Children body fat rate adaptive prediction method and system based on trajectory feature collaboration

Abstract

The invention relates to the technical field of child health assessment, in particular to a child body fat rate adaptive prediction method and system based on trajectory feature collaboration.The method comprises the following steps that a child physical examination data set is obtained, height and weight changes are analyzed, difference ratio screening samples are calculated, waistline change slope assessment stability screening samples are extracted, and a child body fat rate prediction result is obtained; calculating a sign proportion difference to adjust a training weight, identifying an age group residual difference to adjust model parameters, and adjusting a prediction slope generation result in combination with a body fat trend direction and a body weight trend direction. According to the method, samples are screened through height and weight tracks, abnormal data are eliminated in combination with fluctuation continuity of waistline change slope, a feature weighting coefficient is set according to proportion deviation of height, weight and muscle amount, and a response coefficient is adjusted through a distance relation between a grouping mean value of age group residual errors and a concentration interval. And the change path is judged by utilizing the consistency of the body fat and body weight prediction trend, so that non-trend prediction offset is avoided, and the stability of the model and the sample adaptability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of child health assessment, and in particular to a method and system for adaptively predicting children's body fat percentage based on trajectory feature collaboration. Background Art

[0002] The field of child health assessment technology includes various methods and means for quantitative analysis and comprehensive evaluation of the physical health status of children, involving multiple aspects such as child growth and development monitoring, physiological indicator assessment, nutritional status judgment, and potential health risk identification. By collecting basic physical data such as height, weight, waist circumference, BMI, etc. of children, combined with medical standards and health models for assessment and analysis, individualized health reference information is provided for children. In combination with information technology such as electronic health records and health data modeling, continuous tracking and dynamic assessment of children's health status are achieved. Among them, the child body fat prediction method refers to the construction of a child's physical data-oriented method. The body fat percentage prediction model uses a supervised learning mechanism to establish a mapping relationship between variables to estimate body fat levels. It covers the collection of children's physical sign data, training data construction, feature variable selection, model structure setting and prediction output. Specifically, it includes constructing a data sample set based on basic variables such as children's height, weight, age, gender, BMI, etc., using the regression model as the core prediction architecture, combining data standardization processing, multivariate linear fitting, residual analysis optimization and other means to train the model, and completing the model accuracy evaluation through cross-validation. The estimated value of the child's body fat percentage is obtained through the model output, realizing indirect prediction of body fat percentage based on static physical sign data.

[0003] Traditional children's body fat prediction technology uses static physical sign variables to construct a single-cycle sample set and fits the body fat output with a global model structure. It does not dynamically segment the physical sign trajectories of the samples during cyclical growth changes, and is unable to eliminate structurally incoordinated training data during the developmental fluctuation period, resulting in a mixed sample input structure and unstable convergence during training. It also causes systematic errors in the prediction results in groups with unbalanced feature proportions. For example, the BMI input of groups with abnormal muscle proportions amplifies the fat judgment deviation. During continuous monitoring, the lack of a trend adjustment mechanism causes directional distortion of the body fat prediction value, affecting the reliability of the model and the credibility range of the data in health monitoring scenarios. Summary of the Invention

[0004] In order to solve the technical problems existing in the prior art, the embodiments of the present invention provide a method and system for adaptively predicting children's body fat percentage based on trajectory feature collaboration. The technical solution is as follows: To achieve the above objectives, the present invention adopts the following technical solution: a method for adaptively predicting children's body fat percentage based on trajectory feature collaboration, comprising the following steps: S1: Obtain a children's physical examination dataset, analyze the height and weight change trajectories of each sample, calculate the difference ratio between the height and weight growth of each sample in consecutive cycles, and screen the samples based on the distribution characteristics of the difference ratio in multiple intervals to generate a rhythm screening sample set; S2: calling the rhythm screening sample set, analyzing the waist circumference change data of each sample in multiple consecutive detection cycles after screening, calculating the difference in waist circumference change slopes between adjacent cycles, evaluating the stability of the waist circumference change of the sample, further screening the sample, and generating a waist circumference rhythm screening set; S3: Using the waist rhythm screening set, calculating the ratio difference between height, weight and muscle mass of each sample after further screening, calculating the degree of deviation and adjusting the feature weights of the training stage based on the distribution characteristics of the ratio difference in multiple intervals, and generating training weight adjustment data; S4: Extract the training weight adjustment data, train the body fat prediction model, calculate the mean prediction residual of the sample in each age group, identify the distance from the overall sample residual concentration interval, screen the age group with abnormal residuals, adjust the residual response adjustment coefficient, and generate the model parameter adjustment result.

[0005] As a further solution of the present invention, the rhythm screening sample set includes height growth trajectory, weight growth trajectory, and difference ratio distribution interval; the waist circumference rhythm screening set includes waist circumference change slope, slope fluctuation continuity, and waist circumference change trend structure; the training weight adjustment data includes the main interval of proportional difference, sample deviation degree, and feature weighting coefficient; the model parameter adjustment results include prediction residual grouping, age group error level, and residual response adjustment coefficient.

[0006] As a further solution of the present invention, the steps of obtaining the rhythm screening sample set are specifically as follows: S101: Obtain a children's physical examination data set, extract the height data and weight data of each sample in multiple consecutive test cycles, call the height change and weight change of each sample in adjacent cycles, calculate the difference ratio between the height increase and the weight increase, and generate height-weight difference ratio data; S102: Based on the height-weight difference ratio data, the distribution number of samples within each difference ratio interval is identified, and the density difference of the number of samples within each interval is analyzed to generate difference ratio distribution density data; S103: Call the difference ratio distribution density data, identify the interval with the highest distribution density according to the concentration of the sample distribution density in each interval, perform preliminary screening on the samples, and generate a rhythm screening sample set.

[0007] As a further solution of the present invention, the steps for obtaining the waist rhythm screening set are specifically as follows: S201: calling the rhythm screening sample set, extracting waist circumference data of each sample in multiple consecutive detection cycles, calculating the waist circumference change rate of each sample in multiple cycles according to the recording sequence between adjacent detection cycles, and generating waist circumference change slope sequence data; S202: selecting the slope difference between adjacent periods of each sample based on the waist circumference change slope sequence data, analyzing the slope change trend of each sample, evaluating the stability of the waist circumference change of the sample, and generating waist circumference fluctuation structure data; S203: calling the waist circumference fluctuation structure data, further screening the samples according to the stability of the changes in the sample waist circumference data, and generating a waist circumference rhythm screening set.

