Method for preprocessing biological component measurement data for growth prediction

Abstract

The present invention relates to a method, an apparatus, and a computer program for preprocessing input data for growth prediction of children or adolescents. The method for preprocessing biological component measurement data for growth prediction according to an exemplary embodiment of the present invention may comprise the steps of: receiving body data of a subject; generating a first variable, on the basis of the body data of the subject; and determining an error in the body data by comparing the first variable with a preset value.

Description

Preprocessing method of biocomponent measurement data for growth prediction

[0001] The present invention relates to a method, device and computer program for preprocessing input data for predicting the growth of children or adolescents.

[0002] With the recent development of artificial intelligence technology, it is being applied to various fields, and methods for extracting features inherent in data through neural network models and generating additional information are being developed and used as an alternative to conventional data processing methods.

[0003] Recently, artificial intelligence technology has been applied beyond simply tracking and detecting objects, to learning past history and predicting the future, or to deriving current features that reflect time-series changes.

[0004] Furthermore, in AI, which learns from data, data quality is known to be a key factor in AI performance. For example, leading companies with the most advanced AI-based autonomous driving technologies emphasize the importance of securing high-quality data.

[0005] Among these, predictive analytics is a statistical and data mining technique that extracts information from data and uses it to predict trends and behavioral patterns. This predictive analytics can be applied to any area requiring decision-making based on information gleaned from data. The core of predictive analytics is understanding the relationships between variables and then predicting unknown variables.

[0006] For this purpose, various approaches are being used depending on the data characteristics and prediction target.

[0007] Among the various fields requiring predictive analysis is the physical growth of children and adolescents. Parents and adolescents alike are deeply interested in the timing and extent of height growth.

[0008] Conventional predictions of height growth suggest methods for predicting growth plates by X-ray or analyzing the relationship with genetic / environmental factors (Patent Publication No. 10-2075743, Patent Publication No. 10-1866208), or for converting the body data of sample subjects with different measurement periods or measurement counts into a form suitable for learning a growth prediction model (Patent Publication No. 10-2198302).

[0009] Since children and adolescents have different growth stages with different characteristics, taking this into account can increase the reliability of solutions provided through predicted data and analysis.

[0010] In particular, in the case of children and adolescents, since this includes a period of rapid physical changes according to the growth stage, it is necessary to first determine the errors in the input (or measurement) data (biocomponent data) to secure high-quality data to improve the accuracy of growth prediction and the performance of artificial intelligence learning.

[0011] One of the various tasks of the present invention is to provide a method, device and computer program for preprocessing input data for predicting the growth of children or adolescents and providing customized solutions for each stage of growth.

[0012] Various embodiments for solving the problem of the present invention can provide a method for preprocessing biocomponent measurement data for growth prediction, which is performed by a computing device, including a step of receiving body data of a subject, a step of generating a first variable based on the body data of the subject, and a step of comparing the first variable with a preset value to determine an error in the body data.

[0013] The step of inputting the subject's physical data may include a step of inputting the subject's biocomponent data and a step of inputting the subject's identification data.

[0014] It may be characterized by further including a step of generating first data by connecting the subject's biocomponent data and the subject's identification data after receiving the subject's physical data.

[0015] After generating the first data, the method may further include a step of deleting the biocomponent data if a preset value is measured from the biocomponent data.

[0016] The above first variable may be characterized by being generated based on the subject's biocomponent data.

[0017] The above first variable may be characterized by being generated as the sum of soft lean mass, body fat mass, and osseous mineral among the subject's biocomponent data.

[0018] A method for preprocessing biocomponent measurement data for growth prediction, characterized in that the step of determining the error comprises a step of comparing the first variable with a value of any one of the biocomponent data.

[0019] The step of determining the above error may further include a step of determining whether the age of the subject among the identification data corresponds to a preset standard before comparing the first variable with a value of any one of the biocomponent data.

[0020] It may be characterized by further including a step of deleting the biocomponent data when the result of comparing the value of the first variable and any one of the biocomponent data falls within a preset range.

[0021] It may be characterized by further including a step of generating second data when the result of comparing the value of either the first variable or the biocomponent data does not fall within a preset range.

[0022] An exemplary embodiment of the present invention may provide a program stored on a computer-readable recording medium including a program code for executing the method for preprocessing biocomponent measurement data for growth prediction described above.

[0023] An exemplary embodiment of the present invention can provide a computer-readable recording medium having recorded thereon a program for executing the above-described preprocessing method of biocomponent measurement data.

[0024] An exemplary embodiment of the present invention provides a preprocessing device for biocomponent measurement data for growth prediction, comprising an input unit for receiving body data of a subject, a variable generation unit for generating a first variable based on the body data of the subject, and an error determination unit for determining an error in the body data by comparing the first variable with a preset value.

[0025] The above input unit may be characterized by including a first input unit for receiving biocomponent data of the subject and a second input unit for receiving identification data of the subject.

[0026] It may be characterized by further including a connecting portion that connects data input through the first input portion and the second input portion.

[0027] Each feature of the above-described embodiments may be implemented in combination in other embodiments as long as it is not inconsistent with or exclusive of other embodiments.

[0028] According to various embodiments of the present invention, when predicting the growth of a child or adolescent, the accuracy of growth prediction can be improved by determining errors in bio-component data measured for predicting the growth of the child or adolescent.

[0029] According to various embodiments of the present invention, when providing a solution necessary for the growth of a child or adolescent, the accuracy of growth prediction can be improved by eliminating errors in input data.

[0030] According to various embodiments of the present invention, the accuracy of growth prediction of children or adolescents can be improved by determining errors in measured biocomponent data for growth prediction of children or adolescents and generating refined input data.

[0031] According to various embodiments of the present invention, the accuracy of solutions required for children or adolescents can be improved by determining errors in measured bio-component data to generate refined input data for generating solutions required for the growth of children or adolescents.

