System and method for predicting bad posture risk of children and teenagers

Abstract

The invention provides a child adolescent posture adverse risk prediction system and method, and the system comprises a first collection module which is used for collecting the gait data of a child during walking and running; the second acquisition module is used for acquiring plantar pressure distribution data of the child in dynamic and static states; the third acquisition module is used for acquiring human body measurement data; the fourth acquisition module is used for acquiring three-dimensional posture image data of the child; the fifth acquisition module is used for acquiring daily physical activity data of the children; and the processing module is used for collecting the gait data, the plantar pressure distribution data, the human body measurement data, the three-dimensional posture image data and the daily physical activity data to obtain preprocessed data, and inputting the preprocessed data into a pre-constructed LASSO regression model to obtain a corresponding posture adverse risk prediction result. The method has the beneficial effects that the accuracy of poor attitude analysis can be improved, large-scale and real-time attitude risk monitoring is realized, the limitation of environmental conditions is relieved, and an evaluation result is fed back in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of posture assessment and risk prediction, and in particular to a system and method for predicting the risk of poor posture in children and adolescents. Background Art

[0002] Poor posture among children and adolescents is an important concern in the field of global public health. Poor posture not only affects the appearance of the body, but also has a long-term negative impact on the musculoskeletal system. Existing technologies mainly include the following methods:

[0003] (1) Traditional posture assessment methods

[0004] Traditional assessment methods usually rely on manual observation and measurement tools, such as measuring rulers and levels. These methods evaluate posture based on the experience and subjective judgment of professionals. However, the accuracy and consistency of the assessment results are poor, and are greatly affected by the professional level and experience of the assessors. They are time-consuming and labor-intensive, and cannot achieve large-scale, real-time posture monitoring.

[0005] (2) Posture assessment system based on imaging technology

[0006] The posture of the person being evaluated is captured by a camera, and posture assessment is performed through image processing and analysis software. However, the system is expensive, complicated to install and use, requires good environmental conditions (such as lighting and background) to ensure the accuracy of the results, and the data processing and analysis process is complicated, making it difficult to provide real-time feedback on the evaluation results.

[0007] In summary, the existing technology for analyzing poor posture of children and adolescents has the problems of poor accuracy, inability to achieve large-scale and real-time posture monitoring, high environmental conditions requirements, and difficulty in real-time feedback of evaluation results. Summary of the invention

[0008] The technical problem to be solved by the present invention is to improve the accuracy of bad posture analysis, realize large-scale and real-time posture risk monitoring, remove the restrictions of environmental conditions, and provide real-time feedback of evaluation results. In order to overcome the defects of the above-mentioned prior art (or related technology), the present invention provides a system for predicting the risk of bad posture for children and adolescents.

[0009] The present invention provides a system for predicting the risk of poor posture of children and adolescents, comprising:

[0010] a first acquisition module, connected to an external gait analysis system, for acquiring gait data of children walking and running output by the gait analysis system;

[0011] a second acquisition module, connected to an external plantar pressure analysis system, for acquiring plantar pressure distribution data of children in dynamic and static states output by the plantar pressure analysis system;

[0012] a third acquisition module, connected to an external anthropometry device, and configured to acquire anthropometry data output by the anthropometry device;

[0013] a fourth acquisition module, connected to an external depth camera, and configured to acquire three-dimensional posture image data of the child output by the depth camera;

[0014] a fifth acquisition module, connected to an external activity monitoring device, for acquiring the daily physical activity data of the child output by the activity monitoring device;

[0015] A processing module, respectively connected to the first acquisition module, the second acquisition module, the third acquisition module, the fourth acquisition module and the fifth acquisition module, for collecting the gait data, the plantar pressure distribution data, the anthropometric data, the three-dimensional posture image data and the daily physical activity data and preprocessing them to obtain preprocessed data, and inputting the preprocessed data into a pre-built LASSO regression model to obtain a corresponding poor posture risk prediction result.

[0016] In a possible implementation, the gait data collected by the first collection module includes the frequency and stride of the child when walking and running.

[0017] In a possible implementation, the anthropometric data collected by the third collection module includes the child's height, weight, and body mass index.

[0018] In a possible implementation, the daily physical activity data collected by the fifth collection module includes the child's exercise volume and sedentary time.