[0008] As a further embodiment of the present invention, the specific formula for evaluating the stability of waist circumference changes of the sample is: ; Calculate the stability structure value of the waist circumference change slope difference; in, For the The sample in Normalized value of waist circumference change slope in each cycle, For the The sample in Normalized value of waist circumference change slope in each cycle, For the The sample in The horizontal detection frequency weighting factor of each cycle, For the The sample in Normalized value of waist circumference change rate per cycle, For the The age smoothing adjustment factor corresponding to the sample is: For the The number of consecutive detection cycles contained in a sample, is the period index variable, is the sample index variable, For the The stability structure value of the waist circumference change slope difference corresponding to the samples.

[0009] As a further solution of the present invention, the step of obtaining the training weight adjustment data is specifically as follows: S301: Using the waist rhythm screening set, extracting the height, weight, and muscle mass data of each sample after further screening, calculating the ratio of weight to height of each sample, and obtaining the ratio value to muscle mass, and generating body shape ratio distribution data by analyzing the distribution of the ratio values ​​in multiple continuous numerical intervals; S302: Calling the body proportion distribution data, identifying the interval with the largest proportion quantity distribution density and using it as a body proportion benchmark, calculating the difference between each sample proportion value and the benchmark, and generating body structure deviation data; S303: Based on the body structure deviation data, the deviation degree of each sample is evaluated according to the gap between the sample ratio value and the benchmark, the feature weighting coefficient of the training stage is adjusted, the weight parameter set of the training sample is constructed, and the training weight adjustment data is generated.

[0010] As a further solution of the present invention, the steps of obtaining the model parameter adjustment results are specifically as follows: S401: extracting the training weight adjustment data, using the sample to train the body fat prediction model, calculating the residual value between the sample prediction result and the actual result, combining the age group label of each sample, grouping the residual data, and generating age group residual group data; S402: calling the residual grouping data of the age group, calculating the residual mean within each age group, identifying the concentrated interval of the residual of the overall training sample, analyzing the distance between the residual mean of each age group and the overall residual interval, evaluating the deviation range of the residual mean, and generating age group residual deviation data; S403: Based on the residual deviation data of the age groups, according to the distance between the average residual level of each age group and the residual concentration interval of the overall sample, the age groups with abnormal residuals are screened, and the residual response adjustment coefficient of the target age group samples is adjusted to generate the model parameter adjustment results.

[0011] As a further embodiment of the present invention, the method further comprises: S5: Based on the model parameter adjustment results, continuous vital sign data records of the target user are collected, the user's body fat percentage is predicted using the body fat prediction model, the changing trend of the body fat prediction results over multiple periods is identified, and the results are compared with the user's weight data to analyze the consistency of the changing trends. Based on the trend direction relationship, the slope adjustment amplitude of the body fat prediction change path is set to generate a body fat trend prediction result; The body fat trend prediction result specifically includes body fat prediction data, body fat change trend, and trend slope adjustment amplitude.

[0012] As a further embodiment of the present invention, the steps for obtaining the body fat trend prediction result are specifically as follows: S501: Based on the model parameter adjustment result, the target user's physical sign data such as height, weight, waist circumference, etc. for multiple consecutive periods are collected, the body fat prediction model is called, and the body fat prediction result of the user for each period is output to generate body fat percentage sequence data; S502: Based on the body fat percentage sequence data, extract the target user's weight change trend in each cycle, analyze the change direction of the body fat percentage sequence and the weight data, analyze the consistency of the change trend, obtain a trend consistency index, and generate trend relationship data; The specific formula for analyzing the consistency of the change trend is: ; Calculate trend consistency indicators; in, represents the trend consistency indicator, Representative The normalized value of the change in body fat percentage within a cycle, Representative Normalized value of weight change within a cycle, Indicates the total number of consecutive detection cycles, Indicates the current cycle number. Represents a non-zero minimum constant set to prevent the denominator from being zero; S503: Calling the trend relationship data, adjusting the slope adjustment amplitude of the body fat prediction change path according to the relationship between the body fat percentage change trend and the weight change trend, and obtaining the body fat trend prediction result.

[0013] On the other hand, a system for adaptively predicting body fat percentage of children based on trajectory feature collaboration is provided. The system is applied to the method for adaptively predicting body fat percentage of children based on trajectory feature collaboration. The system includes: The physical sign screening module obtains a children's physical examination data set, analyzes the height and weight change trajectory of each sample, calculates the difference ratio between the height and weight growth of each sample in consecutive cycles, and screens the samples based on the distribution characteristics of the difference ratio in multiple intervals to generate a rhythm screening sample set; The waist circumference stability module calls the rhythm screening sample set, analyzes the waist circumference change data of each sample in multiple consecutive detection cycles after screening, calculates the difference in waist circumference change slopes between adjacent cycles, evaluates the stability of the sample waist circumference change, further screens the sample, and generates a waist circumference rhythm screening set; The proportional weight control module uses the waist rhythm screening set to calculate the proportional difference between the height, weight and muscle mass of each sample after further screening, calculates the degree of deviation based on the distribution characteristics of the proportional difference in multiple intervals, adjusts the feature weight of the training stage, and generates training weight adjustment data; The age error correction module extracts the training weight adjustment data, trains the body fat prediction model, calculates the mean prediction residual of the sample in each age group, identifies the distance from the residual concentration interval of the overall sample, screens the age group with abnormal residuals, adjusts the residual response adjustment coefficient, and generates the model parameter adjustment result; The trend output module collects continuous vital sign data records of the target user according to the model parameter adjustment results, uses the body fat prediction model to predict the user's body fat percentage, identifies the changing trend of the body fat prediction results of multiple periods, and compares them with the user's weight data, analyzes the consistency of the changing trend, and uses the trend direction relationship to set the slope adjustment amplitude of the body fat prediction change path to generate a body fat trend prediction result.

