A method and system for pattern recognition of physiological indicators related to lipolipid metabolism and obesity

Abstract

This application relates to the field of physiological data monitoring and risk assessment technology, and discloses a method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators. The method includes acquiring biochemical and morphological indicators of the target subject. When the biochemical indicators are in the warning range and the morphological indicators meet the high-risk warning characteristics, the method determines the correlation constraint parameters based on the correlation between the two, corrects the basic risk score, and generates an equivalent explicit risk assessment score. Based on the temporal characteristics of the biochemical and morphological indicators, the method determines the equivalent implicit risk assessment score and executes risk warning operations. This invention uses long-term abnormal states of biochemical indicators and high-risk warning characteristics to make judgments, clarifies the correlation between biochemical and morphological indicators, determines the correlation constraint parameters, corrects the basic risk score, and analyzes temporal characteristics. This ensures the comprehensiveness and accuracy of risk identification, improves the scientific nature and pertinence of metabolic-related risk warnings, and provides more reliable data support for subsequent health interventions.

Description

Technical Field

[0001] This application relates to the field of physiological data monitoring and risk assessment technology, and more specifically, to a method and system for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators. Background Technology

[0002] During community-based chronic disease follow-up, to continuously identify residents' risks, blood glucose and lipid test results, as well as basic physical characteristics, are typically collected quarterly or semi-annually. Risk levels are then updated based on this data to determine whether residents should be upgraded from general concern to priority intervention targets. In actual follow-up, many residents' blood glucose and lipid levels consistently hover within the borderline range, not yet reaching traditionally considered significantly abnormal levels. However, among these borderline individuals, some exhibit persistent abdominal obesity, while others show no obvious central obesity. When multiple indicators are in the boundary zone, body fat distribution should play a role in determining whether the combination leans towards the high-risk or low-risk side. When blood glucose and lipid levels are consistently in the boundary zone, while body fat distribution exhibits high-risk characteristics, the difference in body fat distribution effectively means that the glucose and lipid combination in the boundary zone is no longer neutral, but rather possesses different risk propensities.

[0003] However, existing follow-up assessment methods often use a uniform threshold judgment template when dealing with borderline glucose and lipid combinations, rarely considering body fat distribution as a constraint factor in the direction of boundary shifts for comprehensive interpretation. If risk escalation judgments are still performed according to a uniform standard, the borderline glucose and lipid combinations in abdominal obesity populations will be treated as ordinary borderline states. The apparent normality or slight deviation of a single indicator will mask the implicit risks of multiple indicator combinations, missing signals that the risk has already shifted to high risk. This will lead to delays in the screening of key individuals and the escalation of early warnings in community chronic disease follow-ups, affecting the accuracy and stability of the ongoing risk identification task. In community chronic disease follow-ups, when residents' blood glucose and blood lipid results are in the borderline zone for a long time while their body fat distribution continues to show a high-risk obesity phenotype, the borderline combination is treated as an ordinary state because the risk escalation judgment does not consider body fat distribution as a constraint factor in the direction of boundary shifts, thus affecting the ongoing identification of high-risk individuals. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this application aims to provide a method and system for recognizing patterns of lipid and glucose metabolism and obesity-related physiological indicators. This method can combine metabolic biochemical data with body shape and posture data to make comprehensive judgments, taking into account both real-time abnormal states and long-term change patterns of biochemical indicators, accurately determining potential metabolic risks, and improving the comprehensiveness of chronic disease risk identification and the accuracy of early warning assessment.

[0005] In a first aspect, this application provides a method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators, the method comprising:

[0006] The biochemical and morphological indicators of the target object are obtained over multiple consecutive monitoring periods. The biochemical indicators are detection values ​​reflecting the metabolic state of the target object, and the morphological indicators are measurement values ​​reflecting the body shape and morphology of the target object.

[0007] It is determined whether the biochemical indicators have been in the preset warning range for a long period of time in the current monitoring cycle and historical monitoring cycles, and at the same time, it is determined whether the morphological indicators in the current monitoring cycle meet the preset high-risk warning characteristics.

[0008] When the biochemical indicator is in the warning range for a long period of time in the current monitoring period and the historical monitoring period, and the morphological indicator in the current monitoring period meets the high-risk warning characteristics, the correlation constraint parameter of the morphological indicator relative to the biochemical indicator is determined according to the correlation between the biochemical indicator and the morphological indicator.

[0009] Obtain the basic risk score corresponding to the biochemical indicator, correct the basic risk score using the correlation constraint parameter, and generate an equivalent explicit risk assessment score characterizing the degree of abnormality of the biochemical indicator based on the corrected basic risk score.

[0010] The equivalent latent risk assessment score is determined based on the temporal characteristics of the biochemical indicators and the temporal characteristics of the morphological indicators; a risk warning operation is performed based on the equivalent explicit risk assessment score and the equivalent latent risk assessment score.

[0011] Secondly, this application provides a pattern recognition system for lipid and glucose metabolism and obesity-related physiological indicators, used to perform the aforementioned method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators. The system includes:

[0012] The acquisition module is used to acquire biochemical and morphological indicators of the target object over multiple continuous monitoring periods. The biochemical indicators are detection values ​​reflecting the metabolic state of the target object, and the morphological indicators are measurement values ​​reflecting the body shape and morphology of the target object.

[0013] The determination module is used to determine whether the biochemical indicators have been in a preset warning range for a long period of time in the current monitoring cycle and historical monitoring cycles, and at the same time determine whether the morphological indicators in the current monitoring cycle meet the preset high-risk warning characteristics.

[0014] The correlation matching module is used to determine the correlation constraint parameters of the morphological indicators relative to the biochemical indicators based on the correlation between the biochemical indicators and the morphological indicators when the biochemical indicators are in the warning range for a long time and the morphological indicators in the current monitoring period meet the high-risk warning characteristics.

[0015] The explicit risk module is used to obtain the basic risk score corresponding to the biochemical indicator, correct the basic risk score using the correlation constraint parameter, and generate an equivalent explicit risk assessment score characterizing the degree of abnormality of the biochemical indicator based on the corrected basic risk score.

[0016] The latent risk module is used to determine the equivalent latent risk assessment score based on the temporal characteristics of the biochemical indicators and the temporal characteristics of the morphological indicators; the early warning output module is used to perform risk early warning operations based on the equivalent explicit risk assessment score and the equivalent latent risk assessment score.

[0017] In summary, this application provides a method and system for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators. By acquiring metabolic biochemical indicators and body shape indicators of the target subject over multiple continuous monitoring periods, and combining the long-term abnormal state and high-risk characteristics of the indicators, a comprehensive judgment is made. This overcomes the shortcomings of traditional risk assessment, which only focuses on a single indicator and lacks multi-dimensional correlation analysis. By clarifying the correlation between biochemical and morphological indicators and determining the correlation constraint parameters, and combining basic risk score correction and time series feature analysis, the defects of one-sided risk assessment and omission of implicit risks are avoided. It can accurately capture the potential risks of metabolic abnormalities, take into account both explicit and implicit risk assessments, ensure the comprehensiveness and accuracy of risk identification, improve the scientific nature and pertinence of metabolic-related risk early warning, and provide more reliable data support for subsequent health interventions. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators provided in an embodiment of this application.

