Method, device, electronic device and storage medium for monitoring spine of adolescents and children

Abstract

This invention provides a method, device, electronic device, and storage medium for monitoring the spine of adolescents and children. The method includes: inputting image data of the adolescent's spine to be tested into a preset parameter encoding module of a preset spine monitoring model to obtain latent posture parameters of the adolescent's spine; inputting the latent posture parameters into a preset posture recognition module of the preset spine monitoring model to obtain visualized posture parameters of the adolescent's spine; and obtaining posture correction suggestions for the adolescent's spine based on the visualized posture parameters and a preset posture correction module of the preset spine monitoring model. The preset spine monitoring model is a model trained based on sample image data containing spinal posture labels. Using this method, posture correction suggestions can be obtained accurately and quickly, improving the efficiency of spine monitoring and also achieving the goal of preventing spinal diseases in adolescents and children caused by lying, sitting, prone, standing, and other postures.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method, device, electronic device and storage medium for monitoring the spine of adolescents and children. Background Art

[0002] With the continuous development of science and technology and the improvement of people's living standards, the forms of entertainment or learning for teenagers and children are becoming more and more diverse. They can play with mobile phones while standing or lying down, or read books while sitting or lying down. However, when standing, lying, sitting or lying down in the same posture for a long time, it is easy to cause spinal problems for teenagers and children, such as scoliosis. Therefore, how to prevent spinal problems for teenagers and children has become a hot research direction.

[0003] In the related technology, the wearable sit-stand reminder system includes a wearable tension sensing module that can sense the tension changes between the human neck and sacrum and between the two shoulder blades. When the signal sensed by the tension sensing module reaches the alarm value, the mobile phone application connected to it sends an alarm signal, thereby achieving the purpose of correcting incorrect sitting posture and protecting spinal health.

[0004] However, since the existing sit-stand reminder system only protects spinal health in the sitting and standing postures, it is not suitable for the prevention of spinal diseases in other postures, and it can only work after being worn by the human body, resulting in a narrow scope of application and low practicality for the prevention of spinal diseases. Summary of the invention

[0005] The present invention provides a method, device, electronic device and storage medium for monitoring the spine of adolescents and children, so as to overcome the defects of the prior art that the spinal health is only protected in the sitting or standing posture and the sitting or standing reminder system can only work after the human body wears the system, resulting in a narrow scope of application and poor practicality of spinal disease prevention, thereby achieving the purpose of preventing spinal diseases caused by adolescents and children in other postures such as lying, sitting, lying down and standing.

[0006] The present invention provides a method for monitoring the spine of adolescents and children, comprising:

[0007] Inputting the image data to be tested of the adolescent child's spine into a preset parameter encoding module of a preset spine monitoring model to obtain the potential posture parameters of the adolescent child's spine;

[0008] Inputting the latent posture parameters into the preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the spine of the adolescent child;

[0009] Based on the visualized posture parameters and the preset posture correction module of the preset spine monitoring model, obtaining posture correction suggestions for the spine of the adolescent or child;

[0010] The preset spine monitoring model is a model trained based on sample image data containing spine posture labels.

[0011] According to a method for monitoring the spine of adolescents and children provided by the present invention, the preset posture correction module based on the visualized posture parameters and the preset spine monitoring model obtains posture correction suggestions for the spine of adolescents and children, including:

[0012] Based on the preset posture correction module of the preset spine monitoring model, obtaining the current posture score of the spine of the adolescent child;

[0013] Based on the preset posture correction module, obtaining the difference between the current posture score and the standard posture score;

[0014] Based on the differences, posture correction recommendations for the adolescent's spine are obtained.

[0015] According to a method for monitoring the spine of adolescents and children provided by the present invention, obtaining posture correction suggestions for the spine of adolescents and children based on the differences includes:

[0016] When it is determined that the difference meets the posture correction condition, obtaining the posture category of the current posture of the spine of the adolescent child;

[0017] The current posture is compared with the standard posture corresponding to the posture category, and a posture correction suggestion for the adolescent's spine is output.