[0014] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least: Based on the difference ratio of height and weight trajectories in children's physical examination data, a change rhythm distribution interval is constructed to screen samples. The fluctuation continuity of the waist circumference change slope is combined to eliminate abnormal data. The feature weighting coefficient is set according to the proportional deviation of height, weight and muscle mass. A training parameter control path for sample structure differences is established. The response coefficient is adjusted by the distance relationship between the group mean of the age group residual and the concentrated interval. The change path is judged based on the consistency of the body fat and weight prediction trends. The slope adjustment amplitude of the body fat percentage result is constructed. The time response performance of the body fat change direction is dynamically enhanced to avoid non-trend prediction offset, improve the stability and sample adaptability of the model in continuous prediction of time periods, enhance the input tolerance range of abnormal body shape samples, and strengthen the rationality of the prediction result output under physical sign disturbance. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 It is a schematic diagram of the workflow of the present invention; Figure 2 This is a detailed flow chart of S1 of the present invention; Figure 3 This is a detailed flow chart of S2 of the present invention; Figure 4 This is a detailed flow chart of S3 of the present invention; Figure 5 This is a detailed flow chart of S4 of the present invention; Figure 6 This is a detailed flow chart of S5 of the present invention; Figure 7 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0017] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0018] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0019] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0020] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0021] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0022] See also Figure 1 The present invention provides a technical solution, a method for adaptively predicting children's body fat percentage based on trajectory feature collaboration, comprising the following steps: S1: Obtain a children's physical examination dataset, analyze the height and weight change trajectories of each sample, calculate the difference ratio between the height and weight growth of each sample in consecutive cycles, and screen the samples based on the distribution characteristics of the difference ratio in multiple intervals to generate a rhythm screening sample set; S2: Call the rhythm screening sample set, analyze the waist circumference change data of each sample in multiple consecutive detection cycles after screening, calculate the difference in waist circumference change slope between adjacent cycles, evaluate the stability of the sample waist circumference change, further screen the sample, and generate a waist circumference rhythm screening set; S3: Using the waist rhythm screening set, calculate the ratio difference between height, weight and muscle mass of each sample after further screening. Based on the distribution characteristics of the ratio difference in multiple intervals, calculate the degree of deviation and adjust the feature weights of the training stage to generate training weight adjustment data; S4: Extract training weight adjustment data, train the body fat prediction model, calculate the mean prediction residual of the sample in each age group, identify the distance from the concentrated interval of the overall sample residual, screen the age group with abnormal residuals, adjust the residual response adjustment coefficient, and generate the model parameter adjustment results; S5: Based on the model parameter adjustment results, collect the target user's continuous vital sign data records, use the body fat prediction model to predict the user's body fat percentage, identify the changing trend of the body fat prediction results of multiple cycles, and compare them with the user's weight data, analyze the consistency of the changing trend, use the trend direction relationship to set the slope adjustment amplitude of the body fat prediction change path, and generate the body fat trend prediction result.

[0023] The rhythm screening sample set includes height growth trajectory, weight growth trajectory, and difference ratio distribution interval; the waist circumference rhythm screening set includes waist circumference change slope, slope fluctuation continuity, and waist circumference change trend structure; the training weight adjustment data includes the main interval of proportional difference, sample deviation degree, and feature weighting coefficient; the model parameter adjustment results include prediction residual grouping, age group error level, and residual response adjustment coefficient; the body fat trend prediction results specifically include body fat prediction data, body fat change trend, and trend slope adjustment amplitude.

[0024] See also Figure 2 ,The specific steps for obtaining the rhythm screening sample set are: S101: Obtain a children's physical examination data set, extract the height data and weight data of each sample in multiple consecutive test cycles, call the height change and weight change of each sample in adjacent cycles, calculate the difference ratio between the height increase and the weight increase, and generate height-weight difference ratio data; Obtain the height and weight data of each sample in the children's physical examination dataset. Taking three consecutive physical examinations of a certain grade in a school as an example, extract some student data to illustrate the specific calculation process of height and weight changes. The example is shown in Table 1: Table 1: Student continuous period physical examination data ; As shown in Table 1, taking student A as an example, the height growth between adjacent cycles is calculated: the subtraction of the second and first test results is 124cm-120cm=4cm, and the subtraction of the third and second test results is 128cm-124cm=4cm. The height growth in the consecutive cycles is 4cm and 4cm respectively. The weight growth in the consecutive cycles is calculated according to the same process. The weight difference between the second and first tests is 25kg-23kg=2kg, and the weight difference between the third and second tests is 27kg-25kg=2kg. Therefore, the weight growth in the consecutive cycles is 2kg. Next, the difference ratio of height growth and weight growth is calculated. Taking student A as an example, the difference ratio calculation process is height growth divided by weight growth. The difference ratio between the first and second test cycles is 4cm÷2kg=2cm / kg. The difference ratio between the second and third test cycles is also 2cm / kg. The same calculation method is used to process other students in the table to obtain the difference ratio of the consecutive cycles of each sample, and obtain the height and weight difference ratio data.

[0025] S102: Based on the height-weight difference ratio data, the distribution number of samples within each difference ratio interval is identified, and the density difference of the number of samples within each interval is analyzed to generate difference ratio distribution density data; According to the height and weight difference ratio data, the continuous cycle difference ratio data of the three students in Table 1 are used as an example for explanation: the difference ratio of student A in both cycles is 2cm / kg, the difference ratio of student B between the first and second cycles is 3cm÷1kg=3cm / kg, and the difference ratio between the second and third cycles is 2cm÷1.5kg≈1.33cm / kg, and the difference ratio between student C between the first and second cycles is 4cm÷2kg=2cm / kg, and the difference ratio between the second and third cycles is 5cm÷3kg≈1.67cm / kg. The difference ratios of each sample are classified into the corresponding distribution interval, as shown in Table 2: Table 2 Statistics of distribution density of difference ratios ; As shown in Table 2, the number of samples appearing in each interval was counted, and the percentage density of the total number of times was calculated. Taking the difference ratio interval of 2 to 2.5 cm / kg as an example, it appeared 3 times in the six difference ratio data of three samples, and the calculated density was 3 times ÷ 6 times × 100% = 50%. Based on this method, statistical analysis was performed on each interval one by one, and the calculation of the sample density in each difference ratio interval was completed to generate the difference ratio distribution density data.

