Individual height prediction method based on multidimensional growth and development data

Abstract

This invention relates to the field of individual height prediction technology, specifically to a method for predicting individual height based on multi-dimensional growth and development data. The method involves determining a child's height and bone age data over several recent years; determining the child's growth rate change, percentile deviation, and the degree of coordinated change in bone age and height over the most recent year relative to historical years based on the height and bone age data; and integrating these factors to obtain a baseline predicted height for the child. Finally, the baseline predicted height is corrected based on the child's genetic factors and current growth hormone secretion levels to obtain a target predicted height. This invention improves the accuracy of children's height prediction through multi-dimensional growth and development data analysis.

Description

Technical Field

[0001] This invention relates to the field of individual height prediction technology, specifically to a method for predicting individual height based on multidimensional growth and development data. Background Technology

[0002] Height is a core indicator for evaluating the growth and development of children and adolescents, and predicting final adult height is a crucial aspect of pediatric endocrinology clinical diagnosis and routine monitoring of children's health. Accurate height prediction not only meets parents' needs for understanding their children's growth and development but also provides analytical and quantitative evidence for early screening, intervention program development, and efficacy evaluation of diseases such as short stature and growth hormone deficiency.

[0003] The current mainstream method for predicting children's height is to use the child's current bone age and current height as the core predictive indicators. The parameters used in the prediction are relatively simple, and the prediction accuracy is low. It does not integrate multiple influencing factors such as genetic potential, longitudinal growth trajectory, puberty development process, endocrine level, environment and lifestyle. Therefore, it cannot fit the synergistic effect of multiple factors on height, and the prediction accuracy is greatly reduced for special groups such as those with early or delayed puberty or growth retardation. Summary of the Invention

[0004] To address the technical problem of low prediction accuracy in existing methods for predicting children's height, the present invention aims to provide an individual height prediction method based on multi-dimensional growth and development data. The specific technical solution adopted is as follows: In a first aspect of the present invention, a method for predicting individual height based on multidimensional growth and development data is provided, comprising: Determine the child's height and bone age data for the most recent several years; the most recent several years include the most recent year and several historical years; the following child growth indicators

[0005] Based on the height data, determine the child's growth rate of change and height percentile deviation in the most recent year relative to the historical year.

[0006] Based on the height and bone age data, determine the degree of coordinated change in bone age and height growth in the child over the most recent year relative to the historical year.

[0007] By integrating the growth rate of height, the percentile deviation of height, and the degree of coordinated change in bone age and height, the child's current basic height prediction value is obtained.

[0008] Based on the child's genetic factors for height and current growth hormone secretion levels, the baseline height prediction is revised to obtain the target height prediction.

[0009] In an exemplary embodiment, the process of obtaining the rate of change of height growth includes: Determine the latest height growth rate for the most recent year, as well as the historical height growth rates for each historical year;

[0010] The overall historical height growth rate for each historical year can be obtained from the historical height growth rate for each historical year.

[0011] The rate of change of the latest height growth rate relative to the overall historical height growth rate is determined to obtain the rate of change of height growth rate.

[0012] In an exemplary embodiment, the process of obtaining the height percentile deviation includes: Determine the child's current height percentile and the historical height percentile at a target historical moment; the target historical moment is within the historical year.

[0013] The average rate of change of the latest height percentile and the historical height percentile is determined to obtain the degree of deviation of the height percentile.

[0014] In an exemplary embodiment, the process of obtaining the degree of coordinated change in bone age and height includes: Determine the child's current height and the target historical height, as well as the child's current bone age and the target historical bone age; the target historical time is within the historical year.

[0015] The deviation of the actual height increase from the preset standard height increase, and the average progression characteristics of the actual bone age are determined; the actual height increase is the difference between the latest height and the historical height.

[0016] The degree of synergistic change in bone age and height is obtained by calculating the ratio of the deviation in height growth to the average progression characteristics of actual bone age.

[0017] In an exemplary embodiment, the process of obtaining the baseline height prediction value includes: By integrating the growth rate of height, the percentile deviation of height, and the degree of coordinated change in bone age and height, a height prediction coefficient is obtained.

[0018] The basic predicted height value is obtained based on the child's current actual height, the percentage increase in adult height corresponding to the child's current bone age, and the height prediction coefficient.

[0019] In an exemplary embodiment, the process of fusing the rate of change in height growth, the degree of height percentile deviation, and the degree of coordinated change in bone age and height growth to obtain a height prediction coefficient includes: The rate of change in height growth rate, the degree of deviation of height percentile, and the degree of coordinated change in bone age and height growth were standardized respectively.

[0020] Calculate the average of the standardized height growth rate change rate, height percentile deviation, and bone age-height co-change, and obtain the height prediction coefficient based on the average value.

