Method for monitoring medium activity intensity of preschool children

Abstract

The invention provides a method for monitoring medium activity intensity of a preschool child. The method comprises the following steps: acquiring a personalized index of the preschool child to be monitored and an acceleration signal in a state to be monitored; and inputting the personalized index and the acceleration signal into a medium activity intensity evaluation model, and outputting medium and strong activity indication information based on a medium activity intensity threshold. And the medium activity intensity threshold range of the preschool children is 3.2-5.3 Mets. The medium activity intensity evaluation model is obtained by fusion construction based on personalized indexes of a plurality of preschool children, acceleration signals in different states and synchronously recorded movement energy consumption label values corresponding to the acceleration signals. When the exercise energy consumption of the preschool children is monitored by using the medium activity intensity evaluation model subsequently, the medium activity intensity indication information of the preschool children can be obtained, so that the indication information of the medium activity intensity for daily evaluation activities of the preschool children can be obtained, and the accuracy of exercise energy consumption monitoring is improved.

Description

Technical Field

[0001] This application relates to the medical field, specifically to the field of measuring physiological parameters of children, and more specifically to a method and related device for monitoring the moderate activity intensity of preschool children. Background Art

[0002] Physical activity (PA) refers to any bodily movement produced by skeletal muscle contractions that results in increased energy expenditure and is one of the important dimensions supporting early childhood growth and development. Physical activity is mainly divided into light physical activity (LPA), moderate physical activity (MPA), and vigorous physical activity (VPA). Increasing evidence shows that lower levels of sedentary behavior (SB) and appropriate physical activity intensity can effectively improve children's physical health. Since the preschool period (which can represent the group of children aged 3 to 6 years old) is a critical period for forming good physical activity habits, the physical activity habits at this stage will affect the physical activity behavior in the later development stage. In order to promote the growth and development of young children, it is necessary to monitor the physical activity intensity of preschool children.

[0003] Under normal circumstances, the corresponding physical activity intensity can be determined by monitoring the exercise energy consumption of preschool children. Exercise energy consumption can be expressed by absolute energy (energy expenditure, EE) or relative energy consumption (metabolicequivalents, METs, also known as metabolic equivalents). That is, the physical activity intensity level can be determined according to the value of EE or METs. Currently, on the premise of determining the physical activity intensity level according to the value of METs, the exercise energy consumption range of moderate-intensity activities is generally recognized as 3 - 6 Mets. However, the objects of data collection and testing in this interval are only applicable to adolescents and adult groups. Affected by growth and development, the relationship between the acceleration signal synchronously recorded during exercise and METs in preschool children is completely different from that of children, adolescents, or adults. In addition, the exercise energy consumption threshold ranges corresponding to moderate-intensity activities for these preschool children aged 3, 4, 5, and 6 years old are not given separately. Therefore, the accuracy of using existing exercise energy consumption monitoring models to monitor the energy consumption of preschool children is relatively low.

[0004] At present, the methods for monitoring exercise energy consumption are mainly constructed based on accelerometer activity counts (AC). AC is the data obtained by processing the acceleration signal through a specific filter, which will lose a lot of information related to exercise, thus affecting the accuracy of the exercise energy consumption monitoring method. The accelerometer sensor device can also record the original acceleration signal data of the test subjects. Existing evidence shows that compared with the exercise energy consumption estimation model constructed based on AC, the estimation accuracy of the exercise energy consumption estimation model constructed based on the characteristics of the original acceleration signal is significantly increased. However, these exercise energy consumption estimation models are all constructed with foreign children as the test subjects, and their reproducibility and accuracy are not ideal yet.

[0005] In addition, in the research on the method for estimating the exercise energy consumption of preschool children constructed based on the characteristics of the original acceleration signal, there is no integration or only partial integration of personalized indicators such as height, weight, gender, age, etc., which will also lead to room for improvement in the estimation accuracy of the model constructed for the exercise energy consumption value of preschool children.

[0006] Combined with the habits of medical institutions and parents of young children, the amount of exercise is generally evaluated by time to determine whether it reaches the standard level. In the previous methods, the energy consumption value was usually output, and there was no model proposed for estimating the moderate activity intensity of preschool children. A model that outputs the length of physical activity time at a moderate activity intensity for preschool children will be more in line with the need to daily evaluate whether the time for preschool children to perform moderate activity intensity meets the requirements. Summary of the Invention

[0007] In view of this, the present application is committed to providing a method and related device for monitoring the moderate activity intensity of preschool children, so as to use this monitoring method to monitor the exercise energy consumption of preschool children, determine the daily physical activity intensity of preschool children and the compliance of the corresponding activity duration. Using this monitoring method helps to improve the accuracy of monitoring the exercise energy consumption of preschool children.

[0008] In the first aspect, the present application provides a method for monitoring the moderate activity intensity of preschool children, and the method includes: Obtain the personalized indicators of the preschool children to be monitored and the acceleration signal in the state to be monitored; Input the personalized indicators and the acceleration signal into the moderate activity intensity evaluation model, and the output result is the medium-strong activity indication information output by the determination system based on the moderate activity intensity threshold. Among them, the moderate activity intensity threshold range for preschool children is 3.2 - 5.3 Mets; Among them, the moderate activity intensity evaluation model is constructed by fusing the personalized indicators of multiple preschool children, the acceleration signals in different states, and the corresponding exercise energy consumption label values of the acceleration signals.

[0009] In a possible implementation, the medium activity intensity threshold range for preschool children aged 3 - 6 is 3.2 - 5.3 Mets.

[0010] In a possible implementation, inputting the personalized index and the acceleration signal into the medium activity intensity evaluation model, and the output result is that the determination system outputs medium - to - high activity indication information based on the medium activity intensity threshold, including: Dividing the acceleration signal according to a preset time window to obtain first acceleration signals corresponding to multiple preset time windows respectively; Based on the first acceleration signal of each preset time window and the personalized index, determining input features corresponding to each preset time window; Inputting the input features into the medium activity intensity evaluation model, and outputting the medium - to - high activity indication information based on the medium activity intensity threshold.

[0011] In a possible implementation, the medium - to - high activity indication information includes the daily average medium - to - high activity index.

[0012] In a possible implementation, inputting the input features into the medium activity intensity evaluation model, and the output result is that the determination system outputs the medium - to - high activity indication information based on the medium activity intensity threshold, including: Inputting the input features corresponding to each preset time window into the medium activity intensity evaluation model to obtain the exercise energy consumption data corresponding to each preset time window; Comparing the exercise energy consumption data with the medium activity intensity threshold to determine the target preset time window corresponding to the exercise energy consumption data that meets the medium activity intensity threshold; Counting the target preset time windows to determine the daily average medium - to - high activity index of the preschool child.

[0013] The so - called daily average medium - to - high activity index of preschool children represents the statistical information of daily activities that meet the medium intensity. This statistical information can include the total duration of daily medium - intensity activities, or the integral corresponding to the time period under certain set conditions of the curve of metabolic equivalent (METs) value changing with time (i.e., the sum of the areas under the METs value curve in a certain time period), etc.

[0014] In a possible implementation, the first acceleration signal includes the x - axis acceleration signal, the y - axis acceleration signal, and the z - axis acceleration signal. Based on the first acceleration signal of each preset time window and the personalized index, determining the input features corresponding to each preset time window includes: Determine the one-dimensional vector sum signal of the x-axis acceleration signal, the y-axis acceleration signal, and the z-axis acceleration signal; Respectively determine the statistical characteristic indexes of the x-axis acceleration signal, the y-axis acceleration signal, the z-axis acceleration signal, and the one-dimensional vector sum signal; Based on the statistical characteristic indexes corresponding to each preset time window and the personalized index, determine the input feature corresponding to each preset time window.

[0015] In a possible implementation manner, the determining the input feature based on the statistical characteristic index and the personalized index includes: Use a feature selection algorithm to select a preset number of indexes from the statistical characteristic indexes and the personalized index; Determine the input feature based on the preset number of indexes.

[0016] In a possible implementation manner, the acceleration signal is measured by an acceleration measuring instrument worn by the preschool child to be monitored, and the acceleration measuring instrument is worn on the right waist.

[0017] In a possible implementation manner, the wearing period of the acceleration measuring instrument is at least 1 day.

[0018] In a possible implementation manner, the threshold range of moderate activity intensity corresponding to 3-year-old preschool children is 3.2 - 4.4 Mets.

[0019] In a possible implementation manner, the threshold range of moderate activity intensity corresponding to 4-year-old preschool children is 3.4 - 4.6 Mets.

[0020] In a possible implementation manner, the threshold range of moderate activity intensity corresponding to 5-year-old preschool children is 3.7 - 4.8 Mets.

[0021] In a possible implementation manner, the threshold range of moderate activity intensity corresponding to 6-year-old preschool children is 3.8 - 5.1 Mets.

[0022] In a possible implementation manner, the age of the preschool child is calculated according to the full years corresponding to the actual date of birth.

[0023] In a possible implementation manner, the obtaining the personalized index of the preschool child to be monitored and the acceleration signal in the state to be monitored includes: Obtain the original personalized index of the preschool child to be monitored and the original acceleration signal in the state to be monitored; Preprocess the original personalized index and the original acceleration signal to obtain the personalized index and the acceleration signal; Among them, the preprocessing process includes one or more of the following: Imputing missing values; or, Performing one-hot encoding on categorical data; or, Normalizing continuous data; or, Normalizing continuous data and converting the normalized data into data with an approximate normal distribution.

[0024] In a possible implementation, the personalized metrics include at least one of gender, age, height, weight, and body mass index.

