Adolescent spine monitoring method, device, equipment, storage medium and program product

Abstract

The present application provides a method, device, equipment, storage medium and program product for monitoring the spine of adolescents. The method includes: obtaining monitoring data of adolescent users in daily life; the monitoring data includes at least one of sitting data, standing data, lying data and backpack data; using a spine analysis model to analyze the monitoring data to obtain analysis results; the analysis results are used to characterize whether the adolescent user has spinal health risks in the current state. The embodiments of the present application obtain monitoring data of adolescents in daily life and analyze the acquired monitoring data, so as to dynamically and in real time analyze the spinal health risks of adolescents, and issue early warning reminders when it is determined that there are spinal health risks.

Description

Technical Field

[0001] The present application relates to the field of data monitoring technology, and more specifically, to a method, device, equipment, storage medium and program product for monitoring the spine of adolescents. Background Art

[0002] During adolescence, growth and development are rapid, and spinal abnormalities are more likely to occur. If not discovered and treated in time, it may seriously affect the individual's growth and development and future quality of life.

[0003] With the rapid development of medical imaging technology, image processing technology has become a common method to assist in the diagnosis of spinal problems. However, this method often requires going to the hospital for professional evaluation, and corrective treatment is not started until the adolescent has already developed spinal problems, which has a certain lag. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method, device, equipment, storage medium and program product for monitoring the spine of adolescents in real time and dynamically.

[0005] In a first aspect, an embodiment of the present application provides a method for monitoring the spine of a teenager, comprising:

[0006] Acquire monitoring data of the adolescent user in daily life; the monitoring data includes at least one of sitting data, standing data, lying data and backpack data;

[0007] The monitoring data is analyzed using a spine analysis model to obtain analysis results; the analysis results are used to characterize whether the adolescent user has spinal health risks in the current state.

[0008] The embodiment of the present application obtains monitoring data of adolescents in their daily lives and analyzes the acquired monitoring data, so as to dynamically and in real time analyze the adolescents' spinal health risks.

[0009] In any embodiment, the sitting posture data includes a first spinal curvature and a first cervical angle in a sitting state;

[0010] The first spinal curvature and the first cervical vertebra angle are acquired through a posture sensor arranged on the adolescent user.

[0011] In the embodiment of the present application, the first spinal curvature and the first cervical vertebra angle of the adolescent in a sitting state can reflect whether the adolescent's sitting posture is correct. The sitting posture will affect the adolescent's spinal health. Therefore, the first spinal curvature and the first cervical vertebra angle collected by the posture sensor can be used as a factor for spinal health monitoring.

[0012] In any embodiment, the sitting posture data further includes pressure data collected by pressure sensors disposed on the seat back and / or the seat cushion.

[0013] In the embodiment of the present application, if the pressure on the seat back and / or seat cushion is unbalanced when the teenager is sitting, it means that the teenager's sitting posture is incorrect, which will also affect the health of the spine. Therefore, the pressure data of the seat back and seat cushion can also be used as a factor in spinal health monitoring.

[0014] In any embodiment, the sitting posture data also includes sitting time and sitting frequency;

[0015] The duration and frequency of sitting are monitored through smart bracelets or smart seat cushions.

[0016] Sitting for a long time or sitting frequently will cause certain pressure on the spine, especially incorrect sitting posture, which will have a greater negative impact on the spine. Therefore, it can be monitored through smart bracelets to provide data support for subsequent spinal health tests.

[0017] In any embodiment, the sitting posture data further includes an image of the back in a sitting posture;

[0018] The back image is acquired by an image acquisition device.

[0019] The back image in the sitting state can reflect whether the sitting posture is standard to a certain extent, and can be used to provide data support for subsequent spinal health testing.

[0020] In any embodiment, the standing posture data includes a second spinal curvature and a second cervical angle in a standing posture;

[0021] The second spinal curvature and the second cervical angle are acquired by collecting the posture sensor arranged on the adolescent user.

[0022] When the adolescent user is in a standing position, the second spinal curvature and the second cervical angle can be sensed by a posture sensor set on the adolescent user. Since long-term scoliosis or hunchback in a standing position will affect spinal health, and standing posture occupies a large part of daily life, the second spinal curvature and the second cervical angle of the adolescent user in a standing position can be collected for monitoring spinal health.

[0023] In any embodiment, the standing posture data further includes an image of the back in the standing posture;

[0024] The back image is obtained by capturing the image of the young user in a standing state through an image acquisition device.

[0025] When adolescent users are standing, the back image can reflect the adolescents' spinal problems and standing posture problems to a certain extent, providing data support for subsequent spinal analysis.

[0026] In any embodiment, the lying posture data includes sleeping posture, sleeping duration and sleeping quality;

[0027] Sleeping posture, sleep duration and sleep quality are collected through smart mattresses and / or smart bracelets.

[0028] Long-term use of inappropriate pillows, sleeping on your stomach and other bad sleeping positions can also affect the spinal development of adolescents. Therefore, sleeping posture data is collected through smart mattresses and / or smart bracelets for subsequent spinal health monitoring.

[0029] In any embodiment, the backpack data includes the weight of the backpack, the pressure difference between the two shoulders, and the time of carrying the backpack;

[0030] The weight of the backpack, the pressure difference between the two shoulders and the length of time the backpack is carried are collected through sensors installed on the shoulder straps of the backpack.

[0031] As young users often need to carry a lot of books, overly heavy bags will put pressure on the spine. When the pressure on both shoulders is unbalanced, or even if the bag is carried on one shoulder, it is more likely to cause scoliosis. Therefore, spinal health can be monitored by collecting backpack data.