[0021] In an exemplary embodiment, obtaining the basic height prediction value based on the child's current actual height, the percentage increase in adult height corresponding to the child's current bone age, and the height prediction coefficient includes: Calculate the ratio of the child's current actual height to the percentage increase in adult height, and then multiply it by the height prediction coefficient to obtain the basic predicted height value.

[0022] In one exemplary embodiment, the height genetic factor is the child's genetic target height, and the current growth hormone secretion level is the current insulin-like growth factor-1 concentration;

[0023] The process of obtaining the height target prediction value includes: Determine the degree of deviation of the genetic target height from the preset standard height of healthy adults of the same sex in terms of adult height, and the degree of deviation of the current insulin-like growth factor-1 concentration from the preset standard concentration of growth hormone secretion.

[0024] Based on the deviations in adult height and growth hormone secretion, a height correction coefficient is obtained;

[0025] Based on the height correction coefficient, the basic predicted height value is corrected to obtain the target predicted height value.

[0026] In an exemplary embodiment, the process of obtaining the height correction factor includes: Based on the relationship between the degree of deviation of adult height and the preset constant, the first correction amplitude characteristic value is obtained;

[0027] The second correction amplitude characteristic value is obtained based on the relationship between the degree of deviation in growth hormone secretion and the preset constant.

[0028] The first correction amplitude feature value and the second correction amplitude feature value are weighted and fused to obtain the height correction coefficient.

[0029] In a second aspect of the present invention, an individual height prediction system based on multidimensional growth and development data is provided, comprising: a memory and a processor; the memory is connected to the processor; the memory is used to store program instructions; the processor is used to implement the above-described individual height prediction method based on multidimensional growth and development data when the program instructions are executed.

[0030] This invention offers the following advantages: It obtains the growth rate of height and the percentile deviation of height based on children's height data from recent years, and combines this with the correlation between these data and bone age data from recent years to determine the degree of coordinated change in bone age and height growth. The growth rate of height accurately represents the dynamic trend of children's height growth, the percentile deviation measures the relative position of a child's current height among their peers, thus revealing any abnormalities in their current height. The degree of coordinated change in bone age and height growth accurately assesses the matching degree between the rate of skeletal maturation and the rate of height growth. By integrating these three parameters—growth rate of height, percentile deviation, and coordinated change in bone age and height growth—a baseline predicted value for a child's height can be obtained. This value reflects the more complex growth regulation mechanisms of children better than a single parameter, and can also suppress abnormal noise to a certain extent, improving the stability and reliability of the baseline predicted value. Furthermore, by introducing genetic factors related to children's height and current growth hormone secretion levels to correct the baseline predicted value, and through multi-dimensional growth and development data analysis, the accuracy of children's height prediction can be further improved. Attached Figure Description

[0031] Figure 1 This is a flowchart of an individual height prediction method based on multi-dimensional growth and development data provided by the present invention. Detailed Implementation

[0032] In this embodiment, the subject whose height needs to be predicted is referred to as the child to be predicted, hereinafter simply referred to as the child. This embodiment provides an individual height prediction method based on multi-dimensional growth and development data, used to predict the height of children by combining their multi-dimensional growth and development data.

[0033] like Figure 1 As shown in the figure, the individual height prediction method based on multi-dimensional growth and development data provided in this embodiment includes the following steps: Step S1: Determine the child's height and bone age data for the most recent years;

[0034] Step S2: Based on the height data, determine the rate of change of the child's height growth rate in the most recent year relative to historical years and the degree of deviation of height percentile.

[0035] Step S3: Based on height and bone age data, determine the degree of coordinated change in bone age and height growth in the child over the most recent year relative to historical years;

[0036] Step S4: Integrate the rate of change in height growth rate, the degree of deviation in height percentile, and the degree of coordinated change in bone age and height growth to obtain the child's current baseline predicted height.

[0037] Step S5: Based on the child's genetic factors for height and current growth hormone secretion levels, the baseline height prediction value is revised to obtain the target height prediction value.

[0038] The following is a detailed description of each step.

[0039] Step S1: Determine the child's height and bone age data for the most recent years.

[0040] Starting from the current moment, we determine the most recent several years in the time dimension. These most recent several years include the most recent year and several historical years. The most recent year refers to the time length between the current moment and the same moment a year ago; that is, the most recent year can also be defined as the first year in the past. The number of historical years included in the "several historical years" is set according to actual needs. Since time periods too far removed from the current moment have a weaker impact on children's height prediction, the number of historical years should not be too large. This example uses two historical years. Therefore, relative to the first year in the past, historical years include the second year in the past and the third year in the past. Taking the current moment as May 1, 2026 as an example, the time period of the most recent year (i.e., the first year in the past) is: May 2, 2025 – May 1, 2026; the time period of the second year in the past is: May 2, 2024 – May 1, 2025; and the time period of the third year in the past is: May 2, 2023 – May 1, 2024. Furthermore, each of the most recent years that coincides with the current time can be considered a historical moment. For example, if the current time is May 1, 2026, the historical moments would be May 1, 2025, May 1, 2024, and May 1, 2023. For clarity, starting from the current time, and moving in the opposite direction of the timeline, the historical moments are defined sequentially as: the first historical moment, the second historical moment, and the third historical moment. The time intervals between the first, second, and third historical moments and the current time gradually increase. Therefore, the most recent year and several historical years can be understood as several time periods enclosed by the current time and the various historical moments.