[0019] Figure 2 This is a schematic diagram of the structure of a pattern recognition system for lipid and glucose metabolism and obesity-related physiological indicators provided in an embodiment of this application.

[0020] Labeling Explanation: 1. Acquisition Module; 2. Judgment Module; 3. Association Matching Module; 4. Explicit Risk Module; 5. Implicit Risk Module; 6. Early Warning Output Module. Detailed Implementation

[0021] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0023] refer to Figure 1 This application proposes a method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators, the method comprising:

[0024] A1. Obtain biochemical and morphological indicators of the target object over multiple consecutive monitoring periods. Biochemical indicators are detection values ​​reflecting the metabolic state of the target object, and morphological indicators are measurement values ​​reflecting the body shape and physique of the target object.

[0025] A2. Determine whether the biochemical indicators have been in the preset warning range for a long period of time in the current monitoring cycle and historical monitoring cycles, and at the same time determine whether the morphological indicators in the current monitoring cycle meet the preset high-risk warning characteristics.

[0026] A3. When biochemical indicators are in the warning range for a long period of time in the current monitoring period and historical monitoring periods, and the morphological indicators in the current monitoring period meet the characteristics of high-risk warning, determine the correlation constraint parameters of morphological indicators relative to biochemical indicators based on the correlation between biochemical indicators and morphological indicators.

[0027] A4. Obtain the basic risk scores corresponding to the biochemical indicators, correct the basic risk scores using the correlation constraint parameters, and generate equivalent explicit risk assessment scores that characterize the degree of abnormality of the biochemical indicators based on the corrected basic risk scores.

[0028] A5. Determine the equivalent latent risk assessment score based on the temporal characteristics of biochemical indicators and morphological indicators;

[0029] A6. Based on the equivalent explicit risk assessment score and the equivalent implicit risk assessment score, execute risk warning operations.

[0030] In this application, biochemical indicators are detection values ​​that characterize the metabolic state of the target object, specifically including at least one physiological detection parameter that can reflect the body's material metabolism, energy metabolism and physiological function level, such as blood glucose, blood lipids, blood uric acid, liver and kidney function indicators, blood routine indicators, endocrine hormones, lactic acid, and ketone bodies.

[0031] Morphological indicators are measurement values ​​that characterize the body shape and posture of the target object. Specifically, they include at least one of the measurement values ​​that can intuitively reflect the external structure and posture characteristics of the body, such as height, weight, various body circumferences, limb length, trunk size, body proportion coefficient, etc.

[0032] The warning range refers to the critical value range of various biochemical indicators in advance. When a biochemical indicator is within this critical value range, it means that it has deviated from the normal level, and the potential risk of metabolic abnormalities can be judged in advance. For example, the warning range for fasting blood glucose can be set at 5.6-6.9 mmol / L.

[0033] High-risk warning characteristics refer to various data and physical changes that reflect a target individual's body shape as it moves towards obesity and imbalance, or worsening abnormal body posture. For example, a continuously increasing waist circumference and a monthly increase in body fat percentage.

[0034] Association constraint parameters refer to reference values ​​used to correct the risk calculation process and prevent unreasonable risk estimates. Examples include weight-related metabolic rate decline coefficients and drug metabolism parameters adjusted based on renal function.

[0035] The baseline risk score refers to an initial risk score calculated solely based on measured values ​​of biochemical indicators, without adjustment or optimization based on body shape and related data. For example, an initial risk score directly assessed based on measured blood glucose and blood lipid values.

[0036] The equivalent overt risk assessment score refers to a score obtained by adjusting the baseline risk score with relevant body shape and posture data, which directly reflects the severity of existing metabolic health problems. For example, a patient diagnosed with type 2 diabetes and accompanied by LDL cholesterol ≥4.9 mmol / L and hypertension would have an equivalent overt risk assessment score of 90.

[0037] The equivalent latent risk assessment score refers to a comprehensive score calculated by combining the long-term trends of various biochemical and morphological indicators. It is used to assess potential metabolic health risks that are not yet apparent and do not currently present with obvious discomfort. For example, potential metabolic health risks include insulin resistance, non-alcoholic fatty liver disease, microalbuminuria, and impaired vascular endothelial function. When the insulin resistance score is 3.5 and accompanied by mild fatty liver, the equivalent latent risk assessment score is 78, indicating a significant risk of insulin resistance and metabolic syndrome.

[0038] The study acquires biochemical and morphological indicators of the target subjects over multiple continuous monitoring periods. Specifically, it collects physiological data of the target subjects at different time points through continuous monitoring to provide basic data for subsequent risk assessment. Biochemical indicators reflect metabolic status, while morphological indicators reflect body shape and physique. The combination of the two can comprehensively characterize the physiological status of the target subjects. Various monitoring devices, such as wearable devices, smart body fat scales, and blood glucose meters, can be used to periodically or in real-time acquire biochemical indicators such as blood glucose, blood lipids, and insulin, as well as morphological indicators such as weight, waist circumference, and body fat percentage.

[0039] The system determines whether biochemical indicators have remained within preset warning ranges for an extended period, both in the current and historical monitoring cycles, and whether morphological indicators in the current monitoring cycle meet preset high-risk warning characteristics. Specifically, by assessing the long-term trends of biochemical indicators and the immediate state of morphological indicators, potential risk subjects can be preliminarily identified. For example, if a person's fasting blood glucose level has been above the upper limit of normal for three consecutive months, and their waist circumference exceeds the warning value of 90cm for men or 85cm for women in the current monitoring cycle, then this condition can be considered met.

[0040] When biochemical indicators remain in the warning range for an extended period within both the current and historical monitoring cycles, and morphological indicators for the current monitoring cycle meet the characteristics of a high-risk warning, the correlation constraint parameters between morphological indicators and biochemical indicators are determined based on the correlation between the two. Specifically, after initially identifying potential risks in the target subject, the degree of correlation between abnormal biochemical indicators and abnormal morphological indicators is further quantified to reflect the mutual influence between different indicators. This is achieved by extracting detection values ​​representing the overall obesity level of the target subject and measurement values ​​representing the degree of localized fat accumulation from the morphological indicators. Correlation analysis is performed on the collected detection and measurement values, and the influence weight of the measurement values ​​on the detection values ​​is determined based on the correlation analysis results, thereby determining the correlation constraint parameters between morphological indicators and biochemical indicators. For example, if there is a strong positive correlation between hyperglycemia symptoms and high body fat percentage in the target subject, the correlation constraint parameter between abnormal body fat percentage and blood sugar risk is increased accordingly.

