Children overweight and obesity detection method and system based on multi-index cooperative judgment

Abstract

The invention discloses a child overweight and obesity detection method and system based on multi-index cooperative determination, and the method comprises the steps: firstly deploying and initializing a special multi-mode detection tool, and inputting the basic information of a child; and then guiding the integrated device and the portable module for children to acquire physique indexes according to prompts, calling intelligent wearing behavior data, verifying the intelligent wearing behavior data, and storing the intelligent wearing behavior data into a local database of the terminal. Extracting multi-modal data, importing the multi-modal data into a built-in model, matching a WHO standard and a metabolic reference range according to age and gender, assigning differential weights according to growth stages, and correcting and outputting an obesity trend probability basic value in combination with a growth rate. The behavior data is quantified, a risk matrix is established, a cause is positioned, then a report result containing obesity probability, influence factors and early warning levels is generated through integration, and suggestions are given to abnormal children on site. And effective intervention of overweight and obesity of children is facilitated.

Description

Technical Field

[0001] This invention belongs to the field of children's overweight and obesity detection technology, and more specifically, relates to a method and system for children's overweight and obesity detection based on multi-indicator synergistic judgment. Background Technology

[0002] In the field of child health management, overweight and obesity have become a significant public health issue affecting children's growth and development. Traditional methods for detecting overweight and obesity in children often rely solely on Body Mass Index (BMI) assessment. This approach has significant limitations, failing to comprehensively reflect the combined impact of children's physiological metabolic status and behavioral habits on overweight and obesity. For example, some children may have a normal BMI but exhibit abnormal metabolic indicators or unhealthy behavioral habits. Relying solely on BMI assessment can easily lead to missed diagnoses of overweight and obesity risk. Furthermore, for children with abnormal BMI, it is difficult to determine whether their overweight and obesity are caused by physiological metabolic disorders, behavioral factors, or a combination of both, thus hindering the provision of precise evidence for intervention.

[0003] Meanwhile, in current methods for detecting overweight and obesity in children, metabolic index testing and behavioral data collection are often disconnected. Metabolic index testing typically utilizes large-scale biochemical equipment, which, due to limitations in facilities and operational complexity, is difficult to implement widely in routine childhood overweight and obesity detection. Behavioral data collection relies heavily on manual interviews, leading to strong subjectivity and low data accuracy. This lack of data dimensions and inadequate collection methods result in a superficial understanding of childhood overweight and obesity, failing to delve into the underlying causes and making health recommendations less targeted, thus significantly reducing the effectiveness of interventions.

[0004] Furthermore, children are in a dynamic process of growth and development, and the criteria for determining overweight and obesity vary across different ages, genders, and growth rates. Traditional testing methods do not fully consider these individual differences, and using uniform assessment standards can easily lead to misjudgments. For example, for children with a rapid growth rate, their weight gain may be due to normal development, but if assessed according to conventional standards, they may be misjudged as being at risk of overweight and obesity, causing unnecessary psychological burden on the child and their family; conversely, inappropriate standards may also miss true risks of overweight and obesity.

[0005] In conclusion, the current field of childhood overweight and obesity detection urgently needs a detection method that can integrate multi-dimensional indicators, adapt to children's growth dynamics, and achieve accurate judgment and causal analysis. This would improve the accuracy of overweight and obesity risk identification, clarify the causes of overweight and obesity, provide a scientific basis for personalized health intervention, and effectively curb the occurrence and development of childhood overweight and obesity, thus ensuring children's healthy growth. Summary of the Invention

[0006] This invention aims to address the problems of traditional childhood obesity detection relying on a single BMI, fragmented multi-dimensional data, and insufficient consideration of individual growth differences. By integrating multimodal detection tools to collect physiological, metabolic, and behavioral data, and through collaborative analysis and dynamic correction, it achieves accurate determination of obesity risk and localization of causes, thus assisting in effective intervention for childhood obesity.

[0007] To address the aforementioned deficiencies or improvement needs of existing technologies, as a first aspect of this invention, the present invention provides a method for detecting overweight and obesity in children based on multi-indicator synergistic judgment, comprising:

[0008] S1. Complete the deployment and initialization of the multimodal detection tool and complete the entry of children's basic information;

[0009] S2. Guide the child to use the integrated device and portable module to complete the collection of physical indicators according to the prompts. At the same time, retrieve the behavioral data of the child's smart wearable device, complete the data verification, and store the multimodal data in the local database of the terminal.

[0010] S3. Extract stored multimodal data, including physical indicators and behavioral data, and import them into the built-in model; assign differentiated weights to different growth stages based on the WHO child growth standards matched to the child's age and gender; and output the baseline value of the probability of overweight and obesity trends in individual children by combining the growth rate correction results.

[0011] S4. Quantify and score behavioral data, use a behavioral risk attribution model, construct a risk contribution matrix based on scores of dimensions including exercise, diet, and rest to locate the causes of overweight and obesity risk; calculate the risk contribution ratio, and adjust the probability value according to the ratio; finally, generate a comprehensive judgment result including the probability of overweight and obesity, the ranking of key influencing factors, and the risk warning level.

