Method of estimating Growth age using artificial intelligence model

Abstract

A method for estimating growth age using an artificial intelligence model according to exemplary embodiments of the present invention may include the steps of receiving biological data of a subject, extracting predicted APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate is expected to reach a maximum value using the biological data of the subject, and calculating the growth age of the subject based on the predicted APHV and a second APHV data, which is the average value of the first APHV data.

Description

Technology Field

[0001] The present invention relates to a method for estimating biological growth age, rather than actual temporal age, based on Age of Peak Height Velocity (APHV) data, which is the age at which the growth rate reaches its maximum value, extracted from biological data of growing children and adolescents. Background Technology

[0002] With the recent advancement of artificial intelligence technology, AI technologies are being applied across various fields, and methods are being developed and used to generate additional information by extracting inherent features from data through neural network models, replacing conventional data processing methods.

[0003] Neural network models used for artificial intelligence can detect and recognize features within input data more quickly and accurately compared to general data processing through learning. Recently, AI technology has moved beyond simply tracking and detecting objects to being applied to learn from past history, predict the future, or derive current features that reflect time-series changes.

[0004] Among these, predictive analytics is a technology within the fields of statistics and data mining that extracts information from data and uses it to predict trends and behavioral patterns. Based on the information obtained from data, such predictive analytics can be applied to all areas requiring decision-making. The core of predictive analytics lies in understanding the relationships between variables and then predicting unknown variables.

[0005] To this end, various approaches are being used depending on the data characteristics and the prediction target.

[0006] Among the various fields requiring predictive analysis, there is the field of physical growth in adolescents. There is significant interest among parents and adolescents regarding when height growth will occur and how much they will grow.

[0007] Conventional methods for predicting height growth involve predicting growth plates by X-ray imaging or analyzing the relationship with genetic and environmental factors (Registered Patent Publication No. 10-2075743, Registered Patent Publication No. 10-1866208), or they also propose methods for converting physical data from sample subjects with different measurement times or frequencies into a format suitable for training a growth prediction model (Registered Patent Publication No. 10-2198302).

[0008] While conventional methods for predicting physical growth in children and adolescents have been presented, research on methods for predicting the risk of occurrence and providing solutions for issues such as precocious puberty and obesity that may accompany growth in children and adolescents, as mentioned above, is currently insufficient.

[0009] Furthermore, since children and adolescents have growth stages with distinct characteristics, considering this can enhance the reliability of solutions provided through predicted data and analysis.

[0010] Meanwhile, regarding conventional technologies disclosed to date, various methods for predicting growth by directly inputting physical or biological data of children and adolescents into a predetermined growth prediction tool have been revealed. However, these conventional technologies have the disadvantage that it is difficult to provide personalized solutions because they predict growth based solely on general indicators such as past and present height, weight, and BMI, without considering individual growth characteristics such as the growth rate of each child or adolescent or whether they have reached a growth stage faster than other sample subjects.

[0011] In addition, the SITAR (SuperImposition by Translation And Rotation) methodology has emerged in relation to height growth rate; however, there is a problem in that only ex post estimation is possible because measurements across the entire growth period are required to estimate the growth rate curve using the SITAR methodology.

[0012] Meanwhile, conventional height growth prediction models use height, weight, body composition, their quantiles, and current quantile-based benchmark growth rates as input variables to predict the height growth curve using an LGBM model. However, there is a possibility of upward bias in the estimation model in specific datasets where the proportion of observations is high due to higher height quantiles in later ages. Additionally, although conventional height growth prediction models have 52 body composition-related variables, they utilize only 8 variables centered on height, weight, and body mass, which has the disadvantage of low prediction accuracy.

[0013] Consequently, although the growth rate trend of sample subjects is important for height prediction in conventional height growth prediction models, information on past measurements is not reflected in the estimation model; therefore, it is necessary to measure tracking growth rate that incorporates past information.

[0014] Furthermore, the importance of a method utilizing these tracking growth rates to estimate biological growth age—that is, the age indicating the biological age of the body based on the state of growth—is emerging, rather than actual age, which is the temporal actual age calculated uniformly from the time of birth. Prior art literature

[0015] Registered Patent Publication No. 10-2075743 Registered Patent Publication No. 10-1866208 Registered Patent Publication No. 10-2198302 The problem to be solved

[0016] One of the various objectives of the present invention is to provide a method, apparatus, and computer program that classifies growth stages based on Age of Peak Height Velocity (APHV) data, which is the age at which the growth rate of children and adolescents reaches its maximum value, predicts growth such as height, obesity, and disease, and provides customized solutions for each growth stage.

[0017] One of the various objectives of the present invention is to provide a method for estimating a tracking growth rate that reflects past information of children and adolescents using an APHV estimation model constructed through artificial intelligence, and based on this, to estimate a biological growth age that indicates the approximate age of the body biologically based on the growth status of the children and adolescents.

[0018] The purpose is to provide a method, device, and computer program that predicts growth, including height, obesity, and disease, based on growth stages and provides customized solutions for each growth stage.