[0041] This method obtains the basic risk scores corresponding to biochemical indicators, corrects these basic risk scores using correlation constraint parameters, and generates equivalent explicit risk assessment scores characterizing the degree of anomaly of the biochemical indicators based on the corrected basic risk scores. Specifically, this method determines the basic risk score of the target object through a pre-established correspondence between biochemical indicators and risk scores, corrects these basic risk scores using correlation constraint parameters, and uses the corrected basic risk scores as equivalent explicit risk assessment scores, thereby accurately characterizing the degree of anomaly of the biochemical indicators. For example, if the correlation constraint parameters indicate a high degree of anomaly in morphological indicators, then even if the basic risk score of the biochemical indicators is not high, it will be corrected to be higher, thus more accurately reflecting the actual risk.

[0042] The equivalent latent risk assessment score is determined based on the temporal characteristics of biochemical and morphological indicators. Specifically, by collecting biochemical indicator data and morphological indicator data of the target subject in the current and historical monitoring periods, the temporal variation characteristics of each indicator are extracted. These variation characteristics are then fused to output an equivalent latent risk assessment score characterizing the degree of latent metabolic risk. For example, if a person's blood glucose level is within the normal range but consistently near the upper limit of normal and their weight continues to increase, then even if the overt risk is not high, the latent risk will be assessed as high.

[0043] Risk warning operations are implemented based on equivalent explicit risk assessment scores and equivalent implicit risk assessment scores. Specifically, a comprehensive judgment is made by integrating equivalent explicit and implicit risk assessment scores to conduct risk warning work. In actual use, various risk scores are pre-set with corresponding warning levels and handling plans. After integrating explicit and implicit risk scores to obtain an overall risk assessment result, the warning level and handling plan are compared to determine the corresponding warning measures. The warning operation may include at least one of risk level prompts, intervention suggestion pushes, and abnormal alarm triggers. This solution, by comprehensively considering explicit and implicit risks, can comprehensively assess the physiological risk level of the target object and issue timely warnings when the risk reaches a preset threshold, thereby achieving early intervention and management of lipid and glucose metabolism and obesity-related physiological risks.

[0044] Through the above technical solution, this application can solve the pattern recognition problem of lipid and glucose metabolism and obesity-related physiological indicators to achieve risk warning operations. This method, by comprehensively analyzing the biochemical and morphological indicators of the target subjects, can more comprehensively assess the physiological risks related to lipid and glucose metabolism and obesity. By introducing correlation constraint parameters to correct the basic risk score, it can more accurately reflect the degree of abnormality of biochemical indicators, avoiding the limitations of single-indicator assessment. By considering the temporal characteristics of physiological indicators, it can capture potential, non-real-time risks, improving the timeliness and accuracy of risk warnings. Finally, by integrating explicit and implicit risks, it can comprehensively assess the physiological risk level of the target subjects and issue timely warnings when the risk reaches a preset threshold, thereby achieving early intervention and management of lipid and glucose metabolism and obesity-related physiological risks.

[0045] In some preferred embodiments, the step of determining whether a biochemical indicator has been within a preset warning range for an extended period in both the current monitoring cycle and historical monitoring cycles includes:

[0046] Extract the values ​​of biochemical indicators within a preset time window covering the current monitoring period and historical monitoring periods, and calculate the moving average and time series variance of the biochemical indicators;

[0047] When the moving average falls into the warning range and the time series variance is less than the preset stability threshold, it is determined that the biochemical indicator has been in the boundary zone for a long time, and thus it is determined that the biochemical indicator has been in the warning range for a long time in the current monitoring cycle and the historical monitoring cycle.

[0048] In this application, the preset stability threshold is a parameter used to judge the volatility of the indicator. For example, a time series variance of less than 0.5 can be considered a relatively stable state.

[0049] This solution extracts biochemical indicator values ​​within a preset time window covering both the current and historical monitoring periods. Specifically, it can obtain biochemical indicator data over a period of time from a database or sensors. For example, it can extract blood glucose, blood lipid, and other biochemical indicator data from the past 3, 6, or 12 months. The length of the preset time window can be adjusted according to the specific application scenario and the characteristics of the biochemical indicator. For example, for blood glucose, the time window can be set to 3 months to reflect the average level of glycated hemoglobin; for blood lipid, the time window can be set to 6 months to observe its long-term trend.

[0050] Calculate the moving average and time-series variance of biochemical indicators. Specifically, the moving average smooths short-term fluctuations and reflects the long-term trend of biochemical indicators, while the time-series variance measures the volatility of biochemical indicators. The moving average can be calculated using methods such as simple moving average, weighted moving average, or exponential moving average. For example, the moving average over the past 7 days or 30 days can be calculated to smooth short-term fluctuations and reflect the long-term trend of biochemical indicators. The time-series variance can be calculated using methods such as standard deviation, moving variance, or exponentially weighted moving variance. For example, the time-series variance over the past 7 days or 30 days can be calculated to measure the volatility of biochemical indicators.

[0051] When the moving average falls within the warning range and the time-series variance is less than a preset stability threshold, it is determined that the biochemical indicator has been lingering in the boundary zone for a long period, thus confirming that the biochemical indicator has been in the warning range for a long period in both the current and historical monitoring periods. Specifically, abnormal states of biochemical indicators are identified by comprehensively judging their long-term trends and volatility. When the moving average is within the warning range, it indicates that the long-term level of the biochemical indicator is at risk; simultaneously, the time-series variance is less than the preset stability threshold, which means that the long-term trend of the biochemical indicator is at risk, and its volatility is small, not a short-term accidental phenomenon, but a relatively stable state remaining at the risk boundary. Therefore, the system can determine that the biochemical indicator has been in the warning range for a long period in both the current and historical monitoring periods. This judgment method, which combines the moving average and the time-series variance, can more accurately identify abnormal states of biochemical indicators, avoid misjudgments or omissions caused by short-term fluctuations, and thus more accurately identify risks related to lipid and glucose metabolism and obesity.

[0052] Through the above technical solution, this application can accurately determine whether biochemical indicators are in the warning range for a long time, avoid misjudgment or missed judgment, and thus more accurately identify the risks related to lipid and glucose metabolism and obesity.

[0053] In some preferred embodiments, the step of determining whether the morphological indicators of the current monitoring period meet the preset high-risk warning characteristics includes:

[0054] When a morphological indicator meets at least one of the following conditions: the morphological indicator value exceeds the limit, the morphological indicator deviates beyond the safe benchmark, the morphological indicator exhibits a central abdominal obesity phenotype, the morphological indicator change rate is positive, and the morphological indicator time series trend continues to deteriorate, the morphological indicator is determined to meet the preset high-risk warning characteristics.