[0012] S5. Based on the comprehensive judgment results, generate a visual detection report including the values ​​of various indicators, reference ranges, and overweight and obesity levels.

[0013] Furthermore, the multimodal detection tool in S1 is an integrated detection device that adapts to children's physical fitness testing and integrates hardware terminals and software systems;

[0014] The hardware terminal integrates a detection unit for detecting core indicators including weight, height / length, body fat percentage, muscle mass, and waist circumference; an interface for connecting to a portable biochemical detection module; a communication unit for connecting to external devices; and an easy-to-use interactive module.

[0015] The software system incorporates multi-dimensional criteria for judging overweight and obesity that are tailored to the age and gender differences of children, as well as the corresponding logic for calculating and processing detection data.

[0016] Furthermore, S2 also includes: completing diet-related data, exercise-related data, and sleep-related data through a simplified questionnaire built into the terminal.

[0017] Furthermore, the differential weights in S3 are specifically as follows:

[0018] For children aged 6-9 in the early school years, construct a function age is the child's actual age, with a value of 6≤age≤9. The result of this function is multiplied by the original detection value of BMI data in the physiological indicators. At the same time, the original detection values ​​of metabolic indicators and behavioral data are multiplied by 1-f1(age). By observing the change of f1(age) with age, the participation of BMI data gradually decreases with age, while the participation of metabolic and behavioral data gradually increases.

[0019] For middle school children aged 10-12, functions are introduced. Δh represents the child's height growth rate over the past six months, and Δh0 represents the average height growth rate during this period. The raw test values ​​of physiological indicators, metabolic indicators, and behavioral data are multiplied by f2(Δh). When Δh approaches Δh0, f2(Δh) approaches 1, and each indicator participates in the calculation with almost equal participation. When Δh deviates from Δh0, f2(Δh) decreases simultaneously. The participation weight of each indicator is adjusted through the stability of height growth.

[0020] Establish a function for adolescents aged 13-17. Δbf represents the monthly change in body fat percentage, Δwaist represents the monthly change in waist circumference, and Δbf0 and Δwaist0 represent the average monthly changes in body fat percentage and waist circumference during the same period, respectively. The raw values ​​of body fat percentage, waist circumference, and metabolic indicators are multiplied by f3(Δbf, Δwaist), and the raw BMI value is multiplied by 1-f3(Δbf, Δwaist). A behavioral data fluctuation function is also introduced. σ is the weekly standard deviation of behavioral data, and σ0 is the average standard deviation of behavioral data during this period. The original detection value of behavioral data is multiplied by f4(σ). Through the synergistic effect of f3(Δbf,Δwaist) and f4(σ), the body fat percentage, waist circumference, metabolic indicators and participation of behavioral data are dynamically adjusted with changes in body composition and behavioral stability.

[0021] Furthermore, the method for calculating the baseline probability value of overweight and obesity trends in S3 is as follows:

[0022] Let Δh be the child's height growth rate over the past 6 months, and Δh0 be the average height growth rate of children of the same age; Δbf be the rate of change in body fat percentage over the past 3 months, and Δbf0 be the average rate of change in body fat percentage of children of the same age; σ be the weekly standard deviation of behavioral data over the past 4 weeks, and σ0 be the average volatility of behavioral data of children of the same age; among the physiological indicators, the standardized scores of BMI, body fat percentage, and waist circumference are denoted as BMI. z BF z Waist z The sum of standardized scores for metabolic indicators is denoted as Met. sumz The sum of standardized scores for behavioral indicators is denoted as Beh. sumz ;

[0023] The formula for calculating the baseline probability P of overweight and obesity trends is:

[0024]

[0025] in, It reflects the deviation of the height growth rate from the average of the same age group; It reflects the deviation of the rate of change in body fat percentage from the average of its age group. It reflects the deviation of behavioral volatility from the average of the same age group, and dynamically adjusts the impact of each dimension on P through the relative deviation relationship between indicators.

[0026] Furthermore, the method for constructing the risk contribution matrix in S4 is as follows:

[0027] Let M be the risk deviation of the exercise dimension; D be the risk deviation of the diet dimension; and S be the risk deviation of the daily routine dimension. The risk contribution matrix R is constructed as follows:

[0028]

[0029] In the matrix: the main diagonal elements M, D, and S represent the independent risk contribution of a single dimension; the off-diagonal elements are cross-product terms of different dimensions, reflecting the synergistic risk effect between dimensions; by using the proportion of each element in the total matrix, the contribution ratio of each dimension and its interaction to the risk of overweight and obesity is located, forming a risk contribution matrix.