[0026] S103: calling the difference ratio distribution density data, identifying the interval with the highest distribution density according to the concentration of the sample distribution density in each interval, performing preliminary screening on the samples, and generating a rhythm screening sample set; The difference ratio distribution density data is called, and the statistical data in Table 2 is continued as an example. The interval with the largest density in each interval is identified through the density comparison method. From the density data shown in Table 2, it is found that the density of the difference ratio interval 2 to 2.5 cm / kg is 50%, which is much higher than the density of other intervals (1 to 1.5 cm / kg is 16.7%, 1.5 to 2 cm / kg is 16.7%, and 3 to 3.5 cm / kg is 16.7%). The 2 to 2.5 cm / kg interval is determined to be the interval with the highest sample density. Subsequently, the student samples in this interval in the difference ratio data are marked and extracted. Taking student A as an example, the difference ratio of all its cycles is 2 cm / kg, which belongs to the highest density interval. The difference ratio of the first and second cycles of student C is also 2 cm / kg, which belongs to this interval. Student B does not enter this interval and is therefore excluded. All samples are screened by this method, and the sample group with the difference ratio in the highest density interval is extracted to generate a rhythm screening sample set.

[0027] See also Figure 3 ,The specific steps for obtaining the waist rhythm screening set are: S201: Calling the rhythm screening sample set, extracting waist circumference data of each sample in multiple consecutive detection cycles, calculating the waist circumference change rate of each sample in multiple cycles based on the recording sequence between adjacent detection cycles, and generating waist circumference change slope sequence data; The rhythm screening sample set was called to extract the waist circumference measurement data of each sample in three consecutive physical examination cycles. Taking the waist circumference data of students A and C in the rhythm screening sample set of the same grade as an example, the waist circumference of student A was 55 cm for the first measurement, 56.5 cm for the second measurement, and 58 cm for the third measurement. The waist circumference of student C was 56 cm for the first measurement, 58 cm for the second measurement, and 61 cm for the third measurement, as shown in Table 3.

[0028] Table 3 Waist circumference data of students in consecutive cycles ; As shown in Table 3, the waist circumference change rate of each sample was calculated. The rate is defined as the difference in waist circumference between adjacent cycles divided by the time interval between the corresponding testing cycles. Taking student A as an example, the difference between the first and second waist circumference data was calculated as 56.5cm-55cm=1.5cm, and the difference between the second and third waist circumference data was calculated as 58cm-56.5cm=1.5cm. Assuming that the testing cycle interval is 6 months, the waist circumference change rate of student A in two adjacent cycles is 1.5cm / 6months=0.25cm / month. The waist circumference change rate of student C was calculated in the same way. The rate between the first and second testing cycles was (58cm-56cm) / 6months≈0.33cm / month, and the rate between the second and third testing cycles was (61cm-58cm) / 6months=0.5cm / month. The waist circumference change rate of all samples in adjacent cycles was calculated and recorded through the above process to generate waist circumference change slope sequence data.

[0029] S202: Based on the waist circumference change slope sequence data, select the slope difference between adjacent periods of each sample, analyze the slope change trend of each sample, evaluate the stability of the sample waist circumference change, and generate waist circumference fluctuation structure data; The specific formula for evaluating the stability of waist circumference changes in samples is: ; Calculate the stability structure value of the waist circumference change slope difference; in, For the The sample in Normalized value of waist circumference change slope in each cycle, For the The sample in Normalized value of waist circumference change slope in each cycle, For the The sample in The horizontal detection frequency weighting factor of each cycle, For the The sample in Normalized value of waist circumference change rate per cycle, For the The age smoothing adjustment factor corresponding to the sample is: For the The number of consecutive detection cycles contained in a sample, is the period index variable, is the sample index variable, For the The stability structure value of the waist circumference change slope difference corresponding to the samples.

[0030] formula: ; Detailed explanation of the formula and the process of formula calculation and derivation: This formula is used to calculate the The dynamic fluctuation index of waist circumference changes of child samples in continuous detection cycles reflects the stability of the waist circumference change trend of the sample during the cyclic change process. The results are used to evaluate whether the sample has structural rhythmic characteristics, so as to further screen samples with good stability to generate waist circumference fluctuation structure data.

[0031] Parameter meaning and setting value: Indicates the The sample in The normalized value of the waist circumference change rate between the cycle and the previous cycle. The normalization basis is the maximum rate change amplitude within the cycle. The unit is a dimensionless value. In the actual test, the waist circumference in the third cycle is 56.8 cm, and the second cycle is 55.2 cm. The cycle time interval is 30 days, and the change rate is , the maximum rate recorded is 0.0667, then the normalized value is ; Indicates the The sample in The horizontal detection frequency weighting factor of the cycle is set based on the number of detections per unit time. With the detection frequency upper limit set to 10 times, the sample is set to 8 times in the third cycle, and the corresponding weighting factor is ; Indicates the The sample in The stability coefficient of the waistline change trend during the cycle is set as the normalized form of the slope standard deviation, and the standard deviation is set to 0.012. After normalization, it is obtained ; Indicates the The age smoothing adjustment factor of the sample is 10 years old, and the mean of the reference sample group is 12 years old. The adjustment factor is defined as ; Indicates that the total number of cycles collected for this sample is 5, so ; Substitute the parameters into the formula for calculation: No. The sample in to The normalized values ​​of the cycle slopes are: , , , ; The differences are: , , ; Substitute the weight factors in turn , , ; ; ; ; ; The result of 0.0682 indicates that the waist circumference change trend of this sample in continuous periods is at a low level. The smaller the value, the more stable the waist circumference change. This value is directly input into the rhythm stability judgment process as a quantitative reference indicator in the subsequent screening process to determine whether it is an acceptable sample and enter the next stage of the calculation process.

[0032] S203: Calling waist circumference fluctuation structure data, further screening samples based on the stability of sample waist circumference data changes, and generating a waist circumference rhythm screening set; Using the waist circumference fluctuation structure data, we continue to perform further screening using students A and C as examples. Specifically, we assess the fluctuation amplitude within the fluctuation structure data. Samples with weak fluctuation amplitudes are directly included in the screening set, samples with moderate fluctuation amplitudes require further analysis, and samples with strong fluctuation amplitudes are directly excluded. For example, since Student A's fluctuation amplitude is weak, we retain it. However, Student C's fluctuation structure data indicates moderate fluctuation amplitudes, so we use additional waist circumference measurement data or growth data to further analyze whether it meets the criteria. Assuming that Student C provides additional waist circumference data from the previous year, which shows a waist circumference of 53.5 cm, we calculate the average annual waist circumference change rate to be (61 cm − 53.5 cm) / 12 months, which is approximately 0.625 cm / month. Comparing this with the single-period data reveals that the annual average fluctuation amplitude exceeds the stable range, indicating a long-term fluctuation trend in waist circumference growth. Therefore, we exclude this sample. After completing this process, all samples with weak fluctuation amplitudes or those that pass further analysis are retained to form the final waist circumference rhythm screening set.

