adaptation domain of the patient and the metabolic adaptation domain of the same type of patient are judged to obtain a patient health data abnormality result; the diabetes health judgment unit is used for performing diabetes health abnormality judgment on the patient molecular physiological interaction index if the patient health data abnormality result is abnormal, and outputting a patient diabetes health risk index; continuing to execute the corresponding diabetes monitoring and management strategy based on the patient's diabetes health risk index; if the patient health data abnormality result is normal, the diabetes patient health data management is continued.

[0012] The present application has the following advantages: 1、The present application not only considers blood glucose fluctuations, but also introduces multi-dimensional signals such as molecular markers, sleep cycles and bioelectric feedback, and through the mapping of a unified time axis and the modification of a causal inference model, a patient molecular physiological interaction index is established; compared with single blood glucose monitoring, this method can capture cross-channel synergistic abnormalities and effectively reduce misjudgments caused by incidental fluctuations; by constructing a patient individual metabolic adaptation domain and a same kind patient metabolic adaptation domain, a double-gated decision mode is adopted, and only when the individual and the group domains simultaneously appear a deviation in the same direction and are significant, the abnormality is determined, which avoids the excessive warning caused by a single threshold standard and ensures the personalized applicability under individual differences.

[0013] 2、The present application identifies the delayed effect of molecular markers, sleep rhythm and neural activity on blood glucose by constructing a causal inference model, and automatically adjusts the time window and intensity, overcoming the defect that traditional real-time monitoring ignores physiological hysteresis, making risk identification more in line with pathological process and human body reaction rules; by dividing the modified interaction index into multiple modified abnormality indexes and calculating the contribution degree respectively, hierarchical output is realized; not only can potential serious complications be captured in time, but also mild abnormalities can be avoided from being over-interpreted, improving the clinical value of health data management. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 A structural schematic diagram of a diabetes patient health data management system used in the embodiments of the present application. DETAILED DESCRIPTION

[0015] In order to enable personnel in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application.

[0016] Embodiment 1, a diabetes patient health data management method, comprising the following steps: According to the set monitoring period, the blood glucose fluctuations and the molecular marker concentration fluctuations of the patient are obtained for correlation calculation, to obtain the molecular regulation ratio parameter in the patient's body; the sleep cycle fluctuation parameter and the bioelectric feedback fluctuation parameter of the patient are obtained in real time; the patient molecular physiological interaction index is constructed according to the molecular regulation ratio parameter, the sleep cycle fluctuation parameter and the bioelectric feedback fluctuation parameter; Specifically, for diabetic patients who need health data management, the continuous blood glucose fluctuation data and the concentration fluctuation data of multiple molecular markers of the patient are obtained according to the set monitoring period. The blood glucose fluctuation refers to the dynamic change of the blood glucose level of the patient within a period of time, including the blood glucose values at different time points such as fasting and postprandial, and also including the rising speed, the falling amplitude and the difference between the peak and the trough, reflecting the immediate response of the body to various factors such as diet, exercise and medication. The concentration fluctuation of the molecular marker refers to the change in the in-vivo concentration of a specific molecule (such as FXNIP, XIAP, ZBP1, etc.) related to glucose metabolism, oxidative stress or inflammatory response in time series. The increase or decrease in the level of these molecules often occurs before the change in blood glucose, which can reveal potential abnormalities in metabolic regulation or cell signaling.