[0030] Furthermore, the method for calculating the risk contribution ratio in S4 and adjusting the probability value according to the ratio is as follows:

[0031] Let the value of each element in the risk contribution matrix R be r. ij Let i and j correspond to the dimensions of exercise, diet, and sleep, respectively. The sum of all elements in the matrix is ​​R. total =∑r ij Then the proportion of independent risk contribution in the i-th dimension is The proportion of collaborative risk contribution between the i-th and j-th dimensions is

[0032] Let p be the maximum value among all contribution percentages. max The second largest value is denoted as p. sub Define the correction factor F = 1 + p max +p max ·p sub ;

[0033] The base probability value P of overweight and obesity trends is adjusted using a correction factor to obtain the corrected probability value P′=P·F; where the dominant risk dimension is determined by p. max The term directly amplifies the probability value, and its synergistic effect with the secondary dominant dimension is achieved through p. max ·p sub Further adjustments are made to achieve dynamic correction based on the natural distribution of risk contribution ratios, with risk factors having a higher ratio having a stronger impact on probability values.

[0034] As a second aspect of the present invention, a system for detecting overweight and obesity in children based on multi-indicator synergistic judgment is also provided, comprising:

[0035] The data acquisition and preparation unit is used to deploy and initialize the multimodal detection tool and to input the child's basic information.

[0036] The data acquisition and verification unit is used to guide children to complete the acquisition of physical indicators using integrated devices and portable modules according to prompts. At the same time, it retrieves behavioral data from children's smart wearable devices, completes data verification, and stores multimodal data in the terminal's local database.

[0037] The probability baseline value calculation unit is used to extract stored multimodal data, including physical indicators and behavioral data, and import them into the built-in model; assign differentiated weights to different growth stages based on the WHO child growth standards matched with the child's age and gender; and output the probability baseline value of the child's overweight and obesity trend by combining the growth rate correction results.

[0038] The comprehensive judgment unit is used to quantify and score behavioral data. It adopts a behavioral risk attribution model and constructs a risk contribution matrix based on scores in dimensions including exercise, diet, and rest to locate the risk factors of overweight and obesity. It calculates the risk contribution ratio and adjusts the probability value according to the ratio. Finally, it generates a comprehensive judgment result that includes the probability of overweight and obesity, the ranking of key influencing factors, and the risk warning level.

[0039] The report production unit is used to generate a visual inspection report based on the comprehensive judgment results, including the values ​​of various indicators, reference ranges, and overweight and obesity levels.

[0040] As a third aspect of the invention, a computer-readable storage medium is also provided, on which a computer program is stored, the computer program being executed by a processor of any step of the method for detecting overweight and obesity in children based on multi-indicator collaborative determination.

[0041] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0042] 1. The present invention provides a method for detecting overweight and obesity in children based on multi-indicator collaborative judgment. This method integrates physiological indicators collected by a device, metabolic indicators collected by a portable module, and behavioral data retrieved from children's smart wearable devices. After data verification, the data is stored in the terminal's local database. This approach integrates multi-source and multi-type data, covering physiological, metabolic, and behavioral dimensions, providing a comprehensive and accurate basis for subsequent analysis. It avoids the limitations of single data dimensions and lays a data foundation for the accuracy of overweight and obesity detection.

[0043] 2. The present invention provides a method for detecting overweight and obesity in children based on multi-indicator collaborative judgment. This method extracts multimodal data and imports it into a built-in model. Based on the child's age and gender, it assigns differentiated weights according to the WHO child growth standards and metabolic indicator reference ranges. Combined with growth rate correction results, it outputs a baseline value for the probability of overweight and obesity trends. Simultaneously, it calculates collaborative abnormality coefficients for metabolic indicators to classify metabolic types, constructs a risk contribution matrix from behavioral data to locate the causes of overweight and obesity risks, and corrects the probability values. Finally, it integrates these factors to generate a comprehensive judgment result and a visualized detection report. This process achieves collaborative analysis and dynamic correction of multiple indicators, accurately outputting the probability of overweight and obesity, clarifying the ranking of key influencing factors and risk warning levels, and providing comprehensive and targeted detection results. Attached Figure Description

[0044] Figure 1 This is a flowchart of a method for detecting overweight and obesity in children based on multi-indicator collaborative determination, according to an embodiment of the present invention.

[0045] Figure 2 This is a schematic diagram of an age-specific BMI detection card for boys under 7 years old, according to an embodiment of the present invention.

[0046] Figure 3 This is a schematic diagram of an age-specific BMI detection card for girls under 7 years old, according to an embodiment of the present invention.

[0047] Figure 4 This is a schematic diagram of the BMI classification standard card for children under 7 years old according to an embodiment of the present invention;

[0048] Figure 5 This is a schematic diagram of a BMI screening test card for children aged 6-18 years according to an embodiment of the present invention;

[0049] Figure 6 This is a system unit diagram of an embodiment of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0051] Example 1

[0052] Please refer to Figure 1 This embodiment 1 provides a method for detecting overweight and obesity in children based on multi-indicator synergistic judgment, including:

[0053] S1. Complete the deployment and initialization of the multimodal detection tool and complete the entry of children's basic information;

[0054] S2. Guide the child to use the integrated device and portable module to complete the collection of physical indicators according to the prompts. At the same time, retrieve the behavioral data of the child's smart wearable device, complete the data verification, and store the multimodal data in the local database of the terminal.