[0032] The effects of the present invention are not limited to those described above, and other effects not mentioned will be clearly recognized by those skilled in the art from the description below.

[0033] FIG. 1 is a diagram showing the structure of a growth prediction system according to an exemplary embodiment of the present invention.

[0034] FIG. 2 is a diagram showing the configuration of a growth prediction and solution generation device according to an exemplary embodiment of the present invention.

[0035] Figure 3 is a diagram showing the data preprocessing unit of Figure 2.

[0036] FIG. 4 is a diagram showing predicted height and target height for each growth stage according to an exemplary embodiment of the present invention.

[0037] FIG. 5 is a diagram showing growth in obesity and normal weight at each growth stage of a male according to an exemplary embodiment of the present invention.

[0038] Figure 6 is a diagram showing the growth of a girl in obesity and normal weight at each growth stage according to an exemplary embodiment of the present invention.

[0039] FIG. 7 is a diagram illustrating the configuration of a neural network that performs growth prediction and solution provision according to an exemplary embodiment of the present invention.

[0040] FIG. 8 is a diagram illustrating a first neural network model according to an exemplary embodiment of the present invention.

[0041] FIG. 9 is a diagram illustrating a second neural network model according to an exemplary embodiment of the present invention.

[0042] FIGS. 10 to 12 are diagrams illustrating a preprocessing method of biocomponent measurement data for growth prediction according to an exemplary embodiment of the present invention.

[0043] Figure 13 is a diagram showing biocomponent data of a subject according to an exemplary embodiment of the present invention.

[0044] FIG. 14 is a diagram showing data linking a subject's biocomponent data and identification data according to an exemplary embodiment of the present invention.

[0045] Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. The following detailed description is provided to facilitate a comprehensive understanding of the methods, devices, and / or systems described herein. However, these are merely examples and the present invention is not limited thereto.

[0046] In describing embodiments of the present invention, detailed descriptions of known technologies related to the present invention will be omitted if they are deemed to unnecessarily obscure the gist of the invention. Furthermore, the terms described below are defined based on their functions in the present invention and may vary depending on the intent or custom of the user or operator. Therefore, their definitions should be based on the overall content of this specification.

[0047] The terminology used in this detailed description is for the purpose of describing embodiments of the present invention only and should not be construed as limiting. Unless expressly stated otherwise, singular forms include plural forms.

[0048] In this description, expressions such as "including" or "having" are intended to indicate certain features, numbers, steps, operations, elements, portions or combinations thereof, and should not be construed to exclude the presence or possibility of one or more other features, numbers, steps, operations, elements, portions or combinations thereof other than those described.

[0049] Additionally, in describing components of embodiments of the present invention, terms such as first, second, A, B, (a), (b), etc. may be used. These terms are only intended to distinguish the components from other components, and the nature, order, or sequence of the components are not limited by the terms.

[0050]

[0051] FIG. 1 is a diagram showing the structure of a growth prediction system according to an exemplary embodiment of the present invention.

[0052] Referring to Fig. 1, the growth prediction system of the present embodiment can input the time-series physical data (1) of a subject and then distinguish the gender (2) of the subject based on the input physical data (1).

[0053] And, based on the distinguished gender, in the case of a boy, growth prediction (5-1) can be performed through a growth prediction model (3-1), and in the case of a girl, growth prediction (5-2) can be performed through a growth prediction model (3-2) different from the growth prediction model (3-1) used in the case of a boy.

[0054] Boys and girls may experience different growth rates at different stages of development. For example, as described below, boys and girls may enter rapid growth stages at different times. Therefore, the rate of growth and the resulting solutions that take this growth rate into account may differ. Therefore, it is desirable to perform growth predictions using models trained with different training data based on the subject's gender.

[0055]

[0056] FIG. 2 is a diagram showing the configuration of a growth prediction and solution generation device according to an exemplary embodiment of the present invention, FIG. 3 is a diagram showing a data preprocessing unit of FIG. 2, FIG. 13 is a diagram showing biocomponent data of a subject according to an exemplary embodiment of the present invention, and FIG. 14 is a diagram showing data linking biocomponent data and identification data of a subject according to an exemplary embodiment of the present invention.

[0057] The following description is provided with reference to FIGS. 2, 3, 13, and 14.

[0058] A growth prediction and solution generation device according to an exemplary embodiment of the present invention may include a data preprocessing unit (100), an input unit (10), a gender determination unit (20), a growth stage determination unit (30), a prediction unit (50), a solution generation unit (70), and a display unit (90).

[0059] Through the input unit (10), the growth prediction and solution generation device can receive time-series physical data on the subject.

[0060] The subject's physical data may include data that can identify the subject (identification data) and the subject's biocomponent data.

[0061] For example, the above identification data may include data for identifying a subject, such as grade (or age), gender, and height, and the above biocomponent data may include data such as weight, protein, mineral content, body fat, body water, muscle mass, body fat 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.

[0062] More specifically, referring to FIGS. 13 and 14, the subject's identification data (1201) may include date of birth (1201-1), gender (1201-2), age (1201-3), and measurement date (1201-4).

[0063] In addition, the subject's biological composition data (1101) may include data number (1101-1), height (1101-2), weight (1101-3), protein mass (1101-4), mineral mass (1101-4), body fat mass (1101-6, Body Fat Mass), muscle mass (1101-7, Soft Lean Mass), bone mineral mass (1101-8, Osseous mineral), and skeletal muscle mass (1101-9).

[0064] This body data is only an example to help understand the present invention, and the present embodiment is not limited thereto, and it is obvious that the types of data constituting the body data may be varied depending on the embodiment.

[0065] Meanwhile, an exemplary embodiment of the present invention may include a data preprocessing unit (100) that determines errors in the subject's body data and generates refined data.

