Method and system for generating foot spine health management scheme for children and adolescents

Abstract

The invention relates to a generation method and system of a child and teenager foot spine health management scheme. The method comprises the following steps: firstly, acquiring a dynamic parameter set of feet of children and adolescents, and inputting the dynamic parameter set into a trained multi-dimensional classification model to obtain an interval division result; then, grouping features are extracted from the interval division result, and a labeling result is obtained; the discretization parameters and the abnormal mode data are integrated according to a marking result, a feature vector set including risk weights is generated, and when arch height data in the feature vector set is lower than a preset interval and the spine curvature exceeds a standard range, a foot health product recommendation table is generated; then, motion demand parameters are extracted from the feature vector set, and an individualized motion scheme is generated; and finally, performing weighted fusion evaluation on the recommendation table and the exercise scheme, and finally generating a health management scheme. Accurate assessment, risk grading and personalized intervention of the health condition of the foot spine are achieved, the health problem of the foot spine is effectively prevented and improved, and a health management strategy is reasonably formulated.

Description

Technical Field

[0001] The present invention belongs to the field of medical and health information technology, and particularly relates to a method for generating a foot and spine health management plan for children and adolescents. Background Art

[0002] With the development of medical and health information technology, there has emerged a technology for generating a foot and spine health management plan for children and adolescents. The period of childhood and adolescence is a critical stage of physical development, and foot and spine health has a profound impact on their normal growth, body posture, and future quality of life. In recent years, due to factors such as increased academic pressure leading to prolonged sedentary time, widespread poor sitting postures, and lack of scientific exercise guidance, the incidence of foot and spine problems among children and adolescents has been on the rise. Traditional foot and spine health assessments mostly rely on manual examinations, which are highly subjective and inefficient, and it is difficult to meet the needs of large-scale screening and precise management. It is difficult to accurately assess the foot and spine health status, conduct risk grading, and provide personalized interventions to assist the healthy growth of children and adolescents. Summary of the Invention

[0003] Based on this, in view of the above technical problems, it is necessary to provide a method for generating a foot and spine health management plan for children and adolescents that can accurately assess the foot and spine health status, conduct risk grading, and provide personalized interventions.

[0004] In a first aspect, the present application provides a method for generating a foot and spine health management plan for children and adolescents, including:

[0005] Obtaining a set of dynamic parameters of the feet of children and adolescents; the set of dynamic parameters includes arch height data and spinal curvature data.

[0006] Dividing the set of dynamic parameters using the statistical quantile method and inputting it into a trained multi-dimensional classification model to obtain an interval division result.

[0007] Performing grouped feature extraction on the interval division result and matching it with abnormal pattern data to obtain a labeled result for judging the risk level.

[0008] Integrating the discretized parameters and abnormal pattern data using the decision tree algorithm according to the labeled result to generate a set of feature vectors including risk weights; if the arch height data in the set of feature vectors is lower than the preset interval and the spinal curvature exceeds the standard range, a recommended list of foot health products is generated.

[0009] Extracting exercise demand parameters from the set of feature vectors and using the random forest algorithm to generate an individualized exercise plan.

[0010] Performing weighted fusion evaluation on the recommended list and the exercise plan, and iteratively adjusting the weight parameters to generate a health management plan.

[0011] In one embodiment, the dynamic parameter set is divided by the statistical quantile method and input into the trained multi-dimensional classification model to obtain the interval division result, including:

[0012] Obtain the quantile threshold of the dynamic parameter set; the quantile threshold is calculated by the statistical quantile method.

[0013] Divide the quantile threshold according to the preset health threshold to obtain the initial division interval; the initial division interval includes the boundary values of the dynamic parameters.

[0014] Obtain the height and weight data of the individual; the height and weight data includes multi-dimensional feature vectors.

[0015] Calculate the feature weight matrix based on the multi-dimensional feature vectors and use the clustering algorithm to iteratively adjust the initial division interval to obtain the adjusted feature weight matrix.

[0016] Fuse the adjusted feature weight matrix with the dynamic parameter boundary values to generate parameter constraint conditions.

[0017] Input the parameter constraint conditions and the multi-dimensional feature vectors into the trained multi-dimensional classification model to obtain the interval division result.

[0018] In one embodiment, calculate the feature weight matrix based on the multi-dimensional feature vectors and use the clustering algorithm to iteratively adjust the initial division interval to obtain the adjusted feature weight matrix, including:

[0019] Calculate the initial weight matrix based on the multi-dimensional feature vectors using matrix operations to obtain the preliminary feature weight matrix.

[0020] The initial weight matrix is calculated using the following formula:

[0021]

[0022] where W ij represents the feature weight of the i-th sample in the j-th dimension, x ik represents the feature value of the i-th sample in the k-th dimension, μ k represents the mean of the k-th dimension, m represents the total number of feature dimensions, and n represents the total number of samples.

[0023] Obtain the data grouping features in the preliminary feature weight matrix, and use the K-means clustering algorithm to group the initial division interval to obtain the grouped data intervals.

[0024] Perform iterative calculations on the grouped data intervals, and judge the interval adjustment direction through the sum of squared errors to obtain the adjusted data intervals.

[0025] Update the preliminary feature weight matrix according to the adjusted data interval, and use matrix factorization technology to extract the adjusted weights to obtain the adjusted feature weight matrix.