[0055] In this application, exceeding the morphological index value means that the target object's current body shape and posture measurement value exceeds the preset upper limit of the normal or safe range. Specifically, this can be achieved by comparing the morphological index value of the current monitoring period with a preset health threshold. For example, if a man's waist circumference exceeds 90cm or a woman's waist circumference exceeds 85cm, the waist circumference value is considered to be exceeding the standard.

[0056] Exceeding the safety benchmark for morphological indicators refers to the deviation of the current value of a morphological indicator from an individual's or group's safety benchmark exceeding a preset limit. Specifically, this can be achieved by calculating the difference or ratio between the current value of the morphological indicator and the individual's historical average or the average value of healthy individuals in the same age group, and determining whether this difference or ratio exceeds a preset deviation threshold. For example, if a person's body mass index (BMI) does not meet the obesity standard, but has increased by more than 10% compared to their historical healthy level, then an deviation exceeding the limit is considered to exist.

[0057] Central abdominal obesity, as defined by morphological indicators, refers to a body type where fat is primarily deposited in the abdominal area, resulting in an apple-shaped figure. This can be achieved by measuring indicators such as waist circumference and waist-to-hip ratio, and comparing them to predefined standards for central obesity. For example, a waist-to-hip ratio greater than 0.90 for men or greater than 0.85 for women can be considered central abdominal obesity.

[0058] A positive rate of change for morphological indicators means that the trend of change in the morphological indicators is increasing within a continuous monitoring period. This can be achieved by calculating the difference between the morphological indicator values ​​of the current monitoring period and the previous monitoring period, and determining whether this difference is greater than zero. For example, if weight has been continuously increasing over the past month, the rate of change for weight is considered positive.

[0059] A sustained deterioration in the trend of a morphological indicator refers to a morphological indicator exhibiting a persistently unfavorable trend over a relatively long period. This can be achieved by analyzing historical data of the morphological indicator over multiple consecutive monitoring periods, using methods such as trend analysis and regression analysis to determine whether the deterioration is continuous. For example, by observing the body mass index (BMI) over the past six months, if the BMI shows a continuous upward trend, even if the individual changes are small, it indicates a sustained deterioration in the trend.

[0060] If any of the above conditions are met, the morphological indicator is determined to meet the preset high-risk warning characteristics. By comprehensively considering these conditions, this solution can more comprehensively and accurately identify the high-risk warning characteristics of the target object's morphological indicators, thereby providing a more reliable basis for subsequent risk warning operations and improving the accuracy and timeliness of pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators.

[0061] Through the above technical solution, this application can address the problem that the lack of specific and comprehensive judgment criteria for determining whether morphological indicators meet the high-risk warning characteristics may lead to inaccurate judgment results and an inability to fully identify the potential obesity and metabolic abnormality risks of the target subject. This solution aims to solve the problem of the lack of specific and comprehensive judgment criteria for determining whether morphological indicators in the current monitoring period meet the preset high-risk warning characteristics by introducing multi-dimensional judgment criteria. When performing the step of determining whether the morphological indicators in the current monitoring period meet the preset high-risk warning characteristics, the morphological indicator data of the target subject in the current monitoring period is first obtained. Subsequently, multiple conditional judgments are performed on these morphological indicators sequentially or in parallel.

[0062] In some preferred embodiments, the step of determining the correlation constraint parameters of morphological indicators relative to biochemical indicators based on the correlation between biochemical indicators and morphological indicators includes:

[0063] Extract the actual values ​​of the first and second morphological indicators; where the actual value of the first morphological indicator is a measured parameter characterizing the overall obesity level of the target object, and the actual value of the second morphological indicator is a measured parameter characterizing the degree of local fat accumulation in the target object.

[0064] The ratio of the actual value of the first form to the preset safety benchmark is calculated to obtain the proportion of the first form; the ratio of the actual value of the second form to the safety benchmark is calculated to obtain the proportion of the second form.

[0065] Determine the correlation coefficients between biochemical indicators and the actual values ​​of the first and second forms, respectively; use the correlation coefficients to weight the proportions of the first and second forms, and then merge the weighted proportions of each form to obtain the correlation constraint parameters.

[0066] In this application, the actual value of the first morphology refers to the measured parameter that characterizes the overall obesity level of the target object. It can directly reflect the degree of obesity of the overall body shape. Specifically, it can be a value such as body mass index, overall body fat percentage, or standard weight difference.

[0067] The second form actual value refers to the measured parameter that characterizes the degree of local fat accumulation in the target object. It is used to reflect the local fat accumulation in the body. Specifically, values ​​such as waist circumference, thigh circumference, upper arm circumference, and neck circumference can be used.

[0068] Safety benchmarks refer to preset health standard values. Specifically, they can adopt the health standard values ​​issued by the National Health Commission. For example, the safety benchmark for body mass index can be set at 18.5-24.9 kg / m², and the safety benchmark for waist circumference can be set at less than 90 cm for men and less than 85 cm for women.

[0069] The correlation coefficient is a numerical value that reflects the degree of correlation between biochemical and morphological indicators. It can be calculated using statistical methods such as Pearson correlation coefficient, Spearman correlation coefficient, or Kendall correlation coefficient. For example, the Pearson correlation coefficient between fasting blood glucose and body mass index, and the Pearson correlation coefficient between fasting blood glucose and waist circumference can be calculated.

[0070] The actual values ​​of the first and second morphological indicators were extracted. The first morphological value is a measured parameter characterizing the overall obesity level of the target subject; the second morphological value is a measured parameter characterizing the degree of localized fat accumulation in the target subject. By distinguishing between these two morphological indicators, the body shape characteristics of the target subject can be assessed more comprehensively, providing richer data dimensions for subsequent correlation analysis.

[0071] The ratio of the actual value of the first morphology to the preset safety benchmark is calculated to obtain the first morphology ratio, and the ratio of the actual value of the second morphology to the safety benchmark is calculated to obtain the second morphology ratio. Specifically, by calculating the ratios of the actual values ​​of the first and second morphologies to the safety benchmark respectively, the actual measured values ​​can be compared with the safety benchmark, and the degree to which the morphological indicators of the target object deviate from the health level can be intuitively reflected, so that morphological indicators of different dimensions can be compared and analyzed on a unified scale.

[0072] Determine the correlation coefficients between biochemical indicators and the actual values ​​of the first and second morphological types, respectively. Specifically, calculate the correlation coefficients between various metabolic data and overall obesity data, as well as localized fat data, to reflect the strength of the relationship and the patterns of change. For example, some biochemical indicators may have a stronger correlation with overall obesity levels, while others may have a stronger correlation with localized fat accumulation. By calculating these correlation coefficients, the components of morphological indicators that have a greater impact on abnormal biochemical indicators can be identified.