[0041] Obtain children's height and bone age data for the most recent several years. Specifically, the height data includes the child's height at the current moment and at various historical points in time. Based on the example of the most recent several years mentioned above, the height data includes: the height at the current moment, the height on May 1, 2025, the height on May 1, 2024, and the height on May 1, 2023. The unit of measurement for height is centimeters.

[0042] Similarly, bone age data specifically includes: a child's bone age values ​​at the current moment and at various historical moments. Based on the examples of the most recent years mentioned above, bone age data includes: the bone age value at the current moment, the bone age value on May 1, 2025, the bone age value on May 1, 2024, and the bone age value on May 1, 2023. Bone age values ​​and chronological age are measured on the same unit.

[0043] Step S2: Based on the height data, determine the rate of change of the child's height growth rate and the degree of height percentile deviation in the most recent year relative to historical years.

[0044] Considering the dynamic characteristics of children's growth and development, height development is not a static indicator at a single point in time, but is deeply influenced by dynamic factors such as continuous growth trends and the matching of bone age and height. Existing static prediction methods do not delve into the key information reflected by these dynamic factors regarding individual growth potential and the prediction of growth inflection points, such as the accelerating trend of individual growth rate and the synergistic matching of bone age progression and height growth. Therefore, step S2 needs to analyze the relevant time-series data and extract the key features reflected therein to provide core support for subsequent accurate predictions in terms of dynamic dimensions.

[0045] First, analyze the height data of children in recent years to obtain the rate of change of height growth rate and the degree of height percentile deviation in the most recent year relative to historical years.

[0046] The overall concept for obtaining the rate of change of height growth rate is as follows: obtain the child's height growth rate in each year, and then, based on the relationship between the latest height growth rate in the most recent year and the overall historical height growth rate in previous years, obtain the rate of change of the child's height growth rate in the most recent year relative to previous years.

[0047] In one exemplary embodiment, the latest height growth rate for the most recent year and the historical height growth rates for each historical year are first determined: ;

[0048] ;

[0049] ;

[0050] in, This indicates the latest height growth rate over the past year. This indicates the historical height growth rate for the second consecutive year. This indicates the historical height growth rate over the past third year. This indicates the current height value. This represents the height at the first historical moment. This represents the height value at the second historical moment. This represents the height value at the third historical moment.

[0051] It should be understood that, according to the normal growth pattern of human height, if the latest height growth rate and all historical height growth rates are greater than 0, it indicates that the child's height has increased to some extent within a year. If at least one of the latest height growth rate and all historical height growth rates is less than or equal to 0, then the child's height growth has become seriously abnormal, or there is a significant error in the data collection. In this case, further data analysis will not be conducted, and an abnormal data indication signal will be output directly. Subsequent data analysis will be performed only if the latest height growth rate and all historical height growth rates are greater than 0.

[0052] The overall historical height growth rate for each historical year is obtained from the historical height growth rate for each historical year. In one exemplary embodiment, the average historical height growth rate for each historical year is calculated as the overall historical height growth rate for that historical year. It should be understood that Greater than 0.

[0053] Determine the latest height growth rate Compared to the overall historical rate of height growth The rate of change of height growth rate is obtained from the rate of change of height growth rate, specifically: ;

[0054] in, This represents the rate of change in height growth rate. By calculating the relative difference between the height growth rate of the most recent year and the average height growth rate of the previous two years, the differences in individual baseline height growth rates can be eliminated, quantifying the relative magnitude of change in height growth rate. (Height growth rate change rate) A positive and larger value indicates that the height growth rate in the most recent year is higher than the average height growth rate in the previous two years, reflecting a significant accelerating trend in the individual child's height growth rate; conversely, a smaller value indicates a lower growth rate. When the value is negative and the absolute value is larger, it indicates that the height growth rate in the most recent year is lower than the average height growth rate in the previous two years, and growth has slowed down significantly.

[0055] Analyzing only the changes in the rate of height growth is not enough to accurately determine the relative development trend of a child's height. The final height of a child is affected by the dynamic changes in their own growth rate and is also closely related to the changes in their height ranking among their peers. The direction and magnitude of the drift in height percentiles are key supplementary evidence for judging whether a child's height growth trajectory deviates from the normal trend and whether their growth potential is stable.

[0056] In one exemplary embodiment, this embodiment determines a target historical moment from various historical moments. The selection of the target historical moment is set according to actual needs. In this embodiment, a historical moment two years ago is used as the target historical moment, that is, the target historical moment is the second historical moment.