[0033] See also Figure 4 , the specific steps for obtaining training weight adjustment data are: S301: Using the waist rhythm screening set, extract the height, weight, and muscle mass data of each sample after further screening, calculate the ratio of weight to height of each sample, and obtain the ratio value to muscle mass, and generate body shape ratio distribution data by analyzing the distribution of the ratio values ​​in multiple continuous value intervals; Using the waist rhythm screening set, the height, weight and muscle mass data of each screening sample were selected for analysis. Specifically, the data of student A who was screened out was used as an example. The height data of student A was 128cm, the weight data was 27kg, and the muscle mass data was 8.5kg. The ratio between weight and height was calculated as follows: weight / height, which is 27kg÷128cm≈0.211kg / cm. The ratio to muscle mass was calculated as follows: muscle mass / (weight / height), which is 8.5kg÷0.21 1kg / cm≈40.28cm. The same method is used to calculate and summarize other samples, and the proportion values ​​are classified into different continuous numerical intervals, with interval ranges of 0 to 20cm, 20 to 40cm, 40 to 60cm, etc. The proportion of the number of samples in each interval is counted, and the distribution of samples in each interval is clarified through the proportion of the number of samples. For example, it is calculated that the interval of 0 to 20cm accounts for 5% of the total samples, 20 to 40cm accounts for 25%, and 40 to 60cm accounts for 70%. Then the distribution proportion statistics of all intervals are completed to generate body proportion distribution data.

[0034] S302: Calling the body proportion distribution data, identifying the interval with the largest proportion quantity distribution density and using it as the body proportion benchmark, calculating the difference between each sample proportion value and the benchmark, and generating body structure deviation data; The body proportion distribution data is retrieved, and the sample density within each proportion interval is used as the basis. Specifically, the sample density of each interval is analyzed, compared, and the interval with the highest density is identified. This interval is then determined as the body proportion benchmark. Taking the data example counted in paragraph S301 as an example, the density of the 0 to 20 cm interval is 5%, the density of the 20 to 40 cm interval is 25%, and the density of the 40 to 60 cm interval is 70%. By directly comparing the density percentages, the 40 to 60 cm interval is identified as the interval with the highest density and is determined as the proportion benchmark interval. The difference between the proportion value of each sample and the median of the benchmark interval is then calculated. The specific difference calculation method is to subtract the median of the interval from the sample proportion value. The median of the interval is (40 cm + 60 cm) / 2 = 50 cm. Taking student A as an example, its proportion value is 40.28 cm, and the calculated difference is 40.28 cm - 50 cm = -9.72 cm, indicating that the sample proportion value is smaller than the median of the benchmark interval. The difference values ​​of other samples are calculated and recorded similarly. After summarizing the difference calculation results of all samples, the body structure deviation data is generated.

[0035] S303: Based on the body structure deviation data and the difference between the sample ratio and the benchmark, the deviation degree of each sample is evaluated, the feature weight coefficients of the training phase are adjusted, a weight parameter set of the training sample is constructed, and training weight adjustment data is generated; Based on the body structure deviation data, the size of the difference between each sample's proportion value and the benchmark median is analyzed. The absolute value of the difference is used to represent the degree of deviation of the sample. The degree of deviation is defined as follows: 0 to 5 cm is a slight deviation, 5 to 10 cm is a moderate deviation, and greater than 10 cm is a significant deviation. The degree of deviation is evaluated and the feature weighting coefficients for the training phase are determined based on the degree of sample deviation. Taking student A as an example, the absolute value of the difference is 9.72 cm, which is a moderate deviation. Therefore, the weighting coefficients of his BMI and waist circumference features are set to 0.8, indicating that the influence of his features is weakened during training compared to samples with slight deviations (coefficients of 1). Similarly, if the absolute value of the difference of other samples is 3 cm, which is a slight deviation, the feature weighting coefficient remains at 1. If the difference is 12 cm, which is a significant deviation, the feature weighting coefficient is further reduced to 0.6. The above evaluation is performed on all samples and the corresponding coefficients are configured. Finally, the feature weighting coefficients of each sample are integrated to construct the weight parameter set of the training sample and generate training weight adjustment data.

[0036] See also Figure 5 , the steps to obtain the model parameter adjustment results are as follows: S401: Extracting training weight adjustment data, using samples to train a body fat prediction model, calculating the residual value between the sample prediction result and the actual result, combining the age group label of each sample, grouping the residual data, and generating age group residual group data; Extract training weight adjustment data. Based on samples with labeled feature weight coefficients, use the height, weight, and waist circumference feature data of each sample to predict body fat percentage. Take the actual body fat measurement as the real data, and calculate the difference between the predicted body fat value and the actual body fat measurement for each sample. This difference is defined as the residual. For example, if the feature weight coefficient of a student sample is set to 0.8, its predicted body fat percentage is 22%, while the actual measurement is 20%. The residual is 22%-20%=2%. Associate the residual value with the age label to which the sample belongs and group them. For example, the residual of the sample aged 8 is 1.5%, and the residual of the sample aged 9 is 2.3%. Record each residual one by one according to the age label into the corresponding age group, forming a residual set by age. Finally, count the residual data in all age groups, complete the age segmentation processing of the residual value, and thus generate age group residual data.