[0066] Since biological component data (1101) may undergo rapid changes depending on the physical characteristics of children or adolescents depending on the growth stage, errors (errors during the measurement process) may occur in other biological component data measurement processes such as the measurement method, measurement time, and measurement environment.

[0067] Additionally, errors (data transmission errors) may occur during the process of transmitting or storing measured biocomponent data.

[0068] Accordingly, the data preprocessing unit (100) of the present embodiment generates a virtual variable (hereinafter referred to as the first variable) based on the measured biocomponent data (1101), and then compares the first variable with a value of any one of the biocomponent data to determine an error in the measurement process and a data transmission error of the measured biocomponent data (1101), thereby generating refined data.

[0069] This is explained in more detail with reference to FIGS. 10 to 12.

[0070] Meanwhile, the data input into the input unit (10) may include the subject's time series body data.

[0071] The above time series body data may be continuous data or discontinuous data, but may be data included in at least one period corresponding to the growth stage of a child or adolescent.

[0072] More specifically, the timing and frequency of collection of the subject's time-series body data may vary.

[0073] For example, the first subject may have measured body data from the age of 8 to 12, which is part of the childhood or adolescent period, and the second subject may have irregularly measured body data, such as from the age of 8, 10 to 12, and 15.

[0074] Additionally, the third subject may have body data measured multiple times over a period of time (a period corresponding to one of the growth stages of a child or adolescent), while the fourth subject may have body data measured only once over a period of time (a period corresponding to one of the growth stages of a child or adolescent).

[0075] As described above, depending on the collection period and number of collections, the subject's physical data may be included in more than one of the growth stages of a child or adolescent (subject 1, subject 2), but may not be included (subject 3, subject 4).

[0076] In the case where there is body data measured multiple times during one of multiple growth stages, such as the third subject, the growth stage determination unit (30) can classify the growth stage including the body data of the third subject through the growth stage classification unit (31) and extract the body data through the body data extraction unit (33). The body data extracted here can generally include all of the body data of the third subject input through the input unit (10).

[0077] However, in cases where the subject's body data, such as that of the fourth subject, corresponds to only one of multiple growth stages and there is body data measured only once during that period, additional body data corresponding to an arbitrary period may be generated before entering the body data of the fourth subject into the growth stage determination unit (30).

[0078] More specifically, the time-series body data of the subject corresponding to an arbitrary period can be generated based on the time-series body growth data of a plurality of previously stored sample subjects based on the body data of the input subject.

[0079] For example, body data can be generated based on a distribution model (similarity) between the body data of the input subject and the time-series body growth data of multiple previously stored sample subjects, or body data can be generated based on a Bayesian inference model (conditional probability).

[0080] The growth stage determination unit (30) can classify the subject's body data input through the input unit (10) into one of a plurality of growth stages through the growth stage classification unit (31) and extract body data corresponding to the classified growth stage through the body data extraction unit (33).

[0081]

[0082] FIG. 4 is a diagram showing predicted height and target height for each growth stage according to an exemplary embodiment of the present invention.

[0083] In Figure 4, the x-axis represents age (months), and the y-axis represents height (cm). The lower dotted line (P) represents the predicted growth rate of the subject, and the upper solid line (G) represents the target growth rate of the subject.

[0084] Referring to Figure 4, the growth stages of children or adolescents described above may include a normal growth period (301), a rapid growth period (303), a decelerated growth period (305), and a growth plate period (307).

[0085] Each growth stage can be distinguished by its growth rate, and the height that grows each year varies depending on the growth stage, and even in the same growth stage, the actual height that grows can vary depending on the growth type.

[0086] The normal growth period (301) usually refers to the period before puberty when secondary sexual characteristics appear, and children or adolescents in this period generally have open growth plates, so they generally grow 4 to 5 cm per year in the case of a short growth type and 6 to 7 cm per year in the case of a tall growth type, depending on the growth environment.

[0087] The rapid growth period (303) is the period when secondary sexual characteristics begin to appear. In women, the breasts swell and pubic hair develops, and in men, the testicles enlarge and pubic hair begins to grow, and the voice changes to a voice stage. The rapid growth period (303) generally lasts about 2 to 3 years after the normal growth period (301), and on average, growth occurs in the range of about 7 to 10 cm per year.

[0088] The deceleration period (305) refers to the period when secondary sexual characteristics are fully developed. During this period, women can be distinguished by the onset of menstruation, and men can clearly see the changes in pubic hair, voice change, and armpit hair. When the deceleration period (305) begins, the growth rate rapidly decreases compared to the rapid growth period (303), and it generally lasts for about 2 to 3 years, and natural growth stops after growing at an average of 5 to 6 cm per year. The growth plate begins to close little by little after the rapid growth period (304), and about 50% of it is closed about 6 months after entering the deceleration period (305).

[0089] The growth plate period (307) is a period when the growth period is not completely over, but natural height growth becomes difficult and the growth plate is closed. Generally, women enter the growth plate period (307) about 1 year and 6 months to 2 years after menarche, and men enter the growth plate period (307) about 1 year and 6 months to 2 years after the time when hair begins to grow in the armpits. During the growth plate period (307), the growth plate closes and natural growth stops, but it is possible to grow in the range of about 1 to 3 cm by changing bad lifestyle habits and improving physical functions through customized exercise, posture correction, and nutrient intake.

[0090] Meanwhile, the prediction unit (50) can be implemented as an artificial intelligence with a recursive neural network (RNN) structure as a prediction model so that it can utilize not only the current value but also time-series values. For example, the prediction model can be implemented with an architecture such as a recurrent neural network such as LSTM (Long Short Term Memory) or GRU (Gated Recurrent Units). Of course, various conventional artificial intelligence architectures can also be applied to the prediction model of the present embodiment, and will be described in detail with reference to FIGS. 7 to 9 described below.