[0057] According to the "Growth Standards for Children Under 7 Years Old" (WS / T 423—2022) and the "Reference Standards for Growth and Development of School-Age Children and Adolescents in China" (WS / T 586-2018), the child's current height percentile is obtained by referring to the table and is defined as the child's latest height percentile. At the same time, the child's height percentile at the target historical moment is obtained by referring to the table and is defined as the historical height percentile at the target historical moment.

[0058] Determine the average rate of change of the current height percentile and the historical height percentile at the target historical time. Specifically, the average rate of change of the height percentile is the average rate of change of the difference between the current height percentile and the historical height percentile at the target historical time, relative to the time span. ;

[0059] in, This indicates the degree of percentile deviation in height. This indicates the current height percentile. The value 2 represents the historical height percentile at the target historical moment, and the value 2 in the denominator represents the time interval between the current moment and the target historical moment, which is 2 years.

[0060] Height percentile deviation Used to reflect the dynamic trend of a child's height relative to their peers, the degree of height percentile deviation. A value greater than 0 indicates that the child's height growth rate is better than the average level of their peers, showing a significant advantage in height growth and a positive growth trend; the degree of height percentile deviation =0 indicates that height growth is relatively stable; the degree of deviation of height percentile A value less than 0 indicates a potential risk of deviation from the growth trajectory of height.

[0061] Step S3: Based on height and bone age data, determine the degree of coordinated change in bone age and height growth in the child over the most recent year relative to historical years.

[0062] Complete the rate of change in height growth rate Degree of deviation from height percentile After obtaining the data, it is necessary to further quantify the relationship between children's bone age development and height growth. In the context of height prediction, bone age is a core indicator reflecting biological maturity. The actual efficiency of height growth during bone age progression cannot be assessed solely through the current bone age difference. Therefore, it is necessary to further analyze the synergistic relationship between bone age and height.

[0063] This embodiment still uses the historical moment two years ago as the target historical moment, that is, the target historical moment is the second historical moment. Based on height data and bone age data, the degree of coordinated change in bone age and height growth of the child in the most recent year relative to the historical year is determined, that is, the degree of coordinated change in bone age and height growth of the child in the present relative to the target historical moment is determined.

[0064] In one exemplary embodiment, the child's height at the current moment is determined as the current latest height, and the child's height at a target historical moment is determined as the historical height at the target historical moment. Simultaneously, the child's bone age at the current moment is determined as the current latest bone age, and the child's bone age at the target historical moment is determined as the historical bone age at the target historical moment.

[0065] Subtract the target historical height from the current height. The difference obtained is the actual height increase, which represents the child's height increase over the past two years. Similarly, subtracting the historical bone age at the target historical moment from the current bone age yields the actual bone age progression, which represents the child's bone age progression over the past two years. It should be understood that the child's height increase and bone age progression over the past two years are normally positive. If the child's height increase or bone age progression over the past two years is less than or equal to 0, it indicates a serious abnormality in the child's height increase or bone age progression over the past two years, possibly due to abnormal data collection. In such cases, further data analysis will be discontinued, and a data anomaly indicator signal will be directly output.

[0066] This embodiment predetermines a standard height growth rate. Since the actual height growth rate mentioned above represents the child's height growth over the past two years, the predetermined standard height growth rate is the normal height growth rate of children of the same sex and age over two years. This predetermined standard height growth rate can be obtained from a large sample of historical children, showing the height growth rate of children of the same sex and age over two years, and then the median or calculated mean can be used as the predetermined standard height growth rate. Alternatively, it can be obtained directly from the "Growth Standards for Children Under 7 Years Old" (WS / T 423—2022) and the "Reference Standards for Growth and Development of School-Age Children and Adolescents in China" (WS / T 586-2018), showing the height growth rate of children of the same sex and age over two years, and then the median can be determined as the predetermined standard height growth rate.

[0067] To determine the degree of deviation between the actual height growth and the preset standard height growth, specifically, calculate the ratio of the actual height growth to the preset standard height growth, and use this ratio as the degree of deviation. A deviation greater than 1 indicates that the child's actual height growth over the past two years is relatively higher than the normal average, suggesting that the child has made good use of their growth potential and has superior growth potential compared to their peers.

[0068] Under standard circumstances, a child's standard bone age increase over the past two years is considered to be two years. Therefore, based on the child's actual bone age progression and the standard bone age increase, the average progression characteristic of the child's actual bone age is obtained. Specifically, the ratio of the actual bone age progression to the standard bone age increase is calculated and used as the average progression characteristic of the child's actual bone age.