[0032] In any embodiment, the monitoring data further includes exercise data, and the exercise data includes exercise type and exercise duration;

[0033] The exercise type and exercise time are collected through the smart bracelet.

[0034] Because insufficient physical activity among teenagers can lead to weak back and core muscle groups, which cannot provide adequate support for the spine and increase the probability of spinal problems, using exercise data as additional data for spinal monitoring improves the accuracy of spinal monitoring.

[0035] In any embodiment, the spine analysis model includes a sub-analysis model constructed for each type of monitoring data; the monitoring data is analyzed using the spine analysis model to obtain analysis results, including:

[0036] Input the monitoring data into the corresponding sub-model, use the corresponding sub-model to perform analysis, and obtain the intermediate analysis results output by each sub-model;

[0037] The intermediate analysis result is analyzed again based on a preset decision strategy to obtain an analysis result; wherein the preset decision strategy includes a softmax gating function or a preset probability distribution function.

[0038] The embodiment of the present application adopts a fine-grained analysis model, inputs the collected monitoring data into the corresponding sub-model, and the sub-model performs professional analysis on the input monitoring data. The analysis results of each sub-model are then integrated to obtain the final result. Through more fine-grained analysis, the accuracy of spinal health analysis is improved.

[0039] In any embodiment, the monitoring data is analyzed using the spine analysis model to obtain analysis results, including:

[0040] Extracting features from the time series data in the monitoring data to obtain time series features; wherein the time series data includes at least one of the sitting time and sitting frequency in the sitting data, the standing time and standing frequency in the standing data, and the sleeping time in the sleeping data;

[0041] Extracting features from the spatial data in the monitoring data to obtain spatial features; wherein the spatial data includes at least one of the spinal curvature, cervical angle and pressure data corresponding to the sitting state and the standing state, and the sleeping posture in the lying state;

[0042] Fuse the temporal features and spatial features to obtain the temporal-spatial features;

[0043] The time-space features are input into the spine analysis model to obtain the analysis results.

[0044] The embodiment of the present application divides the collected monitoring data into time series data and space data, and performs feature extraction respectively to obtain time series features and space features, and then fuses the time series features and space features again, thereby improving the accuracy of spinal monitoring through the complementarity of the time series features and space features.

[0045] In any embodiment, the monitoring data includes a back image of a young user, and the temporal-spatial features are input into a spine analysis model to obtain analysis results, including:

[0046] Extract features from the back image to obtain image features;

[0047] Perform feature fusion on the image features and the temporal-spatial features to obtain fused features;

[0048] The fused features are input into the spine analysis model to obtain the analysis results.

[0049] The embodiment of the present application combines image features with time-space features, uses a spine analysis model for analysis, and improves the accuracy of spinal health analysis by fusing multi-source heterogeneous data.

[0050] In any embodiment, after obtaining the analysis result, the method further comprises:

[0051] If the analysis result indicates that the adolescent user has a spinal health risk, using a spine type recognition model to recognize the back image of the adolescent user to determine the type of spinal problem;

[0052] Recommend exercise programs suitable for adolescent users based on the type of spinal problem.

[0053] The embodiment of the present application subdivides spinal problems and recommends exercise plans suitable for adolescent users based on the specific types of spinal problems, thereby alleviating spinal problems in a targeted manner.

[0054] In any embodiment, an exercise program suitable for a young user is recommended based on the type of spinal problem, including:

[0055] Obtain physical test data and sports preferences of young users;

[0056] Recommend exercise plans suitable for adolescent users based on spinal problem types, physical test data and exercise preferences.

[0057] Since different adolescent users have different physical fitness and sports preferences, the physical fitness and sports preferences of adolescent users can be considered when recommending exercise plans, thereby improving the rationality of the exercise plan recommendations.

[0058] In any embodiment, the method further comprises:

[0059] Based on the monitoring data, the development of spinal problems in adolescent users is predicted after a preset period of time in the future if the adolescent users do not implement the recommended exercise plan.

[0060] The embodiment of the present application can predict the development of spinal problems of teenage users after a preset period of time in the future if they do not exercise, thereby motivating teenage users to persist in exercise and improve spinal problems.

[0061] In any embodiment, after recommending the exercise plan to the teenage user, the method further includes:

[0062] Record historical monitoring data of young users according to preset time intervals;

[0063] Generate changing trends of spinal problems of adolescent users based on historical monitoring data.

[0064] The embodiment of the present application records the changing trends of spinal problems of adolescent users after they persist in exercising. The changing trends of spinal problems can let adolescent users know whether exercise can improve spinal problems, and if it can improve spinal problems, it can play a positive motivational role.

[0065] In any embodiment, after recommending the exercise plan to the teenage user, the method further includes:

[0066] acquiring the monitoring data of the adolescent user again at a preset time interval, and evaluating the spinal problems of the adolescent user using the monitoring data;

[0067] Adjust your exercise plan based on the assessment results.

[0068] The embodiment of the present application enables adolescent users to use a more reasonable exercise plan for exercise by regularly adjusting the exercise plan, thereby improving the efficiency of improving spinal problems.

[0069] In any embodiment, after obtaining the monitoring data of the adolescent user in daily life, the method further includes:

[0070] Based on the monitoring data, if it is detected that the current posture of the adolescent user does not meet the spinal health requirements, an alarm will be issued.

[0071] The embodiment of the present application sends out an alarm in real time when it is found that the posture of the adolescent user does not meet the spinal health requirements, so as to remind the adolescent user to adjust the posture in time and reduce the probability of spinal problems.

[0072] In any embodiment, after obtaining the monitoring data of the adolescent user in daily life, the method further includes:

[0073] Get the age of teenage users;

[0074] Determining the weight of the monitoring data according to age;

[0075] The monitoring data is analyzed using the spine analysis model to obtain analysis results, including:

[0076] The monitoring data and weights are input into the spinal analysis model to obtain the analysis results.