[0026] In one embodiment, use the K-means clustering algorithm to group the initial partitioning intervals to obtain the grouped data intervals, including:

[0027] Use the K-means clustering algorithm to calculate the clustering quality evaluation index using the following formula:

[0028]

[0029] where, Q t represents the clustering quality evaluation index, k represents the number of clusters, n i represents the number of samples in the i-th cluster, d(x,y) represents the distance between samples x and y, and C i represents the i-th clustering set.

[0030] Group the initial partitioning intervals based on the clustering quality evaluation index to obtain the data intervals.

[0031] In one embodiment, according to the annotation results, use the decision tree algorithm to integrate the discretization parameters and abnormal pattern data to generate a set of feature vectors including risk weights, including:

[0032] Obtain the discretization parameters and abnormal pattern data based on the annotation results, and use the decision tree algorithm for preliminary integration to obtain the initial feature set.

[0033] Extract the abnormal patterns from the initial feature set and determine the risk weights to generate weighted pattern data.

[0034] Adjust the discretization parameters according to the weighted pattern data, and use the decision tree algorithm to optimize the integration process to obtain the updated feature set.

[0035] Among them, if the abnormal patterns in the updated feature set exceed the preset threshold, extract the significant abnormalities through pattern recognition to obtain the risk weight distribution.

[0036] Use the updated feature set and the risk weight distribution to generate a set of feature vectors including weight allocation.

[0037] In one embodiment, perform a weighted fusion evaluation on the recommendation table and the exercise plan, and iteratively adjust the weight parameters to generate a health management plan, including:

[0038] Obtain the user health data of the recommendation table and the execution feature data of the exercise plan.

[0039] Generate the initial weight parameters according to the user health data and input them into the exercise plan to obtain the updated exercise plan.

[0040] Use a fusion algorithm to perform weighted calculation on the recommendation list and the updated exercise plan to obtain a set of candidate health management plans.

[0041] Establish an evaluation function based on the execution feature data to score the set of candidate health management plans and obtain the adjustment amount of the weight parameter.

[0042] Input the adjustment amount of the weight parameter into the fusion algorithm for iterative operation until the evaluation function reaches the preset threshold to obtain the final health management plan; the health management plan includes personalized foot health product recommendations, customized exercise rehabilitation plans, and foot and spine health care guidance suggestions.

[0043] In a second aspect, the present application also provides a system for generating a foot and spine health management plan for children and adolescents, the system including:

[0044] A foot data acquisition module for obtaining a set of dynamic parameters of the feet of children and adolescents; the set of dynamic parameters includes arch height data and spinal curvature data.

[0045] A data classification module for using the statistical quantile method to divide the set of dynamic parameters and inputting it into a trained multi-dimensional classification model to obtain an interval division result; it is also used to extract group features from the interval division result and match it with abnormal pattern data to obtain a labeled result for judging the risk level.

[0046] A product recommendation module for integrating discretized parameters and abnormal pattern data using a decision tree algorithm according to the labeled result to generate a set of feature vectors including risk weights; if the arch height data in the set of feature vectors is lower than the preset interval and the spinal curvature exceeds the standard range, a foot health product recommendation list is generated.

[0047] A health management module for extracting exercise demand parameters from the set of feature vectors and using a random forest algorithm to generate an individualized exercise plan; it is also used to perform weighted fusion evaluation on the recommendation list and the exercise plan, and iteratively adjust the weight parameters to generate a health management plan.

[0048] In a third aspect, the present application also provides a computer device, including a memory and a processor, the memory storing a computer program, and the processor implementing the method as described above when executing the computer program.

[0049] In a fourth aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored, and the computer program implements the method as described above when executed by a processor.

[0050] The method and system for generating the above-mentioned foot and spine health management plan for children and adolescents first obtain a set of dynamic parameters of the feet of children and adolescents, including arch height data and spinal curvature data; then use the statistical quantile method to divide the set of dynamic parameters and input it into a trained multi-dimensional classification model to obtain an interval division result; then extract grouping features from the interval division result and match them with abnormal pattern data to obtain a labeling result for judging the risk level; then use the decision tree algorithm to integrate the discretized parameters and abnormal pattern data based on the labeling result to generate a set of feature vectors including risk weights. When the arch height data in the set of feature vectors is lower than the preset interval and the spinal curvature exceeds the standard range, a recommendation form for foot health products is generated; then extract exercise demand parameters from the set of feature vectors and generate an individualized exercise plan by means of the random forest algorithm; finally, perform a weighted fusion evaluation on the recommendation form and the exercise plan, and finally generate a health management plan by iteratively adjusting the weight parameters. It not only realizes the accurate assessment, risk grading and personalized intervention of foot and spine health conditions, but also effectively prevents and improves foot and spine health problems, reasonably formulates health management strategies, comprehensively protects the foot and spine health of children and adolescents, and promotes their normal physical development and growth. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0052] Figure 1 It is a flowchart of the method for generating a foot and spine health management plan for children and adolescents provided by an embodiment of the present invention;

[0053] Figure 2 It is a flowchart of using the statistical quantile method to divide the dynamic parameter set and input it into a trained multi-dimensional classification model to obtain an interval division result provided by an embodiment of the present invention;

[0054] Figure 3 It is a flowchart of using the decision tree algorithm to integrate the discretized parameters and abnormal pattern data based on the labeling result to generate a set of feature vectors including risk weights provided by an embodiment of the present invention;

[0055] Figure 4 It is a flowchart of performing a weighted fusion evaluation on the recommendation form and the exercise plan and iteratively adjusting the weight parameters to generate a health management plan provided by an embodiment of the present invention;

[0056] Figure 5 It is a structural block diagram of the system for generating a foot and spine health management plan for children and adolescents provided by an embodiment of the present invention. Detailed implementation manners

[0057] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0058] First, the implementation environment of the embodiments of the present application will be described. Exemplarily, the implementation environment includes a data acquisition device, a data processing system, and a display device.