[0025] In a second aspect, the present application provides a medium activity intensity monitoring device for preschool children, the device includes: An acquisition unit, configured to acquire the personalized metrics of the preschool children to be monitored and the acceleration signal in the state to be monitored; A monitoring and determination unit, configured to input the personalized metrics and the acceleration signal into a medium activity intensity evaluation model; and the determination system therein outputs medium and strong activity indication information based on a medium activity intensity threshold, where the medium activity intensity threshold range for 3-6-year-old preschool children is 3.2-5.3 Mets; wherein, the medium activity intensity evaluation model is constructed by fusing the personalized metrics of multiple preschool children, the acceleration signals in different states, and the corresponding motion energy consumption label values of the acceleration signals.

[0026] In a third aspect, the present application provides an electronic device, the device includes: a memory and a processor; The memory is used to store relevant program codes; The processor is used to call the program codes to execute the medium activity intensity monitoring method for preschool children according to any one of the implementations in the first aspect above.

[0027] In a fourth aspect, the present application provides a computer-readable storage medium, the computer-readable storage medium is used to store a computer program, and the computer program is used to execute the medium activity intensity monitoring method for preschool children according to any one of the implementations in the first aspect above.

[0028] In a fifth aspect, the present application provides a computer program product, the computer program product includes computer programs / instructions, and when the computer programs / instructions are executed by a processor, the medium activity intensity monitoring method for preschool children according to any one of the implementations in the first aspect above is implemented.

[0029] The inventive point of the present invention lies in: 1. The medium activity intensity threshold range suitable for the characteristics of Chinese preschool children is given; 2. The accurate medium activity intensity threshold ranges suitable for 3-year-old, 4-year-old, 5-year-old, and 6-year-old children respectively are given; 3. Personalized indicators that conform to the development characteristics of Chinese preschool children are integrated; 4. The concept of the daily average moderate-to-vigorous activity index is proposed for the first time. The cumulative duration of moderate-intensity activities can be directly output by the model and the decision-making system, and the integral corresponding to the time period when the curve of the metabolic equivalent of task (METs) value changing with time meets certain set conditions (i.e., the sum of the areas under the METs value curve in a certain time period) and other information can also be directly output.

[0030] In the above implementation manner of this application, the personalized indicators of the preschool children to be monitored and the acceleration signals in the state to be monitored are obtained. First of all, in combination with the development characteristics of preschool children, the research team of this study integrated the personalized physical development indicators of Chinese children, such as height, weight, gender, age, and body mass index (BMI). These seemingly conventional indicators actually have a great impact on the accuracy of the preschool children's motion assessment model. During the research process, five personalized indicators reflecting the characteristics of preschool children are integrated with the original acceleration data and the corresponding motion energy consumption label values to construct a prediction model, and the personalized indicators and acceleration signals are input into the medium activity intensity assessment model, and medium-to-vigorous activity indication information is output based on the medium activity intensity threshold. Among the personalized indicators, the height indicator has the greatest impact on the accuracy of model prediction. These characteristics can be discovered by ranking the importance of feature labels during the training of the model, which directly affects the accuracy of the prediction model.

[0031] Secondly, the medium activity intensity threshold range for 3- to 4-year-old preschool children is 3.2 - 5.3 Mets. Among them, the medium activity intensity assessment model is constructed by integrating the personalized indicators of multiple preschool children, the acceleration signals in different states, and the motion energy consumption label values corresponding to the acceleration signals. That is to say, the medium activity intensity assessment model can be trained by collecting the acceleration signals, personalized indicators, and their motion energy consumption-related data of preschool children. Subsequently, when using the medium activity intensity assessment model to monitor the motion energy consumption of preschool children, the indication information of the medium activity intensity for the daily assessment activities of preschool children can be obtained, and the accuracy of motion energy consumption monitoring can be improved.

[0032] Compared with the prior art, a more accurate threshold range of moderate activity intensity is given through research, and for the first time, accurate threshold ranges of moderate activity intensity applicable to 3-year-old, 4-year-old, 5-year-old, and 6-year-old children are proposed. Since the currently globally applicable range is based on statistical analysis of foreign teenagers, compared with the currently globally applicable range, the data source of the threshold range we have statistically obtained is Chinese preschool children, and the obtained threshold range conforms to the developmental characteristics of Chinese preschool children. Therefore, the corresponding threshold range of moderate activity intensity is more accurate, which also improves the accuracy of the relevant evaluation results of moderate-intensity activities for preschool children.

[0033] For the first time, the concept of the daily average moderate-to-high activity index is proposed. According to the constructed model, at least the cumulative duration of the daily average moderate activity intensity can be directly obtained. By counting the time when preschool children wearing the corresponding accelerometer sensor meet the moderate-intensity activity, and then comparing this time with the data table of the time required for Chinese preschool children of the corresponding age group to meet the moderate-intensity activity for physical development, it can be known whether the preschool child has met the time required for moderate-intensity activity for its physical development. Then, based on this information, the diagnosis in the fields of sports medicine and others for the preschool child can be assisted to avoid the impact caused by insufficient time of moderate-intensity activity for preschool children. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments provided in the present application. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings.

[0035] Figure 1 It is a flowchart of a method for monitoring the moderate activity intensity of preschool children provided by an embodiment of the present application.

[0036] Figure 2 It is a schematic diagram of the test results of a sports energy consumption monitoring model provided by an embodiment of the present application.

[0037] Figure 3 It is a schematic diagram of a system for monitoring the moderate activity intensity of preschool children provided by an embodiment of the present application.

[0038] Figure 4 It is a schematic diagram of a device for monitoring the moderate activity intensity of preschool children provided by an embodiment of the present application.

[0039] Figure 5 It is a schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. The described embodiments are only exemplary implementation manners of the present application, not all implementation manners. Those skilled in the art can obtain other embodiments without creative labor in combination with the embodiments of the present application, and these embodiments are also within the protection scope of the present application.

[0041] Since the preschool period (which can represent the group aged 3 to 6 years old) is a critical period for forming good physical activity habits, the physical activity habits at this stage may affect the physical activity behaviors in the later development stage. In order to promote the growth and development of children in the early stage, it is necessary to monitor the physical activity intensity of preschool children.

[0042] Under normal circumstances, the corresponding physical activity intensity can be determined by monitoring the exercise energy consumption of preschool children. Currently, the exercise energy consumption monitoring method based on the accelerometer sensor is mainly used to monitor the test personnel, and the exercise energy consumption of the test personnel is estimated. However, this exercise energy consumption monitoring method is mainly applicable to adolescents and adult groups, and the physical activity habits of preschool children are quite different from those of adolescents (which can represent the group aged 11 to 17 years old) and adults. Therefore, the accuracy of using the existing exercise energy consumption monitoring model to monitor the energy consumption of preschool children is relatively low.

[0043] In addition, the current exercise energy consumption monitoring methods are all constructed based on the accelerometer activity counts (AC), and AC is the data obtained after processing the acceleration signal through a specific filter, which will lose a lot of information related to movement, thus affecting the accuracy of the exercise energy consumption monitoring method.

[0044] Currently, there are some studies on constructing an exercise energy consumption estimation model for preschool children based on the characteristics of the original acceleration signal. However, these exercise energy consumption estimation models are all constructed with foreign children as the test subjects, and the reproducibility and accuracy are not yet ideal.

[0045] Based on this, the embodiments of the present application provide a method for monitoring the moderate activity intensity of preschool children, so as to improve the accuracy of monitoring the exercise energy consumption of preschool children. Taking Chinese preschool children as the research object, the monitoring results are more in line with the growth and development characteristics of Chinese preschool children. Specifically, when implemented, the personalized indicators of the preschool children to be monitored and the acceleration signals in the state to be monitored are obtained. The personalized indicators and acceleration signals are input into the moderate activity intensity evaluation model, and the output result is the medium-strong activity indication information output by the determination system based on the moderate activity intensity threshold. Among them, the moderate activity intensity threshold range for preschool children is 3.2 - 5.3 Mets. Among them, the moderate activity intensity evaluation model is trained based on the personalized indicators of multiple preschool children, the acceleration signals in different states, and the exercise energy consumption label values corresponding to the acceleration signals. That is, the moderate activity intensity evaluation model can be trained by collecting the acceleration signals, personalized indicators and their exercise energy consumption related data of preschool children. Subsequently, when using the moderate activity intensity evaluation model to monitor the exercise energy consumption of preschool children, it is possible to obtain the indication information of the moderate activity intensity for the daily evaluation activities of preschool children and improve the accuracy of exercise energy consumption monitoring.

[0046] To facilitate the understanding of the technical solutions provided by the embodiments of the present application, the following will be specifically introduced in combination with the accompanying drawings in the embodiments.

[0047] See Figure 1 As shown, it is a flowchart of a method for monitoring the moderate activity intensity of preschool children provided by the embodiments of the present application.

[0048] Optionally, this method can be executed by a processing device. The processing device can be an electronic device or other devices, and the embodiments of the present application do not make limitations in this regard.

[0049] This method may include the following steps: S101: Obtain the personalized indicators of the preschool children to be monitored and the acceleration signals in the state to be monitored.

[0050] In order to monitor the exercise energy consumption status of preschool children, it is first necessary to obtain the personalized indicators of the preschool children to be monitored and the acceleration signals in the state to be monitored. Among them, the preschool children to be detected refer to the children who need to have their exercise energy consumption monitored. In the embodiments of the present application, the exercise energy consumption monitoring is mainly carried out for children between 4 and 6 years old. The personalized indicators of the preschool children to be monitored may include: gender, age, height, weight, body mass index, etc. The acceleration signal can be used to characterize the exercise state of the preschool children to be monitored. The acceleration signals corresponding to the preschool children to be monitored in different states to be monitored are different, which results in different exercise energy consumptions. For example, the states to be monitored of the preschool children to be monitored may include various exercise states such as sitting still, walking slowly, walking fast, running, jumping, etc.