[0017] The specific steps of constructing the patient molecular physiological interaction index according to the molecular regulation ratio parameter, the sleep cycle stability parameter and the bioelectric feedback fluctuation parameter include: The blood glucose fluctuation and the concentration fluctuation of the molecular marker are associated in time series to obtain a plurality of phase segments. The phase segments include pre- and post-meal segments, sleep cycle segments, pre- and post-meal segments and pre- and post-medication segments. The change amplitudes of the blood glucose fluctuation and the concentration fluctuation of the molecular marker in the phase segments are extracted respectively to obtain corresponding molecular regulation ratio parameters. The input blood glucose fluctuation and the concentration fluctuation of the molecular marker are preferentially cleaned, de-noised and time-aligned to eliminate the influence of sampling frequency difference, packet loss and abnormal values. Anchor points are established according to meal, sleep and wake-up, medication and other event time stamps. The two types of sequences are synchronously unfolded along the same time axis, and soft alignment is performed in combination with possible physiological lag, so that the two types of signals can present the linkage characteristics in a comparable manner in the key physiological stages. The phase segments include at least pre- and post-meal segments, sleep cycle segments and pre- and post-medication segments. The pre- and post-meal segments are centered on meal time, and fixed or adaptive observation windows are set forward and backward, covering the pre-meal baseline, post-meal rise, peak and fall stages. The sleep cycle segments are mainly divided into deep sleep, light sleep and rapid eye movement stages, and short awakenings at night are separately marked to avoid diluting the rhythm information. The pre- and post-medication segments are divided by the medication time, and observation windows are set in combination with the onset and maintenance time of the drug action, and are respectively summarized under different dosage forms or doses to ensure that the dose and efficacy environment within the segment are consistent. Each type of phase segment retains event anchor points, duration, data quality scores and alignment methods, etc. metadata, providing a basis for subsequent calculation and verification.

[0018] The amplitude of the blood glucose fluctuation and the concentration fluctuation of the molecular marker and their key kinetic elements, including the rising and falling slopes, the amplitude difference of the peak and valley, the timing of the peak and the falling speed, etc., are extracted from each time phase segment, and the two types of features in the same time window are paired; at the same time, in order to reduce the interference of non-physiological factors, the eating segment is normalized according to the carbohydrate intake, the drug segment is normalized according to the actual dose and dosage form, and the sleep segment is normalized according to the stability of the rhythm of the night; finally, the paired features are compressed into a molecular regulation ratio parameter that can reflect the strength and direction of the response of the molecule to the blood glucose.

[0019] The rhythm consistency of the sleep cycle fluctuation parameter is calculated to obtain a sleep time adjustment factor; the bioelectric feedback fluctuation parameter is subjected to pattern recognition to extract abnormal bioelectric fluctuation characteristics in different neural states; the molecular regulation ratio parameter, the sleep time adjustment factor and the abnormal bioelectric fluctuation characteristics are synchronously mapped on a unified time axis to obtain a multi-dimensional blood glucose-molecule interaction matrix; The rhythm consistency of the patient is evaluated based on the sequence of sleep stages throughout the night and multiple days (including the time of falling asleep and getting up, the proportion of deep sleep, light sleep and rapid eye movement stage, the number and duration of night awakenings), specifically, according to the working attributes or other sleep characteristics of the patient, to avoid misjudgment of abnormality caused by shift work, cross-time zone or weekend schedule changes, the system first identifies the long-term work and rest rhythm and the biological clock phase of the patient, then measures the amplitude of the recent phase shift from the phase, the stability of the stage transition and the disturbance intensity of the night awakening on the stage structure, and models and analyzes the differences between weekdays and weekends respectively, and outputs a sleep time adjustment factor for describing the modulation effect of the sleep rhythm on the metabolic response strength; Meanwhile, the biofeedback fluctuation parameters are subjected to pattern recognition. The data sources of the biofeedback fluctuation parameters include inter-beat interval variation, skin electricity level and instantaneous reaction, peripheral skin temperature and respiratory rhythm signals. The signals are first corrected for motion artifact and baseline drift, and then segmented according to the phase and scene (rest, postprandial, exercise recovery, night sleep). In each phase segment, the characteristic patterns such as sustained high sympathetic activity, slow recovery, high wakefulness burst not matching the sleep stage, and night autonomic imbalance are identified, and compared with the same scene data of the patient in the past, to screen out abnormal fluctuations with sustained or clustered occurrence. Finally, the bioelectricity abnormal fluctuation characteristics are output, which include abnormal type, duration, occurrence frequency and reliability, to mark the potential disturbance of neural activity on metabolism. After the extraction of the three types of multi-dimensional information of molecules, sleep and bioelectricity, the synchronous mapping is performed on the unified time axis. The specific method is to resample each channel to the same time grid, and align them with the events such as eating, taking medicine, falling asleep, waking up, starting and ending exercise as anchor points. If there is a missing interval, it is conservatively interpolated and the weight is reduced. The abnormal segments retain the source and quality label, thereby constructing a multi-dimensional blood glucose-molecule interaction matrix. The rows in the matrix correspond to time steps, and the columns correspond to molecular regulation ratios, sleep time adjustment factors and bioelectricity abnormal fluctuations. The matrix unit also saves the alignment method, event label and data confidence, to ensure the interpretability of subsequent judgment.