[0069] The ratio of the deviation in height growth to the average progression characteristics of actual bone age is calculated to determine the degree of coordinated change in bone age and height growth. ;

[0070] in, Indicates the degree of coordinated change between bone age and height growth. This indicates an individual's actual height increase over the past two years. , This indicates the pre-set standard height growth. Indicates the degree of deviation in height growth. This indicates the actual bone age progression of an individual over the past two years. , Indicates the current latest bone age. This represents the historical bone age at the target historical moment, which is the individual's historical bone age two years ago. The value 2 represents the standard bone age increase, which is 2 years. This indicates the average progression characteristics of actual bone age. It also reflects the degree of coordinated change between bone age and height growth. The higher the value, the higher the efficiency of height growth compared to the rate of bone age development, and the better the coordination and matching between bone age and height growth.

[0071] Step S4: Integrate the rate of change in height growth rate, the degree of deviation in height percentile, and the degree of coordinated change in bone age and height growth to obtain the child's current baseline predicted height.

[0072] Through steps S1-S3, we obtain the child's growth rate of height, percentile deviation in height, and the degree of coordinated change in bone age and height. The higher the values ​​of these three parameters, the more beneficial it is to the child's height growth, and the higher the predicted height. Therefore, by integrating these three parameters, we obtain the child's current baseline predicted height. All three parameters are positively correlated with the child's current baseline predicted height.

[0073] In one exemplary embodiment, these three parameters are first fused to obtain the height prediction coefficient. To facilitate data processing, this embodiment first needs to standardize the rate of change in height growth, the percentile deviation in height, and the degree of coordinated change in bone age and height growth, respectively. This eliminates dimensional differences between different features and ensures the rationality of the prediction correction. This embodiment uses the Z-score standardization algorithm to standardize the rate of change in height growth, the percentile deviation in height, and the degree of coordinated change in bone age and height growth. For any one of these parameters, the parameter value is obtained from a large sample of children of the same sex and age in history. Then, the mean and standard deviation of the parameter value for all samples are calculated, and the Z-score standardization algorithm is used to standardize the parameter. The standardization process for the other two parameters is similar. The large sample size of children of the same sex and age in the past can also be obtained from the publicly available large sample data of the "Growth Standards for Children Under 7 Years Old" (WS / T 423—2022) and the "Reference Standards for Growth and Development of School-Age Children and Adolescents in China" (WS / T 586-2018). It should be noted that when using the Z-score normalization algorithm for normalization, the data used for normalization is from the large sample size of children of the same sex and age in the past. The number of children is the 100 most recently recorded children of the same sex and age as the reference sample. This reference sample can be updated in real time based on the publicly available large sample data of the "Growth Standards for Children Under 7 Years Old" (WS / T 423—2022) and the "Reference Standards for Growth and Development of School-Age Children and Adolescents in China" (WS / T 586-2018), or based on the publicly available sample data in the hospital database.

[0074] The average of the standardized rate of change in height growth, the degree of deviation of height percentile, and the degree of coordinated change in bone age and height growth is calculated. A height prediction coefficient is obtained based on this average. In an exemplary embodiment, a specific calculation method for the height prediction coefficient is given below: ;

[0075] in, This represents the height prediction coefficient. This represents the standardized rate of change in height growth. This indicates the degree of deviation of the standardized height percentile. This indicates the degree of coordinated change in bone age and height growth after standardization. This represents the average value of the standardized rate of change in height growth, the degree of deviation of height percentile, and the degree of coordinated change in bone age and height growth.

[0076] It should be understood that if the calculated height prediction coefficient If the value is less than or equal to a preset lower limit constant (e.g., 0.1), then the height prediction coefficient is forced. It is equal to the preset lower limit constant.

[0077] This embodiment requires determining the percentage increase in adult height corresponding to the child's current bone age. This can be obtained by referring to a table based on the child's current bone age. Specifically, the percentage increase is obtained from the Bayley-Pinneau method bone age-adult height percentage comparison table for Chinese children, which is a clinically published standard value. It should be understood that the adult height increase percentage obtained from the table is calculated in decimal format (e.g., 95% is substituted as 0.95) to prevent the calculation result from being significantly off by a factor of 100. It should be noted that for the BP method, a localized modified PBA comparison table is used (the PBA comparison table is the domestically used clinical standard "Chinese Wrist Bone Development Standard - China 05" (TY / T 3001-2006)). After modification, the racial bias error can be significantly reduced.

[0078] Based on the child's current actual height, the percentage increase in adult height corresponding to the child's current bone age, and the height prediction coefficient, a baseline predicted height is obtained. Specifically: calculate the ratio of the child's current actual height to the percentage increase in adult height, then multiply it by the height prediction coefficient. The result is the child's current baseline predicted height. ;

[0079] in, This represents the child's current baseline height prediction. This indicates the percentage increase in adult height corresponding to the child's current bone age. Baseline height prediction. The higher the value, the better the child's current height base, the more remaining growth space in bone age, the faster the height growth rate, the continuous improvement in height ranking, the better the match between bone age development and height growth, and the significantly higher the height potential in adulthood compared to the average level of the normal population; Predicted height base value The smaller the value, the worse the child's current height base, the smaller the remaining growth space of bone age, and the lower the height potential after adulthood compared to the average level of the normal population.

