Search

CN-114864090-B - Data intelligence-based overweight and obese child intervention effect evaluation system

CN114864090BCN 114864090 BCN114864090 BCN 114864090BCN-114864090-B

Abstract

The application relates to the field of data intelligence, and particularly discloses an overweight and obese child intervention effect evaluation system and an overweight and obese child intervention effect evaluation method based on data intelligence, which are characterized in that the correlation characteristics of physical index data of the overweight and obese child before and after intervention are realized by using artificial intelligence technology based on deep learning and a neural network, and the exercise scheme data and the high-dimensional implicit characteristics of the diet scheme data in the intervention scheme are deeply mined, so that the effect of the overweight and obese child intervention scheme can be accurately and notarized on the basis of getting rid of the limitation and the limitation of artificial cognition. Thus, the overweight and obesity problems of the children and the teenagers can be promoted and solved.

Inventors

  • Wu Yimian

Assignees

  • 浙江大学

Dates

Publication Date
20260512
Application Date
20220523

Claims (10)

  1. 1. An intervention effect evaluation system for overweight and obese children based on data intelligence, comprising: The body index data acquisition unit before and after intervention is used for acquiring body index data of the obese children to be evaluated before intervention and body index data of the obese children to be evaluated after intervention, and the body index data comprises body mass fraction, weight, fat rate, water rate, basal metabolic rate, visceral fat grade, muscle mass, bone salt mass and protein content; An intervention plan acquisition unit configured to acquire an intervention plan including exercise plan data and diet plan data; A pre-intervention index data encoding unit, configured to pass the pre-intervention body index data through a context encoder including an embedded layer to obtain a plurality of feature vectors, and concatenate the plurality of feature vectors to obtain a first feature vector; A dry prognosis index data encoding unit for passing the intervened body index data through the context encoder including the embedded layer to obtain a plurality of feature vectors, and concatenating the plurality of feature vectors to obtain a second feature vector; The body change feature extraction unit is used for calculating a transfer matrix of the second feature vector relative to the first feature vector as a first feature matrix, wherein the transfer matrix is used for representing implicit transformation features of body indexes of obese children to be evaluated before and after intervention; a first intervention scheme encoding unit for passing motion scheme data of the intervention scheme through a temporal encoder of a joint encoder to generate a third feature vector, the temporal encoder consisting of one-dimensional convolutional layers and fully-connected layers alternately arranged; A second intervention scheme encoding unit for passing diet scheme data of the intervention scheme through a semantic encoding model of the joint encoder to generate a fourth feature vector, the semantic encoding model comprising an embedded layer and a two-way long-short-term memory model; an intervention scheme joint coding unit, configured to calculate, using a cross-modal feature fusion module of the joint encoder, an association matrix of the third feature vector and the fourth feature vector as a second feature matrix; a feature fusion unit for fusing the first feature matrix and the second feature matrix based on the response relation between the first feature matrix and the second feature matrix to obtain a classified feature matrix, and And the evaluation result generation unit is used for passing the classification feature matrix through a classifier to obtain a classification result, wherein the classification result is an effect evaluation label of the intervention scheme.
  2. 2. The data-intelligence-based overweight obese child intervention effect assessment system according to claim 1, wherein the pre-intervention index data encoding unit is further adapted to convert the pre-intervention body index data into input vectors using the embedding layers of the encoder model comprising the context of the embedding layers, respectively, to obtain a sequence of input vectors, to encode the sequence of input vectors using the converter of the encoder model comprising the context of the embedding layers, to obtain a plurality of feature vectors based on global context semantics, and to concatenate the plurality of feature vectors to obtain the first feature vector.
  3. 3. The data-intelligence-based overweight obese child's intervention effect assessment system according to claim 2, wherein the body variation feature extraction unit is further configured to calculate a transfer matrix of the second feature vector with respect to the first feature vector as the first feature matrix with the following formula; Wherein, the formula is: Wherein F represents the first eigenvector, T represents the transfer matrix, and S represents the second eigenvector.
  4. 4. The data intelligence based overweight and obese child intervention effect assessment system according to claim 3, wherein the first intervention scheme encoding unit is further configured to arrange motion scheme data of the intervention scheme into one-dimensional input vectors, one-dimensional convolutional encoding the input vectors using a one-dimensional convolutional layer of a temporal encoder of the joint encoder in the following formula to extract high-dimensional implicit correlation features between eigenvalues of respective positions in the input vectors, wherein the formula is: Wherein a is the width of the convolution kernel in the x direction, F is a convolution kernel parameter vector, G is a local vector matrix calculated by a convolution kernel function, and w is the size of the convolution kernel; the fully connected layer of the time sequence encoder of the joint encoder is used for fully connected encoding the input vector in the following formula to extract high-dimensional implicit characteristics of characteristic values of all positions in the input vector, wherein the formula is as follows: wherein Is the input vector to be used for the input of the input device, Is the output vector of the vector, Is a matrix of weights that are to be used, Is the offset vector of the reference signal, Representing a matrix multiplication.
