An integrated growth and development assessment and management system for children

Abstract

The present invention discloses a children's integrated growth and development assessment and management system, specifically relating to the technical field of growth and development assessment; by collecting children's physiological parameters and performing dynamic time series analysis based on machine learning, the growth and development stages of children are divided; the dynamic time warping algorithm is used to process the time series data of children's limb proportions to evaluate the synchrony of bone growth and soft tissue development; by integrating the growth and development stages and synchrony, it is identified whether a child has a risk of developmental deviation; for children with a risk of developmental deviation, combined with genetic information, a genetic algorithm and a dynamic programming model are used to evaluate the individual growth potential of children; the synchrony of children's bone growth and soft tissue development and the individual growth potential and developmental adaptability of children are analyzed in association to evaluate the probability of potential developmental abnormalities and determine whether to initiate personalized intervention measures, realizing comprehensive monitoring and risk warning of children's growth and development status, and having scientificity and practicality.

Description

Technical Field

[0001] The present invention relates to the technical field of growth and development assessment, and more specifically, the present invention relates to a children's integrated growth and development assessment management system. Background Art

[0002] At present, the management of children's growth and development mainly relies on regular physical examinations and doctors' experience judgments, lacking real-time monitoring and accurate risk warning means. In the prior art, multi-modal physiological data, limb proportion information, and genetic information are collected separately, making it difficult to effectively integrate and deeply analyze, resulting in the difficulty of timely detection of potential developmental abnormalities in children. Traditional data processing methods cannot comprehensively capture the dynamic changes and stage characteristics in the growth process of children, and it is difficult to accurately evaluate the synchrony of bone growth and soft tissue development and individual growth potential.

[0003] To solve the above problems, a technical solution is provided now. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a children's integrated growth and development assessment management system to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A children's integrated growth and development assessment management system, including a development stage determination module, a synchrony assessment module, a deviation risk identification module, a growth potential assessment module, and an abnormal probability assessment module;

[0007] The development stage determination module collects real-time multi-modal physiological parameter data of children, uses a dynamic time series analysis algorithm based on machine learning to divide the growth and development status of children, and determines the growth and development stage of children;

[0008] The synchrony assessment module performs a dynamic time warping algorithm on the continuously collected time series data of children's limb proportions to evaluate the synchrony of children's bone growth and soft tissue development;

[0009] The deviation risk identification module, based on the growth and development stage of children and the synchrony of children's bone growth and soft tissue development, uses a decision tree to identify whether there is a development deviation risk in children;

[0010] For children with a development deviation risk, the growth potential assessment module analyzes the genetic information of children and uses a genetic algorithm combined with a dynamic programming model to evaluate the individual growth potential of children;

[0011] The abnormal probability assessment module performs correlation analysis on the synchrony of children's skeletal growth and soft tissue development and the individual growth potential of children, evaluates the probability of potential developmental abnormalities in children, and determines whether to initiate personalized intervention measures.

[0012] In a preferred embodiment, real-time multi-modal physiological parameter data of children is collected, and a dynamic time series analysis algorithm based on machine learning is used to divide the growth and development status of children to determine the growth and development stage of children. Specifically:

[0013] Collect real-time multi-modal physiological data of children;

[0014] Clean the multi-modal physiological data sequence of children and achieve time series synchronization correction;

[0015] Construct a dynamic time series model based on machine learning to extract key features and patterns;

[0016] Use the dynamic time series analysis algorithm to divide the growth status and output the growth stage division index

[0017] Determine the growth and development stage of children according to the growth stage division index.

[0018] In a preferred embodiment, the dynamic time warping algorithm is executed on the continuously collected time series data of children's limb proportions to evaluate the synchrony of children's skeletal growth and soft tissue development. Specifically:

[0019] Obtain continuous time series data of children's limb proportions, covering skeletal and soft tissue parameters;

[0020] Use the dynamic time warping algorithm to align the time series data;

[0021] Calculate the difference index between the skeletal growth rate and the soft tissue development rate;

[0022] Evaluate the synchrony of children's skeletal growth and soft tissue development based on the difference index.