[0018] According to a method for monitoring the spine of adolescents and children provided by the present invention, the step of inputting the latent posture parameters into a preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the spine of adolescents and children comprises:

[0019] Based on the preset posture recognition module of the preset spine monitoring model, the latent posture parameters are three-dimensionally reconstructed to obtain the visualized posture parameters of the spine of the adolescent child.

[0020] According to a spine monitoring method for adolescents and children provided by the present invention, the training process of the preset spine monitoring model includes:

[0021] When determining the visualization posture parameters of the adolescent spine based on the node and degree of freedom representation, sample image data containing spine posture labels and model constraints are obtained, wherein the model constraints are used to limit the rotation angle of the node to not exceed the human body limit;

[0022] Based on the model constraints, the sample image data is used to train an initial neural network model including a parameter encoding module, a parameter decoding module, a posture recognition module and a posture correction module to obtain a preset spinal monitoring model.

[0023] According to a method for monitoring the spine of adolescents and children provided by the present invention, the visualization posture parameters of the adolescent spine include k different nodes and each of the nodes is represented by a posture with n different degrees of freedom; wherein k and n are positive integers respectively.

[0024] The present invention also provides a spine monitoring device for adolescents and children, comprising:

[0025] A parameter encoding unit, used for inputting the image data to be measured of the adolescent child's spine into a preset parameter encoding module of a preset spine monitoring model to obtain the latent posture parameters of the adolescent child's spine;

[0026] A parameter recognition unit, used for inputting the latent posture parameters into a preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the spine of the adolescent child;

[0027] A posture correction unit, configured to obtain posture correction suggestions for the spine of the adolescent or child based on the visualized posture parameters and a preset posture correction module of the preset spine monitoring model;

[0028] The preset spine monitoring model is a model trained based on sample image data containing spine posture labels.

[0029] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned methods for monitoring the spine of adolescents and children is implemented.

[0030] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the spinal monitoring method for adolescents and children as described in any one of the above is implemented.

[0031] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned methods for monitoring the spine of adolescents and children.

[0032] The present invention provides a method, device, electronic device and storage medium for monitoring the spine of adolescents and children. The method for monitoring the spine of adolescents and children accurately and quickly monitors the spine of adolescents and children by inputting the image data to be tested of the spine of adolescents and children into a preset spine monitoring model to first encode the latent posture parameters, then perform visual posture recognition, and finally make posture correction suggestions. This not only improves the efficiency of spine monitoring, but also achieves the purpose of preventing spinal diseases caused by other postures such as lying, sitting, prone, and standing in adolescents and children. Furthermore, since the preset spine monitoring model is a model trained based on sample image data containing spinal posture labels, when the training accuracy of the preset spine monitoring model is improved, the monitoring accuracy of the spine of adolescents and children can also be improved accordingly. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0034] Figure 1 It is a schematic diagram of the flow of the spine monitoring method for adolescents and children provided by the present invention;

[0035] Figure 2 It is a schematic diagram of the structure of the spine monitoring device for adolescents and children provided by the present invention;

[0036] Figure 3 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Combine the following Figure 1-Figure 3The adolescent and child spine monitoring method, device, electronic device and storage medium of the present invention are described, wherein the executor of the adolescent and child spine monitoring method can be an adolescent and child spine monitoring device, and the adolescent and child spine monitoring device can be implemented as part or all of a terminal device through software, hardware or a combination of software and hardware. Optionally, the terminal device can be a personal computer (PC), a portable device, a laptop computer, a smart phone, a tablet computer, a portable wearable device or other electronic device. The present invention does not limit the specific form of the terminal device.

[0039] It should be noted that the execution subject of the following method embodiment may be part or all of the above terminal device. The following method embodiment is described by taking the execution subject as the terminal device as an example.

[0040] Figure 1 The flowchart of the spine monitoring method for adolescents and children provided by the present invention is as follows: Figure 1 As shown, the spinal monitoring method for adolescents and children includes the following steps:

[0041] Step 110: input the image data of the spine of the adolescent or child to be measured into the preset parameter encoding module of the preset spine monitoring model to obtain the latent posture parameters of the spine of the adolescent or child.