  5. 5. The intervention effect evaluation system for overweight and obese children based on data intelligence as claimed in claim 4, wherein the second intervention scheme encoding unit is further used for inputting the diet scheme data of the intervention scheme into an embedding layer of a semantic encoding model of the joint encoder after word segmentation to obtain a sequence of word embedding vectors corresponding to the diet scheme data of each day, and passing the sequence of word embedding feature vectors through a two-way long-short-term memory model of the semantic encoding model of the joint encoder to obtain the fourth feature vector.
  6. 6. The data-intelligence-based overweight and obese child intervention effect assessment system according to claim 5, wherein the intervention scheme joint coding unit is further adapted to calculate the correlation matrix of the third feature vector and the fourth feature vector as the second feature matrix with the following formula using a cross-modal feature fusion module of the joint encoder; Wherein, the formula is: Wherein the method comprises the steps of For the third feature vector to be used, For the fourth feature vector to be used, And the correlation matrix is obtained.
  7. 7. The data intelligence based overweight and obese child intervention effect assessment system according to claim 6, wherein the feature fusion unit is further adapted to fuse the first feature matrix and the second feature matrix to obtain the classification feature matrix in the following formula based on a response relationship between the first feature matrix and the second feature matrix; Wherein, the formula is: Wherein the method comprises the steps of Representing the exponentiation of the matrix, wherein the exponentiation of the matrix represents the exponentiation of the value of each position of the matrix, filling the result into each position of the matrix to obtain the matrix operation result, And Respectively represent the position-wise subtraction and addition of two matrices, and The representation number and the matrix are multiplied by elements, Is a super parameter.
  8. 8. The data intelligence based intervention effect evaluation system of overweight and obese children as claimed in claim 7 wherein said evaluation result generation unit is further configured to process said classification feature matrix by said classifier to generate a classification result with the following formula: wherein Representing the projection of the classification feature matrix as a vector, To the point of For the weight matrix of each full connection layer, To the point of Representing the bias matrix for each fully connected layer.
  9. 9. An evaluation method of an intervention effect evaluation system for overweight and obese children based on data intelligence is characterized by comprising the following steps: Acquiring physical index data of the obese children to be evaluated before intervention and physical index data of the obese children to be evaluated after intervention, wherein the physical index data comprise body mass fraction, body weight, fat rate, water rate, basal metabolic rate, visceral fat level, muscle mass, bone salt mass and protein content; Acquiring an intervention program, wherein the intervention program comprises exercise program data and diet program data; Passing the pre-intervention body marker data through a context encoder comprising an embedded layer to obtain a plurality of feature vectors, and concatenating the plurality of feature vectors to obtain a first feature vector; passing the intervened body-index data through the context encoder comprising an embedded layer to obtain a plurality of feature vectors, and concatenating the plurality of feature vectors to obtain a second feature vector; Calculating a transfer matrix of the second feature vector relative to the first feature vector as a first feature matrix, wherein the transfer matrix is used for representing implicit transformation features of body indexes of obese children to be evaluated before and after intervention; passing the motion scheme data of the intervention scheme through a timing encoder of a joint encoder to generate a third feature vector, the timing encoder consisting of one-dimensional convolution layers and full-connection layers alternately arranged; Passing the diet program data of the intervention program through a semantic coding model of the joint encoder to generate a fourth feature vector, wherein the semantic coding model comprises an embedded layer and a two-way long-short-term memory model; calculating an incidence matrix of the third feature vector and the fourth feature vector as a second feature matrix by using a cross-modal feature fusion module of the joint encoder; fusing the first feature matrix and the second feature matrix based on a response relationship between the first feature matrix and the second feature matrix to obtain a classification feature matrix, and And the classification feature matrix passes through a classifier to obtain a classification result, wherein the classification result is an effect evaluation label of the intervention scheme.
  10. 10. The method of assessing the effects of intervention in a data-intelligent overweight and obese child based intervention assessment system according to claim 9, wherein passing the pre-intervention physical metric data through a context encoder comprising an embedded layer to obtain a plurality of feature vectors and concatenating the plurality of feature vectors to obtain a first feature vector, comprising: The method comprises the steps of obtaining a first feature vector, respectively converting body index data before intervention into input vectors by using an embedding layer of an encoder model containing a context of the embedding layer to obtain a sequence of input vectors, performing global-based context semantic coding on the sequence of input vectors by using a converter of the encoder model containing the context of the embedding layer to obtain a plurality of feature vectors, and cascading the plurality of feature vectors to obtain the first feature vector.