[0055] S3. Extract stored multimodal data, including physical indicators and behavioral data, and import them into the built-in model; assign differentiated weights to different growth stages based on the WHO child growth standards matched to the child's age and gender; and output the baseline value of the probability of overweight and obesity trends in individual children by combining the growth rate correction results.

[0056] S4. Quantify and score behavioral data, use a behavioral risk attribution model, construct a risk contribution matrix based on scores of dimensions including exercise, diet, and rest to locate the causes of overweight and obesity risk; calculate the risk contribution ratio, and adjust the probability value according to the ratio; finally, generate a comprehensive judgment result including the probability of overweight and obesity, the ranking of key influencing factors, and the risk warning level.

[0057] S5. Based on the comprehensive judgment results, generate a visual detection report including the values ​​of various indicators, reference ranges, and overweight and obesity levels.

[0058] This embodiment 1 further elaborates on the above steps.

[0059] (1) Data collection preparation

[0060] Considering the limitations of traditional methods for detecting overweight and obesity in children, such as limited device functionality and fragmented data acquisition and analysis, the deployment and initialization of a dedicated multimodal detection tool was initiated. This tool is an integrated detection device combining hardware and software systems, adapted to children's physical fitness testing scenarios. The hardware terminal integrates detection units for core indicators including weight, height / length, body fat percentage, muscle mass, and waist circumference; an interface for connecting to portable biochemical detection modules; a communication unit for connecting to external devices; and an interactive module for easy operation by testing personnel. The software system incorporates multidimensional overweight and obesity assessment criteria tailored to children's age and gender differences, along with corresponding data calculation and processing logic.

[0061] During deployment, the placement of the equipment should be planned in conjunction with the routine testing process to ensure that subsequent operations conform to the interaction habits of operators and children; during the initialization phase, the functions of each module of the equipment should be verified to ensure stable operation of the hardware and software.

[0062] After preparing the tools, the operator enters the child's basic information on the interactive module of the hardware terminal. Following the interface instructions, the operator accurately inputs key information such as the child's name, gender, and age. This information will serve as the basis for subsequent steps in matching WHO child growth standards and distinguishing individual growth differences. The hardware terminal's integrated detection unit (which can detect weight, height / length, body fat percentage, muscle mass, and waist circumference), interface, and communication unit for connecting to the portable biochemical detection module all rely on this basic information to correlate the corresponding test data. The software system's built-in overweight and obesity judgment standards and data processing logic, which are tailored to the child's age and gender differences, also need to use this information to determine the applicable analysis dimensions, avoiding misjudgments caused by insufficient consideration of individual differences in traditional testing.

[0063] (2) Data collection and verification

[0064] Considering the fragmented nature of traditional childhood overweight and obesity detection methods, which involve the collection of physical and behavioral data (including physiological and metabolic indicators) and suffer from insufficient accuracy due to reliance on manual recording, operators guide children through multi-dimensional data collection following equipment prompts. First, children are guided to use the integrated device's detection unit to collect physiological indicators such as weight, height / length, body fat percentage, muscle mass, and waist circumference. The device acquires and initially records these core physiological data in real time. Simultaneously, operators connect a portable biochemical detection module to a dedicated interface on the hardware terminal. This module collects metabolic indicators such as fasting insulin and low-density lipoprotein cholesterol, ensuring that metabolic and physiological data originate from the same testing environment and reducing data bias.

[0065] Preferably, while collecting physiological and metabolic indicators, the communication unit of the hardware terminal will establish a connection with the child's smart wearable device and automatically retrieve behavioral data such as exercise duration and sleep duration over a recent period of time. In addition, the operator will also use the simplified questionnaire built into the terminal to supplement the collection of children's diet-related data (such as the frequency of high-sugar diets) and details of exercise and daily routines, further improving the dimensions of behavioral data and avoiding data omissions that may occur when relying solely on wearable devices.

[0066] After all data collection is complete, the system will automatically initiate a data verification process, comparing the logical rationality of physiological indicators (such as the matching degree between height and age), the validity of the detection range of metabolic indicators, and the temporal continuity of behavioral data. Abnormal data will be removed, and operators will be prompted to supplement missing information. After successful verification, the integrated physiological, metabolic, and behavioral multimodal data will be uniformly stored in the terminal's local database, providing complete and accurate basic data support for subsequent age- and gender-matched growth standards and multi-indicator collaborative analysis.

[0067] Please refer to Figures 2-5 Preferably, after data collection, the multimodal detection tool uses a BMI stratification standard based on age, gender, and evaluation system to achieve rapid initial screening of overweight and obesity risks in children. Specifically, it integrates age-specific BMI standard deviation tables for children under 7 years old (separate for males and females), gender-specific BMI overweight and obesity cutoff tables for school-aged children and adolescents aged 6-18, and the standard deviation evaluation method for nutritional status of children under 7 years old. It accurately matches corresponding BMI reference intervals according to the child's age and gender (e.g., for children under 7 years old, the ±SD interval is used to determine underweight, overweight, and obesity levels; for children aged 6-18, the gender-specific cutoff values ​​are used to determine overweight / obesity). This allows for the immediate identification of children at risk of overweight and obesity from a BMI perspective after data collection, providing a targeted entry point for subsequent multi-indicator collaborative assessment.