[0019] One of the various objectives of the present invention is to provide a method, apparatus, and computer program that predicts growth, such as height, obesity, and disease, based on the growth stages of children and adolescents using an APHV estimation model constructed through artificial intelligence, and provides customized solutions for each growth stage. means of solving the problem

[0020] A method for estimating growth age using an artificial intelligence model trained on a first Age of Peak Height Velocity (APHV) data, which is the age at which the growth rate of a plurality of sample subjects reaches a maximum value according to exemplary embodiments of the present invention, wherein the growth age estimation method may include: a step of receiving biological data of a subject to be measured; a step of extracting predicted Age of Peak Height Velocity (APHV) data, which is the age at which the growth rate is expected to reach a maximum value, using the biological data of the subject to be measured; and a step of calculating the growth age of the subject to be measured based on the predicted APHV and a second APHV data, which is the average value of the first APHV data.

[0021] The above growth age calculation step can be performed using mathematical formula 1, and

[0022] [Mathematical Formula 1]

[0023] GA i '= (CAi - APHV i ') + APHV G

[0024] Here, GA i ' is the estimated growth age of the subject being measured, and CA i is the chronological age (calendar age) of the subject being measured, and APHV i ' is the predicted APHV data including the estimated and predicted APHV values ​​of the above-mentioned measurement subject, and APHV G The second APHV data may be a representative value or average value of the first APHV data derived from the training data group.

[0025] The predicted APHV extraction step may include: a step of estimating the expected shape of the growth rate using the biological data of the subject being measured; and a step of extracting the predicted AOGS (Age of Onset of Growth Spurt), which is the age at which the increase in the growth rate of the subject being measured begins, and the predicted APHV (Age of Peak Height Velocity), which is the age at which the growth rate of the subject being measured has a maximum value.

[0026] The biometric data of the subject measured above may include at least two time-series biometric data measured at different age groups.

[0027] The above first APHV data may be a set of data obtained by receiving the biometric data of the plurality of sample subjects, estimating a plurality of first growth rate curves representing the growth rate of height relative to the actual age of each of the plurality of sample subjects, and extracting the age at which the growth rate reaches a maximum value for each of the plurality of first growth rate curves.

[0028] A method for estimating growth age using an artificial intelligence model trained with parameters and feature points that determine multiple first growth rate curves representing the growth rate of height relative to the actual age of multiple sample subjects according to exemplary embodiments of the present invention, wherein the growth age estimation method may include: a step of receiving biological data of a subject to measurement; a step of extracting predicted APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate is expected to reach a maximum value, using the biological data of the subject to measurement; and a step of calculating the growth age of the subject to measurement based on the predicted APHV and a second APHV data, which is the average value of the first APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate of the multiple sample subjects reaches a maximum value among the feature points that determine the multiple first growth rate curves.

[0029] The above growth age calculation step can be performed using mathematical formula 1, and

[0030] [Mathematical Formula 1]

[0031] GA i '= (CA i - APHV i ') + APHV G

[0032] Here, GA i ' is the estimated growth age of the subject being measured, and CA i is the chronological age (calendar age) of the subject being measured, and APHV i ' is the predicted APHV data including the estimated and predicted APHV values ​​of the above-mentioned measurement subject, and APHV G The second APHV data may be a representative value or average value of the first APHV data derived from the training data group.

[0033] The biometric data of the subject measured above may include at least two time-series biometric data measured at different age groups.

[0034] The above first APHV (Age of Peak Height Velocity) data may be a set of data in which the age at which the growth rate reaches a maximum value is extracted for each of the plurality of sample subjects' first growth rate curves.

[0035] The above first growth rate curve may be estimated by extracting age data and height data from the time-series biometric data of the above sample subjects and applying them to a curve transformation model that uses the age data and height data as input variables. Effects of the invention

[0036] A growth age estimation model based on biometric data using artificial intelligence according to exemplary embodiments of the present invention estimates a growth rate curve by applying a tracking growth rate extracted from past information to a statistical methodology, and extracts Age of Peak Height Velocity (APHV) data, which is the age at which the growth rate reaches a maximum value, from the growth rate curve. Since the APHV data is utilized in the construction of various prediction models, such as an integrated height growth prediction model, a precocious puberty prediction model, and an obesity prediction model, the reliability of various prediction models, such as an integrated height growth prediction model, a precocious puberty prediction model, and an obesity prediction model, can be maximized.

[0037] In addition, by utilizing the aforementioned tracking growth rate, the biological growth age—that is, the age indicating the biological age of the body based on the growth status—is estimated, rather than the actual age, that is, the actual age calculated uniformly from the time of birth. Since the biological growth age is utilized in the construction of various prediction models, such as an integrated height growth prediction model, a precocious puberty prediction model, and an obesity prediction model, the reliability of these various prediction models can be maximized.

[0038] In addition, according to various embodiments of the present invention, an accurate height can be predicted by considering the growth stages of children and adolescents through a growth age estimation model constructed through artificial intelligence learning, and solutions necessary for height growth can be provided by considering each growth stage. Brief explanation of the drawing

[0039] FIG. 1 is a configuration diagram illustrating the configuration of a bio-data-based growth age estimation model using artificial intelligence according to exemplary embodiments of the present invention. FIG. 2 is a diagram showing predicted elongation and target elongation by growth stage according to exemplary embodiments of the present invention. FIG. 3 is a flowchart illustrating a method for estimating growth age based on biodata using artificial intelligence according to exemplary embodiments of the present invention. Figure 4 is a diagram showing multiple first growth rate curves representing the growth rate of height relative to actual age using time-series biometric data of sample subjects. Figures 5 and 6 are figures showing the APHV estimation error according to the age in months for girls and boys, respectively. Figures 7 and 8 are figures showing the importance of APHV predictors for women and men, respectively. Figure 9 is a diagram showing the importance of variables in a conventional growth prediction model for women and men, and Figure 10 is a diagram showing the importance of variables in a growth prediction model of the present invention. Specific details for implementing the invention

[0040] Specific embodiments of the present invention will be described below. The following detailed description is provided to facilitate a comprehensive understanding of the methods, devices, and / or systems described herein. However, this is merely illustrative and the present invention is not limited thereto.