[0073] The proportions of the first and second morphologies are weighted separately using correlation coefficients, and the weighted proportions of each morphology are then combined to obtain the correlation constraint parameters. By using the previously calculated correlation coefficients as weights to weight the proportions of each morphology, the proportions of morphologies with stronger correlations to biochemical indicators have a greater weight in the final correlation constraint parameters. For example, if a certain biochemical indicator has a much higher correlation with waist circumference than with body mass index, then the morphology proportion corresponding to waist circumference will have a greater impact on the correlation constraint parameters after weighting.

[0074] Through the above technical solutions, this application can more specifically determine the correlation constraint parameters and more precisely quantify the correlation between morphological indicators and biochemical indicators, thereby improving the accuracy and precision of risk assessment.

[0075] In some preferred embodiments, the steps of correcting the baseline risk score using correlation constraint parameters, and generating an equivalent explicit risk assessment score characterizing the degree of abnormality of biochemical indicators based on the corrected baseline risk score, include:

[0076] The risk assessment threshold of biochemical indicators is lowered based on the correlation constraint parameters, and the basic risk score is corrected by using the degree of deviation of the biochemical indicators from the lowered risk assessment threshold, thereby generating an equivalent explicit risk assessment score.

[0077] Alternatively, the original values ​​of biochemical indicators can be weighted and amplified using correlation constraint parameters to obtain amplified physiological values. The basic risk score can then be corrected based on the amplified physiological values ​​to generate an equivalent explicit risk assessment score.

[0078] In this application, the proposed solution provides two specific methods to generate equivalent explicit risk assessment scores by modifying the basic risk scores using correlation constraint parameters, thereby more accurately assessing the degree of abnormality of biochemical indicators.

[0079] The first method involves lowering the risk assessment threshold of biochemical indicators based on correlation constraint parameters, and then using the deviation of the biochemical indicators from the lowered risk assessment threshold to correct the basic risk score, thereby generating an equivalent explicit risk assessment score. Specifically, this scheme first divides the correlation constraint parameters into several levels based on a preset numerical range. A pre-established matching relationship is created between each level of the correlation constraint parameters and the magnitude of the risk assessment threshold reduction, as well as the correction weight for the basic risk score. A specific threshold reduction ratio and score correction weight are assigned to each level. The higher the level of the correlation constraint parameter, the stronger the correlation between the morphological and biochemical indicators, resulting in a larger threshold reduction ratio and a larger score correction weight. After the threshold reduction is completed, the deviation of the actual detected value of the biochemical indicator from the lowered risk assessment threshold is calculated. The corresponding score correction weight is selected based on the level of the correlation constraint parameter, and the initial basic risk score is weighted and corrected using the deviation and correction weight to finally obtain the equivalent explicit risk assessment score. For example, a baseline risk assessment threshold for blood glucose is set at 7.0 mmol / L, the measured blood glucose level is 7.5 mmol / L, and the initial baseline risk score is 30 points. Three levels of correlation are pre-defined and corresponding parameters are assigned: a 5% downgrade coefficient and a 0.8 score correction weight for the low level; a 10% downgrade coefficient and a 1.0 score correction weight for the medium level; and a 15% downgrade coefficient and a 1.2 score correction weight for the high level. When the correlation constraint parameter is at the low level, a new threshold is first obtained through the downgrade coefficient. Then, the deviation between the measured value and the downgrade threshold is calculated. The deviation is multiplied by the corresponding score correction weight to obtain the risk correction score. This correction score is then added to the initial baseline risk score to obtain the final equivalent explicit risk assessment score. This method can simultaneously tighten the anomaly assessment criteria and amplify the risk impact of indicator deviations as the correlation between morphological and biochemical indicators strengthens, thereby identifying potential risks earlier. This method quantifies the impact of morphological indicators on the risk of biochemical indicators, so that the final equivalent explicit risk assessment score not only reflects the abnormality of the biochemical indicators themselves, but also incorporates the risk value brought about by the abnormality of morphological indicators, which can more sensitively reflect the potential risks.

[0080] The second method utilizes correlation constraint parameters to weight and amplify the original values ​​of biochemical indicators, resulting in amplified physiological values. Based on these amplified physiological values, the baseline risk score is adjusted, thereby generating an equivalent explicit risk assessment score. Specifically, the actual measured values ​​of biochemical indicators are multiplied or added according to the magnitude of the correlation constraint parameters to obtain an amplified physiological value. This weighted amplified physiological value is then used as input to recalculate or adjust the baseline risk score of the biochemical indicator. For example, if the original value of the biochemical indicator is X and the correlation constraint parameter is α, the amplified physiological value can be expressed as X*(1+k*α), where k is the amplification coefficient. This amplifies the abnormality of biochemical indicators when morphological indicators have a high risk, thus more fully reflecting the impact of morphological indicators on biochemical risk when calculating the baseline risk score. Adjusting the baseline risk score based on the amplified physiological values ​​directly integrates the risk impact of morphological indicators into the risk assessment of biochemical indicators, enabling the equivalent explicit risk assessment score to more comprehensively reflect the combined risks related to lipid and glucose metabolism and obesity.

[0081] Through the above technical solution, this application solves the problem of lacking specific methodological guidance in the process of correcting the basic risk score using correlation constraint parameters and generating an equivalent explicit risk assessment score characterizing the degree of abnormality of biochemical indicators based on the corrected basic risk score.

[0082] In some preferred embodiments, the step of determining the equivalent latent risk assessment score based on the temporal characteristics of biochemical indicators and morphological indicators includes:

[0083] The duration of biochemical indicators within the warning interval is extracted as the time-series feature of the biochemical indicators, and the duration is linearly weighted to obtain the first risk component.

[0084] The rate of change of morphological indicators is extracted as the temporal feature of morphological indicators. The rate of change is then subjected to nonlinear mapping to obtain the second risk component.

[0085] The first risk component and the second risk component are summed to obtain the equivalent implicit risk assessment score.

[0086] In this application, the duration refers to the length of time that the biochemical indicator values ​​are cumulatively within the preset warning range. Specifically, it can be achieved by using a counter or timestamp recording method. For example, the counter starts counting whenever the biochemical indicator value enters the warning range and stops counting when the value leaves the warning range. Alternatively, the specific time points of entering and leaving the warning range can be recorded, and the duration can be calculated by the time difference.

[0087] The rate of change of morphological indicators refers to the ratio of the magnitude of the change in the value of the morphological indicator within a continuous monitoring period to the time interval. Specifically, it can be achieved by difference calculation or regression analysis. For example, calculate the difference between the morphological indicator value of the current monitoring period and the morphological indicator value of the previous monitoring period, and then divide it by the time interval of the monitoring period to obtain the rate of change.

[0088] Specifically, this solution aims to improve the accuracy of the equivalent latent risk assessment score by extracting and processing more refined temporal features, thereby providing a more comprehensive assessment of lipid and glucose metabolism and obesity-related health risks.