[0020] Based on the degree of synchronous deviation of different dimensional characteristics in the multi-dimensional blood glucose-molecule interaction matrix, a patient molecular physiological interaction index is generated. In the generation process, the same direction or complementary type of collaborative deviation across channels is preferentially identified, the isolated abnormality in a single channel is filtered, and a hysteresis threshold is introduced to avoid alarm jitter. The collaborative deviation with longer duration, larger amplitude and more consistent evidence is given higher contribution. The signals from low-quality segments are automatically weighted. The patient molecular physiological interaction index output includes the overall grade and channel contribution explanation, and is accompanied by a confidence range and a summary of triggered event anchor points. It can be directly used for triggering and grading abnormal coupling effects, or used as a basis for adjusting the evaluation period and intervention timing in subsequent closed-loop optimization.

[0021] Obtain the original parameter monitoring data of the patient; judge the original parameter monitoring data and the individual metabolic adaptation domain of the patient, the metabolic adaptation domain of the same type of patient, to obtain the patient health data abnormal result; if the patient health data abnormal result is abnormal, perform diabetes health abnormality judgment on the patient molecular physiological interaction index, and output the patient diabetes health risk index; continue to execute the corresponding diabetes monitoring and management strategy based on the patient diabetes health risk index; if the patient health data abnormal result is normal, continue to manage the health data of the diabetes patient.

[0022] The specific steps of judging the original parameter monitoring data and the individual metabolic adaptation domain of the patient, the metabolic adaptation domain of the same type of patient, include: Based on the parameter monitoring data of the patient in the recent period, a plurality of phase-state segments are obtained by stratification according to the diurnal phase and physiological state; key event anchor points and patient work-rest medication rhythms are obtained based on the original parameter monitoring data; parameter stability intervals and parameter fluctuation profiles are calculated for each phase-state segment as a key segment adaptation domain, with the key event anchor points as boundaries; data quality weights of each key segment adaptation domain are calculated; each key segment adaptation domain is sorted according to the patient work-rest medication rhythm and data quality weight, to obtain a time-based patient individual metabolic adaptation domain; After obtaining the continuous original parameter monitoring data of the patient, the multi-channel data is preprocessed and aligned based on a unified time axis, including blood glucose sequence, molecular marker sequence, sleep stage sequence and bioelectric signal sequence; in order to ensure the naturalness of the segment boundary, the whole period is first divided into morning, daytime, night and pre-sleep phase according to the diurnal rhythm, and then the information of physiological state is added for detailed segmentation, for example, the signal pattern change of adjacent intervals is identified with eating, exercise, sleep, wake-up, medication and other events as boundaries, and the stability of the cutting point is confirmed by combining variable point detection and inflection point search; for data across days or weeks, segments of the same type are combined into the same category, and short or long segments are subjected to minimum and maximum length constraints, so that each phase-state segment can cover the complete physiological process without fragmentation due to noise; At the same time of forming the phase-state segment, key event anchor points and patient work-rest medication rhythms are automatically extracted based on the original parameter monitoring data; the key event anchor points are derived from the cross-verification of multiple sources of records; for example: wearable monitoring devices provide sleep, wake-up and night wake-up times; diet records and pre-and post-meal blood glucose jumps determine meal times and meal information; medicine boxes or mobile phone reminder records are used to locate the drug administration time and dose with blood drug effect timing; exercise equipment and geographic positioning suggest exercise start and end times and intensity; these events are concatenated in chronological order, and their repeatability and offset in a week and a month are calculated to extract the patient's work-rest rhythm and medication cycle, which are used as a time reference in subsequent segment matching and comparison; With key event anchor points as boundaries, the parameter stable interval and parameter fluctuation profile of each phase-state segment are calculated; the specific steps are to perform robust denoising and short window smoothing on the sequence of blood glucose and molecular markers within the segment, estimate the central level and upper and lower confidence bands of the segment, and then extract the kinetic elements such as rising slope, peak position, peak-valley difference, falling speed, and plateau duration to form a complete fluctuation topography description; sleep and bioelectric information are synchronously included to evaluate the stability of stage transition and the stress level of the autonomic nervous system to correct the modulation background of the segment on metabolic response; the parameter stable interval obtained is used to define the range that the segment should present under normal conditions, and the parameter fluctuation profile is used to describe the reasonable change path that the segment may appear when driven by events, and the two together constitute the key segment adaptation domain of the segment; for each key segment adaptation domain, its corresponding data quality weight is calculated and assigned to reflect its credibility and contribution in subsequent judgment; the data quality weight is formed by multiple factors: for example, the sampling completeness is used to measure the proportion of data missing and continuity, the signal-to-noise ratio and artifact rate are used to measure motion interference and sensor drift, or the cross-device consistency is used to measure the degree of agreement of different source data in the overlapping period, the time alignment confidence measures the synchronization accuracy of multiple channels in the segment, and the event label reliability measures the consistency of the segment and the recorded event; In addition, the influence of external environment and non-regular situation (such as cross-time zone travel, night shift work, acute illness) is also considered, and the related segments are automatically down-weighted or marked as delayed effective to avoid the pollution of abnormal conditions to the regular threshold; the final quality weight is normalized to a continuous quantity between zero and one, and is archived together with the segment adaptation domain.