[0019] In a possible implementation, the processing module includes:

[0020] a data acquisition unit, configured to collect the gait data, the plantar pressure distribution data, the anthropometric data, the three-dimensional posture image data and the daily physical activity data;

[0021] a first processing unit, connected to the data acquisition unit, for sequentially performing data cleaning, normalization and feature extraction on the gait data, the plantar pressure distribution data, the anthropometric data, the three-dimensional posture image data and the daily physical activity data to obtain the pre-processed data;

[0022] A second processing unit is connected to the first processing unit, and is used to call the LASSO regression model, and input the pre-processed data into the LASSO regression model to analyze and obtain a poor posture risk score as the poor posture risk prediction result.

[0023] Another technical solution of the present invention is to provide a method for predicting the risk of poor posture of children and adolescents, which is applied to the above-mentioned system for predicting the risk of poor posture of children and adolescents, and includes the following steps:

[0024] Step S1, collecting the gait data of the child when walking and running output by the gait analysis system, the plantar pressure distribution data of the child in dynamic and static states output by the plantar pressure analysis system, the anthropometric data output by the anthropometric device, the three-dimensional posture image data of the child output by the depth camera, and the daily physical activity data of the child output by the activity monitoring device;

[0025] Step S2, preprocessing the gait data, the plantar pressure distribution data, the anthropometric data, the three-dimensional posture image data and the daily physical activity data to obtain preprocessed data;

[0026] Step S3, inputting the preprocessed data into a pre-built LASSO regression model to obtain a corresponding poor posture risk prediction result.

[0027] In a possible implementation, in step S1, the frequency and stride of the child when walking and running are collected as the gait data.

[0028] In a possible implementation, in step S1, the height, weight and body mass index of the child are collected as the anthropometric data.

[0029] In a possible implementation, in step S1, the amount of exercise and sedentary time of the child are collected as the daily physical activity data.

[0030] In a possible implementation, in step S3, the LASSO regression model is established by the following expression:

[0031]

[0032] in,

[0033] min means minimizing the function;

[0034] n represents the sample size;

[0035] y i Represents the actual output of the i-th sample;

[0036] X i represents the sample set of the i-th category;

[0037] β represents the set of regression coefficients;

[0038] λ represents the regularization parameter;

[0039] β j represents the regression coefficient of the jth feature.

[0040] Compared with the prior art, the system and method for predicting the risk of poor posture of children and adolescents in this application have the following advantages:

[0041] In this application, a variety of advanced evaluation equipment and data processing technologies are integrated, and gait analysis systems, plantar pressure analysis systems, anthropometry equipment, depth cameras and activity monitoring equipment are used as external data source equipment to ensure the reliability of data sources. Then, data of various categories are collected through corresponding acquisition modules. The processing module first improves the accuracy of the data through data preprocessing and deletes invalid data. Then, the LASSO regression model is used to analyze and predict the preprocessed data to obtain the poor posture risk prediction results, so as to assist operators in obtaining targeted posture correction suggestions and intervention measures to meet the specific needs of different children and adolescents. The implementation process of the system and method in this application is not limited by environmental conditions, and can accurately analyze large amounts of data in real time, and provide real-time feedback on poor posture risk prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a structural principle diagram of the present invention;

[0043] Figure 2 is a flow chart of the steps of the present invention;

[0044] Explanation of the accompanying drawings: 1. First acquisition module; 2. Second acquisition module; 3. Third acquisition module; 4. Fourth acquisition module; 5. Fifth acquisition module; 6. Processing module; 61. Data acquisition unit; 62. First processing unit; 63. Second processing unit. DETAILED DESCRIPTION

[0045] First, those skilled in the art should understand that these implementations are only used to explain the technical principles of the embodiments of the present application, and are not intended to limit the protection scope of the embodiments of the present application. Those skilled in the art can make adjustments to them as needed to adapt to specific application scenarios.

[0046] The present application is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] See also Figure 1The embodiment of the present application discloses a system for predicting the risk of poor posture for children and adolescents, which aims to provide an efficient and accurate poor posture assessment and risk prediction system by integrating a variety of advanced evaluation equipment and data processing technologies. The uses of this application include real-time monitoring and evaluation of the posture of children and adolescents, early detection of potential poor posture problems, and providing personalized correction suggestions and intervention measures to promote the healthy growth of children.