[0051] Optionally, the preschool children to be monitored can be made to wear an acceleration measuring instrument, and the acceleration signals of the preschool children in different exercise states can be measured through the acceleration measuring instrument. To objectively and accurately monitor the exercise energy consumption of preschool children, the preschool children to be monitored can wear the acceleration measuring instrument on the right waist to measure the acceleration signal. Among them, the wearing period of the acceleration measuring instrument is at least 1 day. For example, the preschool children can be made to wear the acceleration measuring instrument continuously for 7 days, and the exercise energy consumption of the preschool children to be monitored can be monitored by counting the personalized indicators and acceleration signals of the preschool children to be monitored within 7 days.

[0052] Since there may be misrecorded or missing data in the relevant data of the preschool children obtained, in order to accurately process the data, the collected data can be preprocessed first. In one possible implementation, the original personalized indicators of the preschool children to be monitored and the original acceleration signals in the state to be monitored can be obtained. Then, the original personalized indicators and the original acceleration signals are preprocessed to obtain the personalized indicators and the acceleration signals; among them, the preprocessing process includes one or more of the following: imputing missing values; or, performing one-hot encoding processing on categorical data; or, normalizing continuous data; or, normalizing continuous data and converting the data after normalization into data with an approximate normal distribution.

[0053] When specifically implemented, the missing values in the data can be imputed. That is, when there are missing values in the personalized indicators or acceleration signals of the preschool children obtained, the missing values can be imputed to avoid affecting the subsequent data processing process. For example, the missing values in categorical data (such as boys or girls) are imputed with the value "NaN", and the missing values in numerical data (such as age) are imputed with the mean value.

[0054] Since most models cannot directly process categorical data, it is necessary to perform one-hot encoding on categorical data to convert it into numerical data. Among them, one-hot encoding is also known as one-hot effective encoding. The processing method is to use an N-bit status register to encode N states. Each state has an independent register bit, with a value of 0 or 1, and at any time, only one bit is effective. For example, for the categorical data "boys" and "girls", since there are only two types in this category, so take N = 2, then "01" and "10" can be used to represent "boys" and "girls".

[0055] For continuous data, in order to simplify the data processing process, the continuous data can be normalized. Or, the normalized continuous data can also be converted into data with an approximate normal distribution to improve the data processing efficiency.

[0056] S102: Input the personalized index and the acceleration signal into the moderate activity intensity evaluation model, and the output result is the medium-strong activity indication information output by the determination system based on the moderate activity intensity threshold.

[0057] Among them, the moderate activity intensity evaluation model is trained based on the personalized indexes of multiple preschool children, the acceleration signals in different states, and the motion energy consumption label values corresponding to the acceleration signals. The specific training process can be seen in the subsequent embodiments and will not be introduced here first.

[0058] That is, by collecting the acceleration signals, personalized indexes, and their motion energy consumption-related data of preschool children in different states, the moderate activity intensity evaluation model can be trained. Subsequently, when using this moderate activity intensity evaluation model to monitor the motion energy consumption of the preschool children to be monitored, the accuracy of the motion energy consumption monitoring can be improved, and the medium-strong activity indication information of the preschool children can be obtained more accurately.

[0059] Among them, the medium-strong activity indication information can be used to indicate the state of the preschool children in the moderate activity intensity. For example, the medium-strong activity indication information can include the daily average medium-strong activity index, that is, the time length of the preschool children in the moderate activity intensity.

[0060] In a possible implementation manner, the acceleration signals of the collected preschool children may correspond to the motion states of multiple time periods, and the duration of each time period may be different. In order to unify the data format for convenient processing, the acceleration signals can be divided according to a preset time window to obtain the first acceleration signals corresponding to multiple preset time windows. That is, each preset time window corresponds to the first acceleration signal obtained by division. Among them, the multiple preset time windows can be non-overlapping time windows.

[0061] For each preset time window, an input feature corresponding to the preset time window can be determined based on the first acceleration signal and the personalized index of the preset time window. The input feature corresponding to each preset time window is input into the moderate activity intensity evaluation model, and moderate-to-high activity indication information is output based on the moderate activity intensity threshold.

[0062] Generally, knowing the state of preschool children at a moderate activity intensity is more conducive to understanding the physical health of preschool children. Therefore, by counting the time of preschool children's moderate-intensity physical activity, the exercise energy consumption of preschool children can be better monitored, and the intensity level of physical activity is divided based on metabolic equivalents (METs).

[0063] In specific implementation, the input feature corresponding to each preset time window is input into the moderate activity intensity evaluation model to obtain the exercise energy consumption data corresponding to each preset time window. The determination system compares the exercise energy consumption data with the moderate activity intensity threshold to determine the target preset time window corresponding to the exercise energy consumption data that meets the moderate activity intensity threshold. The total time length of each target preset time window is counted to determine the daily moderate-to-high activity index.

[0064] Among them, the exercise energy consumption data includes metabolic equivalents, that is, the metabolic equivalents corresponding to each preset time window are compared with the moderate activity intensity threshold to determine the target preset time window corresponding to the metabolic equivalents that meet the moderate activity intensity threshold.

[0065] Optionally, the moderate activity intensity threshold can include a first moderate intensity threshold and a second moderate intensity threshold. The first moderate intensity threshold is used as the lower limit of the moderate activity intensity threshold, and the second moderate intensity threshold is used as the upper limit of the moderate activity intensity threshold. The metabolic equivalents are compared with the first moderate intensity threshold and the second moderate intensity threshold. When the metabolic equivalents are greater than or equal to the first moderate intensity threshold and less than or equal to the second moderate intensity threshold, it indicates that the metabolic equivalents meet the moderate activity intensity threshold, and the preset time window corresponding to the metabolic equivalents is determined as the target preset time window.

[0066] In an actual application scenario, the moderate activity intensity thresholds for preschool children of different ages are also different. In the embodiments of the present application, the moderate activity intensity thresholds corresponding to preschool children of different ages can be determined in advance through experimental statistics. For example, the moderate activity intensity threshold for 4-year-old preschool children is [3.4, 4.6] Mets; the moderate activity intensity threshold for 5-year-old preschool children is [3.7, 4.8] Mets; the moderate activity intensity threshold for 6-year-old preschool children is [3.8, 5.1] Mets. Among them, the age of preschool children is calculated according to the full years corresponding to the actual date of birth.

[0067] Among them, according to the digital rule corresponding to the moderate activity intensity threshold of preschool children aged 4 - 6, the moderate activity intensity threshold of 3 - year - old preschool children is estimated to be [3.2, 4.4] Mets; It should be noted that the range of the moderate activity intensity threshold provided in the embodiments of the present application is determined through experimental statistics, and is only an exemplary illustration, not limited to the above implementation manner. The embodiments of the present application do not make any formal limitations on the specific manner of determining the moderate activity intensity threshold.

[0068] Based on the above embodiments, for each preset time window, the input feature corresponding to the preset time window can be determined based on the first acceleration signal and the personalized index of the preset time window. The following will specifically introduce this step.

[0069] In a possible implementation manner, the acceleration signal includes the x - axis acceleration signal, the y - axis acceleration signal, and the z - axis acceleration signal. The one - dimensional vector sum signal can be determined based on the x - axis acceleration signal, the y - axis acceleration signal, and the z - axis acceleration signal. That is, calculate the sum of the squares of the x - axis acceleration signal, the y - axis acceleration signal, and the z - axis acceleration signal, and then calculate the arithmetic square root of the sum of squares as the one - dimensional vector sum signal.

[0070] When the acceleration signal is divided according to the preset time window, the first acceleration signal within each preset time window includes the acceleration signals collected at multiple time points. Therefore, based on the multiple x - axis acceleration signals, multiple y - axis acceleration signals, multiple z - axis acceleration signals, and multiple one - dimensional vector sum signals within the preset time window, the statistical feature indexes of the x - axis acceleration signal, the statistical feature indexes of the y - axis acceleration signal, the statistical feature indexes of the z - axis acceleration signal, and the statistical feature indexes of the one - dimensional vector sum signal can be determined respectively. Then, based on the statistical feature indexes and the personalized index corresponding to each preset time window, the input feature corresponding to each preset time window is determined.

[0071] Among them, the statistical feature index can represent mathematical statistical features. For example, it can include the maximum value, the minimum value, the mean value, the standard deviation, the variance, the percentile, the coefficient of variation, etc. In a possible implementation manner, the statistical feature index of the x - axis acceleration signal at least includes the maximum value of multiple x - axis acceleration signals. The statistical feature index of the y - axis acceleration signal at least includes the peak - to - peak value of multiple y - axis acceleration signals, where the peak - to - peak value represents the difference between the maximum value and the minimum value. The statistical feature index of the z - axis acceleration signal at least includes the minimum value of multiple z - axis acceleration signals. To find the key features representing different motion directions. And the statistical feature index of the one - dimensional vector sum signal at least includes the average value and the variance of multiple one - dimensional vector sum signals.

[0072] It should be noted that in the embodiments of the present application, the specific types and numbers of statistical feature indicators are not limited and can be set according to actual needs. In a possible application scenario, for the x-axis acceleration signal, y-axis acceleration signal, z-axis acceleration signal, and one-dimensional vector sum signal, a total of 91 statistical feature indicators including maximum value, minimum value, mean value, standard deviation, variance, percentile, coefficient of variation, etc. can be determined. Then, in combination with 5 personalized indicators such as gender, age, height, weight, and body mass index, the statistical feature indicators and personalized indicators are fused to obtain 96 indicators to determine the input features.