[0080] Step S5: Based on the child's genetic factors for height and current growth hormone secretion levels, the baseline height prediction value is revised to obtain the target height prediction value.

[0081] After obtaining the child's current basic height prediction value, it is necessary to revise the basic height prediction value based on the child's height genetic factors and current growth hormone secretion level to obtain the final height prediction value, which is defined as the height target prediction value, in order to further improve the accuracy and reliability of height prediction.

[0082] A child's height is determined by genetic factors, which are one of the core foundational factors in determining their height potential. In this embodiment, the genetic factor is the child's genetic target height. The child's genetic target height is an estimated height calculated based on the heights of the child's parents. If the child is a boy, the calculation method for his genetic target height is as follows: ;

[0083] If the child is a girl, the specific method for calculating the child's genetic target height is as follows: ;

[0084] in, Indicates a child's genetic target height. This indicates the height of the child's father. This indicates the height of the child's mother.

[0085] It should be noted that the genetic target height formula is a simplified estimation commonly used in pediatric clinical practice, and it inherently has a large margin of error. However, in this invention, the target height is used as a benchmark reference and as part of the characteristics used to calculate the deviation degree K of adult height. It does not rely solely on parental height to determine final height, which conforms to standard clinical usage.

[0086] This embodiment also determines the preset standard height of healthy adults of the same sex. The preset standard height of healthy adults of the same sex can be the median height of healthy adults of the same sex. For example, the preset standard height of healthy male adults is 172.7 cm and the preset standard height of healthy female adults is 160.6 cm. These two heights are publicly available standard values.

[0087] To determine the degree of deviation of a child's genetic target height from the preset standard height of healthy adults of the same sex in adult height, in an exemplary embodiment, the degree of deviation in adult height is calculated as follows: ;

[0088] in, Indicates the degree of deviation from adult height. This represents the default standard height for healthy adults of the same sex. The degree of deviation from adult height. Used to quantify the deviation of a child's genetic potential from the population average, typically the degree of deviation in adult height. The value ranges around 1. (Degree of deviation from adult height) The higher the value, the better the child's genetic potential for height compared to the average level of the same sex, and the stronger the support of heredity for final height. Conversely, the lower the value, the weaker the genetic potential.

[0089] A child's current growth hormone secretion level is defined as their current insulin-like growth factor-1 concentration (i.e., serum IGF-1 concentration). Aside from genetic factors affecting height, serum IGF-1 concentration is a core indicator reflecting a child's growth hormone secretion level, directly regulating long bone growth and height increase. It should be understood that determining a child's serum IGF-1 concentration using existing external testing equipment is crucial.

[0090] This embodiment predetermines a preset standard concentration related to children, specifically a preset standard concentration of serum IGF-1. This preset standard concentration can be the median serum IGF-1 concentration of various children's samples from a large historical sample (i.e., a large number of children of the same age and sex). For example, the median IGF-1 concentration of children of the same age and sex can be obtained by looking up a table based on the clinically recognized standardized reference range for IGF-1 in Chinese children.

[0091] To determine the degree of deviation of a child's current serum IGF-1 concentration from a preset standard concentration of growth hormone secretion, in an exemplary embodiment, the degree of deviation of growth hormone secretion is calculated as follows: ;

[0092] in, Indicates the degree of deviation in growth hormone secretion. This indicates the child's current serum IGF-1 concentration. This indicates the preset standard concentration. Typically, this represents the degree of deviation in growth hormone secretion. The value ranges around 1. Degree of deviation in growth hormone secretion. The higher the value, the better the child's current basal endocrine level is compared to the average level of normal people of the same age. This reflects a better level of regulation related to the secretion of growth hormone in the child, which will positively correct the height value in the prediction and make the final predicted value more in line with the child's height growth potential. Conversely, it indicates that the level of regulation related to the secretion of growth hormone in the child is weak.

[0093] In addition, the degree of deviation in adult height is usually... Degree of deviation from growth hormone secretion All of these have a certain normal value range, which can be obtained through a large number of historical samples, such as the degree of deviation of adult height. For example, by using a large number of historical samples to determine the degree of adult height deviation for each child sample, the maximum and minimum values ​​are extracted (it should be understood that the minimum value is less than 1 and the maximum value is greater than 1). The range of values ​​formed by these maximum and minimum values ​​is taken as the normal range for the degree of adult height deviation. If the calculated degree of adult height deviation... If the deviation from the maximum value is exceeded, the adult height will be forcibly adjusted. Set to this maximum value; if the calculated adult height deviates from the maximum value... If the value is less than this minimum, the adult height deviation will be forcibly adjusted. Set to this minimum value. Degree of deviation in growth hormone secretion. The same principle applies to the handling of this.