[0077] Since adolescent users of different age groups have different characteristics in their daily study and life, the weight of the monitoring data can be determined according to the age of the adolescent users, and analysis can be performed based on the weight to improve the accuracy of the analysis.

[0078] In a second aspect, an embodiment of the present application provides a spinal monitoring device for adolescents, comprising:

[0079] A data acquisition module, used to acquire monitoring data of the adolescent user in daily life; the monitoring data includes at least one of sitting posture data, standing posture data, lying posture data and backpack data;

[0080] The analysis module is used to analyze the monitoring data using a spinal analysis model to obtain analysis results; the analysis results are used to characterize whether the adolescent user has spinal health risks in the current state.

[0081] In a third aspect, an embodiment of the present application provides an electronic device, including: a processor, a memory, and a bus, wherein:

[0082] The processor and the memory communicate with each other via the bus;

[0083] The memory stores program instructions that can be executed by the processor, and the processor can execute the method of the first aspect by calling the program instructions.

[0084] In a fourth aspect, an embodiment of the present application provides a non-transitory computer-readable storage medium, including:

[0085] The non-transitory computer-readable storage medium stores computer instructions that cause the computer to execute the method of the first aspect.

[0086] In a fifth aspect, an embodiment of the present application provides a computer program product, comprising computer program instructions, wherein when the computer program instructions are read and executed by a processor, the method of the first aspect is executed.

[0087] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or be understood by practicing the embodiments of the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0089] Figure 1 A schematic flow chart of a method for monitoring the spine of adolescents provided in an embodiment of the present application;

[0090] Figure 2 A schematic diagram of sensor layout provided in an embodiment of the present application;

[0091] Figure 3 A schematic diagram of the structure of a spine monitoring device for adolescents provided in an embodiment of the present application;

[0092] Figure 4 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0093] The following embodiments of the technical solution of the present application are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.

[0094] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.

[0095] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0096] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0097] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0098] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0099] In the description of the embodiments of the present application, unless otherwise clearly specified and limited, technical terms such as "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0100] The spinal health of adolescents has become a serious social problem, and large-scale spinal diseases are becoming more and more common among adolescents. The current commonly used medical methods are relatively passive, relying mainly on corrective treatment after the occurrence of spinal pathological problems, which not only increases the health burden of individuals, but also greatly increases the cost of treatment. The effect of waiting until symptoms appear to take treatment is not ideal, and is often accompanied by problems such as difficulty in correction, long treatment cycle, and heavy burden of subsequent rehabilitation for patients. This not only has a significant impact on the daily study and life of adolescent individuals, but also puts great pressure on the allocation of medical resources for families and society. In addition, the limitations of traditional treatment methods also include slow recovery time, high pain levels, and possible postoperative complications, which bring additional psychological and economic burdens to patients.

[0101] Based on this, the embodiments of the present application provide a method, device, equipment, storage medium and program product for monitoring the spine of adolescents. By acquiring the monitoring data of adolescent users in their daily lives and analyzing the monitoring data using a spine analysis model, it is possible to determine whether the adolescent users have spinal health risks. Through daily monitoring, dynamic analysis of spinal health risks can be performed. Before adolescent spines have pathological problems, comprehensive monitoring can be performed based on multiple factors that affect spinal health. Once spinal health risks are identified, early intervention measures can be implemented.

[0102] It can be understood that the adolescent spine monitoring method provided in the embodiment of the present application can be applied to electronic devices, which include terminals and servers; the terminals can specifically be smart phones, tablet computers, computers, personal digital assistants (PDA), etc.; the server can specifically be an application server or a Web server.

[0103] Figure 1 A schematic diagram of a method for monitoring the spine of adolescents provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes:

[0104] Step 101: Acquire monitoring data of the adolescent user in daily life; the monitoring data includes at least one of sitting posture data, standing posture data, lying posture data and backpack data;

[0105] Step 102: Analyze the monitoring data using the spine analysis model to obtain analysis results; the analysis results are used to characterize whether the adolescent user has spinal health risks in the current state.

[0106] In the specific implementation process, the young users may be users within a preset age range, for example, 7-20 years old. The specific age range may also be set according to actual conditions, and the embodiments of the present application do not make specific limitations on this.

[0107] After long-term observation and research, it was found that the factors that affect the spinal health of adolescents are mainly the following:

[0108] 1. Bad posture

[0109] Sedentary learning: Teenagers who sit for a long time to read, do homework or use electronic devices can easily develop problems such as hunchbacks and spinal curvature.

[0110] Incorrect standing posture: Failure to maintain the normal curvature of the spine when standing or walking for a long time may cause spinal deformity.

[0111] Poor sleeping posture: Long-term use of pillows that are too high or too low, sleeping on your stomach, and other bad sleeping postures can also affect the condition of the spine.

[0112] Looking down at your phone: Frequently looking down at your phone or tablet will increase the pressure on your cervical spine, leading to cervical forward bending or cervical spondylosis.

[0113] Lying on the sofa or bed watching mobile phones and other electronic products: Lying down for a long time watching mobile phones and other electronic products will cause deformation of the cervical vertebrae or spinal column.

[0114] 2. Carrying a heavy backpack

[0115] The backpack is too heavy: Teenagers often need to carry a lot of books. A heavy backpack will put pressure on the spine, especially a single-shoulder backpack, which can easily cause scoliosis.

[0116] Asymmetrical use of backpacks: Long-term use of a single-shoulder backpack or asymmetrical adjustment of the backpack straps can cause scoliosis of the spine.