[0059] In the method and system for generating a children and adolescents' foot and spine health management plan, data acquisition devices, such as foot scanners, scoliosis measuring instruments, etc., are connected to the data processing system through wired or wireless transmission protocols, ensuring that a large amount of raw data, such as three-dimensional foot model data, spine morphology measurement data, etc., can be quickly and stably transmitted to the data processing system for operation analysis and modeling; after the data processing system completes the complex processing of the foot and spine health data, the visual data such as analysis results and evaluation reports is transmitted to the display device through a video interface or network, such as a computer monitor, a projector, etc., for medical staff, parents and relevant personnel to view intuitively, assisting in the formulation and implementation of the foot and spine health management plan.

[0060] Combined with the above implementation environment, the application scenarios of the embodiments of the present application will be described.

[0061] The method and system for generating a children and adolescents' foot and spine health management plan provided by the embodiments of the present application first use professional data acquisition devices, such as three-dimensional foot scanners, scoliosis measuring instruments, and intelligent posture monitors, etc., to obtain multi-dimensional data such as the morphology and posture of the feet and spines of children and adolescents. Subsequently, the data processing system uses specific algorithms to deeply analyze and model the data, so as to evaluate the foot and spine health status and identify potential problems. Finally, the processed results are presented in an intuitive visual form on the display device, providing professional personnel with a personalized foot and spine health management plan to promote the healthy development of the feet and spines of children and adolescents. Exemplarily, the method and system for generating a children and adolescents' foot and spine health management plan provided by the embodiments of the present application can be applied to at least one of the following scenarios including but not limited to the following.

[0062] First, the method and system for generating the child and adolescent foot and spine health management plan are applied to the school scenario. In the school environment, data collection devices such as portable foot pressure measuring instruments and simple scoliosis screening instruments can regularly collect foot and spine data from students in the infirmary or physical education classes. Students stand on the foot pressure measuring instrument as required to obtain foot pressure distribution data; the spine is scanned using the screening instrument to record spine morphology parameters. These data are transmitted in real-time to the data processing system equipped in the school. The system analyzes the data using professional algorithms to judge the foot and spine health status of students. The analysis results are presented to teachers and students through display devices such as the school electronic bulletin board and class multimedia display screens. The generated health management plan, such as targeted adjustment of break exercises and seat arrangement suggestions, helps create a good campus foot and spine health management atmosphere.

[0063] Second, the method and system for generating the child and adolescent foot and spine health management plan are applied to the medical institution scenario. The professional data collection devices in the hospital are more accurate and comprehensive, such as high-precision three-dimensional foot scanners and digital X-ray spine imaging devices. When children come to see a doctor, doctors operate these devices to obtain detailed foot and spine data, and the data quickly enters the hospital's data processing system. The system combines medical knowledge bases and past case data for in-depth analysis, evaluates the severity of the condition, and predicts the development trend. Doctors view the analysis results and the generated personalized foot and spine health management plan through the hospital's electronic medical record system. The plan covers treatment methods, rehabilitation training plans, etc., providing strong support for the diagnosis and treatment of foot and spine diseases in children and adolescents.

[0064] Third, the method and system for generating the child and adolescent foot and spine health management plan are applied to the home scenario. With the popularization of home health monitoring devices, parents can use small foot posture sensors and wearable spine monitoring devices to collect daily foot and spine data for their children. These devices transmit the data to the supporting data processing APP. The algorithms in the APP conduct preliminary analysis on the data and generate simple health reports. Parents can intuitively understand the dynamic foot and spine health of their children through the mobile phone screen (display device). If any abnormalities are found, they can take their children to see a doctor in a timely manner. At the same time, the APP pushes daily foot and spine health care suggestions based on the analysis results, such as correct sitting posture reminders and home exercise plans, to help parents manage their children's foot and spine health at home.

[0065] In one embodiment, as Figure 1 shown, the present application provides a method for generating a child and adolescent foot and spine health management plan, which may include the following steps:

[0066] Step S101, obtaining a set of dynamic parameters of the feet of children and adolescents; the set of dynamic parameters includes arch height data and spinal curvature data.

[0067] Specifically, through an advanced foot pressure test system, the changes in the arch height of children and adolescents during dynamic processes such as standing and walking are monitored and recorded in real time, accurately capturing the morphological characteristics of the arch at different movement stages. At the same time, a high-resolution scoliosis measuring instrument is used to carefully measure the curvature of the spine in the sagittal and coronal planes using optical imaging or sensor technology, obtaining the physiological curvature data of the spine. The collected arch height data and spine curvature data together constitute a set of dynamic parameters.