[0042] Among them, adolescent children can be boys or girls aged over six years old and under eighteen years old; the preset spinal monitoring model is a model trained based on sample image data containing spinal posture labels, that is, a model obtained by using sample image data to train an initial neural network containing a parameter encoding module, a parameter decoding module, a posture recognition module and a posture correction module. When the neural network is trained to become the preset spinal monitoring model, the parameter encoding module is trained as a preset parameter encoding module, the posture recognition module is trained as a preset posture recognition module, and the posture correction module is trained as a preset posture correction module.

[0043] Specifically, when it is determined that the adolescent child is in other postures such as lying, standing, sitting or prone, the image data to be tested of the adolescent child's spine can be obtained through a camera, lidar, RGBD camera or other sensors. The image data to be tested can be three-dimensional data that characterizes the current posture of the adolescent child, and the image data to be tested is input into a preset parameter encoding module of a preset spine monitoring model for posture parameter encoding, such as parameter encoding according to a posture parameter representation method, so as to obtain latent posture parameters of the adolescent child's spine, and the latent posture parameters are used to roughly express the current posture of the adolescent child's spine.

[0044] It should be noted that, in an embodiment of the present invention, the visualized posture parameters of the spine of adolescents and children are identified based on nodes and degrees of freedom, that is, the posture parameters include k different nodes and each node is represented by n different degrees of freedom, k and n are positive integers. The definition of nodes is conducive to extracting posture features that can distinguish the correct and incorrect postures of adolescents and children. The more nodes there are, the more accurate the defined posture is. Exemplarily, the number of nodes can be determined based on the number of joints of the human skeleton of adolescents and children, k can be a positive integer greater than or equal to 8 and less than 30, and the minimum value of k can be 8. When k is 8, the visualized posture parameters of the spine of adolescents and children can include 8 nodes of the head, left shoulder, right shoulder, cervical vertebra, thoracic vertebra, lumbar vertebra, left hip, and right hip; each node can be represented by n different degrees of freedom, and when n is 6, the posture is the three-dimensional coordinates of the node in three-dimensional space and 3 rotation angles relative to the three coordinate axes, thereby determining the position and orientation of the joints of the human skeleton of adolescents and children in three-dimensional space.

[0045] Step 120: input the latent posture parameters into the preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the spine of adolescents and children.

[0046] Specifically, the latent posture parameters of the adolescent spine are input into the preset posture recognition module of the preset spine monitoring model for posture recognition processing, that is, the latent posture parameters of the adolescent spine are converted into visual posture parameters, so as to obtain the visualized posture parameters of the adolescent spine. The visualized posture parameters of the adolescent spine can be used to accurately express the current posture of the adolescent spine. The current posture of the adolescent spine can be one fist, one foot, and one inch, high and low shoulders, scoliosis, tilted head, lowered head, etc. One fist, one foot, and one inch means that the upper body is kept upright and slightly leaned forward, the chest is one fist away from the table, the hands are held one inch away from the tip of the pen, the head is upright, naturally leaning forward, and the eyes are about one foot away from the table.

[0047] Step 130: Based on the visualized posture parameters and the preset posture correction module of the preset spine monitoring model, posture correction suggestions for the spine of adolescents and children are obtained.

[0048] Specifically, when the preset spine monitoring model includes a preset posture correction module and the preset posture correction module consists of an m-1-layer mapping layer that hides the spine posture features and an m-th layer of posture correction suggestion layer, the visualized posture parameters of the adolescent's spine are input into the preset posture correction module to first perform the first m-1 layers of layer-by-layer feature mapping operations, and then enter the m-th layer to perform posture correction suggestion operations, so as to obtain corresponding posture correction suggestions such as raising the head, straightening the chest, and leaving the table one fist away.

[0049] It should be noted that when the image data to be tested of the adolescent spine is encoded into latent posture parameters through a preset parameter encoding module, the latent posture parameters can be directly mapped to a preset posture correction module containing dozens to hundreds of eigenvalues, so as to obtain the current posture score of the adolescent spine and perform posture correction recommendations when the current posture score does not meet the standard.