[0041] In describing the embodiments of the present invention, detailed descriptions of known technologies related to the present invention are omitted if it is determined that such detailed descriptions may unnecessarily obscure the essence of the present invention. Furthermore, the terms described below are defined in consideration of their functions within the present invention, and these may vary depending on the intentions or practices of the user or operator. Therefore, such definitions should be based on the content throughout this specification. Terms used in the detailed description are intended merely to describe the embodiments of the present invention and should not be limiting in any way. Unless explicitly stated otherwise, expressions in the singular form include the meaning of the plural form. In this description, expressions such as "include" or "comprise" are intended to refer to certain characteristics, numbers, steps, actions, elements, parts thereof, or combinations thereof, and should not be interpreted to exclude the existence or possibility of one or more other characteristics, numbers, steps, actions, elements, parts thereof, or combinations thereof other than those described.

[0042] In addition, terms such as first, second, A, B, (a), (b), etc. may be used when describing the components of the embodiments of the present invention. These terms are used merely to distinguish the components from other components, and the essence, order, or sequence of the components is not limited by the terms.

[0043] In the exemplary embodiments of the present invention, children and adolescents may be understood as concepts encompassing the period of human growth. More specifically, children and adolescents, referring to the sample subjects in the exemplary embodiments of the present invention, are defined below through the meanings of infant, child, adolescent, toddler, and child.

[0044] Infants are a continuation of the neonatal period, a period during which they grow by sucking on their mother's nipple until two years after birth, and the experiences of nutrition, caress, and excretion during this period influence their general tendencies thereafter. Children are generally defined as individuals in the age group of 6 years or older but under 13 years, and in a broader sense, they may include toddlers (ages 1 to 5).

[0045] Adolescents refer to the intermediate stage between childhood and youth, generally meaning individuals aged 13 to under 19 years based on full age. Infants may refer to individuals from one year of age to five years of full age. Children may generally refer to children up to the age of 15.

[0046] Therefore, the children and adolescents referred to as the sample subjects mentioned above may mean, in a narrow sense, the period of 5 years of age and older and 19 years of age including children and adolescents, and in a broad sense, the period including the general period during which physical growth occurs, from infancy to adolescence.

[0047] FIG. 1 is a configuration diagram illustrating the configuration of a bio-data-based growth age estimation model using artificial intelligence according to exemplary embodiments of the present invention.

[0048] Referring to FIG. 1, a growth age estimation model based on artificial intelligence according to exemplary embodiments of the present invention may include a bio-data input unit (10), a preprocessing unit (20), a growth rate curve estimation unit (30), a first APHV data extraction unit (40), a first APHV data learning unit (50), a predicted APHV data extraction unit (60), a second APHV data extraction unit (70), and a growth age calculation unit (80).

[0049] A growth age estimation model based on bio-data using artificial intelligence according to exemplary embodiments of the present invention can receive time-series physical information about a sample subject through a bio-data input unit (10). The physical information of the sample subject may include not only basic information such as grade (or age), gender, and height, but also additional information such as weight, protein, mineral content, body fat, body water, soft lean mass, fat-free mass, bone tissue, skeletal muscle mass, Body Mass Index (BMI), basal metabolic rate, neck circumference, chest circumference, abdominal circumference, thigh circumference, arm circumference, and hip circumference. Such physical information is merely an example to aid in understanding the present invention, and the present embodiment is not limited thereto; it is understood that the types of information constituting the physical information can be varied according to the embodiment.

[0050] The time-series physical information of the above sample subject may be continuous information or discontinuous information, but may be information included in at least one of the periods corresponding to the growth stages of FIG. 2.

[0051] More specifically, the time-series physical information of the sample subjects may vary in the time and frequency of collection. For example, in the case of the first sample subject, physical information measured up to 8 to 12 years of age, which is part of the childhood and adolescent period, may exist, and in the case of the second sample subject, physical information measured irregularly, such as at 8, 10 to 12, and 15 years of age. Additionally, in the case of the third sample subject, physical information measured multiple times within a certain period (within the period of any one of the multiple growth stages in Fig. 2), whereas in the case of the fourth sample subject, physical information measured only once within a certain period (within the period of any one of the multiple growth stages in Fig. 2).

[0052] As described above, depending on the time and frequency of collection, the physical information of the sample subject may be included in two or more of the growth stages of FIG. 2 (the first sample subject, the second sample subject), but there may also be cases where it is not (the third sample subject, the fourth sample subject).

[0053] The preprocessing unit (20) can be controlled to perform preprocessing on the time-series biometric data of the sample subjects.

[0054] In exemplary embodiments, the preprocessing unit (20) may be performed by determining all biometric data of a sample subject for which biometric data collected during a specific growth period does not exist as noise and removing it. This is intended to improve the prediction accuracy of the biometric data-based APHV estimation model according to the present invention and may not be applied when sufficient data has been accumulated. In this case, instead of determining all biometric data of a sample subject for which biometric data collected during a specific growth period does not exist as noise and removing it, preprocessing may be performed by supplementing or additionally generating some biometric data through various methods such as artificial intelligence models and big data analysis.