[0068] Step S102: Divide the set of dynamic parameters using the statistical quantile method and input it into a trained multi-dimensional classification model to obtain an interval division result.

[0069] First, the statistical quantile method is used to process the set of dynamic parameters obtained previously. The statistical quantile method calculates the values corresponding to specific quantiles (such as the 25%, 50%, and 75% quantiles) based on the distribution characteristics of the data, thereby dividing the set of dynamic parameters into multiple intervals to achieve preliminary classification and sorting of the data. Subsequently, the parameter data of each divided interval is input into a multi-dimensional classification model that has been trained and has stable performance. This multi-dimensional classification model has learned a large number of sample data containing the characteristics of different foot and spine health conditions during the training stage. Through the operation and analysis of the model, detailed interval division results are finally output, and these results intuitively present the categories of foot and spine health characteristics corresponding to different dynamic parameter intervals.

[0070] Step S103: Extract the group characteristics from the interval division result and match them with the abnormal pattern data to obtain a labeled result for judging the risk level.

[0071] Specifically, first, the data within each divided interval is carefully analyzed, and representative characteristic parameters are extracted from multiple dimensions such as arch height and spine curvature. Then, the extracted group characteristics are comprehensively matched with the data in a pre-constructed abnormal pattern database. The abnormal pattern database covers the typical characteristics of various foot and spine health abnormalities summarized in past research and clinical practice. Through efficient algorithm comparison, the similarity between the current group characteristics and the abnormal patterns is identified, and then according to the established risk assessment criteria, a corresponding risk level label is assigned to each interval division result. The finally obtained labeled result for judging the risk level can clearly and intuitively show the degree of foot and spine health risks corresponding to different intervals.

[0072] Step S104: Integrate the discretized parameters and abnormal pattern data using the decision tree algorithm according to the labeled result to generate a set of feature vectors including risk weights; if the arch height data in the set of feature vectors is lower than the preset interval and the spine curvature exceeds the standard range, a recommendation form for foot health products is generated.

[0073] Input the parameters such as the discretized arch height and spinal curvature, as well as the corresponding abnormal pattern data into the decision tree model. Through the node splitting and branch judgment mechanism inside the model, deeply fuse and analyze the data to generate a set of feature vectors containing detailed risk weight information. This set of feature vectors not only comprehensively reflects the importance of various factors related to foot and spine health, but also can accurately locate potential risk points. Subsequently, conduct a detailed screening of the generated set of feature vectors. Once it is found that the arch height data is lower than the pre-set normal range and the spinal curvature exceeds the range specified by medical standards, immediately initiate the process of recommending foot health products. According to the pre-established product adaptation model, comprehensively consider factors such as the degree of risk and individual body parameters to generate a highly targeted and practical foot health product recommendation form, providing a practical product solution for improving the foot and spine health of children and adolescents.

[0074] Step S105: Extract the exercise demand parameters from the set of feature vectors and use the random forest algorithm to generate an individualized exercise plan.

[0075] The extraction of exercise demand parameters is based on a comprehensive consideration of the foot and spine health status of children and adolescents. For example, according to the problems reflected by the arch height data and spinal curvature data, determine information such as the exercise intensity, frequency, and type required to improve foot and spine health. At the same time, factors such as the individual's physical function, age, and gender will also be combined to accurately screen out the parameters that can reflect their unique exercise needs from the set of feature vectors. Subsequently, use the random forest algorithm to generate an individualized exercise plan. Input the extracted exercise demand parameters into the trained random forest algorithm model. This model takes into account various factors such as the impact of different exercises on foot and spine health and the adaptability of individual physical conditions, and finally generates a set of individualized exercise plans specifically for this child or adolescent. This exercise plan will specify in detail the specific content of the exercise, such as suitable aerobic exercises, strength training movements, as well as the time arrangement and cycle plan of the exercise, to maximize the promotion of the recovery and improvement of the foot and spine health of children and adolescents.

[0076] Step S106: Conduct a weighted fusion evaluation of the recommendation form and the exercise plan, and iteratively adjust the weight parameters to generate a health management plan.

[0077] First, for the key indicators such as the efficacy and applicability of various foot health products in the recommendation form, as well as the elements such as exercise intensity, frequency, and expected effects in the exercise plan, corresponding weights are assigned according to their importance in improving the foot and spine health of children and adolescents. The product information in the recommendation form is organically integrated with the specific content of the exercise plan, and the impact of their synergistic effect on foot and spine health is comprehensively considered. During the integration process, the weight parameters are repeatedly adjusted using an iterative algorithm. Each iteration is based on the results of the previous round of integrated evaluation. By analyzing data such as the change trend of foot and spine health-related indicators and individual feedback, the weight allocation is optimized to make the combination of product recommendation and exercise plan more in line with the actual needs of individuals. After multiple rounds of iterative adjustment until the evaluation results reach the preset ideal standard, a comprehensive and accurate health management plan is finally generated. This plan integrates the usage suggestions of foot health products and personalized exercise planning, providing systematic and targeted foot and spine health management strategies for children and adolescents.