[0023] In a preferred embodiment, based on the growth and development stage of children and the synchrony of children's skeletal growth and soft tissue development, a decision tree is used to identify whether children have the risk of developmental deviation. Specifically:

[0024] Obtain the results of the growth and development stage division of children and define the growth and development stage data of children;

[0025] Fuse the growth and development stage data of children with the synchrony evaluation index to construct a standardized feature vector;

[0026] Design a decision tree model, input the feature vector for classification training, and identify whether children have the risk of developmental deviation.

[0027] In a preferred embodiment, by analyzing the genetic information of children and using a genetic algorithm combined with a dynamic programming model, the individual growth potential of children is evaluated. Specifically:

[0028] Collect the genetic information of children at risk of developmental deviation and perform feature extraction.

[0029] Use a genetic algorithm to initialize multiple candidate solutions for the individual growth potential of children.

[0030] Introduce the dynamic programming model into the algorithm iteration to evaluate the adaptability of the candidate solutions for growth potential.

[0031] Output the evaluation result of the individual growth potential of children.

[0032] In a preferred embodiment, the synchrony between children's skeletal growth and soft tissue development is correlated with the individual growth potential of children to evaluate the probability of potential developmental abnormalities in children and determine whether to initiate personalized intervention measures. Specifically:

[0033] Perform a correlation analysis on the synchrony evaluation index and the adaptability score according to the correlation function.

[0034] Calculate the probability value of potential developmental abnormalities based on the correlation result and determine whether to initiate personalized intervention measures.

[0035] In a preferred embodiment, calculate the probability value of potential developmental abnormalities based on the correlation result and determine whether to initiate personalized intervention measures. Specifically:

[0036] Preset a threshold for the probability of developmental abnormalities and compare the probability value of developmental abnormalities with the threshold for the probability of developmental abnormalities:

[0037] When the probability value of developmental abnormalities is greater than or equal to the threshold for the probability of developmental abnormalities, it indicates that the probability of potential developmental abnormalities in children is high.

[0038] When the probability value of developmental abnormalities is less than the threshold for the probability of developmental abnormalities, it indicates that the probability of potential developmental abnormalities in children is low and there is no need to initiate personalized intervention measures.

[0039] The technical effects and advantages of an integrated child growth and development evaluation and management system according to the present invention:

[0040] Through the deep integration of multi-modal physiological parameter data, limb ratio time-series data, and genetic information, and by using dynamic time-series analysis, dynamic time warping, decision trees, genetic algorithms, and dynamic programming models, the accurate division and real-time monitoring of children's growth and development stages are realized. It can not only objectively evaluate the synchrony of bone growth and soft tissue development, but also comprehensively evaluate the growth potential and adaptability at the genetic level for children at risk of developmental deviation, and then calculate the probability value of potential developmental abnormalities, improving the data integration and intelligent analysis capabilities, overcoming the disadvantages of information lag and strong subjectivity in traditional physical examinations, and providing a scientific and accurate basis for early warning and personalized intervention in children's health management. Brief Description of the Drawings

[0041] Figure 1 It is a schematic structural diagram of an integrated children's growth and development evaluation and management system of the present invention. Detailed Embodiments

[0042] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0043] Embodiment 1

[0044] Figure 1 An integrated children's growth and development evaluation and management system of the present invention is given, including a development stage determination module, a synchrony evaluation module, a deviation risk identification module, a growth potential evaluation module, and an abnormal probability evaluation module;

[0045] The development stage determination module collects real-time multi-modal physiological parameter data of children, and uses a dynamic time-series analysis algorithm based on machine learning to divide the growth and development status of children and determine the growth and development stages of children;

[0046] The synchrony evaluation module performs a dynamic time warping algorithm on the continuously collected limb ratio time-series data of children to evaluate the synchrony of bone growth and soft tissue development of children;