[0091] The solution generation unit (70) can generate a growth management solution based on the subject's body data corresponding to the classified growth stage.

[0092] More specifically, when the subject is in the general growth stage (301), a solution for increasing the subject's growth prediction value can be provided. The growth prediction value is a value corresponding to the y-axis in FIG. 4, and the solution for increasing the growth prediction value can be provided to the subject through various solution display units (90) for increasing the target value of the predicted y-axis.

[0093] Meanwhile, examples of solutions provided through the display unit (90) may include current height, predicted height, obesity level, body fat mass, skeletal muscle mass, protein mass, mineral mass, sleep amount, exercise amount, nutritional information, lifestyle habits, posture, etc. Each indicator may be expressed as caution, average, good, etc. based on a preset range, or may be expressed as a level.

[0094] And for each indicator, the current status, customized solutions, and precautions can be displayed. The current status can be displayed in stages or levels based on the target value, and in the case of customized solutions, it can include information on adjusting protein, total minerals, body fat, body water, soft lean mass, fat-free mass, bone tissue, skeletal muscle mass, body mass index (BMI), basal metabolic rate, etc. to reach the current target value based on the entered body data.

[0095] In the case of precautions, it may include information on adjusting the currently deficient protein, total minerals, body fat, body water, soft lean mass, fat free mass, bone tissue, skeletal muscle mass, body mass index (BMI), basal metabolic rate, etc. based on the entered body data.

[0096] In addition, when the subject is in the rapid growth stage (303), a solution for increasing the growth prediction value of the subject's rapid growth stage (303) can be provided. The growth prediction value is a value corresponding to the y-axis in FIG. 2, and the solution for increasing the growth prediction value can provide the subject with various solutions for increasing the target value of the predicted y-axis through the display unit (90).

[0097]

[0098] FIG. 5 is a diagram showing growth in obesity and normal weight at each growth stage of a boy according to an exemplary embodiment of the present invention, and FIG. 6 is a diagram showing growth in obesity and normal weight at each growth stage of a girl according to an exemplary embodiment of the present invention.

[0099] The x-axis of Figure 5 represents age, and the y-axis represents growth rate (cm). The solid line (G1) represents the growth rate by growth stage for a normal-weight male, and the dotted line (P1) represents the growth rate by growth stage for an obese male.

[0100] The x-axis of Figure 6 represents age, and the y-axis represents growth rate (cm). The solid line (G2) represents the growth rate by growth stage for girls of normal weight, and the dotted line (P2) represents the growth rate by growth stage for girls of obesity.

[0101] The following description is provided with reference to Figures 5 and 6.

[0102] Obesity is not simply a simple increase in body weight. It is a disease characterized by excessive accumulation of fat tissue, often accompanied by metabolic disorders. Childhood or adolescent obesity is medically defined as a weight exceeding 20% ​​of the recommended weight for height, typically between infancy and adolescence.

[0103] In most cases, childhood obesity disappears after the first birthday as children become more active and mobile. However, some children remain obese, and in many cases, the condition returns to normal only to relapse during school age.

[0104] Between 75% and 80% of childhood and adolescent obesity progresses to adult obesity. Furthermore, it inhibits growth hormone secretion, which, especially in girls, accelerates puberty and shortens the growth period, hindering growth and causing precocious puberty.

[0105] Therefore, for school-age children and adolescents who are prone to obesity, it is necessary to provide a systematic solution to predict obesity and prevent it if it is predicted.

[0106] Meanwhile, childhood or adolescent obesity can be divided into simple obesity with no clear cause and symptomatic obesity caused by a specific cause disease, and more than 99% of childhood obesity is simple obesity.

[0107] Children with simple obesity are generally characterized by being average or slightly taller than their peers of the same age (multiple sample subjects) during the normal growth period (301) for both boys and girls, but after the rapid growth period (303), they tend to be shorter or have a lower growth rate than their peers of the same age (multiple sample subjects).

[0108] In summary, childhood or adolescent obesity can be understood as a group of diseases that are accompanied by overweight or metabolic disorders caused by a wide variety of causes, and referring to Figures 5 and 6, in general, both boys and girls show an average or slightly higher growth rate than multiple sample subjects of the same age during the normal growth period (301), but after the rapid growth period (303), they show a tendency for a lower growth rate than multiple sample subjects of the same age.

[0109] Therefore, in order to more accurately predict obesity and generate a solution based on growth stage, the present embodiment classifies the gender of the subject through the gender determination unit (20) based on the physical data of the subject input through the input unit (10), and then classifies the growth stage based on the classified gender in the growth stage determination unit (30), extracts the physical data, and then generates a solution in the solution generation unit (70) by considering the gender and growth stage of the subject.

[0110] More specifically, in cases where both boys and girls are commonly obese, it can be confirmed that the growth prediction value in the rapid growth stage (303) is lower than in the normal case. Therefore, if the subject is in the rapid growth stage (303), a solution for increasing the growth prediction value can be provided, and if the subject is in the normal growth stage (301), the content can include information on adjusting indicators that can alleviate cases such as cases where sex hormones increase abnormally, including the physical data considered.

[0111] In addition, when the subject is in the decelerated growth phase (305), a solution for adjusting the duration of the subject's decelerated growth phase (305) can be provided. The growth phase duration adjustment can be divided into a case where the subject's body data is located at the beginning of the decelerated growth phase (305) and a case where the subject's body data is located in the middle or latter part of the decelerated growth phase (305) among the growth stages classified based on the input body data of the subject.

[0112] The criteria for distinguishing between the beginning and the middle and late stages of the aforementioned deceleration growth stage (305) may be distinguished based on a predetermined range corresponding to the rapid growth stage (303) to the deceleration growth stage (305) based on the x-axis in FIGS. 4 to 6, or based on the body data of the input subject, if the secondary sexual characteristics are not completed, it may be distinguished as the beginning of the deceleration growth stage (305), and if completed, it may be distinguished as the middle and late stages of the deceleration growth stage (305).