[0078] The above method for generating a foot and spine health management plan for children and adolescents first obtains a set of dynamic parameters of the feet of children and adolescents, including arch height data and spinal curvature data; then uses the statistical quantile method to divide this set of dynamic parameters and inputs it into a trained multi-dimensional classification model to obtain an interval division result; then extracts grouping features from the interval division result and matches them with abnormal pattern data to obtain a labeled result for judging the risk level; then uses the decision tree algorithm to integrate the discretized parameters and abnormal pattern data based on the labeled result to generate a set of feature vectors including risk weights. When the arch height data in the set of feature vectors is lower than the preset interval and the spinal curvature exceeds the standard range, a recommendation form for foot health products is generated; then the exercise demand parameters are extracted from the set of feature vectors and an individualized exercise plan is generated using the random forest algorithm; finally, a weighted integrated evaluation is performed on the recommendation form and the exercise plan, and through iterative adjustment of the weight parameters, a health management plan is finally generated. It not only realizes the accurate assessment, risk grading, and personalized intervention of foot and spine health conditions, but also effectively prevents and improves foot and spine health problems, reasonably formulates health management strategies, comprehensively safeguards the foot and spine health of children and adolescents, and promotes their normal physical development and growth.

[0079] In one embodiment, as Figure 2 shown, using the statistical quantile method to divide the set of dynamic parameters and input it into a trained multi-dimensional classification model to obtain an interval division result may include the following steps:

[0080] Step S201, obtaining the quantile threshold of the set of dynamic parameters; the quantile threshold is calculated by the statistical quantile method.

[0081] Step S202, dividing the quantile threshold according to the preset health threshold to obtain an initial division interval; the initial division interval includes the boundary values of the dynamic parameters.

[0082] Step S203: Obtain the height and weight data of the individual; the height and weight data includes a multi-dimensional feature vector.

[0083] Step S204: Calculate the feature weight matrix based on the multi-dimensional feature vector and use the clustering algorithm to iteratively adjust the initial division interval to obtain the adjusted feature weight matrix.

[0084] Step S205: Integrate the adjusted feature weight matrix with the dynamic parameter boundary values to generate parameter constraint conditions.

[0085] Step S206: Input the parameter constraint conditions and the multi-dimensional feature vector into the trained multi-dimensional classification model to obtain the interval division result.

[0086] Divide the quantile threshold according to the preset health threshold to obtain the initial division interval including the dynamic parameter boundary values. Then obtain the height and weight data of the individual, which constitute a multi-dimensional feature vector. Since these data reflect the basic physical condition of the individual, they are extremely crucial for subsequent analysis. Immediately afterwards, a feature weight matrix is constructed based on the multi-dimensional feature vector through rigorous calculations, and the initial division interval is iteratively adjusted using the clustering algorithm to finally obtain the adjusted feature weight matrix. This process effectively optimizes the parameter classification structure. Then, the adjusted feature weight matrix is integrated with the dynamic parameter boundary values to successfully generate parameter constraint conditions, providing an accurate basis for data screening. Finally, the parameter constraint conditions and the multi-dimensional feature vector are input into the trained multi-dimensional classification model together to smoothly obtain a scientific and reasonable interval division result.

[0087] This embodiment classifies the dynamic parameters of the foot arch more accurately according to the individual's physical characteristics, laying a solid foundation for accurately judging the health risk level of the foot arch subsequently, greatly improving the scientificity and pertinence of the generation of the health management plan, and facilitating the efficient development of the foot arch health management work for children and adolescents.

[0088] In one of the embodiments, calculating the feature weight matrix based on the multi-dimensional feature vector and using the clustering algorithm to iteratively adjust the initial division interval to obtain the adjusted feature weight matrix may include the following steps:

[0089] Step S301: Calculate the initial weight matrix based on the multi-dimensional feature vector using matrix operations to obtain the preliminary feature weight matrix.

[0090] The initial weight matrix is calculated using the following formula:

[0091]

[0092] where W ij represents the feature weight of the i-th sample in the j-th dimension, and xik represents the eigenvalue of the i-th sample in the k-th dimension, μ k represents the mean of the k-th dimension, m represents the total number of feature dimensions, and n represents the total number of samples.

[0093] Step S302: Obtain the data grouping features in the preliminary feature weight matrix, and use the K-means clustering algorithm to group the initial division intervals to obtain the grouped data intervals.

[0094] Step S303: Perform iterative calculations on the grouped data intervals, and judge the interval adjustment direction through the sum of squared errors to obtain the adjusted data intervals.

[0095] Step S304: Update the preliminary feature weight matrix according to the adjusted data intervals, and use matrix decomposition technology to extract the adjusted weights to obtain the adjusted feature weight matrix.

[0096] First, based on the multi-dimensional feature vector, matrix operations are used to calculate the initial weight matrix, and the preliminary feature weight matrix is obtained according to a specific formula. Then, the data grouping features are extracted from the preliminary feature weight matrix, and the K-means clustering algorithm is used to group the initial division intervals to obtain the grouped data intervals. This algorithm can effectively classify the data by similarity. Subsequently, iterative calculations are carried out on the grouped data intervals, and the sum of squared errors is used to judge the interval adjustment direction, continuously optimizing the interval division to obtain the adjusted data intervals. Finally, the preliminary feature weight matrix is updated according to the adjusted data intervals, and the adjusted weights are extracted by using matrix decomposition technology to successfully obtain the adjusted feature weight matrix.