[0050] The adolescent and child spine monitoring method provided by the present invention accurately and quickly monitors the adolescent and child spine by inputting the image data to be tested of the adolescent and child spine into a preset spine monitoring model to first encode the latent posture parameters, then perform visual posture recognition, and finally make posture correction suggestions. This not only improves the efficiency of spine monitoring, but also achieves the purpose of preventing spinal diseases caused by adolescents and children in other postures such as lying, sitting, prone, and standing. Furthermore, since the preset spine monitoring model is a model trained based on sample image data containing spinal posture labels, when the training accuracy of the preset spine monitoring model is improved, the monitoring accuracy of the adolescent and child spine can also be improved accordingly.

[0051] Optionally, the implementation process of step 120 may include:

[0052] Based on the preset posture recognition module of the preset spine monitoring model, the latent posture parameters are three-dimensionally reconstructed to obtain the visualized posture parameters of the spine of the adolescent child.

[0053] Specifically, the preset posture recognition module is used to decode each latent posture parameter into a visualized three-dimensional posture parameter.

[0054] Specifically, since the visualized posture parameters of the adolescent spine are represented by k different nodes and each node is represented by n postures with different degrees of freedom, in order to improve the accuracy of subsequent posture correction, the latent posture parameters of the adolescent spine can be decoded and reconstructed through a preset posture recognition module, so as to decode and reconstruct the latent posture parameters into k*n visualized posture parameters, thereby helping to identify the relative relationship between the specific position of the adolescent spine and its specific joints.

[0055] The adolescent and child spine monitoring method provided by the present invention improves the accuracy and reliability of subsequent identification of spinal posture characteristics by decoding and reconstructing the latent posture parameters of the adolescent and child spine into visual posture parameters, thereby also providing a guarantee for the monitoring accuracy of the adolescent and child spine.

[0056] Optionally, the implementation process of step 130 may include:

[0057] First, based on the preset posture correction module of the preset spine monitoring model, the current posture score of the adolescent spine is obtained; then, based on the preset posture correction module, the difference between the current posture score and the standard posture score is obtained; further based on the difference between the current posture score and the standard posture score, the posture correction suggestion for the adolescent spine is obtained.

[0058] Specifically, the visualized posture parameters of the adolescent spine are input into a preset posture correction module, and a layer-by-layer feature mapping operation is first performed on the first m-1 layers to complete the mapping operation from the visualized posture parameters to the spinal posture features, thereby identifying the final posture features of the adolescent spine, and the final posture features are input into the mth layer for posture correction suggestion operation, that is, the final posture features are first scored to obtain the current posture score of the adolescent spine, and then the current posture score is compared with the standard posture score to obtain the difference between the current posture score and the standard posture score. For example, when the standard posture score ranges from 0 to 100 points and 100 points is the standard posture score, if the current posture score is less than 90 points, it means that the current posture of the adolescent spine is not good enough and maintaining this posture is likely to induce spinal diseases. Therefore, the terminal device can obtain the posture correction suggestion for the adolescent spine based on the difference between the current posture score and the standard posture score, such as based on the score difference between the current posture score and the standard posture score and the pre-stored mapping relationship between the score difference and the posture correction suggestion.

[0059] It should be noted that if the current posture score of the adolescent's spine is between 90 and 100 points, it means that no correction is needed. If the current posture score is between 80 and 90, it means that the current posture is not too bad and needs correction. If the current posture score is close to 50, it means that the current posture may be high and low shoulders, or lying on the table at a far distance from the table. At this time, posture correction suggestions are output and the correction range is large.

[0060] The adolescent and child spine monitoring method provided in the embodiment of the present invention achieves the purpose of preventing spinal diseases such as scoliosis by scoring the current posture of the adolescent spine and providing posture correction recommendations in a preset posture correction module, thereby effectively improving the possibility and reliability of early prevention, early detection, and early treatment of adolescent and child spinal diseases.