[0113] Preferably, it is possible to determine whether the secondary sexual characteristics have been completed based on the input body data of the subject, and thus determine whether the body data of the current subject is in the beginning or the middle or latter part of the decelerated growth phase (305). In a case where it is not possible to determine whether the secondary sexual characteristics have been completed based on the input body data of the subject, it is possible to determine whether the body data of the subject is in the beginning or the middle or latter part of the decelerated growth phase (305) based on a predetermined range corresponding to the decelerated growth phase (305) from the rapid growth phase (303).

[0114] Meanwhile, if the body information of the input subject is located at the beginning of the deceleration growth phase (305), a period adjustment solution can be provided to delay entry into the deceleration growth phase (305).

[0115] As described above, since secondary sexual characteristics are being completed at the point of transition from the rapid growth phase (303) to the decelerated growth phase (305), a solution for delaying the point of completion of secondary growth can be provided, and various solutions for moving the range of the x-axis corresponding to the decelerated growth phase (305) to the right in FIGS. 4 to 6 can be provided to the subject through the display unit (90). In this case, the range of the decelerated growth phase (305) may increase depending on the subject's physical data, or may decrease as the rapid growth phase (303) increases.

[0116] And, if the body data of the input subject is located in the middle or latter part of the deceleration growth phase (305), a period adjustment solution for increasing the period of the deceleration growth phase (305) can be provided.

[0117] As described above, the decelerated growth phase (305) refers to the time when the subject's growth plate closes. Generally, after entering the decelerated growth phase (305), about 50% of the growth plate closes after 6 months. When the growth plate closes and natural growth stops, the no-growth phase (307) is entered. Therefore, in this case, a solution for increasing the period of the decelerated growth phase (305) can be provided. In other words, various solutions for expanding the range of the x-axis corresponding to the decelerated growth phase (305) in FIGS. 4 to 6 can be provided to the subject through the display unit (90).

[0118] If the subject described above is in the general growth stage (301), the physical data to be considered may include, in particular, content on the adjustment of indicators that can alleviate the degree of closure of the growth plate.

[0119] In addition, if the subject is in the growth plate stage (307), a solution for improving physical function through lifestyle habits, customized exercise, posture correction, nutrient intake, etc. can be provided based on the subject's entered physical data.

[0120] In the case of the growth plate stage (307), the growth plate closes and natural growth stops, so a solution for improving physical function through lifestyle habits, customized exercise, and posture correction can be provided based on the subject's weight, body fat, body water, muscle mass, skeletal muscle mass, body mass index (BMI), basal metabolic rate, neck circumference, chest circumference, abdominal circumference, thigh circumference, arm circumference, and hip circumference, or a solution for improving physical function through nutrient intake, etc. can be provided based on protein, mineral content, bone tissue (bone density), etc.

[0121] Meanwhile, in order to provide a more accurate obesity prediction and solution in this embodiment, the gender of the subject is classified through the gender determination unit (20), and the growth stage classification unit (31) can set the timing of the rapid growth stage differently based on the classified gender.

[0122] This is because, as described above, the timing of entering the rapid growth phase may differ for boys and girls. Furthermore, the subject's body data corresponding to the rapid growth phase, taking into account gender and output from the body data extraction unit (33), is input into the prediction unit (50) to output a predicted value for obesity.

[0123] If the subject is predicted to be obese and is male, a solution for increasing the predicted growth value of the subject in the rapid growth phase (303) can be provided as described above.

[0124] And if the subject is predicted to be obese and is female, a solution for increasing the duration of the rapid growth phase (303) can be provided.

[0125] In more detail, referring to FIGS. 5 and 6, the difference in predicted growth values ​​in the rapid growth stage (303) is smaller for women compared to men, but the decrease in the degree of height growth (growth rate) with age is large as they move from the rapid growth stage (303) to the decelerated growth stage (305).

[0126] For example, in Fig. 5, it can be seen that the difference in growth prediction values ​​in the rapid growth phase (303) for a normal-weight male is significantly different between the growth prediction values ​​in the normal-weight case (G1) and the obese case (P1). Therefore, for obese males, it is necessary to provide a solution for increasing the growth prediction value in the rapid growth phase (303).

[0127] Also, as an example, in Figure 6, a normal weight girl grew 7 cm in height as she grew from age 9 to 10, and grew about 6.5 cm in height as she grew from age 10 to 11.

[0128] In contrast, obese females grew approximately 6.7 cm in height from age 9 to 10, and approximately 5.7 cm from age 10 to 11. Furthermore, as shown in Figure 5, the difference in predicted growth values ​​is relatively small compared to that of males.

[0129] Therefore, for women who are obese, it is necessary to provide a solution to increase the period of the rapid growth phase (303) and reduce the decrease in growth rate that occurs when moving from the rapid growth phase to the decelerated growth phase.

[0130]

[0131] FIG. 7 is a diagram showing the configuration of a neural network that performs growth prediction and solution provision according to an exemplary embodiment of the present invention, FIG. 8 is a diagram showing a first neural network model according to an exemplary embodiment of the present invention, and FIG. 9 is a diagram showing a second neural network model according to an exemplary embodiment of the present invention.

[0132] The following description is provided with reference to Figures 7 to 9.

[0133] An exemplary embodiment of the present invention may include a first model (50) and a second model (13), and may construct a pipeline in which at least a portion of the output of the second model (13) is input to the first model (50).

[0134] More specifically, the first model (50) is a model that learns body data corresponding to at least one growth stage among a plurality of growth stages as learning data based on time-series body data for a plurality of sample subjects.

[0135] The above growth stage may be the rapid growth stage (303), which is the point at which growth retardation due to obesity occurs, but, as described above, it is of course possible to adopt one or more of the multiple growth stages to more accurately predict obesity.