[0048] Continue to see Figure 1 , the existing technology has the following shortcomings in the assessment of poor posture: 1. Limitations of the assessment method: Traditional methods rely on manual observation and subjective judgment, and the results are easily affected by the professional level and experience of the assessor, resulting in poor accuracy and consistency; 2. Inconvenient equipment use: Although the system based on wearable devices and imaging technology has improved the objectivity and accuracy of the assessment, the equipment cost is high, the installation and use are complex, and the environment is highly dependent, making it difficult to widely popularize and apply; 3. Insufficient real-time and feedback: The existing system lags behind in data processing and analysis, making it difficult to achieve real-time monitoring and feedback, and unable to provide personalized correction suggestions and intervention measures in a timely manner; The system of this application has the following advantages: 1. Multi-device integration and synchronous data collection: Realize seamless integration and synchronous data collection of multiple evaluation devices to ensure the comprehensiveness and real-time nature of the evaluation data; 2. Efficient data processing and analysis system: Use advanced data analysis algorithms and models to improve the accuracy and reliability of the evaluation results; 3. Personalized health management plan: Generate personalized posture correction suggestions and intervention measures based on the evaluation results to meet the specific needs of different children.

[0049] Continue to see Figure 1 This application is based on the collection and analysis of multi-dimensional data, and uses the LASSO regression algorithm to achieve efficient and accurate assessment of the risk of poor posture. The model integrates multi-dimensional data such as gait parameters, dynamic and static plantar pressure data, anthropometric data, socio-demographic data, and daily physical activity, aiming to provide personalized posture correction suggestions and intervention measures to promote the healthy growth of children and adolescents. The structure consists of:

[0050] Gait analysis system: used to collect gait parameters of children when walking and running, including data such as step frequency and stride length;

[0051] Plantar pressure analysis system: used to collect the plantar pressure distribution data of children in dynamic and static states, and a foot pressure plate can be used;

[0052] Anthropometric equipment: used to collect basic anthropometric data such as children's height, weight, BMI (body mass index), etc.

[0053] Depth camera: used to capture children's three-dimensional posture image data and analyze the posture changes of various parts of the body;

[0054] Activity monitoring device: used to record children's daily physical activity data, including exercise volume, sedentary time, etc., you can choose a smart bracelet;

[0055] Wireless transmission module: used to transmit the data collected by each evaluation device to the processing module 6 in real time;

[0056] Processing module 6: receiving, processing and analyzing the data from each evaluation device, and executing the LASSO regression model.

[0057] See also Figure 2 The present application embodiment discloses a method for predicting the risk of poor posture of children and adolescents, comprising:

[0058] (1) Data Collection

[0059] Gait data collection: Children walk or run on the gait analysis system, and the system records gait parameters in real time;

[0060] Plantar pressure distribution data collection: Children stand and walk on the plantar pressure analysis system, and the system collects dynamic and static plantar pressure distribution data;

[0061] Anthropometric data collection: basic anthropometric data such as height, weight, and BMI of children are obtained through anthropometric equipment;

[0062] Posture image acquisition: Use a depth camera to capture children’s three-dimensional posture image data and analyze posture changes;

[0063] Daily activity data collection: Use activity monitoring devices to record children’s daily physical activity data such as daily exercise volume and sedentary time;

[0064] (2) Data transmission and preprocessing

[0065] Transmitting the collected gait data, the plantar pressure distribution data, the anthropometric data, the three-dimensional posture image data and the daily physical activity data;

[0066] Preprocess the above data, including data cleaning, normalization and feature extraction, to ensure data consistency and accuracy;

[0067] (3) Construction and application of regression model

[0068] LASSO regression algorithm was used;

[0069] Set the target variable y i For the poor posture risk score, input variable matrix X i Contains all evaluation data features;

[0070] Establish LASSO regression model as a risk prediction model for poor posture:

[0071]

[0072] Among them, λ is the regularization parameter, which is used to control the complexity of the regression model and prevent overfitting;

[0073] The optimal λ value was selected through cross-validation to ensure the predictive performance of the regression model.

[0074] Continue to see Figure 2 The poor posture risk prediction results obtained in this application can help operators give risk assessment and personalized suggestions for children and adolescents. According to the poor posture risk prediction results given by the LASSO regression model, the poor posture risk of each child is scored, and personalized correction suggestions and intervention measures are generated based on the risk score, including posture correction exercises, exercise suggestions and precautions in daily life.

[0075] Continue to see Figure 1 A user interface can be added to the system of the present application, connected to the processing module 6, and the assessment results and correction suggestions can be displayed through the user interface. Users can view detailed reports through the intelligent interactive screen or mobile application, and the system of the present application supports real-time data updates and feedback, so users can view the latest assessment results and health suggestions at any time.

[0076] Continue to see Figure 1 and Figure 2 During the specific operation, the gait analysis system and the plantar pressure analysis system are installed in the test area to ensure that the dynamic and static plantar pressure distribution data and gait data can be collected in real time. The anthropometric measurement equipment and the depth camera are installed in the test area to ensure that the basic anthropometric data such as the height, weight, BMI and three-dimensional posture image data of the children can be collected. The activity monitoring equipment is worn on the children to record the daily physical activity data.