[0037] S402: Calling the residual grouping data of age groups, calculating the residual mean within each age group, and identifying the concentrated interval of the residual of the overall training sample, analyzing the distance between the residual mean of each age group and the overall residual interval, evaluating the deviation range of the residual mean, and generating the residual deviation data of the age group; Call the residual grouping data of age groups and calculate the residual mean within the residual group of each age group. The calculation process is demonstrated with a specific example. For example, there are 3 samples in the 8-year-old age group, and the residuals are 1.5%, 2.0%, and 1.0%, respectively. The mean is calculated as (1.5%+2.0%+1.0%) / 3≈1.5%. Similarly, there are 4 samples in the 9-year-old age group, and the residuals are 2.3%, 1.8%, 2.0%, and 2.5%, respectively. The mean is calculated as (2.3%+1.8%+2.0%+2.5%) / 4=2.15%. Repeat the above calculation method for all age groups to obtain the residual mean within each age group, and further The first step is to conduct an overall analysis of the residual data of all age groups, identify the concentrated distribution interval of the overall sample residuals, and calculate the median and standard deviation of the residual distribution based on all sample residuals. For example, if the median of the overall residual distribution is 2% and the standard deviation is 0.3%, the concentrated interval is defined as the median ± standard deviation, that is, 1.7% to 2.3%. After completion, calculate the distance between the mean residual of each age group and the concentrated interval of the overall residual. For example, the mean residual of the 8-year-old age group is 1.5%, and the distance from the minimum value of the concentrated interval of 1.7% is 0.2%. In this way, calculate and evaluate the degree of deviation of each age group, complete the deviation distance analysis of all groups, and generate the residual deviation data of age group.

[0038] S403: Based on the residual deviation data of the age group, according to the distance between the average residual level of each age group and the residual concentration interval of the overall sample, the age group with abnormal residuals is screened, and the residual response adjustment coefficient of the target age group sample is adjusted to generate the model parameter adjustment result; Based on the residual deviation data of age groups, the deviation distance between the residual mean of each age group and the overall residual concentration interval is clarified. The specific comparison and judgment process is as follows: a deviation distance in the range of 0 to 0.1% is defined as a slight deviation, a distance in the range of 0.1% to 0.3% is defined as a moderate deviation, and a distance greater than 0.3% is defined as a severe deviation. Taking the 8-year-old age group as an example, its distance is 0.2%, which belongs to the moderate deviation category. The distance of the 9-year-old age group is 0.15%, which also belongs to the moderate deviation category. Further, taking an age group with a distance of 0.35% as an example, it belongs to the severe deviation category. Subsequently, the severely deviated age group is determined to be an abnormal group and marked. The residual response adjustment coefficient is specifically adjusted. For example, for the severely deviated group samples, the response adjustment coefficient is reduced from the standard coefficient 1.0 to 0.6, the moderate deviation group is adjusted to 0.8, and the slightly deviated group remains unchanged at 1.0. After completing the adjustment and configuration of the coefficients of all age groups, the residual response adjustment coefficients of each age group are summarized to form the adjusted training parameter set, and the model parameter adjustment results are generated.

[0039] See also Figure 6 , the specific steps for obtaining the body fat trend prediction results are: S501: Based on the model parameter adjustment results, the target user's physical characteristics data such as height, weight, waist circumference, etc. are collected for multiple consecutive periods, the body fat prediction model is called, and the body fat prediction results of the user for each period are output to generate body fat percentage sequence data; According to the results of model parameter adjustment, continuous cycle vital sign data collection operation is carried out on the target users, and the user's height, weight and waist circumference values ​​are recorded in each continuous detection cycle. The collection cycle is set to once every 30 days. It is assumed that the user's height in the 1st to 4th cycles are 125.2cm, 125.8cm, 126.1cm, 126.5cm respectively, the weight is 27.0kg, 27.8kg, 28.5kg, 29.0kg respectively, and the waist circumference is 57.4cm, 57.9cm, 58.5cm, 59.1cm respectively. Based on each The complete physical sign parameters collected during the period are input into the trained body fat prediction model structure. The model uses the feature weighting mechanism and age group adjustment coefficient of the previous training stage to automatically output the body fat prediction value for the corresponding period. Assuming that the prediction results are 19.6%, 20.2%, 21.0%, and 21.7% respectively, the prediction results for the above consecutive periods are arranged in time series to form an ordered body fat percentage numerical sequence. At the same time, the original physical sign data corresponding to each period are retained to ensure the integrity of the data chain structure, facilitate subsequent trend analysis and synchronous processing, and finally generate body fat percentage sequence data.

[0040] S502: Based on the body fat percentage sequence data, extract the target user's weight change trend in each cycle, analyze the change direction of the body fat percentage sequence and the weight data, analyze the consistency of the change trend, obtain a trend consistency index, and generate trend relationship data; The specific formula for analyzing the consistency of the changing trend is: ; Calculate trend consistency indicators; in, represents the trend consistency indicator, Representative The normalized value of the change in body fat percentage within a cycle, Representative Normalized value of weight change within a cycle, Indicates the total number of consecutive detection cycles, Indicates the current cycle number. Represents a non-zero small constant set to prevent the denominator from being zero.

[0041] formula: ; Detailed explanation of the formula and the process of formula calculation and derivation: The formula is used to calculate the directional consistency between the trend of body fat percentage change and the trend of weight change. The results are used to evaluate the stability of the body fat trend relationship data and predict adjustments; Parameter meaning and setting value: is the total number of consecutive testing cycles, which is set to 4, representing 4 consecutive children's physical examination data; is the normalized value of the body fat percentage change in cycle 1, which is obtained by subtracting the sample mean from the change in body fat percentage in each cycle and dividing it by the standard deviation. The actual measured values ​​of body fat percentage are assumed to be 17.0%, 17.8%, 18.2%, and 18.6%, respectively, with changes of 0.8%, 0.4%, and 0.4%, respectively. After normalization, they are 1.27, -0.22, and -0.22, respectively; is the normalized value of weight change in cycle 1, which is obtained by subtracting the sample mean from the weight change in each cycle and dividing it by the standard deviation. The actual measured weight values ​​are assumed to be 31.4 kg, 32.2 kg, 33.0 kg, and 33.8 kg, with changes of 0.8 kg, 0.8 kg, and 0.8 kg, respectively. After normalization, the values ​​are all 1.73; The minimum constant set to prevent the denominator from being zero is set to 0.01.

[0042] Substitute the parameters into the formula for calculation: Cycle 1: , ; ; Cycle 2: , ; ; Cycle 3: , ; ; Cycle 4: , ; ; Average value calculation: ; Substituting into the formula: ; The results show that the trend consistency index is 0.5811, indicating that the changes in body fat percentage and weight show moderate consistency within the four cycles. The closer the value is to 0, the more consistent the trend direction is, and the closer it is to 1, the greater the difference in trend direction. The results can be used to determine body fat trend relationship data and as a basis for subsequent adjustments.