[0117] Therefore, the monitoring data may include at least one of sitting data, standing data, lying data and backpack data. In addition, the monitoring data within a recent period of time may be collected, for example, data within a recent month, data within a recent half month, or data within a recent week.

[0118] Sitting posture data refers to the posture data collected by the relevant data acquisition device when the adolescent user is sitting. Standing posture data refers to the posture data collected by the relevant data acquisition device when the adolescent user is in a standing state or a walking state. Supine posture data refers to the posture data collected by the relevant data acquisition device when the adolescent user is in a lying state. Backpack data refers to the data collected by the relevant data acquisition device when the adolescent user is in a backpack state. It should be noted that these posture data can provide data support for the assessment of adolescent spinal health. The specific content and collection method of the above data are introduced in the following embodiments.

[0119] The spine analysis model is pre-trained and is used to analyze the input monitoring data to determine whether the adolescent user corresponding to the monitoring data has a spine health risk. The spine analysis model can be a logistic regression model, a decision tree model, a random forest model, etc. The spine analysis model can be pre-trained using training samples, which include monitoring data of multiple adolescent users in a historical time period and corresponding labels of whether there are spine health problems or health scores.

[0120] The embodiment of the present application obtains monitoring data of adolescents in their daily lives and analyzes the acquired monitoring data, so as to dynamically and in real time analyze the adolescents' spinal health risks.

[0121] The sitting posture data may include the first spinal curvature and the first cervical angle of the adolescent in the sitting state. The first spinal curvature refers to the specific curvature of the spine on the sagittal plane. The cervical angle refers to the specific curvature of the cervical spine on the sagittal plane. The first spinal curvature and the first cervical angle can be obtained by monitoring sensors installed on the back and cervical spine of the human body. Figure 2 A schematic diagram of sensor layout provided in an embodiment of the present application is shown as follows: Figure 2 As shown. Based on sensor technology, the sensors are arranged at relevant human body positions to monitor the human body for a long time. For example, sensors can be set on the back, left side, right side and front side of the human body. Through the large body model, the data of human body monitoring is analyzed and compared in daily life, so as to supervise the user's behavior norms in daily activities such as correcting posture and sitting posture. Thereby reducing the probability of scoliosis. It should be noted that Figure 2 The sensor monitoring points set in the figure are only examples. In actual applications, monitoring points can be added or deleted according to specific needs.

[0122] Correspondingly, the sensor can be a fiber tension sensor matrix, and a piezoelectric or pressure sensor matrix can be used in individual positions. So as to obtain relevant body posture data. The sensor can be attached to the human body, or it can be set on clothing to form an intelligent wearable device. The intelligent wearable device can be a tight short top underwear + tight shorts, tight short top underwear + tight trousers, tight top underwear + tight shorts, tight top underwear + tight trousers, tight bodysuit.

[0123] The sitting posture data may also include pressure data collected by pressure sensors arranged on the seat back and / or the seat cushion.

[0124] Pressure sensors may be provided on the seat cushion and / or the seat back. When the teenager is in a sitting position and leaning against the seat back, if the sitting position is incorrect, for example, crossing one's legs while sitting or leaning to one side, the pressure on the seat cushion will be uneven. When the back contacts the seat back, the sensor on the seat back will detect the uneven pressure.

[0125] In addition, if a teenage user is doing homework and has an incorrect sitting posture, the pressure of the elbows on the desktop will be different. Therefore, it is also possible to monitor whether the pressure on the left and right sides is balanced through a pressure sensor installed on the desktop.

[0126] Furthermore, the sitting posture data can also include the length of time and frequency of sitting. Maintaining the same sitting posture for a long time, especially when the sitting posture is incorrect, the muscles of the back and neck will continue to be tense, lacking the necessary stretching and relaxation. This continuous state of tension will cause muscle fatigue and pain, which will increase the burden on the spine. The normal spine has a certain physiological curvature, which helps to disperse the pressure on the spine. However, maintaining the same sitting posture for a long time may destroy this physiological curvature, causing the curvature of the spine to become straight or reversed.

[0127] In the embodiment of the present application, being in a sitting state for more than a preset time is called sedentary, where the preset time can be 1 hour, 2 hours or 1.5 hours, etc. Therefore, whether the teenage user sits down can be monitored by a smart bracelet or a smart seat cushion. When the teenage user sits down, the timing starts. If the teenage user continues to sit for more than the preset time, this sitting is called sedentary. When the teenage user is monitored to stand up, the duration of this sedentary time is recorded. The sedentary frequency refers to the number of times the teenage user sits for a long time in a day, which can also be monitored by a smart bracelet or a smart seat cushion. When the smart bracelet or the smart seat cushion detects that the teenage user is sitting for a long time, the teenager is reminded to stand up and move around. A vibration reminder can be set in the corresponding device. After the teenage user sits down again, the next sedentary reminder cycle will be re-entered.

[0128] The sitting posture data may also include a back image in a sitting state; the back image is acquired by an image acquisition device.

[0129] When the sitting posture is correct, the spine should maintain a natural physiological curve, that is, the neck is slightly lordotic, the chest is slightly kyphotic, and the waist is straightened forward. This natural curve helps to disperse the pressure on the spine and reduce fatigue and damage caused by sitting for a long time. If the sitting posture is incorrect, such as hunchback, round shoulders, etc., the back image will show abnormalities in the spinal curve, such as excessive lordosis or kyphosis, which is usually a direct reflection of incorrect sitting posture. Therefore, an image acquisition device can be set behind or obliquely behind the seat of the adolescent user to collect the back image of the adolescent user in a sitting state. The back image provides data support for subsequent spinal health monitoring. It is understandable that the image acquisition device can be a surveillance camera or a camera, etc. In addition, the back image can also be obtained through professional medical testing.