[0097] Through a series of closely related steps, the influence of individual multi-dimensional features on the parameters related to foot arch health can be fully considered, the data intervals can be divided more accurately, the feature weight matrix can be more in line with the actual situation, providing solid data support for the subsequent accurate assessment of the foot arch health risk level and the formulation of personalized health management plans, significantly improving the scientificity and effectiveness of foot arch health management, and effectively protecting the foot arch health of children and adolescents.

[0098] In one embodiment, using the K-means clustering algorithm to group the initial division intervals to obtain the grouped data intervals may include the following steps:

[0099] Step S401: Use the K-means clustering algorithm to calculate the clustering quality evaluation index using the following formula:

[0100]

[0101] where Q t represents the clustering quality evaluation index, k represents the number of clusters, n irepresents the number of samples in the \(i\)-th cluster, \(d(x,y)\) represents the distance between samples \(x\) and \(y\), and \(C\) i represents the \(i\)-th clustering set.

[0102] Step S402: Group the initial partitioning intervals based on the clustering quality evaluation index to obtain data intervals.

[0103] First, apply the K-means clustering algorithm to calculate the clustering quality evaluation index according to a specific formula. Through the operation of this formula, the clustering effect can be quantitatively evaluated from multiple dimensions. Then, apply the calculated clustering quality evaluation index to the grouping of the initial partitioning intervals. With the help of the clustering quality and inferiority information reflected by the index, reasonably group the initial partitioning intervals to finally obtain data intervals.

[0104] This embodiment uses the K-means clustering algorithm and the clustering quality evaluation index to more accurately classify the initial partitioning intervals of the dynamic parameters related to foot arch health, making the data grouping more in line with the internal characteristics and laws of the foot arch health status. It provides a more reliable data basis for accurately analyzing the foot arch health risk level and formulating personalized health management plans in the follow-up, and improves the scientificity and accuracy of foot arch health management work.

[0105] In one of the embodiments, as Figure 3 shown, according to the annotation results, use the decision tree algorithm to integrate the discretized parameters and abnormal pattern data to generate a set of feature vectors including risk weights, which may include the following steps:

[0106] Step S501: Obtain the discretized parameters and abnormal pattern data based on the annotation results, and use the decision tree algorithm for preliminary integration to obtain an initial feature set.

[0107] Step S502: Extract abnormal patterns from the initial feature set and determine the risk weights to generate weighted pattern data.

[0108] Step S503: Adjust the discretized parameters according to the weighted pattern data, and use the decision tree algorithm to optimize the integration process to obtain an updated feature set.

[0109] Among them, if the number of abnormal patterns in the updated feature set exceeds the preset threshold, extract significant abnormalities through pattern recognition to obtain the risk weight distribution.

[0110] Step S504: Use the updated feature set and the risk weight distribution to generate a set of feature vectors including weight allocation.

[0111] First, based on the labeling results of the risk level, the discretization parameters and abnormal pattern data are obtained, and then the decision tree algorithm is used to preliminarily integrate these data to obtain the initial feature set. Then, the abnormal pattern extraction work is carried out in depth for the initial feature set, and the risk weights corresponding to each abnormality are determined, so as to generate weighted pattern data, which highlights the importance of different abnormal situations. With the help of the generated weighted pattern data, the discretization parameters are adjusted, and the decision tree algorithm is used again to optimize the integration process, and finally the updated feature set is obtained. In this process, if the number of abnormal patterns in the updated feature set exceeds the preset threshold, the significant abnormalities are accurately extracted through pattern recognition technology, and the risk weight distribution is obtained, which clearly presents the concentrated area and degree of risk. Finally, the updated feature set is combined with the risk weight distribution to generate a feature vector set containing accurate weight distribution.

[0112] Through repeated data integration, optimization and pattern recognition, we can comprehensively and deeply mine the key information in the foot and spine health data, accurately determine the risk weight, and provide highly targeted and reliable data support for the subsequent generation of personalized foot health product recommendation tables and exercise plans, significantly improving the scientificity and effectiveness of foot and spine health management plans, and effectively helping to prevent and improve foot and spine health problems in children and adolescents.

[0113] In one embodiment, if Figure 4 As shown, weighted fusion evaluation is performed on the recommendation table and the exercise plan, and weight parameters are iteratively adjusted to generate a health management plan. The following steps may be included:

[0114] Step S601, obtaining the user health data and execution characteristic data of the exercise plan in the recommendation table.

[0115] Step S602, generating initial weight parameters according to the user's health data and inputting an exercise plan to obtain an updated exercise plan.

[0116] Step S603: Use a fusion algorithm to perform weighted calculation on the recommendation table and the updated exercise plan to obtain a set of candidate health management plans.

[0117] Step S604: Establish an evaluation function based on the execution feature data to score the candidate health management solution set and obtain a weight parameter adjustment amount.

[0118] Step S605, input the weight parameter adjustment amount into the fusion algorithm for iterative calculation until the evaluation function reaches a preset threshold, and obtains the final health management plan; the health management plan includes personalized foot health product recommendations, customized exercise rehabilitation plans and foot and spine health care guidance suggestions.