[0136] The first model (50) is equipped with LSTM neural networks (50-1, 50-2, 50-3, 50-4) for learning time-series data, and trains the LSTM neural networks (50-1, 50-2, 50-3, 50-4) using past physical information of multiple sample subjects. Then, by inputting the current subject's physical data into the trained LSTM neural networks (50-1, 50-2, 50-3, 50-4), the predicted growth rate and a solution considering the growth rate can be output.

[0137] The above subject's body data may refer to refined data that has gone through an error judgment process through a data preprocessing unit (100), which will be described in more detail with reference to FIGS. 10 and 11 described below.

[0138] LSTM neural networks (50-1, 50-2, 50-3, 50-4) are trained using at least one of the body data of multiple sample subjects as a baseline. For example, for height, they are trained using annual height data from a random period or during a specific growth stage, and then predicted for the next year and compared with the actual data. This comparison allows the training set to learn by moving forward in time by a random period or by a specific growth stage.

[0139] Additionally, LSTM neural networks (50-1, 50-2, 50-3, 50-4) can be trained for each growth stage. Accordingly, they can be trained with past body data of the corresponding growth stages, such as the normal growth stage (301), rapid growth stage (303), decelerating growth stage (305), and growth plate stage (307).

[0140] The subject's body data for the above learning may mean refined data that has gone through an error judgment process through a data preprocessing unit (100), which will be described in more detail with reference to FIGS. 10 and 11 described below.

[0141] Meanwhile, as an example, in this embodiment, time-series physical data of multiple sample subjects can be sequentially input as learning data according to age or an arbitrary period, and the result of calculating the predicted value at a past point in time or growth rate can be transferred to predict the growth rate at the next age or an arbitrary period.

[0142] Therefore, the LSTM neural network (50-1, 50-2, 50-3, 50-4) can learn not only the growth prediction based on the current body data, but also the extent to which the prediction results for various indicators from the past (50-1, 50-2, 50-3, 50-4) affect the current growth prediction, and through this, it can extract items that have a large influence on the change in growth according to age or an arbitrary period among the indicators and reflect them in the growth prediction.

[0143] Furthermore, for time-series learning, it is necessary to secure regular, time-series physical data for multiple sample subjects. However, as described above, it can be difficult to regularly obtain physical data for multiple sample subjects, such as by age or over a given period. Therefore, outliers or discontinuous physical data can be removed for each unit period, allowing for temporal normalization.

[0144] Meanwhile, the second model (13) can derive bone maturity (age) from a bone image using a convolutional neural network trained with the subject's bone maturity data as learning data.

[0145] More specifically, the convolutional neural network includes multiple convolution layers that create a features map for features in the target image of the analysis among the bone images, and a pooling layer that performs sub-sampling between the multiple convolution layers, so that features at different levels for the target area of ​​analysis can be extracted, and the features can be probabilistically inferred through an activation function, or the bone maturity can be derived through weight learning between nodes through regression analysis.

[0146] The bone maturity extracted through the second model (13) can be input into the LSTM neural network (50-1, 50-2, 50-3, 50-4) together with at least some of the subject's body data to increase the accuracy of predicting the subject's growth, thereby further increasing the accuracy of predicting obesity.

[0147]

[0148] FIGS. 10 to 12 are diagrams illustrating a preprocessing method of biocomponent measurement data for growth prediction according to an exemplary embodiment of the present invention.

[0149] As described above, when predicting the growth of a subject (child or adolescent) or providing a solution through growth prediction, the entered physical data may have errors in the measurement process or data transmission.

[0150] Therefore, in order to provide a more accurate growth prediction of the subject or a solution through growth prediction, it is desirable to input refined data that has gone through the process of detecting and judging the above-described error data into the input unit (10).

[0151] In addition, the body data of multiple sample subjects for training the prediction model of the prediction unit (50) described above may also be refined data that has gone through the process of detecting and judging the error data described above.

[0152] The process of detecting and judging errors in body data is described below with reference to FIGS. 3, 10 to 14.

[0153] An exemplary embodiment of the present invention may receive body data (S10) to detect or determine errors in a subject's body data, particularly biocomponent data, and then generate a virtual first variable (S30) based on the input body data. Then, the generated virtual first variable may be compared (S50) with the body data input in step (S10) to determine errors in the input body data (S70).

[0154] More specifically, the subject's body data may include identification data (1201) and biocomponent data (1101) as shown in FIG. 14, and the preprocessing unit (100) may input (S111) the subject's biocomponent data (1101) through the biocomponent data input unit (110), and may input (S112) the subject's identification data (1201) through the identification data input unit (120). In addition, the input biocomponent data (1101) and identification data (1201) may be generated as concatenated data (140d) through the connection unit (140).

[0155] If it is determined that an error has been detected in the input bio-component data (1101) through the process of detecting an error in the bio-component data (S131) ​​through the error detection unit (130) of the connected data (140d) (S131: Yes), at least a portion of the bio-component data constituting the connected data (140d) may be deleted (S132).

[0156] For example, if at least one of the multiple items (1101-1 to 1101-9) constituting the input bio-component data (1101) includes a 'Nan' value, the data preprocessing unit (100) of the present embodiment can delete the bio-component data (1101) through the data deletion unit (180). In this case, all items measured on the measurement date including the 'Nan' value (rows including the measurement date) based on the measurement date (1201-4) of the identification data can be deleted.

[0157] Alternatively, if the age of the identification data (1201-3) is 30 months or less, all items measured on the measurement date that include the 'Nan' value (rows including the measurement date) may be deleted. In more detail, if the age is 30 months or less, it is a period of rapid growth in infants and toddlers, and many errors may be included in the measurement, so the reliability of the biocomponent data is relatively lower than that of biocomponent data with a higher age. Therefore, if the biocomponent data includes the 'Nan' value as described above when the age is 30 months or less, all items measured on the measurement date that includes the 'Nan' value may be deleted.