[0089] By extracting the duration of biochemical indicators within the warning interval as their time-series feature and applying linear weighting, the first risk component can be obtained. Specifically, this method quantifies the duration of abnormal biochemical states; the longer the duration, the greater the cumulative risk effect. Linear weighting can intuitively reflect this cumulative risk. Linear weighting involves multiplying the duration by a fixed weighting coefficient, which can be achieved using the formula: First Risk Component = Duration × Weighting Coefficient. The weighting coefficient can be set based on clinical experience or data analysis results; for example, it can be set to 0.5, 1.0, or 2.0 to reflect the strength of the duration's impact on risk.

[0090] The rate of change of morphological indicators is extracted as its temporal feature and subjected to nonlinear mapping to obtain the second risk component. Specifically, the rate of change of morphological indicators can reflect the trend and speed of body shape deterioration. When the rate of change reaches a certain level, the increase in risk may no longer be linear, but rather exhibit exponential growth. Nonlinear mapping can more realistically simulate this complex relationship. Nonlinear mapping refers to converting the rate of change into a risk component through a nonlinear function to better capture the characteristic value of risk accelerating with the increase of the rate of change. Specifically, exponential functions, logarithmic functions, or sigmoid functions can be used. For example, the second risk component can be expressed as A × exp(B × rate of change), where A and B are adjustable parameters used to control the shape and intensity of the mapping.

[0091] The first risk component and the second risk component are summed to obtain an equivalent latent risk assessment score. Specifically, the summation calculation combines the long-term abnormal states of biochemical indicators and the changing trends of morphological indicators to comprehensively assess the potential health risks of the target population, making the risk assessment more comprehensive and accurate.

[0092] Through the above technical solution, this application can more accurately reflect the degree of hidden risk and solve the problem of how to specifically extract and process the temporal characteristics of biochemical indicators and morphological indicators in the process of determining the equivalent hidden risk assessment score.

[0093] In some preferred embodiments, when it is identified that the biochemical indicators of the current monitoring period have fallen back from the warning interval to the safe zone, and the rate of change of the morphological indicators of the current period is positive, the duration count of the biochemical indicators in the warning interval is frozen, and a fallback observation time window is set.

[0094] Within the observation window of decline, the risk compensation amount is calculated based on the decline amount of biochemical indicators in the previous monitoring period and the associated constraint parameters of the current monitoring period, and the risk compensation amount is added to the equivalent implicit risk assessment score.

[0095] In this application, the safe zone refers to the range of biochemical indicator values ​​that meet the normal metabolic standards of the human body and have no abnormal risks.

[0096] The decline amount refers to the difference between the critical highest value of the biochemical indicator within the warning range and the current safe zone value. The larger the decline amount, the more obvious the improvement of the biochemical indicator. The current safe zone value refers to the measured value of the biochemical indicator that is currently detected and is within the safe range.

[0097] Risk compensation refers to the residual risk value formed after biochemical indicators have returned to normal, but the previous abnormal state still affects the body's metabolism. It can be calculated according to a preset function or model, for example: Risk compensation = f(biochemical indicator decline amount, correlation constraint parameter), where f can be a linear or non-linear function, for example: Risk compensation = (1 - normalized value of decline amount) * correlation constraint parameter * weight coefficient.

[0098] When a biochemical indicator in the current monitoring period falls back from the warning zone to the safe zone, and the rate of change of the morphological indicator in the current period is positive, this scheme specifically receives biochemical and morphological indicator data and compares them with preset warning and safe zones. A fallback judgment is triggered when the biochemical indicator value first enters the safe zone from the warning zone. Simultaneously, the ratio of the change in the morphological indicator within the current monitoring period to the time interval is calculated. If this ratio is positive, the condition that the rate of change of the morphological indicator is positive is met.

[0099] The system freezes the duration count of biochemical indicators within the warning range and sets a fallback observation window. Specifically, when a biochemical indicator falls from the warning range to the safe zone and the rate of change of morphological indicators is positive, the cumulative calculation of the duration of the biochemical indicator within the warning range is paused, and a preset observation period is initiated. This means that even if the biochemical indicator temporarily improves, the risk duration accumulated within the warning range will not be cleared, and the status of the target object can be continuously observed during the observation period to assess the stability of the fallback of biochemical indicators and the subsequent trend of morphological indicators. For example, a Boolean flag can be set to true when the freeze condition is met, and the accumulation operation of the duration counter can be stopped. A timer can be started, and the duration of the timer can be preset based on clinical experience or model training results, such as 7 days, 14 days, or 30 days. Within the observation window, the system will continuously monitor the changes of biochemical and morphological indicators.

[0100] Within the observation window of decline, the risk compensation amount is calculated based on the decline in biochemical indicators from the previous monitoring period and the associated constraint parameters of the current monitoring period. This risk compensation amount is then added to the equivalent implicit risk assessment score. Specifically, within the set observation window, the degree of improvement in biochemical indicators and the potential risks of morphological indicators are comprehensively considered to obtain a risk compensation amount to compensate for potential risks. This risk compensation amount is then added to the equivalent implicit risk assessment score previously determined based on the temporal characteristics of biochemical and morphological indicators. Therefore, even if biochemical indicators temporarily decline, the equivalent implicit risk assessment score can still reflect this comprehensive risk due to the continued deterioration of morphological indicators or the existence of potential risks, avoiding risk assessment distortion caused by the improvement of a single indicator. For example, the system performs a simple summation operation between the calculated risk compensation amount and the current equivalent implicit risk assessment score to update the equivalent implicit risk assessment score.

[0101] Through the above technical solution, this application addresses the problem that simply stopping the counting of the duration of biochemical indicators within the warning interval when biochemical indicators fall back to the safe zone, while the rate of change of morphological indicators remains positive, might overlook the potential risks that still exist after the decline in biochemical indicators, as well as the risk compensation effect that may result from the continued deterioration of morphological indicators, leading to an incomplete and inaccurate risk assessment. This solution freezes the counting of the duration of biochemical indicators within the warning interval and sets a decline observation window, ensuring the preservation of historical risks and continuous monitoring of future risks. Furthermore, by calculating the risk compensation amount and adding it to the equivalent implicit risk evaluation score, effective compensation is achieved for the potential risks after the decline in biochemical indicators and the risks of continued deterioration of morphological indicators, making the risk assessment more comprehensive, accurate, and refined, thereby improving the reliability of risk warning operations.

[0102] In some preferred embodiments, if the biochemical indicators rebound back to the warning range within the observation window of the decline, the frozen state is lifted and the accumulation of the duration is restored;

[0103] If the biochemical indicators are still in the safe zone and the rate of change of the morphological indicators is decreasing at the end of the observation window, then the duration is reset and the equivalent implicit risk assessment score is subjected to smooth decay processing.