Description

Data intelligence-based overweight and obese child intervention effect evaluation system Technical Field The invention relates to the field of intelligent intervention of overweight and obese children, and more particularly relates to an intervention effect evaluation system and an intervention effect evaluation method of overweight and obese children based on data intelligence. Background Over the last 30 years, global childhood overweight obesity rates have shown a growing trend and have become a serious public health problem. The prevalence of overweight and obesity in chinese school-age children increased from 5.3% by 20.5% between 1985 and 2014. In 2020, china 'embodiment for preventing and controlling obesity of children and teenagers' set forth the general goal of reducing overweight rate and obesity rate. The "healthy China 2030" planning schema "proposes to shape the autonomous healthy behavior, and achieves the aim of significantly slowing down the growth rate of overweight and obese people in 2030. Overweight and obesity are risk factors for diabetes and hypertension in children and adolescents, and also place a psychological and cognitive burden on the population. Childhood obesity generally results in adult obesity with associated risks of chronic diseases including heart metabolic disease, non-alcoholic fatty liver, type II diabetes, kidney disease, and the like. Researches show that bad life behaviors such as low dietary diversity, low physical activity, screen-related sedentary behavior and the like are risk factors of obesity, and the screen-time limiting device has reasonable structure, controls diet of caloric intake and regular physical exercise, and is beneficial to reducing body fat. Native land a number of lifestyle intervention studies are carried out inside and outside the body to explore an intervention mode for effectively preventing overweight and obesity. However, searching for low-cost, sustainable, easy-to-popularize intervention means is a current urgent problem to be solved. It should be understood how an accurate and fair assessment of the intervention effect of an intervention is critical prior to the popularization of the intervention. Thus, an evaluation regimen for evaluating the effect of intervention in overweight obese children is desired. Disclosure of Invention The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides an overweight and obese child intervention effect evaluation system and an overweight and obese child intervention effect evaluation method based on data intelligence, which can accurately and equitably evaluate the effect of the overweight and obese child intervention scheme on the basis of getting rid of the limitation and the limitation of artificial cognition by using an artificial intelligence technology based on deep learning and a neural network to perform deep mining on the correlation characteristics of physical index data of the overweight and obese child before and after intervention and the high-dimensional implicit characteristics of exercise scheme data and diet scheme data in the intervention scheme. Thus, the overweight and obesity problems of the children and the teenagers can be promoted and solved. According to one aspect of the present application, there is provided a data intelligence based intervention effect evaluation system for overweight and obese children, comprising: The body index data acquisition unit before and after intervention is used for acquiring body index data of the obese children to be evaluated before intervention and body index data of the obese children to be evaluated after intervention, and the body index data comprises body mass fraction, weight, fat rate, water rate, basal metabolic rate, visceral fat grade, muscle mass, bone salt mass and protein content; An intervention plan acquisition unit configured to acquire an intervention plan including exercise plan data and diet plan data; A pre-intervention index data encoding unit, configured to pass the pre-intervention body index data through a context encoder including an embedded layer to obtain a plurality of feature vectors, and concatenate the plurality of feature vectors to obtain a first feature vector; A dry prognosis index data encoding unit for passing the intervened body index data through the context encoder including the embedded layer to obtain a plurality of feature vectors, and concatenating the plurality of feature vectors to obtain a second feature vector; The body change feature extraction unit is used for calculating a transfer matrix of the second feature vector relative to the first feature vector as a first feature matrix, wherein the transfer matrix is used for representing implicit transformation features of body indexes of obese children to be evaluated before and after intervention; a first intervention scheme encoding unit for passing motion scheme data of th