[0077] Continue to see Figure 1 and Figure 2 During the specific operation, the child stands at the far end of the instrument and uses the depth camera to collect body posture data. The system prompts the child to walk forward, the gait analysis system collects dynamic gait parameters, the plantar pressure analysis system collects dynamic plantar pressure parameters, and the depth camera collects body posture during walking. When the child stands in the test area, the depth camera captures three-dimensional foot data, the plantar pressure analysis system and the anthropometry equipment collect weight, static foot pressure, and body composition, and the ultrasonic height measuring instrument on the top of the head collects height data. Each evaluation device transmits the collected data to the processing module 6 in real time through the wireless transmission module.

[0078] Continue to see Figure 1and Figure 2 The LASSO regression algorithm used in this application can be replaced by a ridge regression algorithm. The ridge regression algorithm is used instead of the LASSO regression algorithm to construct a poor posture risk prediction model. Ridge regression is a linear regression method used to deal with multicollinearity problems. It reduces the estimated value of the regression coefficient by adding a regularization term, thereby improving the stability of the model. Ridge regression works better in dealing with multicollinearity problems and can improve the stability of the model. However, compared with the LASSO regression algorithm, the ridge regression algorithm cannot achieve variable selection and may introduce some irrelevant features, resulting in poor interpretability of the model.

[0079] Continue to see Figure 1 and Figure 2 The LASSO regression algorithm used in this application can be replaced by the principal component analysis method. The principal component analysis method is used to reduce the dimensionality of multidimensional data, and then the reduced dimensionality data is used to construct a poor posture risk prediction model. The principal component analysis method is a commonly used dimensionality reduction technology. It projects high-dimensional data into a low-dimensional space through linear transformation, extracts main features, and reduces data dimensions. The principal component analysis method can effectively reduce data dimensions and reduce the computational complexity of the model. However, some information may be lost in the dimensionality reduction process, affecting the prediction accuracy of the poor posture risk prediction model; the results of the principal component analysis method are not easy to interpret, and it is difficult to understand the actual meaning of each principal component.

[0080] Continue to see Figure 1 and Figure 2 The LASSO regression algorithm used in this application can be replaced by a support vector machine algorithm. The support vector machine algorithm is used instead of the LASSO regression algorithm to perform poor posture risk assessment. The support vector machine algorithm is a machine learning algorithm commonly used in classification and regression analysis. It performs data classification or regression by constructing a hyperplane in a high-dimensional space. The support vector machine algorithm has advantages in processing nonlinear data and high-dimensional data and can find the global optimal solution. However, the support vector machine algorithm has a high computational complexity, low efficiency in processing large-scale data, and complex model parameter selection and tuning.

[0081] Continue to see Figure 1 and Figure 2The LASSO regression algorithm used in this application can be replaced by a personalized suggestion generation method based on a decision tree. The decision tree algorithm is used to generate personalized posture correction suggestions and intervention measures. The decision tree algorithm is a machine learning method based on a tree structure. It divides the data layer by layer and constructs decision rules to achieve classification or regression analysis. The rules generated by the decision tree algorithm are intuitive and easy to understand, and are easy to explain and apply. The decision tree algorithm can automatically select the optimal features for division and improve the accuracy of the model. However, the decision tree algorithm is prone to overfitting and is sensitive to noise and outliers. It is necessary to use pruning and other technologies to improve the generalization ability of the model.

[0082] Continue to see Figure 1 and Figure 2 The LASSO regression model used in this application can be replaced by a Bayesian network model. The Bayesian network is used to fuse and comprehensively analyze multidimensional data and build a poor posture risk prediction model. The Bayesian network is a graphical model based on probability theory. It realizes data fusion and reasoning by constructing conditional dependencies between variables. The Bayesian network can handle uncertainty and complex dependencies and is suitable for the fusion and comprehensive analysis of multidimensional data. However, the construction and reasoning process of the Bayesian network is relatively complicated, relies heavily on prior knowledge, and has high computational complexity.

[0083] Continue to see Figure 1 and Figure 2 Although the above-mentioned alternative solutions can solve the relevant technical problems to a certain extent, they are still insufficient in comprehensive performance and applicability. In contrast, the present application adopts LASSO regression algorithm and multidimensional data analysis technology, and through scientific data preprocessing and comprehensive analysis, it realizes efficient and accurate poor posture risk assessment and personalized suggestion generation, which has obvious technical advantages and application prospects. Therefore, the technical solution of the present invention still has significant advantages in terms of data processing accuracy, efficiency and accuracy of personalized suggestions, and can better meet the needs of poor posture risk assessment and management for children and adolescents.