[0061] Optionally, based on the difference between the current posture score and the standard posture score, a posture correction suggestion for the spine of adolescents and children is obtained, and the implementation process may include:

[0062] First, when it is determined that the difference meets the posture correction conditions, the posture category of the current posture of the adolescent's spine is obtained; then, the current posture is compared with the standard posture corresponding to the posture category, and a posture correction suggestion for the adolescent's spine is output.

[0063] Among them, the posture correction condition is used to characterize that the difference between the current posture score of the spine of adolescent children and the standard posture score is sufficient to explain that the current posture of their spine affects spinal health.

[0064] Specifically, in order to ensure that the spinal posture of adolescents and children can be monitored in different postures such as lying, sitting, prone, and standing, standard spinal postures corresponding to different posture categories can be pre-set, that is, when the posture categories include but are not limited to sitting, standing, prone, and lying, the standard spinal posture corresponding to the sitting posture, the standard spinal posture corresponding to the standing posture, the corresponding standard spinal posture, and the standard spinal posture quasi-spinal posture corresponding to the lying posture can be set; and a standard posture score corresponding to each standard spinal posture is set, and each standard posture score can be a score value or a score range. No specific limitation is made here.

[0065] Based on this, the terminal device can first determine whether the difference between the current posture score and the standard posture score of the adolescent's spine meets the posture correction conditions, that is, determine the degree of deviation of the adolescent's spine from the standard spine posture of the corresponding posture category, that is, the difference or difference range between the current posture score and the standard spine posture score of the corresponding posture category, and when it is determined that the current posture score is less than the standard posture score value of the corresponding posture category or is not within the standard posture score range of the corresponding posture category, it can be determined that the current posture of the adolescent's spine needs to be corrected. At this time, the current posture score of the adolescent's spine can be obtained first. For example, if the posture category of the current posture of the adolescent's spine is sitting, the current posture of the spine is compared with the standard spine posture corresponding to the sitting posture, and based on the comparison result, the posture correction suggestion of the adolescent's spine is output; for example, the current posture of the adolescent's spine is sitting with the head tilted forward, and the standard spine posture corresponding to the sitting posture includes but is not limited to sitting with the buttocks filling the entire chair and the waist and back completely in contact with the chair back, the feet are flat on the ground and as wide as the shoulders, and the back is straight and not bent, so that the posture correction suggestion can be given based on the difference between the two, so that the posture of the spine after correction is the same as the standard spine posture corresponding to the sitting posture.

[0066] It should be noted that it is also possible to give a posture correction suggestion of fine-tuning or major adjustment based on the difference between the current posture score of the spine and the standard posture score of the corresponding posture category. For example, when the standard posture score corresponding to the sitting posture is 100 points, if the current posture score of the spine is 85 points, it can be considered that the current spinal posture is not standard and needs fine-tuning. At this time, a posture correction suggestion that the current posture can be fine-tuned can be given, and the posture correction suggestion also includes how to adjust to the corresponding standard spinal posture; if the current posture score of the spine is 50 points, it can be considered that the current spinal posture is high and low shoulders or lying on the table when far away from the table. At this time, a posture correction suggestion that the current spinal posture must be greatly adjusted can be obtained, and the posture correction suggestion also includes how to adjust to the corresponding standard spinal posture. Adjustment process.

[0067] The adolescent spine monitoring method provided by the embodiment of the present invention determines that the current posture of the adolescent spine needs to be corrected based on the difference between the current posture score and the standard posture score. By comparing the current posture of the spine with the standard spinal posture of the corresponding posture category, the pertinence and accuracy of obtaining spinal posture correction suggestions are improved, thereby preventing spinal diseases caused by adolescents in other postures such as lying, sitting, lying down, and standing.

[0068] Optional, pre-set spine monitoring model training process, including:

[0069] Firstly, when determining the visualization posture parameters of the adolescent spine based on the representation of nodes and degrees of freedom, sample image data containing spinal posture labels and model constraints are obtained. The model constraints are used to limit the rotation angle of the nodes to not exceed the human body limit. Then, based on the model constraints, the sample image data is used to train the initial neural network model containing the parameter encoding module, parameter decoding module, posture recognition module and posture correction module to obtain the preset spinal monitoring model.