[0104] In this application, "unfreezing" refers to stopping the freezing of biochemical indicators for the duration they remain within the warning range. This can be achieved by setting the freeze flag to an unfrozen state.

[0105] The cumulative duration of recovery refers to adding the duration of biochemical indicators within the warning range to the duration before freezing.

[0106] Resetting the duration refers to clearing the duration count of biochemical indicators within the warning range to zero, which can be achieved by resetting the duration counter to zero.

[0107] If the biochemical indicators rebound back into the warning range within the pullback observation window, the freeze will be lifted and the accumulation of duration will resume. Specifically, if the biochemical indicators rebound back into the warning range within the pullback observation window, it indicates that the previous pullback may have been temporary, and the risk has not truly been eliminated. Lifting the freeze and resuming the accumulation of duration ensures that the system continues to track the duration of the biochemical indicators within the warning range, thus accurately reflecting the actual persistence of the risk and avoiding delays or inaccuracies in risk assessment due to misjudgment.

[0108] If, at the end of the observation window, biochemical indicators remain within the safe zone and the rate of change of morphological indicators shows a downward trend, the duration is reset and a smooth decay process is applied to the equivalent implicit risk assessment score. Specifically, if, at the end of the observation window, biochemical indicators remain within the safe zone and the rate of change of morphological indicators shows a downward trend, this indicates that the target's health condition is improving, and the risk may have significantly decreased. In this case, resetting the duration clears the previously accumulated risk duration, reflecting a substantial improvement in the risk status. Simultaneously, applying a smooth decay process to the equivalent implicit risk assessment score gradually reduces the score, making it more accurately reflect the current low risk level and avoiding overestimation of current risk based on historical data, thus making the risk assessment results more reasonable and dynamic.

[0109] Through the above technical solution, this application addresses the issue that during the observation period after biochemical indicators have fallen from the warning range to the safe range, they may rebound back to the warning range, or at the end of the observation window, while biochemical indicators remain in the safe range, the rate of change of morphological indicators may show a downward trend. These situations require further strategies to ensure the accuracy and timeliness of risk assessment. This solution can dynamically adjust the risk assessment strategy based on the dynamic changes of biochemical and morphological indicators, ensuring the timeliness and accuracy of risk assessment, avoiding delays or inaccuracies due to misjudgment, and preventing over-assessment of historical risks, making the risk assessment results more reasonable and dynamic.

[0110] In some preferred embodiments, the equivalent explicit risk assessment score is mapped to the basic risk score, and the equivalent implicit risk assessment score is mapped to the additional risk score.

[0111] Calculate the total score of the basic risk score and the additional risk score. When the total score is higher than the warning trigger threshold, generate a warning instruction.

[0112] In this application, the equivalent explicit risk assessment score is mapped to a basic risk score, and the equivalent implicit risk assessment score is mapped to an additional risk score. Specifically, a pre-defined mapping function or lookup table is used to convert the equivalent explicit risk assessment score into a standardized score, i.e., the basic risk score; and the equivalent implicit risk assessment score is also converted into a standardized score, i.e., the additional risk score. For example, the equivalent explicit risk assessment score can be converted into a score between 0 and 100 according to a certain proportion or piecewise function. The equivalent implicit risk assessment score can be mapped to an additional score between 0 and 50 according to its contribution to potential risks. This step simplifies the multi-dimensional explicit risk assessment results into a quantifiable basic risk value, and also quantifies implicit risks that are not easily observed directly but have potential risks, serving as a supplement to the basic risk, thereby more comprehensively assessing the risk status of the target object.

[0113] The system calculates the total score of the basic risk score and the additional risk score. When the total score exceeds the warning trigger threshold, a warning instruction is generated. Specifically, the total score is the sum of the basic risk score and the additional risk score. The warning trigger threshold is the critical value used to determine whether to generate a warning instruction. A warning instruction is a notification or operational instruction generated by the system when the total score exceeds the warning trigger threshold. This can be implemented through methods such as sending SMS messages, emails, app notifications, or displaying a red alert on the management interface. For example, when the total score reaches the warning trigger threshold, the system can generate a warning instruction stating, "The target object poses a high risk; please pay attention and take intervention measures promptly."

[0114] Through the above technical solution, this application can effectively transform complex equivalent explicit risk assessment scores and equivalent implicit risk assessment scores into operable early warning instructions and determine the conditions for triggering early warnings, thereby ensuring the accuracy and timeliness of risk early warnings and improving the efficiency of identifying and managing risks related to lipid and glucose metabolism and obesity.

[0115] refer to Figure 2 This application provides a pattern recognition system for lipid and glucose metabolism and obesity-related physiological indicators, the system comprising:

[0116] The acquisition module 1 is used to acquire biochemical and morphological indicators of the target object in multiple continuous monitoring periods. The biochemical indicators are detection values ​​that reflect the metabolic state of the target object, and the morphological indicators are measurement values ​​that reflect the body shape and physique of the target object (the specific process can be referred to above).

[0117] The determination module 2 is used to determine whether the biochemical indicators are in the preset warning range for a long period of time in the current monitoring cycle and the historical monitoring cycle, and at the same time determine whether the morphological indicators in the current monitoring cycle meet the preset high-risk warning characteristics (the specific process can be referred to above).

[0118] The association matching module 3 is used to determine the association constraint parameters of the morphological indicators relative to the biochemical indicators based on the correlation between the biochemical indicators and the morphological indicators when the biochemical indicators are in the warning range for a long time and the morphological indicators in the current monitoring period meet the high-risk warning characteristics (the specific process can be referred to above).

[0119] The explicit risk module 4 is used to obtain the basic risk score corresponding to the biochemical indicator, correct the basic risk score using the correlation constraint parameter, and generate an equivalent explicit risk assessment score representing the degree of abnormality of the biochemical indicator based on the corrected basic risk score (the specific process can be referred to above).

[0120] The latent risk module 5 is used to determine the equivalent latent risk assessment score based on the temporal characteristics of the biochemical indicators and the temporal characteristics of the morphological indicators (the specific process can be referred to above); the early warning output module 6 is used to perform risk early warning operations based on the equivalent explicit risk assessment score and the equivalent latent risk assessment score (the specific process can be referred to above).

[0121] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators, characterized in that, The method includes: The biochemical and morphological indicators of the target object are obtained over multiple consecutive monitoring periods. The biochemical indicators are detection values ​​reflecting the metabolic state of the target object, and the morphological indicators are measurement values ​​reflecting the body shape and morphology of the target object. It is determined whether the biochemical indicators have been in the preset warning range for a long period of time in the current monitoring cycle and historical monitoring cycles, and at the same time, it is determined whether the morphological indicators in the current monitoring cycle meet the preset high-risk warning characteristics. When the biochemical indicator is in the warning range for a long period of time in the current monitoring period and the historical monitoring period, and the morphological indicator in the current monitoring period meets the high-risk warning characteristics, the correlation constraint parameter of the morphological indicator relative to the biochemical indicator is determined according to the correlation between the biochemical indicator and the morphological indicator. Obtain the basic risk score corresponding to the biochemical indicator, correct the basic risk score using the correlation constraint parameter, and generate an equivalent explicit risk assessment score characterizing the degree of abnormality of the biochemical indicator based on the corrected basic risk score. The equivalent latent risk assessment score is determined based on the temporal characteristics of the biochemical indicators and the temporal characteristics of the morphological indicators; a risk warning operation is performed based on the equivalent explicit risk assessment score and the equivalent latent risk assessment score.