[0047] The deviation risk identification module uses a decision tree to identify whether a child has a risk of developmental deviation based on the growth and development stage of the child and the synchrony of bone growth and soft tissue development of the child;

[0048] For children at risk of developmental deviation, the growth potential evaluation module analyzes the genetic information of the children and uses a genetic algorithm combined with a dynamic programming model to evaluate the individual growth potential of the children;

[0049] The abnormal probability assessment module performs a correlation analysis between the synchronization of children's skeletal growth and soft tissue development and the individual growth potential of children, evaluates the probability of potential developmental abnormalities in children, and determines whether to initiate personalized intervention measures.

[0050] Specifically, collect real-time multi-modal physiological parameter data of children, and use a dynamic time series analysis algorithm based on machine learning to divide the growth and development status of children to determine the growth and development stages of children, including:

[0051] Real-time collection of children's multi-modal physiological data: To comprehensively monitor the growth and development process of children, multi-modal physiological data is obtained in real time through multiple sensors. The multi-modal physiological data includes but is not limited to: body surface temperature, pulse changes, skin resistance, respiratory rhythm, etc. To ensure the accuracy and continuity of the collected multi-modal physiological data, each sensor needs to be calibrated. During the calibration process, a reference value is used for comparison, and a calibration factor is set. The calibration factor is set as (where represents the calibration coefficient, and its value is determined by the standard reference sample), and the multi-modal physiological data is corrected, and its expression is:

[0052] ; where, represents the corrected multi-modal physiological data sequence; represents time; represents the calibration coefficient; represents the multi-modal physiological data collected by the sensor.

[0053] Since the physiological data collected by each sensor has different sampling frequencies and time windows, data fusion technology is used to integrate the physiological data. Define the physiological data collected by each sensor as (where represents different sensor numbers), and the integrated multi-modal physiological data sequence is denoted as:

[0054] ; where, represents the multi-modal physiological data sequence after fusion processing at time ; represents the data fusion function; represents the number of sensors.

[0055] Clean the multi-modal physiological data sequence of children and implement time series synchronization correction: Use an outlier detection algorithm based on statistical methods to identify and remove abnormal points in the multi-modal physiological data. For example, set the mean within the data window as , the standard deviation as , and use the three-sigma criterion (that is, if a data point exceeds If it is considered abnormal), the multi-modal physiological data sequence after cleaning is denoted as , and its expression is:

[0056] ; where represents the multi-modal physiological data sequence after cleaning; represents the mean value within the data window; represents the standard deviation within the data window; represents the estimated value obtained by interpolating adjacent normal data points when a certain data point is determined to be abnormal.

[0057] For the possible clock drift problem during the multi-modal physiological data acquisition process, data timing alignment is achieved through the timestamp matching algorithm. The cross-correlation algorithm is used to calculate the delay amount between each data sequence, and the data is interpolated and corrected to synchronize each data sequence under a unified time reference. Define the time delay between a certain data sequence and the standard time sequence as , then the correction formula is: ; where represents the multi-modal physiological data sequence after timing synchronization correction; represents the time delay amount between the data sequence and the standard time sequence.

[0058] Construct a dynamic timing model based on machine learning to extract key features and patterns: After completing data preprocessing, construct a machine learning model to perform dynamic modeling on the multi-modal physiological data sequence, and extract the key features and patterns of children's growth and development, specifically:

[0059] Use the sliding window technique to extract local features from the multi-modal physiological data sequence , define the sliding window size as , and calculate the statistical features within each sliding window, such as mean, variance, peak value, etc. Define the multi-modal physiological data within the sliding window as , and the formula for calculating the mean is: ; where represents the average feature value calculated within the sliding window; represents the size of the sliding window; represents the th numerical value of the multi-modal physiological data point within the sliding window.

[0060] Other statistical features (such as variance, peak value, etc.) are calculated similarly and are all components of the feature vector.