[0043] S503: Calling trend relationship data, adjusting the slope adjustment amplitude of the body fat prediction change path based on the relationship between the body fat percentage change trend and the weight change trend, and obtaining a body fat trend prediction result; Call the trend relationship data and adjust the slope amplitude of the body fat percentage change path according to the trend consistency index. If the trend relationship is "high", the original slope trend is maintained. For example, the original slope setting under the T1-T4 cycle is (21.7%-19.6%) / 3≈0.7% / cycle. If the trend consistency is "medium", the slope amplitude is reduced to 0.75 times the original value. If it is "low", it is reduced to 0.5 times. In the current case, the trend is "high", and the slope remains unchanged. The predicted end point is extended by one cycle, and the estimated body fat percentage in the T5 cycle is 21.7%+0.7%=22.4%. The extended predicted value is added to the original body fat percentage sequence to complete the time trend curve and establish a change path, so as to continuously track the dynamic changes of body fat and finally obtain the body fat trend prediction result.

[0044] See also Figure 7 The adaptive prediction system for children's body fat percentage based on trajectory feature collaboration is used to execute the above-mentioned adaptive prediction method for children's body fat percentage based on trajectory feature collaboration. The system includes: The physical sign screening module obtains a children's physical examination data set, analyzes the height and weight change trajectory of each sample, calculates the difference ratio between the height and weight growth of each sample in consecutive cycles, and screens the samples based on the distribution characteristics of the difference ratio in multiple intervals to generate a rhythm screening sample set; The waist circumference stability module calls the rhythm screening sample set, analyzes the waist circumference change data of each sample after screening in multiple consecutive detection cycles, calculates the difference in waist circumference change slopes between adjacent cycles, evaluates the stability of the sample waist circumference change, further screens the samples, and generates a waist circumference rhythm screening set; The proportional weight control module uses the waist rhythm screening set to calculate the proportional difference between the height, weight and muscle mass of each sample after further screening. Based on the distribution characteristics of the proportional difference in multiple intervals, it calculates the degree of deviation and adjusts the feature weights of the training stage to generate training weight adjustment data; The age error correction module extracts training weight adjustment data, trains the body fat prediction model, calculates the mean prediction residual of the sample in each age group, identifies the distance from the concentrated interval of the overall sample residual, screens the age groups with abnormal residuals, adjusts the residual response adjustment coefficient, and generates the model parameter adjustment results; The trend output module collects the continuous vital sign data records of the target user according to the model parameter adjustment results, uses the body fat prediction model to predict the user's body fat percentage, identifies the changing trend of the body fat prediction results of multiple cycles, and compares it with the user's weight data, analyzes the consistency of the changing trend, and uses the trend direction relationship to set the slope adjustment amplitude of the body fat prediction change path to generate the body fat trend prediction result.

[0045] The above embodiments can be implemented in whole or in part via software, hardware (e.g., circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are fully or partially performed. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired means (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0046] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0047] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0048] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0049] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0050] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0051] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of the device or unit, which can be electrical, mechanical or other forms.

[0052] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0053] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0054] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical disks.

[0055] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. The adaptive prediction method of children's body fat percentage based on trajectory feature collaboration is characterized by: The method comprises: S1: Obtain a children's physical examination dataset, analyze the height and weight change trajectories of each sample, calculate the difference ratio between the height and weight growth of each sample in consecutive cycles, and screen the samples based on the distribution characteristics of the difference ratio in multiple intervals to generate a rhythm screening sample set; S2: calling the rhythm screening sample set, analyzing the waist circumference change data of each sample in multiple consecutive detection cycles after screening, calculating the difference in waist circumference change slopes between adjacent cycles, evaluating the stability of the waist circumference change of the sample, further screening the sample, and generating a waist circumference rhythm screening set; S3: Using the waist rhythm screening set, calculating the ratio difference between height, weight and muscle mass of each sample after further screening, calculating the degree of deviation and adjusting the feature weights of the training stage based on the distribution characteristics of the ratio difference in multiple intervals, and generating training weight adjustment data; S4: Extract the training weight adjustment data, train the body fat prediction model, calculate the mean prediction residual of the sample in each age group, identify the distance from the overall sample residual concentration interval, screen the age group with abnormal residuals, adjust the residual response adjustment coefficient, and generate the model parameter adjustment result.

2. The method for adaptively predicting children's body fat percentage based on trajectory feature collaboration according to claim 1 is characterized in that: The rhythm screening sample set includes height growth trajectory, weight growth trajectory, and difference ratio distribution interval; the waist circumference rhythm screening set includes waist circumference change slope, slope fluctuation continuity, and waist circumference change trend structure; the training weight adjustment data includes the main interval of proportional difference, sample deviation degree, and feature weighting coefficient; the model parameter adjustment results include prediction residual grouping, age group error level, and residual response adjustment coefficient.

3. The method for adaptively predicting children's body fat percentage based on trajectory feature collaboration according to claim 1, characterized in that: The steps for obtaining the rhythm screening sample set are specifically as follows: S101: Obtain a children's physical examination data set, extract the height data and weight data of each sample in multiple consecutive test cycles, call the height change and weight change of each sample in adjacent cycles, calculate the difference ratio between the height increase and the weight increase, and generate height-weight difference ratio data; S102: Based on the height-weight difference ratio data, the distribution number of samples within each difference ratio interval is identified, and the density difference of the number of samples within each interval is analyzed to generate difference ratio distribution density data; S103: Call the difference ratio distribution density data, identify the interval with the highest distribution density according to the concentration of the sample distribution density in each interval, perform preliminary screening on the samples, and generate a rhythm screening sample set.

4. The method for adaptively predicting children's body fat percentage based on trajectory feature collaboration according to claim 3 is characterized in that: The steps for obtaining the waist rhythm screening set are specifically as follows: S201: calling the rhythm screening sample set, extracting waist circumference data of each sample in multiple consecutive detection cycles, calculating the waist circumference change rate of each sample in multiple cycles according to the recording sequence between adjacent detection cycles, and generating waist circumference change slope sequence data; S202: selecting the slope difference between adjacent periods of each sample based on the waist circumference change slope sequence data, analyzing the slope change trend of each sample, evaluating the stability of the waist circumference change of the sample, and generating waist circumference fluctuation structure data; S203: calling the waist circumference fluctuation structure data, further screening the samples according to the stability of the changes in the sample waist circumference data, and generating a waist circumference rhythm screening set.