[0068] (3) Calculation of basic probability values

[0069] Furthermore, the system extracts stored physiological, metabolic, and behavioral data from the local database and imports this multimodal data into the built-in model for analysis. During the analysis, the model matches the corresponding WHO child growth standards and metabolic reference ranges based on the previously entered child's age and gender, assigning differentiated weights to each indicator for different growth stages.

[0070] For children aged 6-9 in the early school years, construct a function age is the child's actual age, with a value of 6≤age≤9. The result of this function is multiplied by the original detection value of BMI data in the physiological indicators. At the same time, the original detection values ​​of metabolic indicators and behavioral data are multiplied by 1-f1(age). By observing the change of f1(age) with age, the participation of BMI data gradually decreases with age, while the participation of metabolic and behavioral data gradually increases.

[0071] For middle school children aged 10-12, functions are introduced. Δh represents the child's height growth rate over the past six months, and Δh0 represents the average height growth rate during this period. The raw test values ​​of physiological indicators, metabolic indicators, and behavioral data are multiplied by f2(Δh). When Δh approaches Δh0, f2(Δh) approaches 1, and each indicator participates in the calculation with almost equal participation. When Δh deviates from Δh0, f2(Δh) decreases simultaneously. The participation weight of each indicator is adjusted through the stability of height growth.

[0072] Establish a function for adolescents aged 13-17. Δbf represents the monthly change in body fat percentage, Δwaist represents the monthly change in waist circumference, and Δbf0 and Δwaist0 represent the average monthly changes in body fat percentage and waist circumference during the same period, respectively. The raw values ​​of body fat percentage, waist circumference, and metabolic indicators are multiplied by f3(Δbf, Δwaist), and the raw BMI value is multiplied by 1-f3(Δbf, Δwaist). A behavioral data fluctuation function is also introduced. σ is the weekly standard deviation of behavioral data, and σ0 is the average standard deviation of behavioral data during this period. The original detection value of behavioral data is multiplied by f4(σ). Through the synergistic effect of f3(Δbf,Δwaist) and f4(σ), the body fat percentage, waist circumference, metabolic indicators and participation of behavioral data are dynamically adjusted with changes in body composition and behavioral stability.

[0073] After completing the differentiated weight allocation, the model combines the correction results of the child's height growth rate over the past 6 months, and comprehensively considers the relative deviation of various physiological, metabolic, and behavioral indicators from the average of the same age—including the degree of deviation of height growth rate, body fat percentage change rate, and deviation of behavioral data fluctuation. Through the intrinsic correlation between these indicators, the model dynamically adjusts the influence of each dimension, and finally outputs the basic value of the probability of overweight and obesity trend development for individual children, providing an initial basis for more accurate subsequent judgment.

[0074] Preferably, the method for calculating the baseline value of the probability of overweight and obesity trends is as follows:

[0075] Let Δh be the child's height growth rate over the past 6 months, and Δh0 be the average height growth rate of children of the same age; Δbf be the rate of change in body fat percentage over the past 3 months, and Δbf0 be the average rate of change in body fat percentage of children of the same age; σ be the weekly standard deviation of behavioral data over the past 4 weeks, and σ0 be the average volatility of behavioral data of children of the same age; among the physiological indicators, the standardized scores of BMI, body fat percentage, and waist circumference are denoted as BMI. z BF z Waist z The sum of standardized scores for metabolic indicators is denoted as Met. sumz The sum of standardized scores for behavioral indicators is denoted as Beh. sumz ;

[0076] The formula for calculating the baseline probability P of overweight and obesity trends is:

[0077]

[0078] in, It reflects the deviation of the height growth rate from the average of the same age group; It reflects the deviation of the rate of change in body fat percentage from the average of its age group. It reflects the deviation of behavioral volatility from the average of the same age group, and dynamically adjusts the impact of each dimension on P through the relative deviation relationship between indicators.

[0079] (4) Comprehensive judgment

[0080] Furthermore, in areas with well-equipped metabolic indicator collection devices, optional in-depth analysis of metabolic indicators is also conducted simultaneously. Using the reference range for children of the same age as a benchmark, abnormal values ​​for fasting insulin and low-density lipoprotein cholesterol are calculated. When the measured values ​​of these two indicators exceed the upper limit of their respective reference ranges, the degree of abnormality is quantified according to the extent of the exceedance; otherwise, the abnormal value is recorded as 0. Through these two abnormal values ​​and the cross term reflecting their synergistic effect, a synergistic abnormality coefficient is obtained. Based on this coefficient, children's metabolic types are classified into normal, mildly abnormal, and moderately or severely abnormal types, according to the combination of abnormality levels corresponding to the coefficient magnitude.