[0158] The deletion of the bio-component data described above is exemplary, and if an error (Nan value) is included among the items constituting the bio-component data based on various criteria, it can be detected and deleted.

[0159] Meanwhile, if no error is detected in the biocomponent data (1101) (S131: No), a virtual first variable can be created (S310) through the variable creation unit (150).

[0160] The above first variable can be created as the sum of body fat mass (1101-6, Body Fat Mass), muscle mass (1101-7, Soft Lean Mass), and bone mineral mass (1101-8, Osseous mineral). For reference, muscle mass (1101-7) can be defined as the sum of body water content, protein content (1101-4), and non-osseous mineral excluding bone.

[0161] As described above, by creating a virtual first variable (S310) and comparing the first variable with the data entered in the steps (S111, S112) (S50), an error in the entered biocomponent data (1101) can be determined.

[0162] More specifically, first, it is determined whether the subject's age (1201-3) from the subject's identification data (1201) meets a preset standard, and then, based on the preset standard, it is determined whether to delete the bio-component data based on the degree of difference between the first variable and the body weight (1101-3) among the bio-component data (1101), thereby generating refined data (second data).

[0163] The biocomponent data (1101) input in the above step (S111) can be obtained by measuring through bioelectrical impedance analysis (BIA), which is a method of measuring the composition of the body based on the speed at which an electric current passes through the body.

[0164] The biocomponent data measured in this way has a large range of changes in body composition as the subject's age becomes younger, and there is a high possibility of errors occurring during the measurement process. Therefore, the biocomponent measurement data preprocessing method of this embodiment determines whether the subject's age falls within a plurality of preset criteria, and then determines errors in the biocomponent measurement data under different conditions according to each preset criteria.

[0165] For example, in the present embodiment, if the difference between the first variable generated and the weight (1101-3) is greater than a preset range, it is determined that an error has occurred in the measurement value generating the first variable, and refined data can be generated through a process of deleting a row including the measurement value generating the first variable from the connection data (140d).

[0166] As described above, the reason for comparing the virtual first variable with the weight data is that, theoretically, the weight data (1101-3) is identical to the first variable, but differences may arise between the weight data (1101-3) and the first variable due to various variables. Therefore, if the difference between the weight data (1101-3) and the first variable is outside a preset range, the measurement values ​​of the row containing the measurement values ​​that generate the first variable are judged to be errors and are deleted.

[0167] The process of determining errors in the biocomponent measurement data described above through the error determination unit (170) is described below.

[0168] After creating the first variable in the above step (S310), it can be determined (S511) whether the subject's age is below the first standard, and the first standard can be set to 60 months.

[0169] If the subject's age is below the first standard (S511: Yes), it can be determined whether the difference between the first variable and any one of the biocomponent data is 20% or more (S513). Any one of the biocomponent data values ​​can be set as weight data (1101-3).

[0170] If the difference between the first variable and the weight data (1101-3) is 20% or more (S513: Yes), the biocomponent data can be deleted (S711) through the data deletion unit (180) and refined second data can be generated (S712) through the second data generation unit (190).

[0171] Additionally, if the difference between the first variable and the weight data (1101-3) is not more than 20% (S513: No), refined second data can be generated (S712) through the second data generation unit (190).

[0172] Meanwhile, if the subject's age exceeds the first standard (S511: No), it may be determined whether the subject's age exceeds the first standard and falls within the range of the second standard (S521), and the second standard may be set to 75 months.

[0173] If the subject's age falls between the first and second criteria (S511: Yes), it can be determined whether the difference between the values ​​of the first variable and the biocomponent data is 15% or more (S523).

[0174] Since the age of the subject subject to error judgment in the above step (S523) is higher than the age of the subject subject to error judgment in the above step (S513), the range of change in body composition is relatively small.

[0175] Therefore, the standard for determining an error in the bio-component data entered in the above step (S523) (is the difference between the first variable and the weight data 15% or more?) may be set higher than the standard for determining an error in the bio-component data entered in the above step (S513) (is the difference between the first variable and the weight data 20% or more?).

[0176] A higher setting of the above criteria may mean that the range within which differences between the first variable and weight data in this embodiment are judged as errors is narrower. In other words, the preprocessing method for biocomponent measurement data in this embodiment may set the criteria for judging errors higher as the subject's age increases.

[0177] If the difference between the first variable and the weight data (1101-3) is 15% or more (S523: Yes), the biocomponent data can be deleted (S721) through the data deletion unit (180) and refined second data can be generated (S722) through the second data generation unit (190).

[0178] Additionally, if the difference between the first variable and the weight data (1101-3) is not more than 15% (S523: No), refined second data can be generated (S723) through the second data generation unit (190).

[0179] Meanwhile, if the age of the subject does not fall between the first and second criteria (S521: No), it may be determined whether the age of the subject falls within the range exceeding the second criteria and below the third criteria (S531), and the third criteria may be set to 100 months.

[0180] If the subject's age falls between the second and third criteria (S531: Yes), it can be determined whether the difference between the values ​​of the first variable and the biocomponent data is 10% or more (S533).

[0181] If the difference between the first variable and the weight data (1101-3) is 10% or more (S533: Yes), the biocomponent data can be deleted (S731) through the data deletion unit (180) and refined second data can be generated (S732) through the second data generation unit (190).

[0182] Additionally, if the difference between the first variable and the weight data (1101-3) is not more than 10% (S533: No), refined second data can be generated (S733) through the second data generation unit (190).

[0183] Meanwhile, if the subject's age does not fall between the second and third criteria (S531: No), it can be determined whether the subject's age exceeds the third criteria (S541).