[0130] The standing posture data includes the second spinal curvature and the second cervical vertebra angle in the standing state. The second spinal curvature and the second cervical vertebra angle are acquired through the smart insole and / or the posture sensor arranged on the young user.

[0131] In the specific implementation process, when the human body stands correctly, the spine will present a natural physiological curvature, including the lordosis of the neck, the kyphosis of the chest, the lordosis of the waist and the kyphosis of the sacrum. These curvatures help maintain the balance and stability of the body, while dispersing the pressure of gravity on the spine.

[0132] If you stand incorrectly, such as hunching over or over-exaggerating your chest, the physiological curvature of your spine may change, causing the degree of lordosis or kyphosis to increase. This change may have a negative impact on the health of your spine.

[0133] If you stand incorrectly, such as with your head tilted forward or backward too much, the physiological curvature of the cervical spine may change, causing the degree of cervical lordosis or kyphosis to increase. By measuring the cervical angle or observing the changes in the cervical curvature, you can determine whether your standing posture is correct.

[0134] The standing posture data may also include the second spinal curvature and the second cervical angle of the adolescent in the standing state. The second spinal curvature refers to the specific curvature of the spine in the sagittal plane. The cervical angle refers to the specific curvature of the cervical spine in the sagittal plane. The second spinal curvature and the second cervical angle can be obtained by monitoring the sensors set on the back and cervical spine of the human body. The specific settings of the sensors can be found in Figure 2 , and can also be Figure 2 Based on this, sensors can be added or deleted according to actual needs.

[0135] Correspondingly, the sensor can be a fiber tension sensor matrix, and a piezoelectric or pressure sensor matrix can be used in individual positions. So as to obtain relevant body posture data. The sensor can be attached to the human body, or it can be set on clothing to form an intelligent wearable device. The intelligent wearable device can be a tight short top underwear + tight shorts, tight short top underwear + tight trousers, tight top underwear + tight shorts, tight top underwear + tight trousers, tight bodysuit.

[0136] The standing posture data may also include a back image in the standing state, which can be obtained by collecting images of the back, side, and front of the adolescent user in the standing state through an image acquisition device. By collecting images of the spine in the standing state, it can be determined whether the standing posture is correct. For example, if the spine is observed from the side, if its curve is smooth and conforms to the natural physiological curvature, it means that the standing posture is correct; if the curve is abnormal or excessively lordotic / kyphotic, it may indicate that there is a problem with the standing posture.

[0137] In addition, if the teenager is in a normal standing state, but the spine in the back image is in an abnormal state, it means that the teenager user may have health problems with his spine, such as scoliosis, hunchback, etc.

[0138] The embodiment of the present application can reflect the adolescent's spinal problems and standing posture problems to a certain extent by collecting back images of adolescent users in a standing state, thereby providing data support for subsequent spinal analysis.

[0139] The lying posture data includes sleeping posture, sleeping duration and sleeping quality; sleeping posture, sleeping duration and sleeping quality are collected through smart mattresses and / or smart bracelets.

[0140] In the specific implementation process, long-term use of inappropriate pillows, sleeping on one's stomach and other bad sleeping positions can also affect the development of adolescents' spine. For example, sleeping in a prone position makes it difficult to maintain a neutral natural position of the spine, which will put pressure on the back and neck and cause the spine to arch. Excessive curling: The back will be in an arched state, the back muscles will be overstretched and unable to stretch and relax, which is not good for the thoracic spine, lumbar spine, cervical spine and even the entire spine.

[0141] Long-term lack of sleep may cause physical fatigue and affect the normal physiological function of the spine, while too long a sleep period may also cause body stiffness, which is not conducive to the flexibility and health of the spine.

[0142] High-quality sleep helps the spine rest and repair. When the body is in a deep sleep state, the spine can fully relax and recover. On the contrary, low-quality sleep may lead to insufficient rest and repair of the spine, increasing the risk of spinal diseases.

[0143] Therefore, the embodiments of the present application can monitor the sleeping posture, sleep duration and sleep quality of teenagers. Specifically, a smart mattress and / or a smart bracelet can be used to collect their sleeping posture data.

[0144] There are many sensors installed on the smart mattress, such as pressure sensors, heart rate monitoring sensors, blood pressure monitoring sensors, etc. When a teenager lies on the smart mattress to sleep, the smart mattress can determine the teenager's sleeping posture through the pressure sensor, and the sleep quality through the heart rate monitoring sensor and the blood pressure monitoring sensor. The smart mattress can determine the sleep duration through the time information of the monitored pressure data. In addition, the sleep duration can also be determined based on smart bracelets and smart watches.

[0145] Smart bracelets can collect lying posture data of adolescent users by monitoring heart rate, blood pressure, heart rate variability and respiratory variability.

[0146] The collected lying posture data will be used for subsequent monitoring of spinal health.

[0147] There is also a special scenario for adolescent users, that is, they carry school bags when going to and from school every day. Some adolescent users carry double backpacks, and some carry single-shoulder backpacks (including cross-body backpacks). Carrying school bags may also affect spinal development. Therefore, for the application scenario of spinal health monitoring for adolescent users, their backpack data is also one of the influencing factors. And the backpack data includes the weight of the school bag, the pressure difference between the two shoulders, and the length of time the backpack is worn.

[0148] Long-term use of a heavy backpack may cause uneven pressure and stress on the spine, which may lead to scoliosis. Carrying a backpack for a long time, especially when the weight is too heavy or the pressure difference between the two shoulders is large, will significantly increase the pressure on the spine and even increase the risk of spinal deformity.