[0084] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "in the present embodiment", "specific example", or "some examples" etc. means that the specific features, mechanisms, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, mechanisms, materials or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0085] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A system for predicting the risk of poor posture among children and adolescents, characterized in that: include: a first acquisition module (1), connected to an external gait analysis system, and used to acquire gait data of children walking and running output by the gait analysis system; a second acquisition module (2), connected to an external plantar pressure analysis system, and used to acquire plantar pressure distribution data of the child in dynamic and static states output by the plantar pressure analysis system; a third acquisition module (3), connected to an external anthropometry device, and used for acquiring anthropometry data output by the anthropometry device; a fourth acquisition module (4), connected to an external depth camera, and used for acquiring three-dimensional posture image data of the child output by the depth camera; a fifth collection module (5), connected to an external activity monitoring device, and used to collect the daily physical activity data of the child output by the activity monitoring device; A processing module (6) is respectively connected to the first acquisition module (1), the second acquisition module (2), the third acquisition module (3), the fourth acquisition module (4) and the fifth acquisition module (5), and is used to collect the gait data, the plantar pressure distribution data, the anthropometric data, the three-dimensional posture image data and the daily physical activity data, and perform preprocessing to obtain preprocessed data, and input the preprocessed data into a pre-constructed LASSO regression model to obtain a corresponding poor posture risk prediction result.

2. The system for predicting the risk of poor posture of children and adolescents according to claim 1 is characterized in that: The gait data collected by the first collection module (1) includes the frequency and stride of the child when walking and running.

3. The system for predicting the risk of poor posture of children and adolescents according to claim 1 is characterized in that: The anthropometric data collected by the third collection module (3) includes the child's height, weight and body mass index.

4. The system for predicting the risk of poor posture of children and adolescents according to claim 1, characterized in that: The daily physical activity data collected by the fifth collection module (5) includes the amount of exercise and sedentary time of the child.

5. The system for predicting the risk of poor posture of children and adolescents according to claim 1 is characterized in that: The processing module (6) comprises: a data acquisition unit (61), used for collecting the gait data, the plantar pressure distribution data, the anthropometric data, the three-dimensional posture image data and the daily physical activity data; a first processing unit (62), connected to the data acquisition unit (61), and configured to sequentially perform data cleaning, normalization and feature extraction processing on the gait data, the plantar pressure distribution data, the anthropometric data, the three-dimensional posture image data and the daily physical activity data to obtain the pre-processed data; A second processing unit (63) is connected to the first processing unit (62) and is used to call the LASSO regression model and input the pre-processed data into the LASSO regression model to analyze and obtain a poor posture risk score as the poor posture risk prediction result.

6. A method for predicting the risk of poor posture in children and adolescents, characterized in that: The system for predicting the risk of poor posture of children and adolescents as claimed in any one of claims 1 to 5 comprises the following steps: Step S1, collecting the gait data of the child when walking and running output by the gait analysis system, the plantar pressure distribution data of the child in dynamic and static states output by the plantar pressure analysis system, the anthropometric data output by the anthropometric device, the three-dimensional posture image data of the child output by the depth camera, and the daily physical activity data of the child output by the activity monitoring device; Step S2, preprocessing the gait data, the plantar pressure distribution data, the anthropometric data, the three-dimensional posture image data and the daily physical activity data to obtain preprocessed data; Step S3, inputting the preprocessed data into a pre-built LASSO regression model to obtain a corresponding poor posture risk prediction result.

7. The method for predicting the risk of poor posture of children and adolescents according to claim 6, characterized in that: In the step S1, the step frequency and stride length of the child when walking and running are collected as the gait data.

8. The method for predicting the risk of poor posture of children and adolescents according to claim 6, characterized in that: In the step S1, the height, weight and body mass index of the child are collected as the anthropometric data.

9. The method for predicting the risk of poor posture of children and adolescents according to claim 6, characterized in that: In the step S1, the amount of exercise and sedentary time of the child are collected as the daily physical activity data.

10. The method for predicting the risk of poor posture of children and adolescents according to claim 6, characterized in that: In step S3, the LASSO regression model is established by the following expression: in, min means minimizing the function; n represents the number of sample data; y i Represents the actual output of the i-th sample data; X i Represents the sample data set of the i-th category; β represents the set of regression coefficients; λ represents the regularization parameter; β j represents the regression coefficient of the jth feature.

Images