[0023] After identifying the key segment adaptation domain and quality weight, the individual metabolic adaptation domain is constructed according to the patient's work and rest and drug rhythm, which is time-sequenced and oriented to the daily schedule; the implementation is to arrange the segments within a day and a week according to the sequence of events and biological priority, the priority rules include that drug effects are prior to postprandial effects, postprandial effects are prior to general exercise, and sleep stages impose monitoring background adjustment on all processes; when there is overlap or conflict between segments, the one with higher quality weight is dominant, or it is divided into two consecutive sub-segments in time sequence to retain their respective characteristics; then the continuous adaptation domain time band is generated on the time axis, so that any time can quickly retrieve the matching segment adaptation domain and its quality weight, and automatically inherit the differences between weekdays and weekends in the time axis; the time-sequenced patient individual metabolic adaptation domain formed in this way can be used as the immediate control of the original monitoring data and the first layer of reference for subsequent abnormal gating, which not only reflects the individual long-term stable law, but also can adaptively update with the changes of the patient's recent state.

[0024] Based on the publicly available diabetes patient parameter database; match the corresponding fingerprint features of the patient, output the behavior-physiological twin patient group; based on the behavior-physiological twin patient group, construct the same patient metabolic adaptation domain; the corresponding fingerprint features of the patient are unique feature patterns after compression and combination of the multi-dimensional health data of the patient, the individualized fingerprint features of the patient, which can reflect the metabolic characteristics and life rhythm, and can be similar to the features of other patients in the database; The sources of the diabetes patient parameter database mainly include various publicly available or ethically approved medical and scientific data resources; for example, long-term follow-up data from clinical trials and cohort studies, which often record the patient's blood glucose level, medication, complication development and lifestyle characteristics; based on the diabetes patient parameter database, the behavior-physiological fingerprint features of the target patient (such as sleep structure ratio, postprandial blood glucose response delay pattern, medication compliance rhythm and bioelectric feedback morphology) are matched, and a group of highly similar behavior-physiological twin patient groups in multi-modal characteristics are screened out as group data indicators for reference; Within the behavior-physiological twin patient group, further secondary division and adaptation domain reconstruction are carried out according to the phase-state segment, the stable interval and fluctuation profile of each segment are weighted and integrated, and the bias introduced by region, season or equipment difference is corrected; the finally formed same patient metabolic adaptation domain can reflect the overall law of the group, and can also retain the key phase dependence, providing a dynamic and personalized group reference for the abnormal deviation judgment of the target patient.