[0081] Specifically, let the measured value of fasting insulin be Ins, and its upper limit of the age-related reference range be Ins. max The measured value of low-density lipoprotein cholesterol was LDL-C, and its upper limit of the age-related reference range was LDL-C. max ; Calculate outliers separately:

[0082] If Ins>Ins max ,but Otherwise, Instagram abn =0; Ins abn This represents an abnormal value for fasting insulin.

[0083] If LDL-C > LDL-C max ,but Otherwise, LDL-C abn =0; LDL-C abn This indicates an abnormal value for low-density lipoprotein cholesterol.

[0084] Next, calculate the co-anomaly coefficient Coeff:

[0085] Coeff=Ins abn +LDL-C abn +k×Ins abn ×LDL-C abn

[0086] The classification is based on the co-abnormality coefficient. If Coeff < α, it is classified as "normal metabolism", meaning that neither of the two indicators is significantly abnormal or only one indicator is slightly abnormal. If α ≤ Coeff < β, it is classified as "mild metabolic abnormality", meaning that one indicator is moderately abnormal or both indicators are slightly abnormal. If Coeff ≥ β, it is classified as "moderate or above metabolic abnormality", meaning that one indicator is severely abnormal or both indicators are moderate or above abnormal. Here, k is the co-abnormality coefficient, and α and β are the preset metabolic type classification thresholds.

[0087] Simultaneously, the system quantifies and scores behavioral data, determining risk deviation from three dimensions: exercise, diet, and lifestyle, and constructs a risk contribution matrix. The elements on the main diagonal of the matrix represent the independent risk contribution of each dimension, while the off-diagonal elements are cross-product terms of different dimensions, reflecting the synergistic risk effects between dimensions. By calculating the proportion of each element in the matrix's sum, the risk contribution percentage of each dimension and its interactions is obtained, thereby pinpointing the causes of overweight and obesity risk.

[0088] In a preferred embodiment, the risk contribution matrix is ​​constructed as follows:

[0089] Let M be the risk deviation of the exercise dimension; D be the risk deviation of the diet dimension; and S be the risk deviation of the daily routine dimension. The risk contribution matrix R is constructed as follows:

[0090]

[0091] In the matrix: the main diagonal elements M, D, and S represent the independent risk contribution of a single dimension; the off-diagonal elements are cross-product terms of different dimensions, reflecting the synergistic risk effect between dimensions; by using the proportion of each element in the total matrix, the contribution ratio of each dimension and its interaction to the risk of overweight and obesity is located, forming a risk contribution matrix.

[0092] Simultaneously, based on the risk contribution ratio, the largest and second largest ratios are selected, and a correction factor is constructed using the two and their product; specifically, let the value of each element in the risk contribution matrix R be r. ij Let i and j correspond to the dimensions of exercise, diet, and sleep, respectively. The sum of all elements in the matrix is ​​R. total =∑r ij Then the proportion of independent risk contribution in the i-th dimension is The proportion of collaborative risk contribution between the i-th and j-th dimensions is

[0093] Let p be the maximum value among all contribution percentages. max The second largest value is denoted as p. sub Define the correction factor F = 1 + p max +p max ·p sub ;

[0094] The base probability value P of overweight and obesity trends is adjusted using a correction factor to obtain the corrected probability value P′=P·F; where the dominant risk dimension is determined by p. max The term directly amplifies the probability value, and its synergistic effect with the secondary dominant dimension is achieved through p. max ·p sub Further adjustments are made to achieve dynamic correction based on the natural distribution of risk contribution ratios, with risk factors having a higher ratio having a stronger impact on probability values.

[0095] (5) Report production

[0096] Based on the comprehensive assessment results obtained from previous integration, a visual test report can be automatically generated. The report clearly presents the specific values ​​of various physiological, metabolic, and behavioral data of children, while also indicating the age-appropriate reference range for each indicator, making the comparison between the data and standards immediately clear. In addition, the report also clearly indicates the child's overweight and obesity level, as well as the ranking of key influencing factors and the risk warning level, ensuring that the information is presented comprehensively and intuitively, making it easy for testing personnel to quickly grasp the core testing information related to overweight and obesity in children.

[0097] After obtaining the visualized test report, the report content is first reviewed. Combining the child's specific test data, overweight and obesity levels, and risk factors, a targeted analysis is conducted on children with abnormal initial screening results or those diagnosed with overweight or obesity. For children with abnormal results, the test results are explained in detail to the child and their accompanying guardian, explaining the degree to which the current indicators deviate from the reference range, the potential impact, and the correlation between major risk factors in the behavioral dimension (such as insufficient exercise and high-sugar diets) and overweight and obesity. Based on this, specific health recommendations are further provided, including suitable types and durations of exercise for children, directions for dietary adjustments, and plans to optimize sleep patterns. These recommendations are ensured to be actionable and relevant to the child's actual life, helping the child improve their health through targeted interventions.

[0098] Example 2

[0099] Please refer to Figure 6 This embodiment 2 provides a system for detecting overweight and obesity in children based on multi-indicator collaborative judgment, including:

[0100] The data acquisition and preparation unit is used to deploy and initialize the multimodal detection tool and to input the child's basic information.