[0073] When there are many indicators in the input features, it will increase the complexity of data processing and may affect the accuracy of model processing. Based on this, in a possible implementation manner, a feature selection algorithm can be used to select a preset number of indicators from the above-determined statistical feature indicators and personalized indicators, and then determine the input features based on the selected preset number of indicators. That is, some indicators in the statistical feature indicators and personalized indicators can be selected as the input features, which can not only ensure the diversity of features, but also appropriately reduce the data volume (the number of indicators), reduce the complexity of data processing, and improve the speed of model processing.

[0074] In a possible implementation manner, the feature selection algorithm can be the minimum redundancy maximum correlation method. The main idea of this method is to select features from the dataset that are highly correlated with the target variable but have low redundancy with each other. That is, select indicators from the statistical feature indicators and personalized indicators that are highly correlated with the exercise energy consumption data and have low redundancy among the selected indicators.

[0075] In a possible implementation manner, the moderate activity intensity assessment model can include an exercise energy consumption monitoring model and a moderate activity intensity monitoring and determination unit. Among them, the personalized indicators and acceleration signals of the preschool children to be monitored are input into the exercise energy consumption monitoring model, and the exercise energy consumption data of the preschool children to be monitored can be output. Moreover, the exercise energy consumption monitoring model is constructed by fusing the personalized indicators of multiple preschool children, the acceleration signals in different states, and the exercise energy consumption label values corresponding to the acceleration signals. That is, the exercise energy consumption monitoring model is trained using the personalized indicators of multiple preschool children, the acceleration signals in different states, and the exercise energy consumption label values corresponding to the acceleration signals.

[0076] The moderate activity intensity monitoring and determination unit is also used to compare the exercise energy consumption data with the moderate activity intensity threshold and output medium and high activity indication information.

[0077] In specific implementation, the acceleration signal is divided according to a preset time window to obtain first acceleration signals corresponding to multiple preset time windows respectively. Based on the first acceleration signal and the personalized index of each preset time window, the input feature corresponding to each preset time window is determined. The input feature corresponding to each preset time window is input into the motion energy consumption monitoring model to output the corresponding motion energy consumption data. The medium activity intensity monitoring and determination unit obtains the motion energy consumption data of each preset time window, and compares the motion energy consumption data with the medium activity intensity threshold to determine the target preset time window corresponding to the motion energy consumption data that meets the medium activity intensity threshold. The target preset time windows are counted to determine the daily average medium-intensity activity index.

[0078] To facilitate understanding of the implementation principle of the motion energy consumption monitoring model, the training process of the motion energy consumption monitoring model will be introduced first below.

[0079] To train the motion energy consumption monitoring model, it is first necessary to obtain sample data for training. Since the motion energy consumption monitoring model is designed to accurately monitor the motion energy consumption of preschool children, it is necessary to collect data related to the motion energy consumption of preschool children. The process of training the motion energy consumption monitoring model mainly includes the following steps A1 - A4: A1: Obtain the personalized index, three-axis acceleration signal, and the motion energy consumption label value corresponding to the three-axis acceleration signal of preschool children.

[0080] Among them, preschool children mainly include children aged 3 to 6 years old, and the specific age range for data collection can be determined according to actual needs. The collected samples can be made to cover as many age groups as possible, so as to improve the diversity of the samples and make the trained motion energy consumption monitoring model more accurate.

[0081] The personalized indicators for preschool children include: gender, age, height, weight, and body mass index, etc. The three-axis acceleration signals include the x-axis acceleration signal, the y-axis acceleration signal, and the z-axis acceleration signal, which can be used to characterize the motion state of preschool children. For example, a preschool child can be equipped with an acceleration measuring device to measure the three-axis acceleration signals of the preschool child in different motion states. And the preschool child wears a gas metabolism analyzer. While measuring the three-axis acceleration signals with the acceleration measuring device, the gas metabolism analyzer is used to measure the exercise energy consumption in different motion states as the exercise energy consumption label value, that is, the sample true value, so that each three-axis acceleration signal corresponds to an exercise energy consumption label value. Among them, the gas metabolism analyzer is measured by exhalation analysis, and the exercise energy consumption label value is calculated by measuring oxygen uptake, carbon dioxide output, etc. Among them, the personalized indicators and the three-axis acceleration signals can be used as inputs when training the model, and the exercise energy consumption label value is used as the sample true value. The training goal is to make the exercise energy consumption data output by the exercise energy consumption monitoring model closer to the exercise energy consumption label value.

[0082] Optionally, the exercise energy consumption label value may include energy expenditure (EE) or metabolic equivalent. When the obtained exercise energy consumption label value includes energy consumption, the trained exercise energy consumption monitoring model can output the energy consumption of the preschool child to be monitored. When the obtained exercise energy consumption label value includes metabolic equivalent, the trained exercise energy consumption monitoring model can output the metabolic equivalent of the preschool child to be monitored. That is, different exercise energy consumption monitoring models can be trained with different samples (exercise energy consumption label values) to monitor different exercise energy consumption data.

[0083] A2: Determine sample features based on the personalized indicators and three-axis acceleration signals of preschool children.

[0084] Since there may be incorrect records or missing data in the relevant data of the obtained preschool children, in order to simplify the model's data processing process, in a possible implementation, the collected data can be preprocessed first to obtain preprocessed personalized indicators and preprocessed three-axis acceleration signals. Then, based on the preprocessed personalized indicators and preprocessed three-axis acceleration signals, sample features are determined.

[0085] Specifically, interpolation can be performed on the missing values in the data. That is, when there are missing values in the personalized indicators or three-axis acceleration signals of the obtained preschool children, interpolation can be performed on the missing values to avoid affecting the subsequent data processing process. For example, the missing values in categorical data (such as boys or girls) are interpolated with "NaN" values, and the missing values in numerical data (such as age) are interpolated with the mean value.

[0086] Since most models cannot directly process categorical data, it is necessary to perform one-hot encoding on categorical data to convert it into numerical data. Among them, one-hot encoding is also known as one-hot effective encoding. Its processing method is to use an N-bit status register to encode N states. Each state has an independent register bit, with a value of 0 or 1, and at any time, only one bit is effective. For example, for the categorical data "boys" and "girls", since there are only two types in this category, so take N = 2, then "01" and "10" can be used to represent "boys" and "girls".

[0087] For continuous data, in order to simplify the data processing process, the continuous data can be normalized. Or, the normalized continuous data can also be converted into data with an approximate normal distribution to improve the data processing efficiency.

[0088] In a possible implementation, the personalized metrics and the three-axis acceleration signals can be fused as sample features.

[0089] In a possible implementation, the three-axis acceleration signals include the x-axis acceleration signal, the y-axis acceleration signal, and the z-axis acceleration signal. To improve the diversity and richness of the samples and enhance the ability of the training model, a one-dimensional vector sum signal can be determined based on the x-axis acceleration signal, the y-axis acceleration signal, and the z-axis acceleration signal. That is, calculate the sum of the squares of the x-axis acceleration signal, the y-axis acceleration signal, and the z-axis acceleration signal, and then calculate the arithmetic square root of the sum of squares as the one-dimensional vector sum signal.

[0090] According to the above embodiments, the three-axis acceleration signals of the collected preschool children may correspond to the motion states of multiple time periods, and the duration of each time period may be different. To unify the format of the sample features and facilitate the processing of the sample features, the obtained three-axis acceleration signals can be divided according to a preset time window to obtain the first acceleration signals corresponding to multiple preset time windows. That is, each preset time window corresponds to the first acceleration signal obtained by division. Among them, the first acceleration signal within each preset time window includes multiple x-axis acceleration signals, multiple y-axis acceleration signals, and multiple z-axis acceleration signals. For each preset time window, based on the first acceleration signal and the personalized metrics of this preset time window, the first sample feature corresponding to this preset time window can be determined.

[0091] Specifically, for any preset time window, based on multiple x-axis acceleration signals, multiple y-axis acceleration signals, multiple z-axis acceleration signals, and corresponding multiple one-dimensional vector sum signals within the preset time window, calculate the statistical feature indicators of the x-axis acceleration signals, the statistical feature indicators of the y-axis acceleration signals, the statistical feature indicators of the z-axis acceleration signals, and the statistical feature indicators of the one-dimensional vector sum signals respectively. Among them, the statistical feature indicators can represent mathematical statistical features. For example, they can include the maximum value, the minimum value, the mean value, the standard deviation, the variance, the percentile, the coefficient of variation, and so on. Then, based on the statistical feature indicators and the personalized indicators corresponding to the preset time window, determine the first sample feature corresponding to the preset time window. For example, the various statistical feature indicators and the personalized indicators can be fused to determine the first sample feature.

[0092] In a possible implementation manner, the statistical feature indicators of the x-axis acceleration signals at least include the maximum value of the multiple x-axis acceleration signals. The statistical feature indicators of the y-axis acceleration signals at least include the peak-to-peak value of the multiple y-axis acceleration signals, where the peak-to-peak value represents the difference between the maximum value and the minimum value. The statistical feature indicators of the z-axis acceleration signals at least include the minimum value of the multiple z-axis acceleration signals. The statistical feature indicators of the one-dimensional vector sum signals at least include the average value and the variance of the multiple one-dimensional vector sum signals.