[0025] Based on the original parameter monitoring data, the patient health data abnormality result is obtained in the patient individual metabolic adaptation domain and the same patient metabolic adaptation domain. The raw parameter monitoring data newly entered is subjected to data normalization processing, and first, a first round of comparison is completed in the patient individual metabolic adaptation domain. Specifically, the key indicators (including the rising and falling characteristics of blood glucose patterns, the change intensity and direction of molecular regulation ratio parameters, the modulation degree of sleep rhythm on metabolism, and the neural activity abnormalities shown by bioelectric signals) around the current segment are matched with the stable interval and fluctuation profile corresponding to the segment one by one, the deviation direction, amplitude and duration are calculated, and the individual domain deviation score and its confidence are formed in combination with the data quality weight of the segment. Then, a second round of comparison is performed in the same patient metabolic adaptation domain. In order to eliminate environmental and equipment differences, the raw data is first corrected in terms of region, season and caliber. Then, under the same phase-state, the patient indicators are compared with the distribution band, typical fluctuation curve and typical response time of the twin population one by one, and the group domain deviation score and its confidence are obtained. The patient individual metabolic adaptation domain is set as the first gate, and the same patient metabolic adaptation domain is set as the second gate. Only when the two gates are opened at the same time in the same observation window, the deviation directions are consistent, the amplitudes reach the respective minimum effective thresholds, and the shortest duration requirements are met, can it be determined as a qualified trigger, that is, the patient health data abnormality result is normal. If only one gate is triggered or the directions of the two gates are opposite, it is marked as non-abnormal, that is, the patient health data abnormality result is abnormal.

[0026] The specific steps for the patient molecular physiological interaction index to judge the diabetes health abnormality include: Based on the patient molecular physiological interaction index and the causal inference model, the fluctuation lag intensity is identified. The patient molecular physiological interaction index is corrected according to the fluctuation lag intensity to obtain a corrected patient molecular physiological interaction index. The corrected patient molecular physiological interaction index is used to determine the diabetes health abnormality by using the abnormal coupling effect judgment mechanism to obtain the patient diabetes health risk index. Specifically, the corrected patient molecular physiological interaction index is divided into N types of corrected diabetes abnormality indicators. The contribution degree of each type of corrected diabetes abnormality indicator is calculated. If the N types of corrected diabetes abnormality indicators exist in the same time window and the contribution degrees all exceed the preset threshold, it is determined that the patient diabetes health risk index is a diabetes compound risk. If only part of the corrected diabetes abnormality indicators exist in the same time window or any contribution degree is lower than the preset threshold, it is determined that the patient diabetes health risk index is a diabetes observation risk. The preset threshold is set artificially by a professional technician. The revised patient molecular physiological interaction index is grouped by three layers of tags, wherein the first group of tags is molecule, sleep, and nerve; the second group of tags is post-meal, night, exercise recovery, and post-drug; and the third group of tags is fast-acting, slow-accumulating, and repeated fluctuation. Under the same tag, similar segments are aggregated into N types of revised diabetes abnormal indicators according to the amplitude, duration, and frequency of occurrence. Then, the contribution degree of each type in the selected time window is calculated. First, the average amplitude and coverage duration of the abnormality of the type are evaluated. Then, the consistency of the synchronous deviation from other channels, the deviation strength relative to the same group of people, the physiological matching degree with the event anchor points (eating, taking medicine, falling asleep, etc.), and the data quality and delayed alignment credibility of the segment are superimposed to form a standardized score as the contribution degree. In the same time window, the scores of all types are summarized and checked whether they all exceed the preset threshold. If all of them meet the requirement, it is judged as a composite risk. If only part of them meets the requirement or any contribution degree is insufficient, it is judged as an observation risk. At the same time, the ranking of each type of score is retained for subsequent intervention priority.

[0027] The causal inference model is obtained based on historical data training. The patient molecular physiological interaction index is input into the trained causal inference model to determine the fluctuation lag intensity of the current monitoring period. The fluctuation lag intensity refers to the strength of the delayed effect of a certain type of physiological signal (such as a molecular marker level, a sleep rhythm parameter, or a bioelectric feedback feature) change on blood glucose fluctuation. It not only indicates the causal relationship between the signal and blood glucose, but also quantifies the lag degree and influence amplitude of the relationship in time. For example, the rise of a certain molecular marker may not trigger a significant increase in blood glucose until several hours later. If this effect occurs consistently and stably in multiple time windows, the fluctuation lag intensity is determined to be high. Conversely, if the effect is weak or unstable across different windows, the intensity is low. The fluctuation lag intensity aims to avoid misjudgment based solely on immediate values and more accurately reveal potential pathological mechanisms and individual differences.