[0149] Frequently carrying a backpack on the same shoulder will cause the soft tissue around the shoulder joint to be in a state of tension for a long time. Over time, the lower shoulder muscles will tighten and the upper arm muscles will lengthen to form a slanted shoulder, resulting in uneven shoulders. If the pressure difference between the two shoulders is too large, it will pull the spine to cause curvature, which may lead to scoliosis in the long run. Although a backpack can disperse the pressure of the backpack to a certain extent, the two shoulders also have different load-bearing capacity due to the different lengths of the two shoulder straps. In the long run, it may also cause problems with the spine.

[0150] When monitoring backpack data, a sensor may be provided on the shoulder strap of the backpack, specifically a pressure sensor, etc. When a young user carries a backpack, the pressure data on the shoulder strap and the backpack wearing time may be collected.

[0151] In addition, young users can also wear wearable devices. The smart wearable devices can be tight short top underwear + tight shorts, tight short top underwear + tight trousers, tight upper underwear + tight shorts, tight upper underwear + tight trousers, tight jumpsuit. The shoulders of the above-mentioned clothes are equipped with sensors, which can be used to detect the pressure on the shoulders.

[0152] Based on the above embodiment, the monitoring data also includes exercise data, and the exercise data includes exercise type and exercise duration;

[0153] The exercise type and exercise time are collected through the smart bracelet.

[0154] In the specific implementation process, some sports can enhance the strength of muscles around the spine, such as swimming, jogging, yoga, etc., which can provide better support and protection for the spine by enhancing the strength of muscles around the spine. Some sports can improve spinal flexibility, such as yoga and Pilates, which can help relieve spinal stiffness caused by long-term sitting or bad posture. Therefore, sports data can be used as a factor affecting spinal health.

[0155] The exercise type and exercise time in the exercise data can be collected through the smart bracelets worn by young users.

[0156] Because insufficient physical activity among teenagers can lead to weak back and core muscle groups, which cannot provide adequate support for the spine and increase the probability of spinal problems, using exercise data as additional data for spinal monitoring improves the accuracy of spinal monitoring.

[0157] It should be noted that the smart bracelet in the above embodiment includes: a piezoelectric film PVDF sensor for collecting physiological data. That is to say, the sensor used in the above intelligent monitoring device may include: a sensor made of polyvinylidene fluoride PVDF material, also known as a piezoelectric film PVDF sensor, or simply a PVDF sensor. Due to the small size of this PVDF sensor, it is not only easy to be woven, but also easy to hide in, for example, the strap of a smart watch or the wristband of a smart bracelet. Since the PVDF material has a very high piezoelectric coefficient, it means that it is very sensitive to small pressure changes. This high sensitivity enables the PVDF sensor to capture subtle pulse waveforms and blood pressure changes, providing more accurate data.

[0158] And / or, the flexible sweat sensor is used to collect potassium ion concentration and / or sodium ion concentration. When the flexible sweat sensor collects the biochemical ion concentration of sweat, the potential signal data caused by the change of this concentration. The above-mentioned sweat biochemical ion concentration is, for example, sodium ion concentration or potassium ion concentration. After the concentration of any one of these two ions changes, the signal processing circuit in the flexible sweat sensor fiber can convert the potential difference analog signal caused by the change of ion concentration into a digital signal. Then, the flexible sweat sensor can send the digital signal to an intelligent monitoring device (such as a smart watch or a smart bracelet).

[0159] Optionally, the flexible sweat sensor can be designed as a filament sensor, such as a flexible sweat sensor fiber. Since the flexible sweat sensor fiber is very thin, it is easy to weave and make it into a smart watch strap or a smart bracelet wristband. Of course, it can also be woven with other natural fiber lines (such as cotton fiber lines or wool fiber lines, etc.) and / or artificial chemical fiber lines (such as polyester and nylon) and made into a smart watch strap or a smart bracelet wristband.

[0160] Based on the above embodiments, good nutritional intake can support bone and muscle health and reduce the risk of spinal problems. Therefore, the monitoring data can also include nutritional intake, the data source of which is the diet log input into the electronic device, and specifically may include protein intake, calcium intake, and vitamin D intake.

[0161] On the basis of the above embodiment, after the monitoring data is acquired, the monitoring data is analyzed using the spine analysis model, and the specific process of obtaining the analysis result includes:

[0162] Input the monitoring data into the corresponding sub-model, use the corresponding sub-model to perform analysis, and obtain the intermediate analysis results output by each sub-model;

[0163] The intermediate analysis result is analyzed again based on a preset decision strategy to obtain an analysis result; wherein the preset decision strategy includes a softmax gating function or a preset probability distribution function.

[0164] In a specific implementation process, the spine analysis model may be composed of a hybrid model, i.e., a sub-model, each of which is used to analyze the corresponding type of monitoring data. Assume that the monitoring data includes sitting data, standing data, lying data, and backpack data. Then the spine analysis model may include a sub-model for analyzing sitting data, a sub-model for analyzing standing data, a sub-model for analyzing lying data, and a sub-model for analyzing backpack data.

[0165] Therefore, each type of monitoring data can be input into the corresponding sub-model, and the sub-model can extract features from the input data. For sitting data, at least one feature of the maximum, minimum, and average values ​​of the spinal curvature; the maximum, minimum, and average values ​​of the cervical angle; the maximum, minimum, and average values ​​of the sitting time; the sitting frequency; and the maximum, minimum, and average values ​​of the backrest pressure can be extracted. For standing data, the maximum, minimum, and average values ​​of the spinal curvature; the maximum, minimum, and average values ​​of the cervical angle can be extracted. For lying data, the maximum, minimum, and average values ​​of sleeping posture and sleep time; and the maximum, minimum, and average values ​​of sleep quality can be extracted. For backpack data, the maximum, minimum, and average values ​​of the backpack weight and backpack time and the maximum, minimum, and average values ​​of the pressure difference between the two shoulders can be extracted. Each sub-model performs analysis based on the extracted features to output the corresponding intermediate analysis results. The intermediate analysis results can be used to characterize the possibility of determining whether there is a spinal health risk based on the input data.