[0093] It should be noted that in the embodiments of the present application, the specific types and numbers of the statistical feature indicators are not limited and can be set according to actual needs. In a possible application scenario, for the x-axis acceleration signals, the y-axis acceleration signals, the z-axis acceleration signals, and the one-dimensional vector sum signals, 91 statistical feature indicators including the maximum value, the minimum value, the mean value, the standard deviation, the variance, the percentile, the coefficient of variation, etc. can be determined. Then, in combination with 5 personalized indicators of gender, age, height, weight, and body mass index, the statistical feature indicators and the personalized indicators are fused to obtain 96 indicators to determine the first sample feature.

[0094] Although including more indicators in the sample feature can improve the diversity of the sample, it will also increase the complexity of data processing and the difficulty of training the motion energy consumption monitoring model, which may lead to overfitting in training and affect the accuracy of the motion energy consumption monitoring model. Based on this, in a possible implementation manner, a feature selection algorithm can be used to select a preset number of indicators from the above-determined statistical feature indicators and personalized indicators, and then determine the first sample feature based on the selected preset number of indicators. That is, some indicators in the statistical feature indicators and the personalized indicators can be selected as the first sample feature, which can not only ensure the diversity of the sample but also appropriately reduce the data volume (the number of indicators) of the sample, reduce the complexity of data processing, and improve the speed of model training.

[0095] For example, the feature selection algorithm can be the minimum redundancy maximum relevance method. The main idea of this method is to select features from the dataset that are highly correlated with the target variable but have low redundancy with each other. That is, select indicators that are highly correlated with exercise energy consumption from statistical feature indicators and personalized indicators, and the selected indicators have low redundancy with each other.

[0096] Optionally, the preset number of selected indicators can be determined through multiple experiments. That is, multiple different numbers can be determined in advance. After selecting multiple indicators for each number, it is determined as a sample feature. After training the exercise energy consumption monitoring model using the sample feature, judge which sample feature with which number of indicators results in a higher accuracy of the exercise energy consumption monitoring model, and then it can be determined as the preset number. Subsequently, when applying the exercise energy consumption monitoring model to determine the exercise energy consumption data of preschool children, the preset number of indicators can be selected to determine the sample feature, making the exercise energy consumption data predicted by the exercise energy consumption monitoring model more accurate.

[0097] Based on this, after determining the first sample feature corresponding to each preset time window, since there are multiple three-axis acceleration signals in this preset time window corresponding to multiple exercise energy consumption label values, at this time, it can be determined that the sample true value corresponding to this first sample feature is the average value of the multiple exercise energy consumption label values in this preset time window. That is, for each preset time window, obtain the multiple exercise energy consumption label values in this preset time window, and calculate the average exercise energy consumption label value of the multiple exercise energy consumption label values as the sample true value corresponding to the first sample feature. Thus, using the first sample feature as the input and the average exercise energy consumption label value corresponding to the first sample feature as the sample true value to train the exercise energy consumption monitoring model.

[0098] A3: Input the sample feature and the exercise energy consumption label value corresponding to the sample feature into a machine learning algorithm to obtain an exercise energy consumption prediction value.

[0099] After determining the sample feature for training, the sample feature can be used as the independent variable and the exercise energy consumption label value corresponding to the sample feature can be used as the target variable and input into a machine learning algorithm, and the machine learning algorithm outputs an exercise energy consumption prediction value.

[0100] In a possible implementation manner, the machine learning algorithm can include an extremely randomized tree regression algorithm, a random forest regression algorithm, a light gradient boosting machine (LGBM) algorithm, a CatBoost algorithm, or a linear regression algorithm.

[0101] A4: Train the machine learning algorithm based on the exercise energy consumption prediction value and the exercise energy consumption label value until the training termination condition is met, and obtain a trained exercise energy consumption monitoring model.

[0102] Among them, the training objective is to make the predicted value of the exercise energy consumption output by the exercise energy consumption monitoring model closer to the exercise energy consumption label value.

[0103] After obtaining the predicted value of the exercise energy consumption, since there is an exercise energy consumption label value corresponding to the triaxial acceleration signal in the sample features, that is, the true sample value, and the predicted value of the exercise energy consumption represents the predicted value of the machine learning algorithm, the machine learning algorithm can be trained based on the predicted value of the exercise energy consumption and the exercise energy consumption label value, and iteratively trained until the training cut-off condition is met. Among them, the training objective is to reduce the error between the predicted value of the exercise energy consumption and the exercise energy consumption label value.

[0104] In a specific implementation, the estimation error can be determined based on the predicted value of the exercise energy consumption and the exercise energy consumption label value, and this estimation error is used to represent the error between the predicted value of the exercise energy consumption and the exercise energy consumption label value. When the estimation error is larger, it indicates that the error between the predicted value of the exercise energy consumption and the exercise energy consumption label value is also larger, and the accuracy of the machine learning algorithm is lower. When the estimation error is greater than or equal to the preset value, it indicates that the error between the predicted value of the exercise energy consumption and the exercise energy consumption label value is large and does not meet the requirements. It is necessary to re-adjust the parameters of the machine learning algorithm and re-execute step A3 and step A4, that is, re-execute the process of inputting the sample features into the machine learning algorithm with adjusted parameters and the subsequent process of determining the estimation error until the training cut-off condition is met.

[0105] In a possible implementation, the training cut-off condition can be that the estimation error is less than the preset value, or the number of iterative training reaches the preset number (the number of times of adjusting the parameters of the machine learning algorithm reaches the preset number). At this time, the training process can be stopped to obtain the trained exercise energy consumption monitoring model.

[0106] In a possible implementation, the exercise energy consumption monitoring model can be trained using the k-fold cross-validation method. Among them, the main principle of the k-fold cross-validation method is to randomly divide the sample data into k parts, where k - 1 parts are used as the training set and the remaining 1 part is used as the test set. When training and testing, the training set and the test set corresponding to the training set are selected in turn. For example, the ten-fold cross-validation method is to divide the sample data into 10 sub-samples, and a single sub-sample is used as the test set to verify the model, and the remaining 9 sub-samples are used for training. The cross-validation is repeated 10 times, and each sub-sample is used as the test set to verify once. That is, the sample features can be divided into a training set and a test set, and the training set is used to train the exercise energy consumption monitoring model, and the test set is used to test the exercise energy consumption monitoring model. Then, the mean value of the results obtained from the 10 tests is calculated as the final test result.

[0107] When testing the motion energy consumption monitoring model, the test set is input into the motion energy consumption monitoring model to obtain the motion energy consumption test value. Based on the motion energy consumption test value and the motion energy consumption label value corresponding to the test set, the test error is determined. Optionally, the root mean square error (RMSE) between the motion energy consumption test value and the motion energy consumption label value can be calculated as the test error. For example, when using the ten-fold cross-validation method for training, after each test process outputs the motion energy consumption test value, the difference between the motion energy consumption test value and the motion energy consumption label value can be calculated. Then, the mean of the squares of the ten differences is calculated, and the arithmetic square root of the mean of the squares is calculated as the test error. The ten-fold cross-validation method can obtain 10 test errors, and the average of these 10 test errors is calculated as the final test result of the motion energy consumption monitoring model.

[0108] Based on the embodiments of determining sample features introduced in step A2 above, it can be known that the obtained three-axis acceleration signal can be divided according to a preset time window, and the corresponding first sample feature is determined for each preset time window. And the average value of the multiple motion energy consumption label values (average motion energy consumption label value) within the preset time window is calculated as the sample true value corresponding to the first sample feature. Then, the first sample feature corresponding to each preset time window is used as the independent variable, and the average motion energy consumption label value corresponding to the first sample feature is used as the target variable and input into the machine learning algorithm to obtain the first motion energy consumption prediction value corresponding to each preset time window. Then, based on the first motion energy consumption prediction value and the average motion energy consumption label value corresponding to each preset time window, the machine learning algorithm is trained to obtain a trained motion energy consumption monitoring model.

[0109] Among them, the length of the preset time window for division may affect the accuracy of training the motion energy consumption monitoring model. Therefore, multiple different time windows can be determined in advance, and for each time window, a process of training the motion energy consumption monitoring model is completed, and the accuracy of the motion energy consumption monitoring model is tested. The time window corresponding to the motion energy consumption monitoring model with the highest accuracy is determined as the preset time window. Subsequently, when applying the motion energy consumption monitoring model to monitor the motion of preschool children, the acceleration signal is divided according to the preset time window.

[0110] In specific implementation, multiple different time windows are first determined. Among them, the different lengths of the multiple time windows can be divided in combination with experience and actual requirements. For example, the length of the time window can be set to 15 seconds, 30 seconds, 60 seconds, etc. For the first time window, the obtained three-axis acceleration signal is divided according to the first time window, so as to obtain second acceleration signals corresponding to multiple first time windows respectively. Among them, the first time window represents any one of the multiple different time windows. Based on the second acceleration signal and the personalized index, the second sample feature corresponding to the first time window is determined. Based on the second sample features corresponding to multiple first time windows and the average motion energy consumption label value corresponding to the second sample feature of each first time window, a training set and a test set are determined. For example, the second sample features of all first time windows and the average motion energy consumption label value corresponding to the second sample feature of each first time window can be fused to obtain a data set, and then the data set is divided into a training set and a test set according to a preset ratio. For example, the fused data set can be divided into a training set and a test set according to 8:2. Among them, the training set includes multiple second sample features and the average motion energy consumption label value corresponding to each second sample feature, and the test set also includes multiple second sample features and the average motion energy consumption label value corresponding to each second sample feature.