[0055] Meanwhile, the biometric data input unit (10) may be controlled to receive more body information of the subject for measurement, in addition to receiving time-series body information of the sample subject, for which APHV is to be estimated. At this time, the body information of the subject for measurement may include not only basic information such as grade (or age), gender, and height, but also additional information such as weight, protein, mineral content, body fat, body water, soft lean mass, fat-free mass, bone tissue, skeletal muscle mass, Body Mass Index (BMI), basal metabolic rate, neck circumference, chest circumference, abdominal circumference, thigh circumference, arm circumference, and hip circumference.

[0056] The growth rate curve estimation unit (30) can be controlled to estimate multiple first growth rate curves representing the growth rate of height relative to the actual age of each of the multiple sample subjects using time-series biometric data of the sample subjects.

[0057] In exemplary embodiments, the estimation of the first growth rate curve can be performed by extracting age data and height data from the time-series biometric data of the sample subjects and applying them to a curve transformation model that uses the age data and the height data as input variables.

[0058] At this time, the type of the curve transformation model is not particularly limited. That is, the curve transformation model may be a model constructed using a polynomial function, for example, the curve transformation model may be a model constructed using Equation 2.

[0059] [Mathematical Formula 2]

[0060]

[0061] Here, y it is α i , β i and γ iθ, θ, and θ are random correction values ​​for adjusting the vertical shift, horizontal shift, and slope of the curve, respectively, h is a specific function related to height data, and t can be age data.

[0062] However, the concept of the present invention is not necessarily limited thereto, and may be a module constructed using a polynomial function other than the above mathematical formula 2.

[0063] The first APHV data extraction unit (40) can be controlled to extract first APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate reaches a maximum value for each of the plurality of first growth rate curves.

[0064] The above first APHV data can be used to construct a growth stage diagnosis and estimation model and to derive the growth age, and the above growth age and tracking growth rate can ultimately be used to construct various prediction models, such as an integrated height growth prediction model, a precocious puberty prediction model, and an obesity prediction model.

[0065] Meanwhile, conventional height growth prediction models use height, weight, body composition, their quantiles, and current quantile-based benchmark growth rates as input variables to predict the height growth curve using an LGBM model. However, there is a possibility of upward bias in the estimation model in specific datasets where the proportion of observations is high due to higher height quantiles in later ages. Additionally, although conventional height growth prediction models have 52 body composition-related variables, they utilize only 8 variables centered on height, weight, and body mass, which has the disadvantage of low prediction accuracy.

[0066] In other words, although the growth rate trend of a sample is important for height prediction in conventional height growth prediction models, information on past measurements is not reflected in the estimation model; therefore, it is necessary to measure the tracking growth rate that incorporates past information. Accordingly, the bio-data-based APHV estimation model using artificial intelligence according to exemplary embodiments of the present invention is characterized by estimating a growth rate curve by applying a statistical methodology to the tracking growth rate extracted from past information, extracting APHV (Age of Peak Height Velocity) data—which is the age at which the growth rate reaches its maximum value—from the growth rate curve, and utilizing the APHV data in constructing various prediction models, such as an integrated height growth prediction model, a precocious puberty prediction model, and an obesity prediction model.

[0067] The first APHV data learning unit (50) can be controlled to input the first APHV data into an artificial intelligence model that uses the target data for learning.

[0068] The predicted APHV data extraction unit (60) can be controlled to extract predicted APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate of the subject is expected to reach its maximum value, by applying the biological data of the subject to be measured to an artificial intelligence model trained with the first APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate of the plurality of sample subjects reaches its maximum value, as target data.

[0069] The second APHV data extraction unit (70) can be controlled to extract the second APHV data, which is the average value of the first APHV data. In exemplary embodiments, the second APHV data extraction unit (70) can be controlled to extract the second APHV data through an average value calculation algorithm that adds all observations and divides by the number of observations.

[0070] The growth age calculation unit (80) can be controlled to calculate the growth age of the subject to measurement based on the second APHV data, which is the average value of the first APHV data, and the predicted APHV.

[0071] At this time, the growth age calculation unit (80) can be controlled to calculate the growth age of the subject to measurement using the following mathematical formula 1, and

[0072] [Mathematical Formula 1]

[0073] GA i '= (CA i - APHV i ') + APHV G

[0074] Here, GA i ' is the estimated growth age of the subject being measured, and CA i is the chronological age (calendar age) of the subject being measured, and APHV i ' is the predicted APHV data including the estimated and predicted APHV values ​​of the above-mentioned measurement subject, and APHV G The second APHV data may be a representative value or average value of the first APHV data derived from the training data group.

[0075] In other words, conventional growth age estimation models generally calculate growth age using bone age, the degree of opening / closing of the growth plate, etc., but this has the disadvantage that objectivity is not guaranteed and accuracy is low because X-ray measurement of the bone or growth plate is absolutely necessary, and the growth age may be measured differently depending on the entity analyzing the X-ray of the bone or growth plate.

[0076] In contrast, the growth age estimation model based on bio-data using artificial intelligence according to exemplary embodiments of the present invention estimates a growth rate curve by applying a tracking growth rate extracted from past information to a statistical methodology, extracts Age of Peak Height Velocity (APHV) data—the age at which the growth rate reaches its maximum value—for the growth rate curve, and utilizes the APHV data to construct a growth age estimation model. Therefore, there is no need to perform additional processes such as X-ray measurement. Furthermore, since the growth age is determined by applying accurate numerical values ​​to a pre-set mathematical formula, there is no possibility that the growth age will be measured differently depending on the subject. Accordingly, the objectivity of the growth age estimation can be guaranteed, and there is an advantage of high accuracy.