[0101] The data acquisition and verification unit is used to guide children to complete the acquisition of physical indicators using integrated devices and portable modules according to prompts. At the same time, it retrieves behavioral data from children's smart wearable devices, completes data verification, and stores multimodal data in the terminal's local database.

[0102] The probability baseline value calculation unit is used to extract stored multimodal data, including physical indicators and behavioral data, and import them into the built-in model; assign differentiated weights to different growth stages based on the WHO child growth standards matched with the child's age and gender; and output the probability baseline value of the child's overweight and obesity trend by combining the growth rate correction results.

[0103] The comprehensive judgment unit is used to quantify and score behavioral data. It adopts a behavioral risk attribution model and constructs a risk contribution matrix based on scores in dimensions including exercise, diet, and rest to locate the risk factors of overweight and obesity. It calculates the risk contribution ratio and adjusts the probability value according to the ratio. Finally, it generates a comprehensive judgment result that includes the probability of overweight and obesity, the ranking of key influencing factors, and the risk warning level.

[0104] The report production unit is used to generate a visual inspection report based on the comprehensive judgment results, including the values ​​of various indicators, reference ranges, and overweight and obesity levels.

[0105] Example 3

[0106] This embodiment 3 also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it can implement any step of a method for detecting overweight and obesity in children based on multi-indicator collaborative judgment.

[0107] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0108] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments; further details will not be repeated here.

[0109] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for detecting overweight and obesity in children based on multi-indicator synergistic judgment, characterized in that, include: S1. Complete the deployment and initialization of the multimodal detection tool and complete the entry of children's basic information; S2. Guide the child to use the integrated device and portable module to complete the collection of physical indicators according to the prompts. At the same time, retrieve the behavioral data of the child's smart wearable device, complete the data verification, and store the multimodal data in the local database of the terminal. S3. Extract stored multimodal data, including physical indicators and behavioral data, and import them into the built-in model; assign differentiated weights to different growth stages based on the WHO child growth standards matched to the child's age and gender; and output the baseline value of the probability of overweight and obesity trends in individual children by combining the growth rate correction results. S4. Quantify and score behavioral data, and use a behavioral risk attribution model to construct a risk contribution matrix based on scores in dimensions including exercise, diet, and rest to pinpoint the causes of overweight and obesity risk. Calculate the risk contribution ratio, and adjust the probability value according to the ratio; finally, generate a comprehensive judgment result that includes the probability of overweight and obesity, the ranking of key influencing factors, and the risk warning level. S5. Based on the comprehensive judgment results, generate a visual detection report including the values ​​of various indicators, reference ranges, and overweight and obesity levels.

2. The method for detecting overweight and obesity in children based on multi-indicator synergistic judgment according to claim 1, characterized in that, The multimodal detection tool in S1 is an integrated detection device that is adapted to children's physical fitness testing and combines hardware terminals and software systems; The hardware terminal integrates a detection unit for detecting core indicators including weight, height / length, body fat percentage, muscle mass, and waist circumference; an interface for connecting to a portable biochemical detection module; a communication unit for connecting to external devices; and an easy-to-use interactive module. The software system incorporates multi-dimensional criteria for judging overweight and obesity that are tailored to the age and gender differences of children, as well as the corresponding logic for calculating and processing detection data.

3. The method for detecting overweight and obesity in children based on multi-indicator synergistic judgment according to claim 1, characterized in that, The S2 also includes: completing diet-related data, exercise-related data, and daily routine-related data through a simplified questionnaire built into the terminal.

4. The method for detecting overweight and obesity in children based on multi-indicator synergistic judgment according to claim 1, characterized in that, The specific differential weights in S3 are as follows: For children aged 6-9 in the early school years, construct a function age is the child's actual age, with a value of 6≤age≤9. The result of this function is multiplied by the original detection value of BMI data in the physiological indicators. At the same time, the original detection values ​​of metabolic indicators and behavioral data are multiplied by 1-f1(age). By observing the change of f1(age) with age, the participation of BMI data gradually decreases with age, while the participation of metabolic and behavioral data gradually increases. For middle school children aged 10-12, functions are introduced. Δh represents the child's height growth rate over the past six months, and Δh0 represents the average height growth rate during this period. The raw test values ​​of physiological indicators, metabolic indicators, and behavioral data are multiplied by f2(Δh). When Δh approaches Δh0, f2(Δh) approaches 1, and each indicator participates in the calculation with almost equal participation. When Δh deviates from Δh0, f2(Δh) decreases simultaneously. The participation weight of each indicator is adjusted through the stability of height growth. Establish a function for adolescents aged 13-17. Δbf represents the monthly change in body fat percentage, Δwaist represents the monthly change in waist circumference, and Δbf0 and Δwaist0 represent the average monthly changes in body fat percentage and waist circumference during the same period, respectively. The raw values ​​of body fat percentage, waist circumference, and metabolic indicators are multiplied by f3(Δbf, Δwaist), and the raw BMI value is multiplied by 1-f3(Δbf, Δwaist). A behavioral data fluctuation function is also introduced. σ is the weekly standard deviation of behavioral data, and σ0 is the average standard deviation of behavioral data during this period. The original detection value of behavioral data is multiplied by f4(σ). Through the synergistic effect of f3(Δbf,Δwaist) and f4(σ), the body fat percentage, waist circumference, metabolic indicators and participation of behavioral data are dynamically adjusted with changes in body composition and behavioral stability.