[0061] A long short-term memory network is used to construct a time series model to capture the temporal correlation in the process of children's growth and development. The constructed long short-term memory network model is denoted as , the input of the long short-term memory network model is a sequence of feature vectors , and the output is a sequence of hidden states . The expression is: ; where represents the output hidden state; represents the long short-term memory network model.

[0062] Based on the sequence of hidden states , the hidden states are classified using a clustering algorithm (such as K-Means or DBSCAN) to identify the patterns corresponding to different growth and development stages. The clustering result is defined as , and each cluster center represents a typical growth and development pattern. The calculation formula for the cluster center is: ; where represents the center vector of the th cluster; represents the total number of data samples in the th cluster; represents the hidden layer state vector of the th sample in the th cluster.

[0063] Through clustering, the key feature patterns related to different growth stages can be automatically identified.

[0064] A dynamic time series analysis algorithm is used to divide the growth state and output the growth stage division index: Based on the extracted sequence of hidden states and the clustering result , a state division model is designed. The model uses a hidden Markov model to perform state transition modeling on the multi-modal physiological data sequence after time series synchronization correction. The hidden Markov model is denoted as , where the state set is defined, the transition probability matrix is denoted as , and the observation probability is . The formula for the hidden Markov model is: ; where represents the conditional probability that the child is in the growth state (i.e., the output hidden state) given the observed data ; represents a state in the hidden Markov model, corresponding to a certain growth and development stage of the child; represents the hidden Markov model; The state transition probability matrix in the Hidden Markov Model; Indicates the observation probability of the hidden state observed in the growth state .

[0065] Solve the Hidden Markov Model to obtain the optimal growth state sequence . Each growth state corresponds to a division index of the growth and development stage.

[0066] The solution formula for the optimal state sequence is: ; where, represents the optimal growth state sequence determined by the Hidden Markov Model after optimal path search; is a mathematical symbol indicating the growth state that maximizes the probability expression among all growth states ; represents the joint probability of the growth state under the conditions of the hidden state , the state transition probability matrix , and the observation probability .

[0067] The output can be used as the division index of the growth stage.

[0068] Determine the child's growth and development stage based on the division index of the growth stage: According to the optimal growth state sequence , establish the mapping relationship between the state and the growth stage. Define the mapping function as , and its expression is: ; where, represents the child's growth and development stage, including the initial development stage, the accelerated development stage, and the stable development stage; is the mapping function.

[0069] Specifically, perform the dynamic time warping algorithm on the continuously collected time series data of the child's limb ratio to evaluate the synchronization of the child's bone growth and soft tissue development, including:

[0070] Obtain the continuously collected time series data of the child's limb ratio, covering bone and soft tissue parameters: Use multi-sensor fusion to obtain the key parameters of the child's bone and soft tissue development. The sensors include but are not limited to optical measurement devices, ultrasonic scanners, etc., which are respectively used to obtain growth data such as bone length, joint angle, muscle thickness, etc. All data are recorded according to the unified time stamp to ensure the time continuity and synchronization of the data. Define the continuous time series data as: ; where, represents the limb ratio data sequence collected at the sampling time ; Indicates the sampling time.

[0071] The collected limb proportion data sequence contains two main components: bone parameters and soft tissue parameters. The signal separation algorithm is used to preliminarily separate the original data to obtain two subsequences. For example, through the signal separation algorithm, is decomposed into and ; among them, Indicates the bone growth data sequence at the sampling time ; Indicates the soft tissue development data sequence at the sampling time ;

[0072] The dynamic time warping algorithm is used to achieve time series data alignment: Since the bone growth and soft tissue development may have time series deviations during the actual acquisition process due to reasons such as sensor sampling frequency, data loss, or clock drift, it is necessary to use the dynamic time warping algorithm to achieve the time series alignment of the two sets of data and ensure the best match between the data sequences on the time axis.