[0077] FIG. 2 is a diagram showing predicted elongation and target elongation by growth stage according to exemplary embodiments of the present invention.

[0078] Referring to FIG. 2, the growth stages may include a general growth period (301), a rapid growth period (303), a decelerated growth period (305), a slowed growth period, and a growth plate-free period (309).

[0079] Each growth stage can be classified according to the degree of growth, and the height gained each year varies depending on the stage; even within the same growth stage, the actual height gained can differ depending on the growth type.

[0080] The general growth period (301) usually refers to the period prior to puberty when secondary sexual characteristics appear. Children and adolescents in this period generally have open growth plates, and depending on the growth environment, they generally grow 4 to 5 cm per year in the case of a short growth type and grow 6 to 7 cm per year in the case of a tall growth type. In exemplary embodiments, the general growth period (301) may be defined as the period prior to the onset of rapid pubertal growth in children and adolescents, and may be a growth stage with an end point that is earlier than the predicted APHV by the first growth period. In one embodiment, the first growth period may be a period between 1 year and 2 years, for example, about 1.5 years.

[0081] The rapid growth period (303) is a period when secondary sexual characteristics begin to appear, during which the breasts of women swell and buds form, while the testicles of men enlarge, pubic hair begins to grow, and voice changes occur. The rapid growth period (303) generally lasts for about 2 to 3 years after the general growth period (301) and grows in the range of about 7 to 10 cm per year on average. In exemplary embodiments, the rapid growth period (303) may be defined as a period during which the growth rate increases from the onset of pubertal rapid growth until the APHV, and may be a growth stage with a starting point equal to the first growth period preceding the predicted APHV and an ending point at the predicted APHV. In one embodiment, the first growth period may be a period between 1 and 2 years, for example, about 1.5 years.

[0082] The deceleration period (305) refers to the period when secondary sexual characteristics are being completed. During this period, women can be distinguished by the onset of menstruation, while men can clearly see the changes through pubic hair, voice change, and underarm hair. When the deceleration period (305) begins, the rate of growth drops rapidly compared to the rapid growth period (303). It generally lasts for about 2 to 3 years, and natural growth stops as growth averages within a range of about 5 to 6 cm per year. The growth plates begin to close gradually after the rapid growth period (304), and about 50% close after about 6 months of entering the deceleration period (305). In exemplary embodiments, the decelerated growth period (305) may be defined as a period during which the growth rate gradually decreases from the APHV until the end of the pubertal rapid growth, and may be a growth stage with the predicted APHV as the starting point and a second growth period after the predicted APHV as the ending point. In one embodiment, the second growth period may be a period between one and two years, for example, about 1.5 years.

[0083] The slowing growth period (307) may refer to a period after the rapid growth of puberty and before entering the growth plate period, during which the growth rate approaches zero. In exemplary embodiments, the slowing growth period (307) may be defined as a period during which the growth rate gradually converges to zero after the rapid growth of puberty of children and adolescents has ended, and may be a growth stage with a starting point that is two years past the predicted APHV and a ending point that is three years past the predicted APHV. In one embodiment, the third growth period may be a period between two and four years, for example, about three years.

[0084] The growth plate period (309) refers to a time when the growth plate has closed, meaning that although the growth period has not completely ended, natural height growth has become difficult. Generally, for women, the growth plate period (307) is entered approximately 1 year and 6 months to 2 years after the onset of menstruation, and for men, the growth plate period (307) is entered approximately 1 year and 6 months to 2 years after the time when body hair grows in the armpits. During the growth plate period (307), the growth plate closes and natural growth stops, but growth may be possible within a range of about 1 to 3 cm by changing bad lifestyle habits and improving physical function through customized exercise, posture correction, and nutrient intake. In exemplary embodiments, the growth plate period (309) may be defined as a period when the growth rate is close to zero after the growth plate has closed, and it may be a growth stage starting at a point that is 3 years past the predicted APHV. In one embodiment, the third growth period may be a period between 2 and 4 years, for example, about 3 years.

[0085] FIG. 3 is a flowchart illustrating a method for estimating growth age based on biodata using artificial intelligence according to exemplary embodiments of the present invention.

[0086] Referring to FIG. 3, in a method for estimating growth age using an artificial intelligence model trained with a first APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate of a plurality of sample subjects reaches a maximum value according to exemplary embodiments of the present invention, the growth age estimation method may include the step of receiving biological data of a subject to measurement (S10), the step of extracting predicted APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate is expected to reach a maximum value using the biological data of the subject to measurement (S20), and the step of calculating the growth age of the subject to measurement based on the second APHV data, which is the average value of the first APHV data, and the predicted APHV (S30).

[0087] First, regarding an artificial intelligence model trained using first APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate of multiple sample subjects reaches a maximum value, the artificial intelligence model may be constructed through the following steps: receiving time-series biometric data of multiple sample subjects; estimating multiple first growth rate curves representing the growth rate of height relative to the actual age of each of the multiple sample subjects using the time-series biometric data of the sample subjects; extracting first APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate reaches a maximum value for each of the multiple first growth rate curves; and inputting the first APHV data into an artificial intelligence model using the target data for training.