[0119] First, collect the user health data associated with the recommendation form, including foot and spinal health conditions, past medical history, etc. At the same time, obtain the execution characteristic data of the exercise plan, such as exercise intensity, frequency, and individual adaptability performance. Then, based on the obtained user health data, generate initial weight parameters through rigorous calculation and input them into the exercise plan to optimize and adjust the exercise plan, obtaining an updated exercise plan that better fits the individual's health needs. Subsequently, use a fusion algorithm to perform weighted calculation on the foot health product information in the recommendation form and the updated exercise plan, comprehensively considering the synergistic effect of the two on improving foot and spinal health, thereby obtaining a set of candidate health management plans to provide diverse plan options. Construct an evaluation function based on the execution characteristic data to score the set of candidate health management plans. Through quantitative analysis, accurately obtain the adjustment amount of the weight parameters to indicate the direction for plan optimization. Finally, input the adjustment amount of the weight parameters into the fusion algorithm for iterative operation to continuously optimize the plan combination until the evaluation function reaches the preset threshold. At this time, generate the final health management plan. This plan includes personalized foot health product recommendations to match appropriate auxiliary products for individuals; customized exercise rehabilitation plans to design exclusive exercise plans based on physical conditions; and foot and spinal health care guidance suggestions to comprehensively guide the health care key points in daily life.

[0120] In one embodiment, as Figure 5 shown, the present application also provides a system for generating a foot and spinal health management plan for children and adolescents, the system including:

[0121] A foot data collection module 701, configured to obtain a set of dynamic parameters of the feet of children and adolescents; the set of dynamic parameters includes arch height data and spinal curvature data.

[0122] A data classification module 702, configured to divide the set of dynamic parameters using the statistical quantile method and input it into a trained multi-dimensional classification model to obtain an interval division result; it is also configured to perform group feature extraction on the interval division result and match it with abnormal pattern data to obtain a labeled result for judging the risk level.

[0123] A product recommendation module 703, configured to integrate the discretized parameters and abnormal pattern data using a decision tree algorithm according to the labeled result to generate a set of feature vectors including risk weights; if the arch height data in the set of feature vectors is lower than the preset interval and the spinal curvature exceeds the standard range, generate a foot health product recommendation form.

[0124] A health management module 704, configured to extract exercise demand parameters from the set of feature vectors and use a random forest algorithm to generate an individualized exercise plan; it is also configured to perform weighted fusion evaluation on the recommendation form and the exercise plan, and iteratively adjust the weight parameters to generate a health management plan.

[0125] The above-mentioned generation system for the foot and spine health management plan for children and adolescents. The foot data collection module obtains the set of dynamic parameters of the feet of children and adolescents, which includes arch height data and spinal curvature data. The data classification module uses the statistical quantile method to divide the set of dynamic parameters, and inputs it into the trained multi-dimensional classification model to obtain the interval division result. Then, it extracts the group characteristics of this result, matches it with the abnormal pattern data, and finally obtains the annotation result for judging the risk level. The product recommendation module, based on the above-mentioned annotation result, integrates the discretized parameters and abnormal pattern data with the decision tree algorithm to generate a set of feature vectors including risk weights. When the arch height data in the set of feature vectors is lower than the preset interval and the spinal curvature exceeds the standard range, it generates a foot health product recommendation form. The health management module extracts the exercise requirement parameters from the set of feature vectors, uses the random forest algorithm to generate an individualized exercise plan, and conducts a weighted fusion evaluation on the recommendation form and the exercise plan. By iteratively adjusting the weight parameters, it finally generates a comprehensive health management plan. It not only realizes the accurate assessment, risk grading, and personalized intervention of the foot and spine health status, but also effectively prevents and improves foot and spine health problems, reasonably formulates health management strategies, comprehensively safeguards the foot and spine health of children and adolescents, and promotes their normal physical development and growth.

[0126] It should be understood that although the steps in the flowcharts involved in the above-mentioned embodiments are shown in sequence according to the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless there is a clear indication in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily have to be executed at the same time, but can be executed at different times. The execution order of these steps or stages does not necessarily have to be sequential, but can be executed alternately or in turn with at least a part of the steps or stages in other steps or other steps.

[0127] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the above-mentioned method and system for generating the foot and spine health management plan for children and adolescents.

[0128] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps in the above-mentioned method embodiments.

[0129] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to the descriptions in the method embodiments. The device embodiments described above are merely illustrative. The components described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the present disclosure solution. A person of ordinary skill in the art can understand and implement it without creative efforts.

[0130] The above embodiments only represent several implementation manners of the embodiments of the present application. The descriptions are relatively specific and detailed, but should not be construed as a limitation on the patent scope of the embodiments of the application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the embodiments of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the embodiments of the present application.

Claims

1. A method for generating a health management plan for the foot spine of children and adolescents, characterized in that: The method comprises: Acquire a dynamic parameter set of the foot of children and adolescents; the dynamic parameter set includes arch height data and spinal curvature data; Dividing the dynamic parameter set by using the statistical quantile method and inputting the trained multidimensional classification model to obtain interval division results; Extracting group features from the interval division results and matching them with abnormal pattern data to obtain a labeling result for determining the risk level; Integrate the discretization parameters and abnormal pattern data using a decision tree algorithm according to the labeling results to generate a feature vector set including risk weights; if the arch height data in the feature vector set is lower than a preset interval and the spinal curvature exceeds a standard range, generate a foot health product recommendation table; Extracting motion demand parameters from the feature vector set and using a random forest algorithm to generate an individualized motion plan; A weighted fusion evaluation is performed on the recommendation table and the exercise plan, and the weight parameters are iteratively adjusted to generate a health management plan.