5. The method for adaptively predicting children's body fat percentage based on trajectory feature collaboration according to claim 4 is characterized in that: The specific formula for evaluating the stability of waist circumference changes of the sample is: ； Calculate the stability structure value of the waist circumference change slope difference; in, For the The sample in Normalized value of waist circumference change slope in each cycle, For the The sample in Normalized value of waist circumference change slope in each cycle, For the The sample in The horizontal detection frequency weighting factor of each cycle, For the The sample in Normalized value of waist circumference change rate per cycle, For the The age smoothing adjustment factor corresponding to the sample is: For the The number of consecutive detection cycles contained in a sample, is the period index variable, is the sample index variable, For the The stability structure value of the waist circumference change slope difference corresponding to the samples.

6. The method for adaptively predicting children's body fat percentage based on trajectory feature collaboration according to claim 4, characterized in that: The steps for obtaining the training weight adjustment data are specifically as follows: S301: Using the waist rhythm screening set, extracting the height, weight, and muscle mass data of each sample after further screening, calculating the ratio of weight to height of each sample, and obtaining the ratio value to muscle mass, and generating body shape ratio distribution data by analyzing the distribution of the ratio values ​​in multiple continuous numerical intervals; S302: Calling the body proportion distribution data, identifying the interval with the largest proportion quantity distribution density and using it as a body proportion benchmark, calculating the difference between each sample proportion value and the benchmark, and generating body structure deviation data; S303: Based on the body structure deviation data, the deviation degree of each sample is evaluated according to the gap between the sample ratio value and the benchmark, the feature weighting coefficient of the training stage is adjusted, the weight parameter set of the training sample is constructed, and the training weight adjustment data is generated.

7. The method for adaptively predicting children's body fat percentage based on trajectory feature collaboration according to claim 6, characterized in that: The steps for obtaining the model parameter adjustment results are specifically as follows: S401: extracting the training weight adjustment data, using the sample to train the body fat prediction model, calculating the residual value between the sample prediction result and the actual result, combining the age group label of each sample, grouping the residual data, and generating age group residual group data; S402: calling the residual grouping data of the age group, calculating the residual mean within each age group, identifying the concentrated interval of the residual of the overall training sample, analyzing the distance between the residual mean of each age group and the overall residual interval, evaluating the deviation range of the residual mean, and generating age group residual deviation data; S403: Based on the residual deviation data of the age groups, according to the distance between the average residual level of each age group and the residual concentration interval of the overall sample, the age groups with abnormal residuals are screened, and the residual response adjustment coefficient of the target age group samples is adjusted to generate the model parameter adjustment results.

8. The method for adaptively predicting children's body fat percentage based on trajectory feature collaboration according to claim 1, characterized in that: The method further comprises: S5: Based on the model parameter adjustment results, continuous vital sign data records of the target user are collected, the user's body fat percentage is predicted using the body fat prediction model, the changing trend of the body fat prediction results over multiple periods is identified, and the results are compared with the user's weight data to analyze the consistency of the changing trends. Based on the trend direction relationship, the slope adjustment amplitude of the body fat prediction change path is set to generate a body fat trend prediction result; The body fat trend prediction result specifically includes body fat prediction data, body fat change trend, and trend slope adjustment amplitude.

9. The method for adaptively predicting children's body fat percentage based on trajectory feature collaboration according to claim 8, characterized in that: The steps for obtaining the body fat trend prediction result are specifically as follows: S501: Based on the model parameter adjustment result, the target user's physical sign data such as height, weight, waist circumference, etc. for multiple consecutive periods are collected, the body fat prediction model is called, and the body fat prediction result of the user for each period is output to generate body fat percentage sequence data; S502: Based on the body fat percentage sequence data, extract the target user's weight change trend in each cycle, analyze the change direction of the body fat percentage sequence and the weight data, analyze the consistency of the change trend, obtain a trend consistency index, and generate trend relationship data; The specific formula for analyzing the consistency of the change trend is: ； Calculate trend consistency indicators; in, represents the trend consistency indicator, Representative The normalized value of the change in body fat percentage within a cycle, Representative Normalized value of weight change within a cycle, Indicates the total number of consecutive detection cycles, Indicates the current cycle number. Represents a non-zero minimum constant set to prevent the denominator from being zero; S503: Calling the trend relationship data, adjusting the slope adjustment amplitude of the body fat prediction change path according to the relationship between the body fat percentage change trend and the weight change trend, and obtaining the body fat trend prediction result.

10. Children's body fat percentage adaptive prediction system based on trajectory feature collaboration, characterized by: The system is used to implement the method for adaptively predicting children's body fat percentage based on trajectory feature collaboration according to any one of claims 1 to 9, and the system includes: The physical sign screening module obtains a children's physical examination data set, analyzes the height and weight change trajectory of each sample, calculates the difference ratio between the height and weight growth of each sample in consecutive cycles, and screens the samples based on the distribution characteristics of the difference ratio in multiple intervals to generate a rhythm screening sample set; The waist circumference stability module calls the rhythm screening sample set, analyzes the waist circumference change data of each sample in multiple consecutive detection cycles after screening, calculates the difference in waist circumference change slopes between adjacent cycles, evaluates the stability of the sample waist circumference change, further screens the sample, and generates a waist circumference rhythm screening set; The proportional weight control module uses the waist rhythm screening set to calculate the proportional difference between the height, weight and muscle mass of each sample after further screening, calculates the degree of deviation based on the distribution characteristics of the proportional difference in multiple intervals, adjusts the feature weight of the training stage, and generates training weight adjustment data; The age error correction module extracts the training weight adjustment data, trains the body fat prediction model, calculates the mean prediction residual of the sample in each age group, identifies the distance from the residual concentration interval of the overall sample, screens the age group with abnormal residuals, adjusts the residual response adjustment coefficient, and generates the model parameter adjustment result; The trend output module collects continuous vital sign data records of the target user according to the model parameter adjustment results, uses the body fat prediction model to predict the user's body fat percentage, identifies the changing trend of the body fat prediction results of multiple periods, and compares them with the user's weight data, analyzes the consistency of the changing trend, and uses the trend direction relationship to set the slope adjustment amplitude of the body fat prediction change path to generate a body fat trend prediction result.

Images