5. The method for detecting overweight and obesity in children based on multi-indicator synergistic judgment according to claim 1, characterized in that, The method for calculating the baseline value of the probability of overweight and obesity trends in S3 is as follows: Let Δh be the child's height growth rate over the past 6 months, and Δh0 be the average height growth rate of children of the same age; Δbf be the rate of change in body fat percentage over the past 3 months, and Δbf0 be the average rate of change in body fat percentage of children of the same age; σ be the weekly standard deviation of behavioral data over the past 4 weeks, and σ0 be the average volatility of behavioral data of children of the same age; among the physiological indicators, the standardized scores of BMI, body fat percentage, and waist circumference are denoted as BMI. z BF z Waist z The sum of standardized scores for metabolic indicators is denoted as Met. sumz The sum of standardized scores for behavioral indicators is denoted as Beh. sumz ; The formula for calculating the baseline probability P of overweight and obesity trends is: in, It reflects the deviation of the height growth rate from the average of the same age group; It reflects the deviation of the rate of change in body fat percentage from the average of its age group. It reflects the deviation of behavioral volatility from the average of the same age group, and dynamically adjusts the impact of each dimension on P through the relative deviation relationship between indicators.

6. The method for detecting overweight and obesity in children based on multi-indicator synergistic judgment according to claim 1, characterized in that, The method for constructing the risk contribution matrix in S4 is as follows: Let M be the risk deviation of the exercise dimension; D be the risk deviation of the diet dimension; and S be the risk deviation of the daily routine dimension. The risk contribution matrix R is constructed as follows: In the matrix: the main diagonal elements M, D, and S represent the independent risk contribution of a single dimension; the off-diagonal elements are cross-product terms of different dimensions, reflecting the synergistic risk effect between dimensions; by using the proportion of each element in the total matrix, the contribution ratio of each dimension and its interaction to the risk of overweight and obesity is located, forming a risk contribution matrix.

7. The method for detecting overweight and obesity in children based on multi-indicator synergistic judgment according to claim 6, characterized in that, The method for calculating the risk contribution ratio in S4 and adjusting the probability value according to the ratio is as follows: Let the value of each element in the risk contribution matrix R be r. ij Let i and j correspond to the dimensions of exercise, diet, and sleep, respectively. The sum of all elements in the matrix is ​​R. total =∑r ij Then the proportion of independent risk contribution in the i-th dimension is The proportion of collaborative risk contribution between the i-th and j-th dimensions is Let p be the maximum value among all contribution percentages. max The second largest value is denoted as p. sub Define the correction factor F = 1 + p max +p max ·p sub ; The base probability value P of overweight and obesity trends is adjusted using a correction factor to obtain the corrected probability value P′=P·F; where the dominant risk dimension is determined by p. max The term directly amplifies the probability value, and its synergistic effect with the secondary dominant dimension is achieved through p. max ·p sub Further adjustments are made to achieve dynamic correction based on the natural distribution of risk contribution ratios, with risk factors having a higher ratio having a stronger impact on probability values.

8. A system for detecting overweight and obesity in children based on multi-indicator synergistic judgment, characterized in that, include: The data acquisition and preparation unit is used to deploy and initialize the multimodal detection tool and to input the child's basic information. The data acquisition and verification unit is used to guide children to complete the acquisition of physical indicators using integrated devices and portable modules according to prompts. At the same time, it retrieves behavioral data from children's smart wearable devices, completes data verification, and stores multimodal data in the terminal's local database. The probability baseline value calculation unit is used to extract stored multimodal data, including physical indicators and behavioral data, and import them into the built-in model; assign differentiated weights to different growth stages based on the WHO child growth standards matched with the child's age and gender; and output the probability baseline value of the child's overweight and obesity trend by combining the growth rate correction results. The comprehensive judgment unit is used to quantify and score behavioral data. It adopts a behavioral risk attribution model and constructs a risk contribution matrix based on scores of dimensions including exercise, diet, and rest to locate the causes of overweight and obesity risk. Calculate the risk contribution ratio, and adjust the probability value according to the ratio; finally, generate a comprehensive judgment result that includes the probability of overweight and obesity, the ranking of key influencing factors, and the risk warning level. The report production unit is used to generate a visual inspection report based on the comprehensive judgment results, including the values ​​of various indicators, reference ranges, and overweight and obesity levels.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is executed by a processor as described in any one of claims 1-7: a method for detecting overweight and obesity in children based on multi-indicator collaborative determination.

Images