[0070] The parameter encoding module and parameter decoding module contained in the initial neural network constitute an autoencoder, and the training of the initial neural network includes further training the posture recognition module and the posture correction module on the basis of training the autoencoder. In addition, when the parameter encoding module is used to obtain latent posture parameters after sparse encoding the input sample image data, and the parameter decoding module is used to three-dimensionally decode and reconstruct the latent posture parameters output by the parameter encoding module into new sample image data, the training goal of the autoencoder is to minimize the difference between the input sample image data and the new sample image data.

[0071] Specifically, when it is determined that the visualization posture parameters of the spine of adolescents and children are represented by k different nodes and each node is represented by n different degrees of freedom, sample image data and model constraints containing spine posture labels are further obtained. The spine posture label represents the posture category of the corresponding sample image data, such as sitting, prone, lying or standing, and the number of sample image data of each posture category can be multiple. The model constraints are constraints on parameter encoding, parameter decoding and multi-layer perception, so that the latent posture parameters obtained by encoding, the image data obtained by decoding, the recognized visualization posture parameters and the posture correction suggestions given are all within a reasonable range. The model constraints can be that the rotation of each node cannot exceed the limit of the human body, such as the head cannot rotate 180° when the neck, head and body rotate relative to each other; for the relative position of the n degrees of freedom obtained by rotation, the relative angle between the two nodes of the head and the body cannot exceed 180°. Furthermore, the model constraints not only help to express the current posture of the spine, but also make the parameters obtained by encoding and decoding have certain constraints, thereby providing a guarantee for the reliability of subsequent multi-layer perception.

[0072] In addition, the parameter encoding module is used to encode the latent posture parameters of the input sample image data, the parameter decoding module is used to perform three-dimensional decoding and reconstruction of the latent posture parameters encoded by the parameter encoding module, the posture recognition module is used to perform posture feature mapping on the visualized posture parameters decoded by the parameter decoding module, and the posture correction module is used to perform posture scoring and posture correction suggestions on the current posture of the spine obtained by the posture recognition module. Based on this, a large amount of sample image data is used to train the initial neural network model, and in the training process, a large amount of sample image data is first used to train the autoencoder in the initial neural network, until the training goal reaches the preset requirements, it is determined that the autoencoder has been trained and the parameter encoding module and the parameter decoding module have been trained, the trained parameter encoding module is the preset parameter encoding module, and the trained parameter decoding module is the preset parameter decoding module, and then a large amount of sample image data is used to train the posture recognition module and the posture correction module, that is, the latent posture parameters output by the preset parameter encoding module are input to the posture recognition module for posture feature mapping, and the visualized posture parameters output by the posture recognition module are output to the posture correction module for posture scoring and posture correction suggestion training, until the training is completed, and the corresponding neural network model at the end of the training is determined as the preset spinal monitoring model, at this time the trained posture recognition module is the preset posture recognition module, and the trained posture correction module is the preset posture correction module.

[0073] The adolescent and child spine monitoring method provided by the present invention obtains a preset spine monitoring model for adolescent and child spine monitoring by using sample image data containing spine posture labels to train an initial neural network model containing a parameter encoding module, a parameter decoding module, a posture recognition module and a posture correction module. This combines neural network learning and encoding and decoding principles to improve the accuracy of spine monitoring, thereby effectively preventing spinal diseases caused by adolescents and children in other postures such as lying, sitting, lying prone, and standing.

[0074] The adolescent and child spine monitoring device provided by the present invention is described below. The adolescent and child spine monitoring device described below and the adolescent and child spine monitoring method described above can be referenced to each other.