[0088] At this time, the first APHV data may be a set of data obtained by receiving the biometric data of the plurality of sample subjects, estimating a plurality of first growth rate curves representing the growth rate of height relative to the actual age of each of the plurality of sample subjects, and extracting the age at which the growth rate reaches a maximum value for each of the plurality of first growth rate curves.

[0089] Here, the step of estimating the first growth rate curve can be performed by extracting age data and height data from the time-series biometric data of the sample subjects and applying them to a curve transformation model that uses the age data and height data as input variables.

[0090] In exemplary embodiments, the curve transformation model may be a model constructed using a polynomial function.

[0091] In one embodiment, the curve transformation model may be a model constructed using Equation 2.

[0092] [Mathematical Formula 2]

[0093]

[0094] In the above mathematical formula 2, y it is α i , β i and γ i is a random correction value for adjusting the vertical shift, horizontal shift, and slope of the curve, respectively, h is a specific function related to height data, and t can be age data.

[0095] Meanwhile, the artificial intelligence model according to the present invention is performed after the step of inputting time-series biometric data of the sample subjects, and may further include a step (not shown) of preprocessing the time-series biometric data of the sample subjects.

[0096] The above preprocessing step can be performed by determining all biometric data of sample subjects for whom biometric data collected during a specific growth period does not exist as noise and removing them.

[0097] In exemplary embodiments, the preprocessing step may be performed by classifying the growth periods of the plurality of sample subjects into a general growth period, a rapid growth period, and a decelerated growth period, and then removing all data of sample subjects for which biometric data corresponding to each of the classified growth stages does not exist as noise.

[0098] In one embodiment, the preprocessing step may be performed by further dividing the growth periods of the plurality of sample subjects into a slowed growth phase and a growth-free phase, and then removing all data of sample subjects for which biodata corresponding to each of the divided growth phases does not exist as noise.

[0099] For example, the above preprocessing step can be performed by classifying the age groups of the plurality of sample subjects into ages 8 to 18, and then removing all data of sample subjects as noise when the difference in input period between adjacent biometric data is 4 years or more.

[0100] Alternatively, the above preprocessing step may be performed by removing all data as noise from sample subjects for whom no biometric data was input during the period between 9 and 14 years among the age groups of the plurality of sample subjects.

[0101] The bio-data input step (S10) can be performed by receiving body information about the subject of measurement through the bio-data input unit (10). At this time, the body information about the subject of measurement may include not only basic information such as grade (or age), gender, and height, but also additional information such as weight, protein, mineral content, body fat, body water, soft lean mass, fat-free mass, bone tissue, skeletal muscle mass, Body Mass Index (BMI), basal metabolic rate, neck circumference, chest circumference, abdominal circumference, thigh circumference, arm circumference, and hip circumference. However, such body information is merely an example to aid in understanding the present invention, and the present embodiment is not limited thereto, and it is understood that the types of information constituting the body information can be varied according to the embodiment.

[0102] In addition, the body information of the subject being measured may be single body information measured once, or body information measured multiple times, that is, information that is continuous in a time series. However, the concept of the present invention is not necessarily limited thereto, and the body information of the subject being measured may be discontinuous in a time series. That is, the collection time and the number of times the body information of the subject being measured is collected may vary.

[0103] In one embodiment, the biometric data of the subject being measured may include at least two time-series biometric data measured at different age groups.

[0104] The predicted APHV extraction step (S20) may include a step of estimating the expected shape of the growth rate using the biological data of the subject being measured, and a step of extracting a predicted AOGS (Age of Onset of Growth Spurt), which is the age at which the growth rate of the subject being measured begins to increase, and a predicted APHV (Age of Peak Height Velocity), which is the age at which the growth rate of the subject being measured has a maximum value.

[0105] In exemplary embodiments, in the predicted APHV extraction step (S20), the step of estimating the predicted growth rate of the subject being measured can be performed by generating a graph (hereinafter referred to as the second growth rate curve) representing the predicted growth rate of the subject being measured using a single or multiple data as shown in FIG. 4, and then extracting predicted AOGS data representing the predicted AOGS value and predicted APHV data representing the predicted APHV value from the second growth rate curve.

[0106] In the above description, the predicted APHV extraction step (S20) was described based on the fact that it is performed by generating the second growth rate curve based on the biological data of the subject being measured, and then extracting the predicted AOGS data and the predicted APHV data from the second growth rate curve; however, the concept of the present invention is not necessarily limited thereto. That is, the predicted APHV extraction step (S20) may be controlled to immediately extract the predicted AOGS data and the predicted APHV data through a preset algorithm without the need to generate the second growth rate curve, simply by inputting the biological data of the subject being measured into a growth age estimation model using an artificial intelligence model.

[0107] In exemplary embodiments, the step of estimating the first growth rate curve (S2) can be performed by extracting age data and height data from the time-series biometric data of the sample subjects and then applying the age data and height data to a curve transformation model that uses the age data and height data as input variables.

[0108] The growth age calculation step (S30) can be performed using the following mathematical formula 1, and

[0109] [Mathematical Formula 1]

[0110] GA i '= (CA i - APHV i ') + APHV G

[0111] Here, GA i ' is the estimated growth age of the subject being measured, and CA i is the chronological age (calendar age) of the subject being measured, and APHV i ' is the predicted APHV data including the estimated and predicted APHV values ​​of the above-mentioned measurement subject, and APHV G The second APHV data may be a representative value or average value of the first APHV data derived from the training data group.