[0028] The specific steps for training the causal inference model include: preprocessing the long-term collected blood glucose sequence, molecular marker concentration curve, sleep quality record, and bioelectric feedback signal, including time alignment, missing value interpolation, and outlier removal; constructing a candidate causal structure diagram, taking different physiological variables as nodes, and setting possible causal paths according to event anchor points such as eating, taking medicine, and falling asleep; using statistical learning methods to fit the causal relationship on historical data, identifying the sequence and lag between signals; introducing cross-validation and counterfactual testing during training to ensure that the model not only fits the data but also maintains interpretability in unseen time periods; the model outputs the causal strength and time delay distribution between variables, and continuously updates the parameters through new data to achieve adaptive optimization, thereby more accurately depicting the physiological regulation mechanisms of diabetes patients in different states.

[0029] The time and weight of the patient molecular physiological interaction index are double modified according to the fluctuation lag strength; the specific steps are as follows: the time offset is applied to different channels in the identified delay window, and the truly causal significance linkage segment is realigned; the synchronous segment which does not match the physiological time sequence or has low reliability is given weight or removed; the weight of the evidence segment which is in the high-quality sampling interval and matches the event anchor point is moderately increased; after obtaining the modified molecular physiological interaction index, the abnormal coupling effect judgment mechanism is called to carry out the diabetes health abnormality judgment; the abnormal coupling effect judgment mechanism is set based on the causal inference algorithm; the abnormal coupling effect judgment mechanism is used to identify the abnormal linkage between multiple physiological signals (such as molecular markers, blood glucose, sleep rhythm and neural activity) in the patient's body; when multiple signals simultaneously occur abnormally and interact under a certain time delay, the mechanism can judge whether it is a single indicator fluctuation or a composite risk with clinical significance; first, the causal inference method is used to identify the lag relationship between signals, determine the time sequence and causal strength of different channels; then, the real-time monitoring data are compared with the individual and group health reference domain to screen whether there are multiple signals in the same time window that are consistent in direction, significant in amplitude and sufficient in duration; if the conditions are met, the abnormality judgment is triggered, and further classification is made into observation risk or composite risk; if the signals are inconsistent or insufficient to support the causal relationship, it remains in the observation state; the setting of the abnormal coupling effect judgment mechanism is based on the causal inference algorithm, combined with historical data and prior physiological knowledge, to construct a causal graph of multiple signals and estimate their delay effect.

[0030] The specific steps of continuing to execute the corresponding diabetes monitoring and management strategy based on the patient's diabetes health risk indicators include: When the patient is in the diabetes composite risk, the high-frequency data acquisition mode is executed for the patient, and the length and density of the monitoring cycle are shortened; wherein the setting of the monitoring cycle is manually set by professional technicians; When the patient is in the diabetes observation risk, at least two set monitoring cycles are shortened; if the patient is not in the diabetes composite risk or the diabetes observation risk in two consecutive monitoring cycles, the set monitoring cycle is restored; otherwise, the high-frequency data acquisition mode is executed for the patient.

[0031] Specifically, in one embodiment, patient A is determined to be a diabetes compound risk in a continuous monitoring period; the system automatically starts a high-frequency data acquisition mode, adjusts the original once-a-day blood glucose monitoring (i.e., the set monitoring period) to once every four hours, simultaneously collects molecular markers and bioelectric feedback signals in real time, and shortens the statistical interval of sleep cycle evaluation to capture subtle metabolic abnormalities and hysteresis effects; in this high-density mode, the patient's risk state is quickly verified and updated, and the professional technical personnel synchronously receive an early warning report containing an evidence chain for developing a personalized intervention plan; in another embodiment, patient B is determined to be a diabetes observation risk in a certain monitoring period; the system does not immediately enter high-frequency acquisition, but shortens the evaluation window from once a day to twice a day in the next monitoring period, and performs more intensive statistics on sleep rhythm fluctuations and blood glucose responses; if the patient's risk results do not trigger compound risk or observation risk in the next two monitoring periods, the system automatically returns to the regular monitoring period, i.e., once-a-day blood glucose acquisition and nightly sleep parameter recording; if the risk signal persists or intensifies during this period, the system immediately upgrades to a high-frequency data acquisition mode to ensure that potential disease deterioration can be captured and managed in a timely manner.