[0075] Reference Figure 2 , is a schematic diagram of the structure of a spine monitoring device for adolescents and children provided by an embodiment of the present invention, Figure 2 The adolescent spine monitoring program includes:

[0076] The parameter encoding unit 210 is used to input the image data to be tested of the adolescent and child spine into the preset parameter encoding module of the preset spine monitoring model to obtain the latent posture parameters of the adolescent and child spine; the parameter recognition unit 220 is used to input the latent posture parameters into the preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the adolescent and child spine; the posture correction unit 230 is used to obtain the posture correction suggestions for the adolescent and child spine based on the visualized posture parameters and the preset posture correction module of the preset spine monitoring model; wherein the preset spine monitoring model is a model trained based on sample image data containing spine posture labels.

[0077] Optionally, the posture correction unit 230 can be specifically used to obtain the current posture score of the adolescent's spine based on the preset posture correction module of the preset spine monitoring model; obtain the difference between the current posture score and the standard posture score based on the preset posture correction module; and obtain posture correction suggestions for the adolescent's spine based on the difference.

[0078] Optionally, the posture correction unit 230 can also be used to obtain the posture category of the current posture of the adolescent's spine when determining that the difference meets the posture correction conditions; compare the current posture with the standard posture corresponding to the posture category, and output posture correction suggestions for the adolescent's spine.

[0079] Optionally, the parameter recognition unit 220 can be specifically used to perform three-dimensional reconstruction of the latent posture parameters based on a preset posture recognition module of the preset spine monitoring model to obtain visualized posture parameters of the adolescent's spine.

[0080] Optionally, the device also includes a training unit, which can be specifically used to determine that when the visualization posture parameters of the adolescent spine are represented based on nodes and degrees of freedom, sample image data containing spinal posture labels and model constraints are obtained, and the model constraints are used to limit the rotation angle of the node to not exceed the limit of the human body; based on the model constraints, the sample image data is used to train an initial neural network model containing a parameter encoding module, a parameter decoding module, a posture recognition module and a posture correction module to obtain a preset spinal monitoring model.

[0081] Optionally, in the parameter identification unit 220, the visualization posture parameters of the adolescent spine include k different nodes and each of the nodes is represented by a posture with n different degrees of freedom; wherein k and n are positive integers respectively.

[0082] Figure 3 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 3 As shown, the electronic device 300 may include: a processor 310, a communication interface 320, a memory 330 and a communication bus 340, wherein the processor 310, the communication interface 320 and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the adolescent and child spine monitoring method, which includes:

[0083] Inputting the image data to be tested of the adolescent child's spine into a preset parameter encoding module of a preset spine monitoring model to obtain the potential posture parameters of the adolescent child's spine;

[0084] Inputting the latent posture parameters into the preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the spine of the adolescent child;

[0085] Based on the visualized posture parameters and the preset posture correction module of the preset spine monitoring model, obtaining posture correction suggestions for the spine of the adolescent or child;

[0086] The preset spine monitoring model is a model trained based on sample image data containing spine posture labels.

[0087] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0088] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the adolescent and child spine monitoring method provided by the above methods, the method includes:

[0089] Inputting the image data to be tested of the adolescent child's spine into a preset parameter encoding module of a preset spine monitoring model to obtain the potential posture parameters of the adolescent child's spine;

[0090] Inputting the latent posture parameters into the preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the spine of the adolescent child;

[0091] Based on the visualized posture parameters and the preset posture correction module of the preset spine monitoring model, obtaining posture correction suggestions for the spine of the adolescent or child;

[0092] The preset spine monitoring model is a model trained based on sample image data containing spine posture labels.

[0093] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to perform the spinal monitoring method for adolescents and children provided by the above methods, the method comprising:

[0094] Inputting the image data to be tested of the adolescent child's spine into a preset parameter encoding module of a preset spine monitoring model to obtain the potential posture parameters of the adolescent child's spine;

[0095] Inputting the latent posture parameters into the preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the spine of the adolescent child;

[0096] Based on the visualized posture parameters and the preset posture correction module of the preset spine monitoring model, obtaining posture correction suggestions for the spine of the adolescent or child;

[0097] The preset spine monitoring model is a model trained based on sample image data containing spine posture labels.