[0166] It should be noted that the sub-model may adopt a convolutional neural network model or a recurrent neural network, or may adopt an expert model.

[0167] The spine analysis model also includes a fusion analysis module, which is used to receive the intermediate analysis results output by each submodule and re-analyze the intermediate analysis results based on a preset decision strategy to obtain a final analysis result. The preset decision strategy may include a softmax gating function or a preset probability distribution function.

[0168] The softmax gating function, also known as the normalized exponential function, is a mapping that converts any real number vector into a probability distribution. It represents the predicted probability of each category, that is, the probability that the adolescent user's spine has health risks.

[0169] The preset probability distribution function is a probability distribution function predefined according to a specific application scenario and prior knowledge. It can also be used to calculate the probability of the adolescent user's spine having health risks based on the intermediate analysis results.

[0170] It should be noted that before using the spine analysis model to analyze the monitoring data, the acquired monitoring data can also be cleaned to remove invalid data, and interpolate or delete the confirmed data. The data can also be normalized and smoothed.

[0171] The embodiment of the present application adopts a fine-grained analysis model, inputs the collected monitoring data into the corresponding sub-model, and the sub-model performs professional analysis on the input monitoring data. The analysis results of each sub-model are then integrated to obtain the final result. Through more fine-grained analysis, the accuracy of spinal health analysis is improved.

[0172] In another analysis method provided by the present application, the monitoring data is analyzed using a spine analysis model to obtain analysis results, including:

[0173] Extracting features from the time series data in the monitoring data to obtain time series features; wherein the time series data includes at least one of the sitting time and sitting frequency in the sitting data, the standing time and standing frequency in the standing data, and the sleeping time in the sleeping data;

[0174] Extracting features from the spatial data in the monitoring data to obtain spatial features; wherein the spatial data includes at least one of the spinal curvature, cervical angle and pressure data corresponding to the sitting state and the standing state respectively;

[0175] Fuse the temporal features and spatial features to obtain the temporal-spatial features;

[0176] The time-space features are input into the spine analysis model to obtain the analysis results.

[0177] In the specific implementation process, the acquired monitoring data can be divided into time series data and spatial data in advance; among which, time series data refers to data related to time, for example, it can include the sedentary time and sedentary frequency in the sitting data; the standing time and standing frequency in the standing data; at least one of the sleep time in the sleep data; spatial data refers to data related to the posture of the adolescent user, for example, it can include at least one of the spinal curvature, cervical angle and pressure data corresponding to the sitting state and the standing state. These data reflect the spatial characteristics of adolescents in different postures, which are very critical for evaluating the specific health status of the spine.

[0178] It should be noted that the collection method of the above-mentioned time series data and spatial data can be referred to the above-mentioned embodiment, which will not be repeated here. The time series data can also include the exercise duration and exercise frequency in the exercise data. These data reflect the changes and trends over time, which are of great significance for evaluating the long-term poor posture and lack of exercise of adolescents.

[0179] After obtaining the time series data and spatial data, feature extraction can be performed on them respectively. Before feature extraction, the acquired time series data and spatial data can be preprocessed, including:

[0180] Preprocessing of time series data: Normalize data such as sitting time, sitting frequency, standing time, standing frequency, sleep time, exercise time, and exercise frequency to ensure that the data are at the same level.

[0181] Interpolation: Missing data points can be filled using linear interpolation, polynomial interpolation, or time series-based interpolation methods (such as moving average).

[0182] Fill: Fill with the value of the previous or next data point, or with a specific constant value.

[0183] Filtering: Use low-pass, high-pass, or band-pass filters to remove high-frequency noise or low-frequency trends in time series data.

[0184] Smoothing: Use methods such as moving average and exponential smoothing to smooth time series data and reduce the impact of random fluctuations.

[0185] Time alignment: If the timestamps of different types of time data (such as sedentary time, standing time, sleeping time, etc.) are not consistent, the model may not be able to effectively capture the relationship between these data. Therefore, time alignment is necessary.

[0186] Interpolation alignment: Align data with different timestamps to the same sampling frequency through interpolation method.

[0187] Resampling alignment: resample all time series data to a fixed frequency (such as hourly, daily, etc.).

[0188] Time window alignment: Split the data into time windows of fixed length and then align within each window.

[0189] Preprocessing of spatial data: Standardize the spinal curvature, cervical angle, pressure data, etc. to eliminate the impact of dimension, such as converting pressure data into relative pressure values.

[0190] The feature extraction method of time series data includes: extracting features using time series analysis methods (such as time series feature extraction, frequency domain analysis, etc.). For example, the average, maximum, minimum and standard deviation of sitting time can be calculated, the average, maximum, minimum and standard deviation of standing time can be calculated, and the average, maximum, minimum and standard deviation of sleeping time can be calculated.

[0191] The feature extraction method of spatial data includes: extracting features using spatial feature extraction methods (such as cluster analysis, principal component analysis, etc.). For example, the average, maximum, minimum and standard deviation of the spinal curvature, the average and maximum of the cervical angle, etc. can be calculated. The pressure data can be the pressure distribution of the backrest and / or the pressure distribution of the desktop in a sitting state. It can also be the pressure distribution of the smart insole in a standing state.