[0111] Then, the training set is input into the machine learning algorithm to obtain the predicted value of motion energy consumption output by the machine learning algorithm. The machine learning algorithm is trained based on the predicted value of motion energy consumption and the average motion energy consumption label value corresponding to the first time window until the training cut-off condition is met, and a trained motion energy consumption monitoring model is obtained. The process of determining the average motion energy consumption label value corresponding to the first time window can refer to the above embodiments and will not be elaborated here.

[0112] After obtaining the trained motion energy consumption monitoring model, the accuracy of the motion energy consumption monitoring model can be verified by using the test set. Specifically, based on the second sample features corresponding to multiple first time windows, a test set is determined. For example, a certain proportion is divided from the second sample features of multiple first time windows as the test set. The test set is input into the motion energy consumption monitoring model to obtain the test value of motion energy consumption. Based on the test value of motion energy consumption and the motion energy consumption label value corresponding to the test set, the test error is determined. The above process is executed for multiple different time windows, and then the test error corresponding to each time window can be determined. Based on the test errors corresponding to multiple different time windows, the minimum test error is determined, so that the time window corresponding to the minimum test error can be determined as the preset time window. Subsequently, when using the motion energy consumption monitoring model to monitor the motion energy consumption of preschool children, the acceleration signal can be divided according to the preset time window.

[0113] Among them, the method for determining the test error can be to calculate the root mean square error between the measured value of the motion energy consumption and the motion energy consumption label value corresponding to the test set, or to calculate the mean absolute error between the measured value of the motion energy consumption and the motion energy consumption label value corresponding to the test set, etc. The embodiments of the present application do not limit this.

[0114] After multiple training and testing processes, when the preset time window is determined to be 15 seconds in the embodiments of the present application, the accuracy of the trained motion energy consumption monitoring model is relatively high.

[0115] According to the above embodiments, the three-axis acceleration signal can include the x-axis acceleration signal, the y-axis acceleration signal, the z-axis acceleration signal, and their vector sum signal, and the statistical feature indexes of the x-axis acceleration signal, the statistical feature indexes of the y-axis acceleration signal, the statistical feature indexes of the z-axis acceleration signal, and the statistical feature indexes of the vector sum signal can be determined respectively. When determining the sample features through each statistical feature index and the personalized index, since there are a relatively large number of indexes in the sample features, it will increase the complexity of data processing and affect the accuracy of training the motion energy consumption monitoring model. Therefore, some of the indexes can be selected from each statistical feature index and the personalized index to train the motion energy consumption monitoring model. The accuracy of the trained motion energy consumption monitoring model is different when the number of selected indexes is different. Based on this, multiple different numbers can be determined in advance to select indexes, determine the sample features for each number of indexes to train the motion energy consumption monitoring model, and test the accuracy of the motion energy consumption monitoring model, so as to select the number of indexes corresponding to the motion energy consumption monitoring model with the highest accuracy as the preset number.

[0116] In specific implementation, using the feature selection algorithm, select indexes with multiple different numbers from the statistical feature indexes and the personalized index, and determine the sample features corresponding to the indexes with multiple different numbers. For the third sample feature corresponding to the indexes with the first number, determine the training set and the test set. Here, the first number represents any one of the multiple different numbers. That is, for any first number, the feature selection algorithm can be used to select the indexes with the first number from the statistical feature indexes and the personalized index, and fuse the selected indexes to determine the third sample feature. Fuse the third sample feature and the motion energy consumption label value corresponding to the third sample feature, and then divide the fused data set into a training set and a test set. Use the training set to train the machine learning algorithm to obtain the motion energy consumption monitoring model. The specific training process will not be elaborated here.

[0117] Then, input the test set into the motion energy consumption monitoring model to obtain the motion energy consumption test value. Based on the motion energy consumption test value and the motion energy consumption label value corresponding to the test set, determine the test error. Based on the test errors corresponding to multiple different numbers of metrics respectively, determine the minimum test error. Determine the number corresponding to the minimum test error as the preset number. Subsequently, when applying the motion energy consumption monitoring model to monitor motion energy consumption, select the preset number of metrics as the features for inputting into the motion energy consumption monitoring model from the statistical feature metrics and personalized metrics.

[0118] The motion energy consumption label value in the embodiment of the present application includes energy consumption or metabolic equivalent, and two motion energy consumption monitoring models can be trained for energy consumption and metabolic equivalent respectively. In order to obtain a motion energy consumption monitoring model for monitoring energy consumption, after obtaining the personalized metrics, three-axis acceleration signals, and energy consumption label values of preschool children, multiple numbers of metrics can be determined to form sample features, and then the motion energy consumption monitoring model can be trained. After completing the training and testing processes using multiple numbers of metrics, the embodiment of the present application determines that when the sample features include 80 metrics, the motion energy consumption monitoring model obtained has a relatively high accuracy in monitoring energy consumption.

[0119] Similarly, in order to obtain a motion energy consumption monitoring model for monitoring metabolic equivalent, after obtaining the personalized metrics, three-axis acceleration signals, and metabolic equivalent label values of preschool children, multiple numbers of metrics can be determined to form sample features, and then the motion energy consumption monitoring model can be trained. After completing the training and testing processes using multiple numbers of metrics, the embodiment of the present application determines that when the sample features include 96 metrics, the motion energy consumption monitoring model obtained has a relatively high accuracy in monitoring metabolic equivalent.

[0120] Table 1

[0121] Specifically, as shown in Table 1 above, it is an example for determining sample features. Among them, the total number of samples represents the number of sample features. Whether feature selection is performed indicates whether only some metrics are selected to form sample features from all the statistical feature metrics and personalized metrics.

[0122] According to the above embodiments, in order to enable the trained motion energy consumption monitoring model to more accurately monitor the motion energy consumption of preschool children, multiple machine learning algorithms can be used to train the motion energy consumption monitoring model. For example, multiple machine learning algorithms commonly used for data modeling in this field may include the extremely randomized tree regression algorithm, random forest regression algorithm, LGBM, CatBoost algorithm, or linear regression algorithm. By first training these multiple motion energy consumption monitoring models and testing their accuracy, select the most accurate motion energy consumption monitoring model among these multiple motion energy consumption monitoring models.

[0123] In specific implementation, the sample features can be divided into a training set and a test set. For each motion energy consumption monitoring model, the training set is used to train the motion energy consumption monitoring model to obtain a trained motion energy consumption monitoring model. Among them, the specific training process can refer to the above embodiments and will not be elaborated here. After obtaining the trained motion energy consumption monitoring model, the test set is input into the corresponding motion energy consumption monitoring model of the motion energy consumption monitoring model to obtain motion energy consumption test values. Based on the motion energy consumption test values and the motion energy consumption label values corresponding to the test set, the test error is determined. Since each machine learning algorithm-trained corresponding machine learning model corresponds to a test error, as the accuracy test result of the motion energy consumption monitoring model, the minimum test error can be determined based on the test errors corresponding to multiple motion energy consumption monitoring models respectively, and the motion energy consumption monitoring model corresponding to the minimum test error is determined as the preferred motion energy consumption monitoring model. That is, the motion energy consumption monitoring model corresponding to the minimum test error is used as the motion energy consumption monitoring model for subsequent monitoring of preschool children.

[0124] Among them, the method for determining the test error can be to calculate the root mean square error between the motion energy consumption test value and the motion energy consumption label value, or calculate the mean absolute error between the motion energy consumption test value and the motion energy consumption label value, etc. The embodiments of the present application do not limit this.

[0125] In the embodiments of the present application, the 5 motion energy consumption monitoring models obtained by training are respectively tested. It can be known that among the 5 motion energy consumption monitoring models, the motion energy consumption monitoring model obtained by training with the extremely randomized tree regression algorithm has the smallest test error and the highest accuracy. Therefore, when subsequently monitoring the motion energy consumption of preschool children, the motion energy consumption monitoring model obtained by training with the extremely randomized tree regression algorithm can be selected for monitoring.

[0126] In the embodiments of the present application, the Mets of the preschool children to be monitored estimated by the trained motion energy consumption monitoring model can also be compared with the Mets of the preschool children to be monitored measured in advance to test the accuracy of the motion energy consumption monitoring model. For specific reference, see Figure 2As shown, it is a schematic diagram of the test results of a motion energy consumption monitoring model provided by an embodiment of the present application. The embodiment of the present application can train five motion energy consumption monitoring models based on five algorithms. Among them, et represents the extremely randomized tree regression model, catboost represents the CatBoost regression model, lightgbm represents the light gradient boosting machine model, rf represents the random forest regression model, and lr represents the linear regression model. Then, the Mets of the preschool children to be monitored predicted by the five motion energy consumption monitoring models are respectively compared with the Mets of the preschool children to be monitored measured in advance. The five comparison test results are respectively shown in (a), (b), (c), (d), and (e). According to Figure 2 It can be seen that the Mets predicted by the extremely randomized tree regression model are the closest to the measured Mets.

[0127] After introducing the above process of training the motion energy consumption monitoring model, the motion energy consumption of preschool children can be monitored by using the moderate activity intensity evaluation model, where the moderate activity intensity evaluation model includes the motion energy consumption monitoring model. According to the above embodiment, the motion energy consumption monitoring model can be trained by the extremely randomized tree regression algorithm.

[0128] In a possible implementation manner, the moderate activity intensity evaluation model may include a motion energy consumption monitoring model and a moderate activity intensity monitoring and determination unit.

[0129] In a possible implementation manner, after obtaining the personalized indicators and acceleration signals of the preschool children to be monitored, the acceleration signals can be divided according to a preset time window to obtain first acceleration signals corresponding to multiple preset time windows respectively. For each preset time window, based on the first acceleration signal and personalized indicators of the preset time window, the input features corresponding to the preset time window are determined. Then, the input features corresponding to each preset time window are input into the motion energy consumption monitoring model to obtain motion energy consumption data corresponding to each preset time window respectively.