[0098] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0099] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring the spine of adolescents and children, characterized in that: include: Inputting the image data to be tested of the adolescent child's spine into a preset parameter encoding module of a preset spine monitoring model to obtain the potential posture parameters of the adolescent child's spine; The latent posture parameters are used to roughly express the current posture of the spine of the adolescent child; The latent posture parameters are input into the preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the adolescent spine; the visualized posture parameters of the adolescent spine are used to accurately express the current posture of the adolescent spine, and the visualized posture parameters of the adolescent spine include k different nodes and each node is represented by a posture with n different degrees of freedom; wherein k and n are positive integers respectively; Based on the visualized posture parameters and the preset posture correction module of the preset spine monitoring model, a posture correction suggestion for the spine of the adolescent child is obtained; the preset posture correction module is composed of an m-1 layer of mapping layers with hidden spine posture features and an m-th layer of posture correction suggestion layer; The preset spine monitoring model is a model trained based on sample image data containing spine posture labels.

2. The spine monitoring method for adolescents and children according to claim 1, characterized in that: The preset posture correction module based on the visualized posture parameter and the preset spine monitoring model obtains the posture correction suggestion for the spine of the adolescent child, including: Based on the preset posture correction module of the preset spine monitoring model, obtaining the current posture score of the spine of the adolescent child; Based on the preset posture correction module, obtaining the difference between the current posture score and the standard posture score; Based on the differences, posture correction recommendations for the adolescent's spine are obtained.

3. The spine monitoring method for adolescents and children according to claim 2, characterized in that: Based on the differences, obtaining the posture correction suggestions for the adolescent's spine includes: When it is determined that the difference meets the posture correction condition, obtaining the posture category of the current posture of the spine of the adolescent child; The current posture is compared with the standard posture corresponding to the posture category, and a posture correction suggestion for the adolescent's spine is output.

4. The spine monitoring method for adolescents and children according to claim 1, characterized in that: The step of inputting the latent posture parameters into the preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the spine of the adolescent or child includes: Based on the preset posture recognition module of the preset spine monitoring model, the latent posture parameters are three-dimensionally reconstructed to obtain the visualized posture parameters of the spine of the adolescent child.

5. The spine monitoring method for adolescents and children according to any one of claims 1 to 4, characterized in that: The training process of the preset spine monitoring model includes: When determining the visualization posture parameters of the adolescent spine based on the node and degree of freedom representation, sample image data containing spine posture labels and model constraints are obtained, wherein the model constraints are used to limit the rotation angle of the node to not exceed the human body limit; Based on the model constraints, the sample image data is used to train an initial neural network model including a parameter encoding module, a parameter decoding module, a posture recognition module and a posture correction module to obtain a preset spinal monitoring model.

6. A spinal monitoring device for adolescents and children, characterized in that: include: A parameter encoding unit, used for inputting the image data to be measured of the adolescent child's spine into a preset parameter encoding module of a preset spine monitoring model to obtain the latent posture parameters of the adolescent child's spine; The latent posture parameters are used to roughly express the current posture of the spine of the adolescent child; A parameter recognition unit, used for inputting the latent posture parameters into the preset posture recognition module of the preset spine monitoring model to obtain the visualized posture parameters of the spine of the adolescent child; the visualized posture parameters of the spine of the adolescent child are used to accurately express the current posture of the spine of the adolescent child, and the visualized posture parameters of the spine of the adolescent child include k different nodes and each of the nodes is represented by a posture with n different degrees of freedom; wherein k and n are positive integers respectively; A posture correction unit, for obtaining posture correction suggestions for the spine of the adolescent child based on the visualized posture parameters and the preset posture correction module of the preset spine monitoring model; the preset posture correction module is composed of an m-1 layer of mapping layers hiding spine posture features and an m-th layer of posture correction suggestion layer; The preset spine monitoring model is a model trained based on sample image data containing spine posture labels.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, it implements the spine monitoring method for adolescents and children as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for monitoring the spine of adolescents and children as described in any one of claims 1 to 5 is implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for monitoring the spine of adolescents and children as described in any one of claims 1 to 5 is implemented.

Images