[0112] That is, the growth age of the subject being measured (E) calculated through the growth age calculation step (S30). ga ) is the above second APHV data (S AA ) That is, with respect to the average value of the above first APHV data, the actual age (E) of the subject being measured ra ) and the predicted APHV data (E) of the above measurement subject PA It can be calculated by correcting for the difference of ).

[0113] Accordingly, the biological growth age of the subject is estimated by applying the actual age of the subject, that is, the temporal actual age calculated uniformly from the time of birth, the predicted APHV data calculated using the subject's tracking growth rate, and the second APHV data corresponding to the group average APHV of the sample subjects. Since the biological growth age can be utilized in constructing various prediction models such as an integrated height growth prediction model, a precocious puberty prediction model, and an obesity prediction model, the reliability of various prediction models such as the integrated height growth prediction model, the precocious puberty prediction model, and the obesity prediction model can be maximized.

[0114] Figure 4 is a diagram showing multiple first growth rate curves representing the growth rate of height relative to actual age using time-series biometric data of sample subjects.

[0115] FIG. 4 illustrates a plurality of first growth rate curves generated through a first growth rate curve estimation unit (30), and the first APHV data extraction unit (40) can be controlled to extract first APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate reaches a maximum value, for each of the plurality of first growth rate curves. The first APHV data may represent data indicating the point at which the growth rate reaches a maximum during pubertal rapid growth, that is, the age at which the growth acceleration is 0.

[0116] Meanwhile, the first APHV data extraction unit (40) may be controlled to extract additional data in addition to the first APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate reaches a maximum value for each of the plurality of first growth rate curves.

[0117] For example, the first APHV data extraction unit (40) can be controlled to further extract data such as Age of Onset of Growth Spurt (AOGS) data, which is the point where pubertal rapid growth begins, and Age of End of Growth Spurt (AEGS) data, which is the point where pubertal rapid growth ends.

[0118] The above AOGS data may represent the point where the growth rate slope increases rapidly and the growth acceleration reaches its maximum, and may represent data indicating an age approximately 1.5 years prior to the above first APHV data.

[0119] The above AEGS data may represent the point where the pubertal growth spurt ends, which is the point where the growth rate after APHV becomes equal to the growth rate (Onset of growth spurt velocity) at AOGS. It may also represent data indicating an age approximately 1.5 years after the first APHV data.

[0120] Figures 5 and 6 are figures showing the APHV estimation error according to the age in months for girls and boys, respectively.

[0121] Referring to FIGS. 5 and 6, predicted APHV data of subjects of various ages were extracted using a bio-data-based APHV estimation model using artificial intelligence according to exemplary embodiments of the present invention, and then the predicted APHV data and the actual APHV data of the subjects were compared.

[0122] Specifically, through the experimental results described above, it was confirmed that for girls aged 10 years or older, the average error between the predicted APHV data value and the actual APHV data value was 0.35 years, and for boys aged 12 years or older, the average error between the predicted APHV data value and the actual APHV data value was 0.39 years. In other words, as a result, it was confirmed that the bio-data-based APHV estimation model using artificial intelligence according to the exemplary embodiments of the present invention has very high APHV prediction accuracy, as the average error between the predicted data value and the actual data value at the average age at which secondary sexual characteristics appear for both girls and boys is within 0.5 years.

[0123] In addition, through the experiment described above, it was confirmed that the prediction accuracy of the APHV estimation model improves as the actual age of the subjects approaches the actual APHV data value.

[0124] Figures 7 and 8 are figures showing the importance of APHV predictors for women and men, respectively.

[0125] Referring to FIGS. 7 and 8, various variables for constructing a biodata-based APHV estimation model using artificial intelligence according to exemplary embodiments of the present invention were classified according to their importance.

[0126] Specifically, as shown in Fig. 7, the variable of highest importance in predicting APHV for women was found to be Previous Velocity, whereas as shown in Fig. 8, the variable of highest importance in predicting APHV for men was found to be Height Quantile.

[0127] Meanwhile, in predicting APHV for women, current growth rate (Velocity) was found to be the fourth most important variable, while in predicting APHV for men, previous growth rate (Previous Velocity) and current growth rate (Velocity) were found to be the second and sixth most important variables, respectively.

[0128] In other words, it was confirmed that previous velocity and current velocity act as very important variables for predicting APHV in women and men, respectively. Specifically, previous velocity was found to be the most important variable for women and the second most important variable for men. Consequently, it was confirmed that the accuracy of predicted APHV data extracted using the APHV estimation model can be significantly improved when there are two or more sets of biological data measured at different time points of the subject, allowing for the verification of the previous velocity.

[0129] In addition, it was confirmed that additional variables such as height quantile, waist-to-hip ratio (WHR), and body fat mass are required to increase the accuracy of the predicted APHV data extracted through the bio-data-based APHV estimation model using artificial intelligence according to exemplary embodiments of the present invention.

[0130] Figure 9 is a diagram showing the importance of variables in a conventional growth prediction model for women and men, and Figure 10 is a diagram showing the importance of variables in a growth prediction model of the present invention.

[0131] Referring to Figures 9 and 10, the most important variable in both the conventional growth prediction model for women and men and the growth prediction model of the present invention was found to be the growth rate, whereas in the case of the conventional growth prediction model, the second most important variable was the height quantile, and in the case of the growth prediction model of the present invention, the second most important variable was the age of peak height velocity (APHV).