2. The method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators according to claim 1, characterized in that, The step of determining whether the biochemical indicator has been within a preset warning range for an extended period in the current monitoring cycle and historical monitoring cycles includes: Extract the values ​​of the biochemical indicators within a preset time window covering the current monitoring period and historical monitoring periods, and calculate the moving average and time series variance of the biochemical indicators; When the moving average falls into the warning interval and the time series variance is less than the preset stability threshold, it is determined that the biochemical indicator has been in the boundary zone for a long time, and thus it is determined that the biochemical indicator has been in the warning interval for a long time in the current monitoring cycle and the historical monitoring cycle.

3. The method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators according to claim 1, characterized in that, The step of determining whether the morphological indicators of the current monitoring period meet the preset high-risk warning characteristics includes: When the morphological indicator meets at least one of the following conditions: the morphological indicator value exceeds the standard, the morphological indicator deviates beyond the limit relative to the safety benchmark, the morphological indicator exhibits a central abdominal obesity phenotype, the morphological indicator change rate is positive, and the morphological indicator time series trend continues to deteriorate, the morphological indicator is determined to meet the preset high-risk warning characteristics.

4. The method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators according to claim 1, characterized in that, The step of determining the correlation constraint parameter of the morphological index relative to the biochemical index based on the correlation between the biochemical index and the morphological index includes: Extract the first morphological actual value and the second morphological actual value from the morphological indicators; wherein, the first morphological actual value is a measured parameter characterizing the overall obesity level of the target object, and the second morphological actual value is a measured parameter characterizing the degree of local fat accumulation in the target object; The ratio of the actual value of the first form to the preset safety benchmark is calculated to obtain the first form ratio; the ratio of the actual value of the second form to the safety benchmark is calculated to obtain the second form ratio. Determine the correlation coefficients between the biochemical indicators and the actual values ​​of the first and second morphologies, respectively; use the correlation coefficients to weight the proportions of the first and second morphologies, respectively, and then merge the weighted proportions of each morphology to obtain the correlation constraint parameters.

5. The method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators according to claim 4, characterized in that, The steps of correcting the basic risk score using the correlation constraint parameters and generating an equivalent explicit risk assessment score characterizing the degree of abnormality of the biochemical indicator based on the corrected basic risk score include: The risk assessment threshold of the biochemical indicator is lowered according to the associated constraint parameters, and the basic risk score is corrected by using the degree of deviation of the biochemical indicator from the lowered risk assessment threshold, thereby generating the equivalent explicit risk assessment score. Alternatively, the original values ​​of the biochemical indicators can be weighted and amplified using the associated constraint parameters to obtain amplified physiological values. The basic risk score can then be corrected based on the amplified physiological values ​​to generate the equivalent explicit risk assessment score.

6. The method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators according to claim 1, characterized in that, The step of determining the equivalent latent risk assessment score based on the temporal characteristics of the biochemical indicators and the temporal characteristics of the morphological indicators includes: The duration of the biochemical indicator within the warning interval is extracted as the time-series feature of the biochemical indicator, and the duration is linearly weighted to obtain the first risk component. The rate of change of the morphological index is extracted as the temporal feature of the morphological index, and the rate of change is subjected to nonlinear mapping processing to obtain the second risk component. The equivalent implicit risk assessment score is obtained by summing the first risk component and the second risk component.

7. The method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators according to claim 6, characterized in that, The method also includes: When it is detected that the biochemical indicator in the current monitoring period has fallen from the warning interval to the safe zone, and the rate of change of the morphological indicator in the current period is positive, the duration count of the biochemical indicator in the warning interval is frozen, and a fallback observation time window is set. Within the observed decline time window, the risk compensation amount is calculated based on the decline amount of the biochemical indicators in the previous monitoring period and the associated constraint parameters in the current monitoring period, and the risk compensation amount is added to the equivalent implicit risk assessment score.

8. The method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators according to claim 7, characterized in that, The method also includes: If the biochemical indicators rebound back to the warning range within the observed decline time window, the frozen state will be lifted and the accumulation of the duration will resume. If the biochemical indicators are still in the safe zone and the rate of change of the morphological indicators is decreasing at the end of the observation window, then the duration is reset and the equivalent latent risk assessment score is subjected to smooth decay processing.

9. The method for pattern recognition of lipid and glucose metabolism and obesity-related physiological indicators according to claim 8, characterized in that, The step of performing risk warning operations based on the equivalent explicit risk assessment score and the equivalent implicit risk assessment score includes: The equivalent explicit risk assessment score is mapped to a basic risk score, and the equivalent implicit risk assessment score is mapped to an additional risk score. Calculate the total score of the basic risk score and the additional risk score. When the total score is higher than the warning trigger threshold, generate a warning instruction.

10. A pattern recognition system for lipid and glucose metabolism and obesity-related physiological indicators, used to perform the steps of the method according to any one of claims 1 to 9, characterized in that, include: The acquisition module is used to acquire biochemical and morphological indicators of the target object over multiple continuous monitoring periods. The biochemical indicators are detection values ​​reflecting the metabolic state of the target object, and the morphological indicators are measurement values ​​reflecting the body shape and morphology of the target object. The determination module is used to determine whether the biochemical indicators have been in a preset warning range for a long period of time in the current monitoring cycle and historical monitoring cycles, and at the same time determine whether the morphological indicators in the current monitoring cycle meet the preset high-risk warning characteristics. The correlation matching module is used to determine the correlation constraint parameters of the morphological indicators relative to the biochemical indicators based on the correlation between the biochemical indicators and the morphological indicators when the biochemical indicators are in the warning range for a long time and the morphological indicators in the current monitoring period meet the high-risk warning characteristics. The explicit risk module is used to obtain the basic risk score corresponding to the biochemical indicator, correct the basic risk score using the correlation constraint parameter, and generate an equivalent explicit risk assessment score characterizing the degree of abnormality of the biochemical indicator based on the corrected basic risk score. The latent risk module is used to determine the equivalent latent risk assessment score based on the temporal characteristics of the biochemical indicators and the temporal characteristics of the morphological indicators; the early warning output module is used to perform risk early warning operations based on the equivalent explicit risk assessment score and the equivalent latent risk assessment score.

Images