[0130] It should be noted that the determination method of the preset time window can refer to the preset time window determined in the above training process. For example, the length of the preset time window can be set to 15 seconds, and the specific process will not be elaborated here.

[0131] In a possible implementation, when the exercise energy consumption data includes metabolic equivalents, the moderate activity intensity monitoring and determination unit can obtain the metabolic equivalents corresponding to each preset time window, compare the metabolic equivalents with the moderate activity intensity threshold, and determine the target preset time window corresponding to the metabolic equivalents that meet the moderate activity intensity threshold. Then, all eligible target preset time windows are counted, so that the daily average moderate-to-high activity index of the preschool children to be monitored can be determined. That is, the length of time that the preschool children to be monitored are in the moderate activity intensity.

[0132] In a possible implementation, the acceleration signal can include the x-axis acceleration signal, the y-axis acceleration signal, and the z-axis acceleration signal. When determining the input features, the one-dimensional vector sum signal of the x-axis acceleration signal, the y-axis acceleration signal, and the z-axis acceleration signal can be determined first. Then, the statistical feature indexes of the x-axis acceleration signal, the y-axis acceleration signal, the z-axis acceleration signal, and the one-dimensional vector sum signal are determined respectively.

[0133] Optionally, after dividing the acceleration signal by using the preset time window, for each preset time window, based on the multiple x-axis acceleration signals, multiple y-axis acceleration signals, multiple z-axis acceleration signals, and multiple one-dimensional vector sum signals within the preset time window, the statistical feature indexes of the x-axis acceleration signal, the y-axis acceleration signal, the z-axis acceleration signal, and the one-dimensional vector sum signal corresponding to the preset time window are calculated respectively. Then, based on the statistical feature indexes and the personalized indexes, the input features are determined. For example, the statistical feature indexes and the personalized indexes can be fused to determine the input features.

[0134] When there are a relatively large number of indexes in the sample features, it will increase the complexity of data processing. In a possible implementation, a feature selection algorithm can be used to select a preset number of indexes from the statistical feature indexes and the personalized indexes, and then the input features are determined based on the selected preset number of indexes.

[0135] Currently, on the premise of determining the physical activity intensity level according to the value of METs, the exercise energy consumption range of moderate-intensity activities is generally recognized as 3 - 6 Mets. However, the objects for data collection and testing in this range are only applicable to adolescents and adult groups. Therefore, 20 preschool children samples aged 3 to 6 were randomly selected from the group of preschool children participating in the training exercise energy consumption monitoring model, and the feasibility of this method was verified for these 20 children. When we use the more accurate moderate activity intensity threshold ranges for Chinese preschool children aged 3, 4, 5, and 6 proposed by the present invention to determine the daily physical activity intensity of preschool children, rather than using the moderate activity intensity threshold ranges for Chinese children and adolescents that have been proposed, the accuracy of the assessment of the daily moderate activity intensity of this group of preschool children and the activity duration of the corresponding moderate-intensity activities has been improved. In particular, it reduces the error when using the Mets threshold range to count the cumulative duration of Chinese preschool children's daily moderate activity intensity, and avoids the error caused by directly using the Mets threshold range corresponding to the moderate activity intensity used by Chinese children, adolescents, and adults.

[0136] The exercise energy consumption monitoring method based on the accelerometer sensor monitors the testers to estimate the Mets of preschool children. Then, combined with the moderate activity intensity threshold range, the daily average moderate-intensity activity index is obtained through automated monitoring and determination. Finally, according to the exercise schedule recommended in the domestic preschool children (aged 3 - 6) exercise guidelines led by this team, which states that "preschool children should accumulate no less than 60 minutes of moderate and above-intensity exercise within 24 hours a day", a conclusion is further given on whether the duration of moderate-intensity exercise meets the requirements.

[0137] It should be noted that the method for determining the preset number can refer to the preset number determined in the above training process, and when the monitored exercise energy consumption data is different, the number of indicators in the input features corresponding to different exercise energy consumption monitoring models is also different. For example, when it is necessary to obtain the energy consumption of the preschool children to be monitored, 80 indicators can be selected from the statistical feature indicators and personalized indicators to determine the input features, and input them into the exercise energy consumption monitoring model for monitoring energy consumption to output the energy consumption of the preschool children to be monitored. When it is necessary to obtain the metabolic equivalent of the preschool children to be monitored, 96 indicators can be selected from the statistical feature indicators and personalized indicators to determine the input features, and input them into the exercise energy consumption monitoring model for monitoring the metabolic equivalent to output the metabolic equivalent of the preschool children to be monitored.

[0138] Through the method provided by the embodiments of the present application, a pre-trained medium activity intensity evaluation model can be used to monitor the exercise energy consumption of preschool children, determine the medium-intensity activity indication information for the daily evaluation activities of preschool children, and improve the accuracy of exercise energy consumption monitoring.

[0139] Based on the above method embodiments, the embodiments of the present application also provide a medium activity intensity monitoring system for preschool children. Refer to Figure 3 As shown, it is a schematic diagram of a medium activity intensity monitoring system for preschool children provided by the embodiments of the present application.

[0140] The medium activity intensity monitoring system 400 for preschool children includes: a data preprocessing module 401, a data modeling module 402, and an exercise energy consumption monitoring module 403; Among them, the data preprocessing module 401 is used to obtain the personalized indicators of preschool children and the acceleration signals in the state to be monitored, and process the personalized indicators and acceleration signals to determine the training set and the test set; It is also used to obtain the personalized indicators of the preschool children to be monitored and the acceleration signals in the state to be monitored, and process the personalized indicators and acceleration signals to determine the input features.

[0141] The data modeling module 402 is used to train multiple exercise energy consumption monitoring models using the training set, and test the trained multiple exercise energy consumption monitoring models using the test set. Select the optimal exercise energy consumption monitoring model according to the test results of the multiple exercise energy consumption monitoring models. Thus, a medium activity intensity evaluation model is determined based on the exercise energy consumption monitoring model.

[0142] The exercise energy consumption monitoring module 403 is used to monitor the exercise energy consumption of the preschool children to be monitored using the medium activity intensity evaluation model, and output the medium-intensity activity indication information.

[0143] Optionally, the exercise energy consumption monitoring module 403 can also visualize the exercise energy consumption data or the medium-intensity activity indication information of the preschool children to be monitored. For example, it can be organized into formats such as tables and bar charts for display.

[0144] Among them, the specific implementation principles of the data preprocessing module 401, the data modeling module 402, and the exercise energy consumption monitoring module 403 can refer to the above method embodiments, and will not be elaborated here.

[0145] Based on the above method embodiments and system embodiments, the embodiments of the present application also provide a medium activity intensity monitoring device for preschool children. Refer to Figure 4 As shown, it is a schematic diagram of a medium activity intensity monitoring device for preschool children provided by the embodiments of the present application.

[0146] The device 500 includes: An acquisition unit 501, configured to acquire personalized indicators of a preschool child to be monitored and an acceleration signal in a state to be monitored; A monitoring and determination unit 502, configured to input the personalized indicators and the acceleration signal into a moderate activity intensity evaluation model, and output medium-strong activity indication information based on a moderate activity intensity threshold, where the moderate activity intensity threshold range for preschool children is 3.2 - 5.3 Mets; Wherein, the moderate activity intensity evaluation model is constructed by fusing personalized indicators of multiple preschool children, acceleration signals in different states, and the corresponding motion energy consumption label values of the acceleration signals.

[0147] In a possible implementation manner, the moderate activity intensity threshold range for 4 - 6-year-old preschool children is 3.4 - 5.1 Mets.

[0148] In a possible implementation manner, the monitoring and determination unit 502 is specifically configured to divide the acceleration signal according to a preset time window to obtain first acceleration signals corresponding to multiple preset time windows respectively; determine input features corresponding to each preset time window based on the first acceleration signal of each preset time window and the personalized indicators; input the input features into the moderate activity intensity evaluation model, and output the medium-strong activity indication information based on the moderate activity intensity threshold.

[0149] In a possible implementation manner, the medium-strong activity indication information includes a daily average medium-strong activity index.

[0150] In a possible implementation manner, the monitoring and determination unit 502 is specifically configured to input the input features corresponding to each preset time window into the moderate activity intensity evaluation model to obtain motion energy consumption data corresponding to each preset time window; compare the motion energy consumption data with the moderate activity intensity threshold to determine a target preset time window corresponding to the motion energy consumption data that meets the moderate activity intensity threshold; count the target preset time windows to determine the daily average medium-strong activity index.

[0151] In a possible implementation manner, the first acceleration signal includes an x-axis acceleration signal, a y-axis acceleration signal, and a z-axis acceleration signal. The monitoring and determination unit 502 is specifically configured to determine a one-dimensional vector sum signal of the x-axis acceleration signal, the y-axis acceleration signal, and the z-axis acceleration signal; respectively determine statistical feature indexes of the x-axis acceleration signal, the y-axis acceleration signal, the z-axis acceleration signal, and the one-dimensional vector sum signal; determine input features corresponding to each preset time window based on the statistical feature indexes corresponding to each preset time window and the personalized indicators.

[0152] In a possible implementation manner, the monitoring and determination unit 502 is specifically configured to use a feature selection algorithm to select a preset number of indicators from the statistical feature indicators and the personalized indicators; and determine the input features based on the preset number of indicators.