2. The method according to claim 1, characterized in that: The method of dividing the dynamic parameter set by using the statistical quantile method and inputting the trained multidimensional classification model to obtain the interval division result includes: Obtaining a quantile threshold of the dynamic parameter set; the quantile threshold is calculated by a statistical quantile method; The quantile threshold is divided according to a preset health threshold to obtain an initial division interval; the initial division interval includes a boundary value of a dynamic parameter; Acquire the height and weight data of the individual; the height and weight data include a multidimensional feature vector; Calculating a feature weight matrix based on the multidimensional feature vector and iteratively adjusting the initial partition interval using a clustering algorithm to obtain an adjusted feature weight matrix; The adjusted feature weight matrix is ​​merged with the dynamic parameter boundary value to generate parameter constraint conditions; The parameter constraints and the multidimensional feature vector are input into the trained multidimensional classification model to obtain an interval division result.

3. The method according to claim 2, characterized in that The step of calculating a feature weight matrix based on the multidimensional feature vector and iteratively adjusting the initial partition interval using a clustering algorithm to obtain an adjusted feature weight matrix includes: Calculate an initial weight matrix based on the multidimensional feature vector using matrix operations to obtain a preliminary feature weight matrix; The initial weight matrix is ​​calculated using the following formula: Among them, W ij represents the feature weight of the i-th sample in the j-th dimension, x ik represents the eigenvalue of the i-th sample in the k-th dimension, μ k represents the mean of the kth dimension, m represents the total number of feature dimensions, and n represents the total number of samples; Obtaining data grouping features in a preliminary feature weight matrix, and grouping the initial partition intervals using a K-means clustering algorithm to obtain grouped data intervals; Iteratively calculate the grouped data intervals, determine the interval adjustment direction by the error square sum, and obtain the adjusted data interval; The preliminary feature weight matrix is ​​updated according to the adjusted data interval, and the adjustment weights are extracted using matrix decomposition technology to obtain the adjusted feature weight matrix.

4. The method according to claim 3, characterized in that The K-means clustering algorithm is used to group the initial partition intervals to obtain grouped data intervals, including: The clustering quality evaluation index is calculated using the K-means clustering algorithm using the following formula: Among them, Q t represents the clustering quality evaluation index, k represents the number of clusters, and n i represents the number of samples in the i-th cluster, d(x,y) represents the distance between samples x and y, and C i represents the i-th cluster set; The initial partition intervals are grouped based on the clustering quality evaluation index to obtain data intervals.

5. The method according to claim 1, characterized in that The step of integrating the discretization parameters and the abnormal pattern data using a decision tree algorithm according to the labeling results to generate a feature vector set including risk weights includes: Obtaining discretized parameters and abnormal pattern data based on the annotation results, and performing preliminary integration using a decision tree algorithm to obtain an initial feature set; Extract abnormal patterns from the initial feature set and determine risk weights to generate weighted pattern data; The discretization parameters are adjusted according to the weighted pattern data, and the integration process is optimized using a decision tree algorithm to obtain an updated feature set; Wherein, if the abnormal pattern in the updated feature set exceeds a preset threshold, significant abnormalities are extracted through pattern recognition to obtain a risk weight distribution; The updated feature set and risk weight distribution are used to generate a feature vector set including weight allocation.

6. The method according to claim 1, characterized in that The step of performing a weighted fusion evaluation on the recommendation table and the exercise plan and iteratively adjusting the weight parameters to generate a health management plan includes: Acquire the user health data of the recommendation table and the execution characteristic data of the exercise plan; Generate initial weight parameters according to the user health data and input the exercise plan to obtain an updated exercise plan; Using a fusion algorithm to perform weighted calculation on the recommendation table and the updated exercise plan to obtain a set of candidate health management plans; Establishing an evaluation function according to the execution characteristic data to score the candidate health management solution set and obtain a weight parameter adjustment amount; The weight parameter adjustment amount is input into the fusion algorithm for iterative calculation until the evaluation function reaches a preset threshold, thereby obtaining a final health management plan; the health management plan includes personalized foot health product recommendations, customized exercise rehabilitation plans, and foot and spine health care guidance suggestions.

7. A system for generating a health management plan for the foot spine of children and adolescents, characterized in that: The system comprises: A foot data collection module is used to obtain a dynamic parameter set of the feet of children and adolescents; the dynamic parameter set includes arch height data and spinal curvature data; A data classification module is used to divide the dynamic parameter set by using the statistical quantile method and input the trained multidimensional classification model to obtain an interval division result; it is also used to extract grouping features from the interval division result and match it with abnormal pattern data to obtain a labeling result for judging the risk level; A product recommendation module is used to integrate the discretization parameters and abnormal pattern data using a decision tree algorithm according to the labeling results to generate a feature vector set including risk weights; if the arch height data in the feature vector set is lower than a preset interval and the spinal curvature exceeds a standard range, a foot health product recommendation table is generated; The health management module is used to extract exercise demand parameters from the feature vector set and generate an individualized exercise plan using a random forest algorithm; it is also used to perform a weighted fusion evaluation on the recommendation table and the exercise plan, and iteratively adjust the weight parameters to generate a health management plan.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images