[0073] The goal of the dynamic time warping algorithm is to find the optimal matching path to minimize the cumulative distance between the two data sequences. The alignment cost function is defined as: ; among them, Indicates the cumulative alignment cost calculated by the dynamic time warping algorithm; Represents a specific alignment path, that is, a set of mapping relationships, which determines the corresponding relationship between the bone growth data and the soft tissue development data on the time axis; Represents the set composed of all possible alignment paths; Represents the distance metric function, which calculates the difference between the bone growth data at the sampling time and the soft tissue development data at the alignment time , and the Euclidean distance is commonly used; Represents the value of the bone growth data at the sampling time ; Represents the value mapped to the corresponding time in the soft tissue development data sequence through the alignment path; Represents the sampling time in the bone growth data sequence; Represents the sampling time corresponding to the mapping to the soft tissue development data sequence in the alignment path.

[0074] Calculate the difference index of the bone growth rate and the soft tissue development rate: Differentiate the aligned bone growth data and soft tissue development data respectively:

[0075] The calculation formula for the bone growth rate is: ; among them, Indicates the instantaneous rate of bone growth at the sampling time.

[0076] Similarly, the calculation formula for the soft tissue development rate is: ; where Indicates the instantaneous rate of soft tissue development at the sampling time.

[0077] Define the difference index between the bone growth rate and the soft tissue development rate, which is expressed using the absolute value function as:

[0078] ; where Indicates the difference between the bone growth rate and the soft tissue development rate at the sampling time.

[0079] By comparing the differences in the bone growth rate and the soft tissue development rate at each time, the asynchrony phenomenon between bone growth and soft tissue development is quantified.

[0080] Evaluate the synchrony of children's bone growth and soft tissue development based on the difference index: conduct an overall statistics on the difference index, and construct a synchrony evaluation index based on the similarity calculation method to evaluate the synchrony of children's bone growth and soft tissue development.

[0081] Define the calculation formula for the synchrony evaluation index as: ; where Indicates the synchrony evaluation index, and its value range is between 0 and 1; Indicates the sampling time total number; Indicates the difference between the bone growth rate and the soft tissue development rate at the sampling time.

[0082] The closer the synchrony evaluation index is to 1, the higher the synchrony of children's bone growth and soft tissue development, that is, it indicates that children's bone growth and soft tissue development maintain a basically consistent growth rate and change trend during the growth process, reflecting a coordinated and stable growth process, and predicting healthy and stable overall development of children.

[0083] Specifically, based on the children's growth and development stage and the synchrony of children's bone growth and soft tissue development, use a decision tree to identify whether children have the risk of developmental deviation, including:

[0084] Obtain the result of the classification of children's growth and development stages: The children's growth and development stages include the initial development stage, the accelerated development stage, and the stable development stage.

[0085] Define the data of children's growth and development stages as: ; where Indicates at the time variable , the growth and development stage of children. For example, Indicates that the child is in the initial development stage; Indicates that the child is in the accelerated development stage; Indicates that the child is in the stable development stage.

[0086] Fuse the data of children's growth and development stages with the synchrony evaluation index to construct a standardized feature vector: Normalize the data of children's growth and development stages. Define the normalized data of children's growth and development stages as . Based on the synchrony evaluation index and the normalized data of children's growth and development stages, construct a standardized feature vector denoted as: ; where Represents a two-dimensional feature vector that fuses the normalized data of children's growth and development stages and the synchrony evaluation index; Represents the synchrony evaluation index.

[0087] Design a decision tree model, input the feature vector for classification training, and identify whether there is a risk of developmental deviation in children: Based on the constructed standardized feature vector , use the decision tree classification model to judge the risk of children, and screen and identify children at risk of developmental deviation. Define the decision tree model as ; where Represents the judgment result of the risk of developmental deviation output by the decision tree model (the output is usually binary, 1 indicates that the child has a risk of developmental deviation, and 0 indicates normal); Represents the decision tree classification function.