[0184] If the subject's age exceeds the third criterion (S541: Yes), it can be determined whether the difference between the values ​​of the first variable and the biocomponent data is 3% or more (S543).

[0185] If the difference between the first variable and the weight data (1101-3) is 3% or more (S543: Yes), the biocomponent data can be deleted (S741) through the data deletion unit (180) and refined second data can be generated (S742) through the second data generation unit (190).

[0186] Additionally, if the difference between the first variable and the weight data (1101-3) is not more than 3% (S543: No), refined second data can be generated (S743) through the second data generation unit (190).

[0187] Meanwhile, if the subject's age does not exceed the third criterion (S541: No), this means that there is an error in the subject's age data (1201-3) or an error occurred in the process of determining whether the subject's age data (1201-3) falls within the range defined by the first to third criteria described above, and thus the process can be terminated. Alternatively, the process of determining whether the subject's age is below the first criterion (S511) may be repeatedly performed.

[0188] The second data generated as described above is refined data that has gone through a process of determining errors based on the input subject's biocomponent data (1101) and identification data (1201), and can be used as input data for growth prediction or for generating solutions through growth prediction. In this case, the second data can be used as data input to the input unit (10) of FIG. 2.

[0189] Additionally, the second data can be used as training data for a growth prediction or growth prediction model. In this case, the second data can be used as input data for the prediction model of FIG. 7.

[0190] The present invention has been described so far, focusing on preferred embodiments thereof. All embodiments and conditional examples disclosed herein are intended to assist those skilled in the art in understanding the principles and concepts of the present invention. Those skilled in the art will appreciate that the present invention can be implemented in modified forms without departing from its essential characteristics.

[0191] Therefore, the disclosed embodiments should be considered illustrative rather than restrictive. The scope of the present invention is determined by the claims, not the foregoing description, and all differences within the scope equivalent thereto should be construed as being encompassed by the present invention.

[0192] Meanwhile, the methods according to the various embodiments of the present invention described above can be implemented as programs and provided to servers or devices. Accordingly, each device can access the server or device where the program is stored and download the program.

[0193] In addition, the methods according to the various embodiments of the present invention described above may be implemented as programs and stored and provided on various non-transitory computer-readable media. A non-transitory computer-readable medium refers to a medium that permanently stores data and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, the various applications or programs described above may be stored and provided on non-transitory computer-readable media, such as a CD, DVD, hard disk, Blu-ray disk, USB, memory card, or ROM.

[0194] In addition, although the preferred embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above, and various modifications can be made by a person having ordinary skill in the art to which the invention pertains without departing from the gist of the present invention claimed in the claims, and such modifications should not be understood individually from the technical idea or prospect of the present invention.

Claims

1. In a preprocessing method of biocomponent measurement data for growth prediction performed by a computing device, Step of receiving the subject's physical data; A step of generating a first variable based on the physical data of the subject; and A preprocessing method of biocomponent measurement data for growth prediction, comprising a step of comparing the first variable with a preset value to determine an error in the body data.

2. In paragraph 1, The step of entering the subject's physical data is as follows: A step of inputting the subject's biocomponent data; and A preprocessing method for biocomponent measurement data for growth prediction, comprising: a step of inputting identification data of the subject; 3. In paragraph 2, A preprocessing method of biocomponent measurement data for growth prediction, characterized in that it further comprises a step of generating first data by connecting the biocomponent data of the subject and the identification data of the subject after receiving the physical data of the subject.

4. In paragraph 3, A preprocessing method of biocomponent measurement data for growth prediction, characterized in that it further includes a step of deleting the biocomponent data when a preset value is measured from the biocomponent data after generating the first data.

5. In paragraph 3, A preprocessing method for biocomponent measurement data for growth prediction, characterized in that the first variable is generated based on the biocomponent data of the subject.

6. In paragraph 5, A preprocessing method for biocomponent measurement data for growth prediction, characterized in that the first variable is generated as the sum of body fat mass, soft lean mass, and osseous mineral among the biocomponent data of the subject.

7. In paragraph 3, The steps for judging the above error are: A preprocessing method of biocomponent measurement data for growth prediction, characterized in that it comprises a step of comparing the first variable with a value of any one of the biocomponent data.

8. In paragraph 7, The steps for judging the above error are: A preprocessing method of biocomponent measurement data for growth prediction, characterized in that it further comprises a step of determining whether the age of the subject among the identification data corresponds to a preset standard before comparing the first variable with any value of the biocomponent data.

9. In paragraph 7, A preprocessing method of biocomponent measurement data for growth prediction, characterized in that it further includes a step of deleting the biocomponent data if the result of comparing the value of one of the first variable and the biocomponent data falls within a preset range.

10. In paragraph 7, A preprocessing method of biocomponent measurement data for growth prediction, characterized in that it further comprises a step of generating second data when the result of comparing the value of either the first variable or the biocomponent data does not fall within a preset range.

11. A program stored on a computer-readable recording medium including a program code for executing a method for preprocessing biocomponent measurement data for growth prediction described in any one of paragraphs 1 to 10.

12. A computer-readable recording medium having recorded thereon a program for executing a method for preprocessing biocomponent measurement data for growth prediction described in any one of claims 1 to 10.

13. In a preprocessing device for biocomponent measurement data for growth prediction, An input section for receiving the subject's physical data; A variable generation unit that generates a first variable based on the subject's physical data; and A preprocessing device for biocomponent measurement data for growth prediction, comprising an error determination unit that determines an error in the body data by comparing the first variable with a preset value.

14. In paragraph 13, The above input section, A first input unit for receiving the subject's biocomponent data; and A preprocessing device for biocomponent measurement data for growth prediction, characterized by including a second input unit for receiving identification data of the subject.

15. In paragraph 14, A preprocessing device for biocomponent measurement data for growth prediction, characterized in that it further includes a connecting unit for connecting data input through the first input unit and the second input unit.

Images