[0032] Embodiment 2, a diabetes patient health data management system, see Figure 1 as shown, comprising: a health data monitoring module, comprising a data monitoring unit; the data monitoring unit is used for obtaining the blood glucose fluctuation and the molecular marker concentration fluctuation of the patient according to the set monitoring period for correlation calculation, obtaining the molecular regulation ratio parameter in the patient's body; real-time acquisition of sleep cycle fluctuation parameters and bioelectric feedback fluctuation parameters of the patient; constructing the patient's molecular physiological interaction index according to the molecular regulation ratio parameter, the sleep cycle fluctuation parameter and the bioelectric feedback fluctuation parameter; a health data management module, comprising a diabetes abnormality judgment unit and a diabetes health judgment unit; the diabetes abnormality judgment unit is used for obtaining the original parameter monitoring data of the patient; the original parameter monitoring data and the individual metabolic adaptation domain of the patient, the metabolic adaptation domain of the same type of patient are judged to obtain the patient health data abnormality result; the diabetes health judgment unit is used for if the patient health data abnormality result is abnormal, the diabetes health abnormality judgment of the patient's molecular physiological interaction index is carried out, and the patient's diabetes health risk index is output; based on the patient's diabetes health risk index, the corresponding diabetes monitoring and management strategy is continued to be executed; if the patient health data abnormality result is normal, the diabetes patient health data management is continued.

[0033] It should be understood that those skilled in the art can make improvements or changes according to the above description, and all these improvements and changes shall belong to the protection scope of the appended claims of the present application. The parts not described in detail in the specification belong to the prior art known to those skilled in the art.

Claims

1. A method for managing health data of diabetic patients, characterized in that, Includes the following steps: The patient's blood glucose fluctuations and molecular marker concentration fluctuations are acquired according to a set monitoring period and correlated to obtain the molecular regulation ratio parameter in the patient's body; the patient's sleep cycle fluctuation parameters and bioelectrical feedback fluctuation parameters are acquired in real time; and the patient's molecular physiological interaction index is constructed based on the molecular regulation ratio parameter, sleep cycle fluctuation parameter, and bioelectrical feedback fluctuation parameter. Obtain the patient's raw parameter monitoring data; The raw parameter monitoring data, along with the individual metabolic adaptation domain of the patient and the metabolic adaptation domain of similar patients, are used to determine abnormal results in the patient's health data. If the patient's health data results are abnormal, the patient's molecular physiological interaction index is used to determine the abnormality of diabetes health and output the patient's diabetes health risk index. Continue to implement the corresponding diabetes monitoring and management strategy based on the patient's diabetes health risk indicators; If the patient's health data shows abnormal results but is otherwise normal, then the management of the diabetic patient's health data will continue.

2. The method for managing health data of diabetic patients according to claim 1, characterized in that, The specific steps for constructing a patient's molecular physiological interaction index based on molecular regulation ratio parameters, sleep cycle stability parameters, and bioelectrical feedback fluctuation parameters include: Blood glucose fluctuations and molecular marker concentration fluctuations were correlated by time series to obtain several time segments; these time segments included pre- and post-meal segments, sleep cycle segments, pre- and post-meal segments, and pre- and post-medication segments; the amplitude of changes in blood glucose fluctuations and molecular marker concentration fluctuations in each time segment was extracted to obtain the corresponding molecular regulation ratio parameters. The rhythmic consistency of sleep cycle fluctuation parameters was calculated to obtain the sleep time regulation factor; pattern recognition was performed on the bioelectric feedback fluctuation parameters to extract abnormal bioelectric fluctuation features under different neural states; molecular regulation ratio parameters, sleep time regulation factor and abnormal bioelectric fluctuation features were synchronously mapped on a unified time axis to obtain a multidimensional blood glucose-molecular interaction matrix. Based on the degree of synchronization deviation of different dimensions of features in the multidimensional glucose-molecular interaction matrix, a patient molecular physiological interaction index is generated.