[0088] During the training process of the decision tree model, a large amount of labeled data is used for supervised learning. The tree structure is constructed using the children's growth and development stages and the synchrony evaluation index in the feature vector, and the branch judgment conditions are automatically generated to form risk judgment nodes. The decision tree model has the advantages of intuitiveness, strong interpretability, and low computational complexity, and is suitable for quickly screening the growth and development status of children.

[0089] After the decision tree model is trained, input the constructed standardized feature vector , and identify children at risk of developmental deviation according to the judgment result of the risk of developmental deviation output by the decision tree model; where Indicates that the child has a risk of developmental deviation, Indicates normal.

[0090] Specifically, by analyzing the genetic information of children, using a genetic algorithm combined with a dynamic programming model to evaluate the individual growth potential of children, including:

[0091] Collect the genetic information of children at risk of developmental deviation and perform feature extraction: Obtain the genetic information of children at risk of developmental deviation, defined as ; among which, represents the genetic information data of the child numbered .

[0092] Deeply analyze the genetic information data of each child, extract the feature sequences related to growth potential, and define the feature extraction formula as: ; among which, represents the feature vector extracted from the genetic information data of the child numbered , reflecting the gene sequence pattern and variation information related to growth potential; represents the function for performing feature extraction on the genetic information data, using machine learning models, statistical analysis, or pattern recognition methods.

[0093] Initialize multiple candidate solutions for children's individual growth potential using the genetic algorithm: Based on the feature vector , sample the genetic algorithm to generate multiple candidate solutions, representing different prediction results of children's individual growth potential. Define the candidate solution set as: ; among which, represents the th growth potential candidate solution of the child numbered in the candidate solution set; represents the total number of growth potential candidate solutions generated by the genetic algorithm.

[0094] The generation of candidate solutions uses the genetic algorithm function: ; among which, represents the initialization function of the genetic algorithm, used to generate growth potential candidate solutions from the feature vector.

[0095] The initialization function includes basic operations such as selection, crossover, and mutation to ensure the diversity of growth potential candidate solutions.

[0096] Introduce the dynamic programming model into the algorithm iteration to evaluate the fitness of growth potential candidate solutions: In order to optimize the growth potential candidate solutions and accurately evaluate the growth potential of children, introduce the dynamic programming model to iteratively optimize the growth potential candidate solutions generated by the genetic algorithm. Define the fitness evaluation function of the growth potential candidate solutions as:

[0097] ;

[0098] Among which, represents the fitness score of the th growth potential candidate solution of the child numbered ; represents the evaluation of the growth potential candidate solution The score obtained from the partial fitness evaluation of the genetic algorithm; Indicates the additional score obtained after iterative optimization of the candidate solutions for growth potential by the dynamic programming model ; Indicates the weight coefficient of the evaluation function in the genetic algorithm part; Indicates the weight coefficient of the evaluation part of the dynamic programming model. The weight coefficients satisfy .

[0099] During the iterative process of the dynamic programming model, multi-stage optimal decision-making calculations are performed on each candidate solution for growth potential, and the fitness score is optimized through a recursive method. Specifically, the dynamic programming model updates the state of the candidate solution using the following iterative relationship: ; where Indicates the cumulative cost of the -th candidate solution for growth potential of the child numbered ; Indicates the immediate cost generated from the candidate solution for growth potential to the candidate solution for growth potential ; Indicates the cumulative cost of the -th candidate solution for growth potential of the child numbered ; Indicates the index set of candidate solutions in the dynamic programming model. The set contains all candidate solutions for growth potential to be evaluated, and the optimal conversion path is found by traversing.

[0100] The dynamic programming model is combined with the genetic algorithm, and an iterative update and optimal path search method is used to refine the fitness score of the candidate solutions for growth potential to ensure that each candidate solution for growth potential can accurately reflect the growth potential of the child.