[0094] Based on the degree of deviation from adult height Degree of deviation from growth hormone secretion This yields the height correction factor for children. This is based on the degree of deviation in adult height. Degree of deviation from growth hormone secretion All of these factors positively influence children's height predictions; therefore, the degree of deviation in adult height... Degree of deviation from growth hormone secretion The higher the value, the higher the height correction factor for the child. In one exemplary embodiment, this embodiment presets a constant, based on the degree of deviation in adult height. Degree of deviation from growth hormone secretion The values ​​all fluctuate around the value 1, therefore, the preset constant is set to the value 1.

[0095] Based on the degree of deviation from adult height The relationship between the magnitude of 1 and 1 yields the first correction amplitude characteristic value, which is then used to determine the degree of deviation in growth hormone secretion. The relationship between the magnitude of the first correction amplitude and 1 is used to obtain the second correction amplitude characteristic value. In an exemplary embodiment, the first correction amplitude characteristic value is: , representing the individual relative correction magnitude of genetic potential for height prediction; the second correction magnitude characteristic value is: This represents the individual relative correction magnitude of basal endocrine levels to height prediction. Therefore, a first correction magnitude eigenvalue greater than 0 (i.e., a positive value) indicates the degree of deviation from the predicted adult height. A higher value indicates that the child's genetic height potential is better than the average level for their sex; a first correction magnitude eigenvalue less than 0 (i.e., a negative value) indicates the degree of deviation in adult height. A lower value indicates that the child's genetic height potential is lower than the average level of the same sex; when the first correction magnitude characteristic value is equal to 0, it indicates that the child's genetic height potential is equal to the average level of the same sex.

[0096] Similarly, when the second correction amplitude characteristic value is greater than 0 (i.e., the value is positive), it indicates the degree of deviation in growth hormone secretion. A higher level indicates that the child's current baseline endocrine level is better than the average level of the normal population of the same age; when the second correction amplitude characteristic value is less than 0 (i.e., the value is negative), it indicates the degree of deviation in growth hormone secretion. A lower value indicates that the child's current baseline endocrine level is lower than the average level of the normal population of the same age; when the second correction amplitude characteristic value is equal to 0, it indicates that the child's current baseline endocrine level is the same as the average level of the normal population of the same age.

[0097] For any of the first and second correction magnitude feature values, subtracting 1 is used to convert absolute correction into relative correction, eliminate benchmark conflicts, and retain only the pure variation range that deviates from the population average. When the corresponding deviation is greater than 1, it indicates that the child is better than the population average. In this case, the value after subtracting 1 is greater than 0, which means that there will be a positive correction magnitude for the child's height, which will raise the predicted value. Conversely, the value after subtracting 1 is less than 0, which means that there will be a negative correction magnitude for the child's height, which will lower the predicted value.

[0098] The first and second correction amplitude feature values ​​are weighted and fused to obtain the height correction coefficient, as follows: ;

[0099] in, This represents the height correction factor. and These are the weight values ​​of the first and second correction amplitude characteristic values, respectively. Both weight values ​​range from 0 to 1. In an exemplary embodiment, to suppress the correction amplitude and avoid amplifying prediction distortion, these weight values ​​are typically small, and can be understood as empirical attenuation coefficients for each correction amplitude characteristic value. The specific values ​​of these two weight values ​​can be determined empirically, for example, both set to 0.1. It should be noted that setting α and β to 0.1 represents a low-damping, weak-attenuation design: on the one hand, single-item deviations only produce small corrections, avoiding excessive interference from a single factor in the baseline prediction; on the other hand, when multiple deviations are superimposed, the total correction amplitude is controllable, stably limiting K to the safe range of 0.7~1.3, while weakening prediction fluctuations caused by measurement noise, aligning with the clinical assessment logic of "basic bone age prediction as the main factor, with only minor adjustments from genetic, developmental, and acquired factors."

[0100] Height Correction Factor The higher the value, the more pronounced the combined advantages of the child's genetic potential and endocrine levels, indicating a superior innate foundation for height growth and a greater positive correction to the baseline prediction value; conversely, a lower height correction coefficient indicates a lower overall height. The smaller the value, the weaker the overall level of a child's genetic potential and endocrine level, and the greater the negative correction to the baseline prediction value.

[0101] Finally, based on the obtained height correction coefficient Predicted baseline height After correction, the target height prediction value is obtained. Baseline height prediction value. and height correction factor The larger the value, the larger the corresponding predicted height. In an exemplary embodiment, a height correction factor is calculated. Compared with the baseline predicted height The product of: To obtain the target height prediction value for children. It should be understood that if the obtained height target prediction value... If the value is too high and exceeds the normal height range for adults, an early warning signal can be issued to facilitate investigation of the cause.

[0102] Thus, by combining multi-dimensional growth and development data of individual children, a multi-dimensional feature system was constructed, their growth characteristics were analyzed, and individual height prediction of children was completed, thereby improving the accuracy of children's height prediction.