[0153] In a possible implementation manner, the acceleration signal is measured by an acceleration measuring instrument worn by the preschool child to be monitored, and the acceleration measuring instrument is worn on the right waist.

[0154] In a possible implementation manner, the wearing period of the acceleration measuring instrument is at least 1 day.

[0155] In a possible implementation manner, the medium activity intensity threshold range corresponding to 3-year-old preschool children is 3.2 - 4.4 Mets.

[0156] In a possible implementation manner, the medium activity intensity threshold range corresponding to 4-year-old preschool children is 3.4 - 4.6 Mets.

[0157] In a possible implementation manner, the medium activity intensity threshold range corresponding to 5-year-old preschool children is 3.7 - 4.8 Mets.

[0158] In a possible implementation manner, the medium activity intensity threshold range corresponding to 6-year-old preschool children is 3.8 - 5.1 Mets.

[0159] In a possible implementation manner, the age of the preschool child is calculated according to the full years corresponding to the actual date of birth.

[0160] In a possible implementation manner, the obtaining unit 501 is specifically configured to obtain the original personalized indicators of the preschool child to be monitored and the original acceleration signal in the state to be monitored; perform preprocessing on the original personalized indicators and the original acceleration signal to obtain the personalized indicators and the acceleration signal; wherein, the preprocessing process includes one or more of the following: imputing missing values; or, performing one-hot encoding processing on categorical data; or, performing normalization processing on continuous data; or, performing normalization processing on continuous data and converting the data after normalization processing into data with an approximate normal distribution.

[0161] In a possible implementation manner, the personalized indicators include at least one of gender, age, height, weight, and body mass index.

[0162] Based on the above method embodiments and apparatus embodiments, an embodiment of the present application further provides an electronic device. This will be introduced below with reference to the accompanying drawings.

[0163] See Figure 5 , Figure 5 which is a schematic diagram of an electronic device provided by an embodiment of the present application.

[0164] The device 600 includes: a memory 601 and a processor 602; The memory 601 is used to store relevant program codes; The processor 602 is used to call the program codes to execute the monitoring method for moderate activity intensity of preschool children described in the above method embodiment.

[0165] In addition, an embodiment of the present application also provides a computer-readable storage medium, which is used to store a computer program, and the computer program is used to execute the monitoring method for moderate activity intensity of preschool children described in the above method embodiment.

[0166] An embodiment of the present application also provides a computer program product, which includes computer programs / instructions, and when the computer programs / instructions are executed by a processor, the monitoring method for moderate activity intensity of preschool children described in the above method embodiment is implemented.

[0167] It should be noted that the above computer-readable medium of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0168] The computer program product can be written in any combination of one or more programming languages to write program codes for performing the operations of the embodiments of the present application. The programming languages include object-oriented programming languages such as Java, C++, etc., and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program codes can be executed completely on the user computing device, partially on the user device, executed as an independent software package, partially on the user computing device and partially on a remote computing device, or completely on a remote computing device or server.

[0169] It should be noted that the various embodiments in this specification are described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts among the embodiments, reference can be made to each other. In particular, for system or device embodiments, since they are basically similar to method embodiments, the description is relatively simple. For relevant parts, reference can be made to the corresponding descriptions in the method embodiments. The device embodiments described above are merely illustrative. The units or modules described as separate components may or may not be physically separated. The components shown as units or modules may or may not be physical modules, that is, they may be located in one place or distributed to multiple network units. Some or all of the units or modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement this without creative efforts.

[0170] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of methods, devices, and equipment according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and this module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0171] It should be understood that in the present application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that there can be three relationships. For example, "A and / or B" can represent: only A exists, only B exists, and both A and B exist at the same time. Here, A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after. "At least one (one)" or its similar expression below refers to any combination of these items, including any combination of single item (one) or plural items (ones). For example, at least one (one) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0172] It should also be noted that in this application, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0173] The steps of the methods or algorithms described in connection with the embodiments disclosed in this application can be implemented directly in hardware, in software modules executed by a processor, or in a combination of both. The software modules can be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0174] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined in this application can be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed in this application.

Claims

1. A method for monitoring the intensity of moderate activities of preschool children, characterized in that: The method comprises: Obtaining personalized indicators of the preschool children to be monitored and acceleration signals in the state to be monitored; The personalized index and the acceleration signal are input into a moderate activity intensity assessment model, and the output result is output by a determination system of moderate-strong activity indication information based on a moderate activity intensity threshold, wherein the moderate activity intensity threshold range for preschool children aged 3-6 is 3.2-5.3 Mets; wherein the moderate activity intensity assessment model is constructed based on a fusion of personalized indicators of multiple preschool children, acceleration signals in different states, and motion energy consumption label values ​​corresponding to the acceleration signals.

2. The method according to claim 1, characterized in that The step of inputting the personalized index and the acceleration signal into a moderate activity intensity assessment model, and outputting the moderate to strong activity indication information by a determination system based on a moderate activity intensity threshold, includes: Dividing the acceleration signal according to preset time windows to obtain first acceleration signals corresponding to a plurality of preset time windows respectively; Determine the input feature corresponding to each preset time window based on the first acceleration signal of each preset time window and the personalized index; The input feature is input into the moderate activity intensity assessment model, and according to the output result of the assessment model, the determination system outputs the moderate-strong activity indication information based on the moderate activity intensity threshold.

3. The method according to claim 2, characterized in that The moderate-to-vigorous activity indication information includes the average daily moderate-to-vigorous activity index of preschool children.

4. The method according to claim 3, characterized in that: The inputting of the input feature into the moderate activity intensity assessment model, and the outputting of the moderate to strong activity indication information by the determination system based on the moderate activity intensity threshold, comprises: Inputting the input features corresponding to each preset time window into the moderate activity intensity assessment model to obtain the exercise energy consumption data corresponding to each preset time window; The determination system compares the exercise energy consumption data with the moderate activity intensity threshold, and determines a target preset time window corresponding to the exercise energy consumption data that meets the moderate activity intensity threshold; The number of the target preset time windows is counted, and the daily average moderate to strong activity index is determined in combination with the length of the time windows.

5. The method according to claim 3, characterized in that: The first acceleration signal includes an x-axis acceleration signal, a y-axis acceleration signal, and a z-axis acceleration signal. The determining of the input feature corresponding to each preset time window based on the first acceleration signal of each preset time window and the personalized index includes: Determine a one-dimensional vector sum signal of the x-axis acceleration signal, the y-axis acceleration signal, and the z-axis acceleration signal; respectively determining statistical characteristic indicators of the x-axis acceleration signal, the y-axis acceleration signal, the z-axis acceleration signal, and the one-dimensional vector and signal; Based on the statistical feature index corresponding to each preset time window and the personalized index, the input feature corresponding to each preset time window is determined.

6. The method according to claim 5, characterized in that The determining the input feature based on the statistical feature indicator and the personalized indicator includes: Using a feature selection algorithm, selecting a preset number of indicators from the statistical feature indicators and the personalized indicators; The input feature is determined based on the preset number of indicators.

7. The method according to claim 1, characterized in that The acceleration signal is measured by the preschool child to be monitored wearing an accelerometer, where the accelerometer is worn on the right waist.

8. The method according to claim 7, characterized in that The accelerometer is worn for at least 1 day.

9. The method according to claim 1, characterized in that The moderate activity intensity threshold range for 3-year-old preschoolers is 3.2-4.4 Mets.

10. The method according to claim 1, characterized in that The moderate activity intensity threshold range for 4-year-old preschoolers is 3.4-4.6 Mets.

11. The method according to claim 1, characterized in that The moderate activity intensity threshold range for 5-year-old preschoolers is 3.7-4.8 Mets.

12. The method according to claim 1, characterized in that The corresponding moderate activity intensity threshold range for 6-year-old preschoolers is 3.8-5.1 Mets.

13. The method according to claim 1, characterized in that The age of preschool children is calculated based on the age corresponding to their actual date of birth.

14. The method according to any one of claims 1 to 13, characterized in that: The step of obtaining the personalized index of the preschool child to be monitored and the acceleration signal in the state to be monitored includes: Acquiring the original personalized index of the preschool child to be monitored and the original acceleration signal in the state to be monitored; Preprocessing the original personalized index and the original acceleration signal to obtain the personalized index and the acceleration signal; The pretreatment process includes one or more of the following: impute missing values; or, One-hot encode categorical data; or, Normalize continuous data; or, Normalize the continuous data and convert the normalized data into data with approximate normal distribution.

15. The method according to any one of claims 1 to 13, characterized in that The personalized indicator includes at least one of gender, age, height, weight and body mass index.

16. A device for monitoring the intensity of moderate activities of preschool children, characterized in that: The device comprises: An acquisition unit, used for acquiring personalized indicators of the preschool child to be monitored and an acceleration signal in a state to be monitored; A monitoring and determination unit, configured to input the personalized index and the acceleration signal into a moderate activity intensity assessment model; wherein the determination system therein outputs moderate to strong activity indication information based on a moderate activity intensity threshold, wherein the moderate activity intensity threshold range for preschool children aged 3-6 is 3.2-5.3 Mets; Among them, the moderate activity intensity assessment model is constructed based on the fusion of personalized indicators of multiple preschool children, acceleration signals under different states, and motion energy consumption label values ​​corresponding to the acceleration signals.

17. An electronic device, characterized in that: The device comprises: a memory and a processor; The memory is used to store relevant program codes; The processor is used to call the program code to execute the method for monitoring moderate activity intensity for preschool children as described in any one of claims 1 to 15.

18. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and the computer program is used to execute the method for monitoring moderate activity intensity for preschool children as described in any one of claims 1 to 15.

Images