[0101] Output the evaluation result of the individual growth potential of the child: After generating candidate solutions for growth potential by the genetic algorithm and iterative optimization by the dynamic programming model, compare all candidate solutions for growth potential and their fitness scores , and select the candidate solution for growth potential with the highest fitness score. Define the final output result as:

[0102] ; where Indicates the final determined evaluation result of the growth potential of the child numbered , that is, the candidate solution for growth potential with the highest fitness score; Indicates the -th candidate solution for growth potential of the child numbered , that is, the candidate solution for growth potential with the highest fitness score; Indicates the index of the candidate solution for growth potential with the highest adaptability score determined after screening.

[0103] Specifically, conduct a correlation analysis between the synchrony of children's skeletal growth and soft tissue development and the individual growth potential of children, evaluate the probability of potential developmental abnormalities in children, and determine whether to initiate personalized intervention measures, including:

[0104] Conduct a correlation analysis on the synchrony evaluation index and the adaptability score according to the correlation function: Conduct a correlation analysis on the synchrony evaluation index corresponding to the synchrony of children's skeletal growth and soft tissue development and the adaptability score corresponding to the individual growth potential of children. Define the correlation function as:

[0105] ; where Indicates the association strength between the synchrony evaluation index and the adaptability score; Indicates the calculation function for the correlation analysis; Indicates the synchrony evaluation index; Indicates numbered of the th candidate solution for growth potential of the child.

[0106] Calculate the probability value of potential developmental abnormalities based on the association result and determine whether to initiate personalized intervention measures: Calculate the probability value of potential developmental abnormalities in children, specifically: ; where Indicates the probability value of potential developmental abnormalities in children; Indicates the function for calculating the probability of developmental abnormalities.

[0107] Set a preset threshold for the probability of developmental abnormalities and compare the probability value of developmental abnormalities with the threshold for the probability of developmental abnormalities:

[0108] When the probability value of developmental abnormalities is greater than or equal to the threshold for the probability of developmental abnormalities, it indicates that the probability of potential developmental abnormalities in children is high, and the growth potential of children deviates from the normal range in terms of developmental adaptability; it is necessary to initiate personalized intervention measures, through early monitoring, further diagnosis, and providing targeted treatment plans to reduce the possible risk of developmental abnormalities;

[0109] When the probability value of developmental abnormalities is less than the threshold for the probability of developmental abnormalities, it indicates that the probability of potential developmental abnormalities in children is low, indicating that the growth and development process of children meets expectations and there is no risk of deviating from the normal developmental trajectory, and there is no need to initiate personalized intervention measures.

[0110] The setting of the dysplasia probability threshold is based on large-scale clinical data statistical analysis and expert consensus, aiming to scientifically distinguish normal development from potential abnormal risks. Usually, by collecting and training a large amount of children's growth data, the probability distributions in different growth stages and development states are determined, and then the key percentiles (such as the 90th percentile or 95th percentile) in the statistical distribution are selected as the threshold. In addition, the threshold is corrected and optimized by combining the actual clinical situation, the results of genetic research, and long-term follow-up data.

[0111] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0112] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that the computer can access, or a data storage device such as a server or data center containing one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0113] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present application.

[0114] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and modules described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0115] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there can be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or modules can be in electrical, mechanical, or other forms.

[0116] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules. They can be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0117] In addition, in each embodiment of the present application, the functional modules can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0118] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0119] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the technical field can easily think of changes or substitutions within the technical scope disclosed by the present application, and all of them should be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the said claims.

[0120] Finally, the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An integrated growth and development assessment and management system for children, characterized in that, It includes a development stage determination module, a synchrony assessment module, a deviation risk identification module, a growth potential assessment module, and an abnormal probability assessment module; The development stage determination module collects real-time multi-modal physiological parameter data of children, uses a machine learning-based dynamic time series analysis algorithm to divide the growth and development status of children, and determines the growth and development stage of children; The synchrony assessment module executes a dynamic time warping algorithm on the continuously collected time series data of children's limb proportions to evaluate the synchrony of children's bone growth and soft tissue development; The time series data of children's limb proportions includes bone parameters and soft tissue parameters; Based on the growth and development stage of children and the synchrony of children's bone growth and soft tissue development, the deviation risk identification module uses a decision tree to identify whether children have a risk of developmental deviation; For children with a risk of developmental deviation, the growth potential assessment module analyzes the genetic information of children, and uses a genetic algorithm combined with a dynamic programming model to evaluate the individual growth potential of children; The abnormal probability assessment module performs a correlation analysis on the synchrony of children's bone growth and soft tissue development and the individual growth potential of children, evaluates the potential probability of developmental abnormalities in children, and determines whether to initiate personalized intervention measures.