[0103] This embodiment also provides an individual height prediction system based on multi-dimensional growth and development data, including: a memory and a processor; the memory is connected to the processor, and the memory is used to store program instructions; the processor is used to implement the steps in the above-described embodiment of the individual height prediction method based on multi-dimensional growth and development data when the program instructions are executed.

[0104] In one exemplary embodiment, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the embodiment of the individual height prediction method based on multidimensional growth and development data.

[0105] It should be noted that all data and information collected in this application have been obtained with full consent and authorization.

Claims

1. A method for predicting individual height based on multidimensional growth and development data, characterized in that, include: Determine the child's height and bone age data for the most recent several years; the most recent several years include the most recent year and several historical years; Based on the height data, determine the child's growth rate of change and height percentile deviation in the most recent year relative to the historical year. Based on the height and bone age data, determine the degree of coordinated change in bone age and height growth in the child over the most recent year relative to the historical year. By integrating the growth rate of height, the percentile deviation of height, and the degree of coordinated change in bone age and height, the child's current basic height prediction value is obtained. Based on the child's genetic factors for height and current growth hormone secretion levels, the baseline height prediction is revised to obtain the target height prediction.

2. The individual height prediction method based on multi-dimensional growth and development data as described in claim 1, characterized in that, The process of obtaining the rate of change of height growth includes: Determine the latest height growth rate for the most recent year, as well as the historical height growth rates for each historical year; The overall historical height growth rate for each historical year can be obtained from the historical height growth rate for each historical year. The rate of change of the latest height growth rate relative to the overall historical height growth rate is determined to obtain the rate of change of height growth rate.

3. The individual height prediction method based on multi-dimensional growth and development data as described in claim 1, characterized in that, The process of obtaining the percentile deviation of height includes: Determine the child's current height percentile and the historical height percentile at a target historical moment; the target historical moment is within the historical year. The average rate of change of the latest height percentile and the historical height percentile is determined to obtain the degree of deviation of the height percentile.

4. The individual height prediction method based on multi-dimensional growth and development data as described in claim 1, characterized in that, The process of obtaining the degree of coordinated changes in bone age and height includes: Determine the child's current height and the target historical height, as well as the child's current bone age and the target historical bone age; the target historical time is within the historical year. The deviation of the actual height increase from the preset standard height increase, and the average progression characteristics of the actual bone age are determined; the actual height increase is the difference between the latest height and the historical height. The degree of synergistic change in bone age and height is obtained by calculating the ratio of the deviation in height growth to the average progression characteristics of actual bone age.

5. The individual height prediction method based on multi-dimensional growth and development data as described in claim 1, characterized in that, The process of obtaining the basic predicted height value includes: By integrating the growth rate of height, the percentile deviation of height, and the degree of coordinated change in bone age and height, a height prediction coefficient is obtained. The basic predicted height value is obtained based on the child's current actual height, the percentage increase in adult height corresponding to the child's current bone age, and the height prediction coefficient.

6. The individual height prediction method based on multi-dimensional growth and development data as described in claim 5, characterized in that, The height prediction coefficient is obtained by integrating the growth rate of height, the percentile deviation of height, and the degree of coordinated change in bone age and height growth, including: The rate of change in height growth rate, the degree of deviation of height percentile, and the degree of coordinated change in bone age and height growth were standardized respectively. Calculate the average of the standardized height growth rate change rate, height percentile deviation, and bone age-height co-change, and obtain the height prediction coefficient based on the average value.

7. The individual height prediction method based on multi-dimensional growth and development data as described in claim 5, characterized in that, The process of obtaining the baseline height prediction value based on the child's current actual height, the percentage increase in adult height corresponding to the child's current bone age, and the height prediction coefficient includes: Calculate the ratio of the child's current actual height to the percentage increase in adult height, and then multiply it by the height prediction coefficient to obtain the basic predicted height value.

8. The individual height prediction method based on multi-dimensional growth and development data as described in claim 1, characterized in that, The height genetic factors refer to the child's genetic target height, and the current growth hormone secretion level refers to the current insulin-like growth factor-1 concentration. The process of obtaining the height target prediction value includes: Determine the degree of deviation of the genetic target height from the preset standard height of healthy adults of the same sex in terms of adult height, and the degree of deviation of the current insulin-like growth factor-1 concentration from the preset standard concentration of growth hormone secretion. Based on the deviations in adult height and growth hormone secretion, a height correction coefficient is obtained; Based on the height correction coefficient, the basic predicted height value is corrected to obtain the target predicted height value.

9. The individual height prediction method based on multi-dimensional growth and development data as described in claim 8, characterized in that, The process of obtaining the height correction coefficient includes: Based on the relationship between the degree of deviation of adult height and the preset constant, the first correction amplitude characteristic value is obtained; The second correction amplitude characteristic value is obtained based on the relationship between the degree of deviation in growth hormone secretion and the preset constant. The first correction amplitude feature value and the second correction amplitude feature value are weighted and fused to obtain the height correction coefficient.

Images