[0132] In other words, in conventional growth prediction models prior to the application of the APHV concept, the Height Quantile appeared as the variable of second-highest importance after the Growth Rate; however, in the growth prediction model according to the present invention after the application of the APHV (Age of Peak Height Velocity) concept, the APHV was calculated to be more important than the Height Quantile, appearing as the variable of second-highest importance after the Growth Rate. Accordingly, it was confirmed that the importance of APHV is very high among the variables affecting the predicted final height.

[0133] As described above, the bio-data-based APHV estimation model using artificial intelligence according to exemplary embodiments of the present invention estimates a growth rate curve by applying a statistical methodology to tracking growth rates extracted from past information, and extracts APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate reaches a maximum value, from the growth rate curve. Since the APHV data is utilized in the construction of various prediction models, such as an integrated height growth prediction model, a precocious puberty prediction model, and an obesity prediction model, the reliability of various prediction models, such as an integrated height growth prediction model, a precocious puberty prediction model, and an obesity prediction model, can be maximized.

[0134] In addition, according to various embodiments of the present invention, an APHV estimation model constructed through artificial intelligence learning can accurately predict height by considering the growth stages of children and adolescents, and solutions necessary for height growth can be provided by considering each growth stage.

[0135] However, the concept of the present invention is not necessarily limited thereto, and the apparatus / method / system according to the exemplary embodiments of the present invention may be applied to various product / technology fields in addition to the aforementioned product / technology fields.

[0136] Although various embodiments of the present invention have been described in detail above, those skilled in the art will understand that various modifications can be made to the above-described embodiments without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the claims set forth below as well as equivalents thereof.

Claims

Claim 1 A method for estimating growth age using an artificial intelligence model, wherein the growth age estimation method comprises: a step of receiving biological data of a subject to measurement; a step of extracting predicted APHV (Age of Peak Height Velocity) data, which is the age at which the growth rate is expected to reach a maximum value, using the biological data of the subject to measurement; and a step of calculating the growth age of the subject to measurement based on the predicted APHV and a second APHV data, which is the average value of the first APHV data; wherein the first APHV data is a set of data from which the age at which the growth rate reaches a maximum value is extracted for a first growth rate curve representing the growth rate of height relative to the actual age of each of a plurality of sample subjects, and wherein the artificial intelligence model is trained using the first APHV data as target data for each of the first growth rate curves estimated from the biological data of the plurality of sample subjects. Claim 2 In claim 1, the growth age calculation step is performed using Mathematical Formula 1, and [Mathical Formula 1] GA i '= (CA i - APHV i ') + APHV G Here, GA i ' is the estimated growth age of the subject being measured, and CA i is the chronological age (calendar age) of the subject being measured, and APHV i ' is the predicted APHV data including the estimated and predicted APHV values ​​of the above-mentioned measurement subject, and APHV G A method for estimating growth age using an artificial intelligence model, characterized in that the second APHV data is a representative value or average value of the first APHV data derived from a training data group. Claim 3 A method for estimating growth age using an artificial intelligence model according to claim 1, wherein the predicted APHV extraction step comprises: a step of estimating the expected shape of the growth rate using the biological data of the subject being measured; and a step of extracting a predicted AOGS (Age of Onset of Growth Spurt), which is the age at which the growth rate of the subject being measured begins to increase, and a predicted APHV (Age of Peak Height Velocity), which is the age at which the growth rate of the subject being measured has a maximum value. Claim 4 A method for estimating growth age using an artificial intelligence model according to claim 1, characterized in that the biometric data of the subject being measured includes at least two time-series biometric data measured at different age groups. Claim 5 delete Claim 6 A growth age estimation model using an artificial intelligence model, comprising: an input unit for receiving biometric data of a subject to measurement; a predicted APHV (Age of Peak Height Velocity) data extraction unit that extracts predicted APHV data, which is the age at which the growth rate is expected to reach a maximum value from the biometric data of the subject to measurement using an artificial intelligence model; and a growth age calculation unit that calculates the growth age of the subject to measurement based on a second APHV data, which is the average value of the predicted APHV and the first APHV (Age of Peak Height Velocity) data; wherein the first APHV data is a set of data from which the age at which the growth rate reaches a maximum value is extracted for a first growth rate curve representing the growth rate of height relative to the actual age of each of a plurality of sample subjects, and the artificial intelligence model is characterized by being trained using the first APHV data as target data in the growth rate curve estimated from the biometric data of the plurality of sample subjects. Claim 7 In Clause 6, the growth age calculation unit is performed using Mathematical Formula 1, and [Mathical Formula 1] GA i '= (CA i - APHV i ') + APHV G Here, GA i ' is the estimated growth age of the subject being measured, and CA i is the chronological age (calendar age) of the subject being measured, and APHV i ' is the predicted APHV data including the estimated and predicted APHV values ​​of the above-mentioned measurement subject, and APHV G A growth age estimation model using an artificial intelligence model characterized by the second APHV data being a representative value or average value of the first APHV data derived from a training data group. Claim 8 A growth age estimation model using an artificial intelligence model, characterized in that, in claim 6, the biometric data of the subject being measured includes at least two time-series biometric data measured at different age groups. Claim 9 delete Claim 10 A growth age estimation model using an artificial intelligence model, characterized in that, in claim 6, the first growth rate curve is estimated by extracting age data and height data from the time-series biometric data of the sample subjects and applying them to a curve transformation model that uses the age data and height data as input variables.

Images