2. The integrated growth and development assessment and management system for children according to claim 1, wherein, Collect real-time multi-modal physiological parameter data of children, use a machine learning-based dynamic time series analysis algorithm to divide the growth and development status of children, and determine the growth and development stage of children. Specifically: Collect children's multi-modal physiological data in real time; Clean the time series data of children's multi-modal physiological data and achieve time series synchronization correction; Construct a machine learning-based dynamic time series model, and extract key features and patterns; Use a dynamic time series analysis algorithm to divide the growth status and output the growth stage division index Determine the growth and development stage of children based on the growth stage division index.

3. The integrated growth and development assessment and management system for children according to claim 2, wherein Execute a dynamic time warping algorithm on the continuously collected time series data of children's limb proportions to evaluate the synchrony of children's bone growth and soft tissue development. Specifically: Obtain the continuous time series data of children's limb proportions, covering bone and soft tissue parameters; Use a dynamic time warping algorithm to align the time series data; Calculate the difference index between the bone growth rate and the soft tissue development rate; Evaluate the synchrony of children's bone growth and soft tissue development based on the difference index.

4. The integrated child growth and development assessment and management system according to claim 3, characterized in that, Based on the growth and development stage of children and the synchrony of children's bone growth and soft tissue development, use a decision tree to identify whether children have a risk of developmental deviation. Specifically: Obtain the results of the growth and development stage division of children and define the data of the growth and development stage of children; Fuse the data of the growth and development stage of children and the synchrony evaluation index to construct a standardized feature vector; Design a decision tree model, input the feature vector for classification training, and identify whether children have a risk of developmental deviation.

5. The integrated growth and development assessment and management system for children according to claim 4, characterized in that, By analyzing the genetic information of children, use a genetic algorithm combined with a dynamic programming model to evaluate the individual growth potential of children. Specifically: Collect the genetic information of children with a risk of developmental deviation and perform feature extraction; Use a genetic algorithm to initialize multiple candidate solutions for the individual growth potential of children; Introduce a dynamic programming model into the algorithm iteration to evaluate the adaptability of the candidate solutions for growth potential; Output the evaluation results of the individual growth potential of children.

6. The integrated child growth and development assessment and management system according to claim 5, characterized in that Correlate the synchrony of children's skeletal growth and soft tissue development with the individual growth potential of children, evaluate the probability of potential developmental abnormalities in children, and determine whether to initiate personalized intervention measures, specifically as follows: Conduct a correlation analysis on the synchrony evaluation index and the adaptability score according to the correlation function; Calculate the probability value of potential developmental abnormalities based on the correlation results, and determine whether to initiate personalized intervention measures.

7. A child integrated growth and development assessment and management system according to claim 6, characterized in that, Calculate the probability value of potential developmental abnormalities based on the correlation results, and determine whether to initiate personalized intervention measures, specifically as follows: Preset a threshold for the probability of developmental abnormalities, and compare the probability value of developmental abnormalities with the threshold for the probability of developmental abnormalities: When the probability value of developmental abnormalities is greater than or equal to the threshold for the probability of developmental abnormalities, it indicates that the probability of potential developmental abnormalities in children is high; When the probability value of developmental abnormalities is less than the threshold for the probability of developmental abnormalities, it indicates that the probability of potential developmental abnormalities in children is low, and there is no need to initiate personalized intervention measures.

Images