3. The method for managing health data of diabetic patients according to claim 2, characterized in that, The specific steps for determining the raw parameter monitoring data and the individual metabolic adaptation domain of a patient, as well as the metabolic adaptation domain of similar patients, include: Based on the patient's parameter monitoring data over a recent period, the data is stratified according to diurnal time phases and physiological states to obtain several time-state segments. Key event anchors and patient circadian rhythms are obtained based on the original parameter monitoring data. Using the key event anchors as boundaries, parameter stability intervals and parameter fluctuation profiles are calculated for each time-state segment, serving as the key segment adaptation domain. Data quality weights are calculated for each key segment adaptation domain. Each key segment adaptation domain is then sorted according to the patient's circadian rhythm and data quality weights to obtain time-series-based individual metabolic adaptation domains for each patient. Based on a publicly available database of diabetic patient parameters; matching patients with their corresponding fingerprint features to output a behavioral-physiological twin patient group; and constructing a metabolic adaptation domain for similar patients based on the behavioral-physiological twin patient group. Based on the original parameter monitoring data, the abnormal results of the patient's health data are obtained by judging the individual metabolic adaptation domain of the patient and the metabolic adaptation domain of similar patients.

4. The method for managing health data of diabetic patients according to claim 3, characterized in that, The specific steps for assessing diabetic health abnormalities using molecular physiological interaction indices in patients include: The fluctuation lag strength is identified based on the patient molecular physiological interaction index and causal inference model; the patient molecular physiological interaction index is corrected according to the fluctuation lag strength to obtain the corrected patient molecular physiological interaction index; the corrected patient molecular physiological interaction index is used to judge diabetic health abnormalities using the abnormal coupling effect judgment mechanism to obtain the patient's diabetic health risk index. Specifically, the modified patient molecular physiological interaction index is divided into N types of modified diabetes abnormality indicators; the contribution of each type of modified diabetes abnormality indicator is calculated; if the N types of modified diabetes abnormality indicators are superimposed in the same time window and the contribution of each type exceeds the preset threshold, the patient's diabetes health risk indicator is determined to be a composite diabetes risk; if only some of the modified diabetes abnormality indicators are superimposed in the same time window or any one of the contribution is lower than the preset threshold, the patient's diabetes health risk indicator is determined to be a diabetes observation risk.

5. A method for managing health data of diabetic patients according to claim 4, characterized in that, The specific steps for continuing to implement the corresponding diabetes monitoring and management strategy based on the patient's diabetes health risk indicators include: When patients are at multiple risk of diabetes, a high-frequency data collection mode is implemented to shorten the length and frequency of monitoring cycles. If a patient is at risk of diabetes observation, the monitoring period should be shortened by at least two consecutive monitoring periods. If the patient is not at risk of diabetes complex or diabetes observation in two consecutive monitoring periods, the monitoring period should be restored to the set period. Otherwise, the patient should be subjected to high-frequency data collection mode.

6. A method for managing health data of diabetic patients according to claim 5, characterized in that, The abnormal coupling effect determination mechanism is based on a causal inference algorithm.

7. A health data management system for diabetic patients, characterized in that, The system employs a method for managing health data of diabetic patients as described in any one of claims 1-6, comprising: The health data monitoring module includes a data monitoring unit. The data monitoring unit is used to acquire the patient's blood glucose fluctuations and molecular marker concentration fluctuations according to a set monitoring cycle, perform correlation calculations, and obtain the molecular regulation ratio parameters in the patient's body. It also acquires the patient's sleep cycle fluctuation parameters and bioelectrical feedback fluctuation parameters in real time. Based on the molecular regulation ratio parameters, sleep cycle fluctuation parameters, and bioelectrical feedback fluctuation parameters, it constructs the patient's molecular physiological interaction index. The health data management module includes a diabetes anomaly judgment unit and a diabetes health judgment unit. The diabetes anomaly judgment unit acquires the patient's raw parameter monitoring data; it then judges the raw parameter monitoring data against the patient's individual metabolic adaptation domain and the metabolic adaptation domain of similar patients to obtain the patient's health data anomaly result. The diabetes health judgment unit, if the patient's health data anomaly result is abnormal, performs a diabetes health anomaly judgment on the patient's molecular physiological interaction index and outputs the patient's diabetes health risk index; based on the patient's diabetes health risk index, it continues to execute the corresponding diabetes monitoring and management strategy; if the patient's health data anomaly result is normal, it continues to manage the diabetes patient's health data.

Images