[0192] After obtaining the temporal features and spatial features, the two are fused. For example, the sitting time and frequency can be combined with the spinal curvature and cervical angle to form a comprehensive temporal-spatial feature vector. Recurrent neural networks (RNN) or long short-term memory networks (LSTM) are used to process this multimodal data. These networks can capture temporal changes and spatial features, namely, temporal-spatial features.

[0193] The time-space features are input into a pre-trained spinal analysis model, and the analysis results are output to characterize whether the adolescent user has spinal health risks in the current state.

[0194] Among them, the spine analysis model can choose a deep learning model (such as a multimodal CNN-LSTM model) or a traditional machine learning model (such as a random forest, support vector machine, etc.). During the training process, methods such as cross-validation can be used to evaluate the performance of the model and perform hyperparameter tuning.

[0195] The embodiment of the present application divides the collected monitoring data into time series data and space data, and performs feature extraction respectively to obtain time series features and space features, and then fuses the time series features and space features again, thereby improving the accuracy of spinal monitoring through the complementarity of the time series features and space features.

[0196] Based on the above embodiment, the monitoring data includes the back image of the adolescent user, and the time-space features are input into the spine analysis model to obtain the analysis results, including:

[0197] Extract features from the back image to obtain image features;

[0198] Perform feature fusion on the image features and the temporal-spatial features to obtain fused features;

[0199] The fused features are input into the spine analysis model to obtain the analysis results.

[0200] In a specific implementation process, the back image may include a back image collected when the adolescent user is sitting or standing, and may include an image of the front and back and an image of the side and back. After obtaining the image data, the back image may be preprocessed, including cropping, scaling, grayscale and other preprocessing. After obtaining the back image, image processing and computer vision techniques (such as convolutional neural networks, feature point detection, etc.) may be used to extract features. For example, the curvature points of the spine, the angle points of the cervical vertebrae, etc. may be extracted.

[0201] Before inputting the time-space features into the spine analysis model, the time-space features can be fused with the image features again. Specifically, the image feature vector can be spliced ​​with the comprehensive time-space feature vector to form the final multimodal feature vector, i.e., the fused feature. The fused feature is input into the spine analysis model to obtain the final analysis result.

[0202] It should be noted that before fusing temporal-spatial features with image features, the following factors need to be considered:

[0203] (1) Consistency of data sampling frequency

[0204] The sampling frequencies of time series data and spatial data may be different. For example, the sedentary duration and frequency recorded by a smart bracelet may be updated once a minute, while the spinal curvature and cervical angle recorded by a posture sensor may be updated once a second. Based on this, data of different frequencies can be aligned to the same sampling frequency through resampling or interpolation methods. For example, high-frequency data can be downsampled to a low frequency, or low-frequency data can be upsampled to a high frequency.

[0205] (2) Time alignment of data

[0206] Data recorded by different sensors may have different timestamps, and direct fusion may result in inconsistent data. Therefore, the following alignment operation can be performed:

[0207] Interpolation alignment: interpolation is performed in time to fill in missing time points and align the data in time.

[0208] Time window alignment: Split the data into time windows of fixed length and ensure that the data in each window is time aligned. For example, you can split the data into 1-minute time windows and then process them in each window.

[0209] (3) Data standardization and normalization

[0210] Time series data and spatial data may have different dimensions and distributions. Direct fusion may result in data from one modality dominating the results. Therefore, before fusion, the data of each modality is standardized and normalized to ensure that the data are at the same level. Commonly used standardization methods include Z-score standardization and Min-Max standardization.

[0211] (4) Temporal and spatial consistency of data

[0212] To ensure the consistency and logic of the fusion of time series data and spatial data, a time consistency check can be performed to ensure that the time series data and spatial data at the same time point or time window are synchronized. A logical consistency check can also be performed to ensure that the logical relationship of the data is reasonable. For example, the spinal curvature data during sitting should be valid during the sitting time, not during the standing time.

[0213] The embodiment of the present application combines image features with time-space features, uses a spine analysis model for analysis, and improves the accuracy of spinal health analysis by fusing multi-source heterogeneous data.

[0214] Based on the above embodiment, after obtaining the analysis result, the method further includes:

[0215] If the analysis result indicates that the adolescent user has a spinal health risk, using a spine type recognition model to recognize the back image of the adolescent user to determine the type of spinal problem;

[0216] Recommend exercise programs suitable for adolescent users based on the type of spinal problem.

[0217] In the specific implementation process, if the spine analysis model is used to determine that the adolescent user has spinal health risks, then the spine type recognition model can be further used to identify the adolescent user's back image. Specifically, an image acquisition device can be used to capture the adolescent user's back image when standing naturally, including a front back image, a left back image, and a right back image. The captured back image is input into the spine type recognition model to obtain the specific spinal problem type of the adolescent user. The spine type recognition model can be obtained by pre-training using training samples, and the training samples include images corresponding to each spinal problem type. The specific structure of the spine type recognition model can be a support vector machine, a decision tree model, a convolutional neural network model, etc.

[0218] Among them, the types of spinal problems may include scoliosis, kyphosis and lordosis, etc. Moreover, scoliosis also specifically includes various degrees of scoliosis, namely, mild scoliosis, moderate scoliosis and severe scoliosis.

[0219] Different types of spinal problems are suitable for different types of exercises. For example, for mild scoliosis without obvious deformity and functional limitation, some exercises within one's ability can be performed, such as brisk walking and jogging. For moderate scoliosis with obvious deformity or functional limitation, strenuous exercise should be avoided, such as basketball, football and other confrontational sports. For severe scoliosis, especially for patients with severe functional limitation, strenuous exercise should be avoided as much as possible. It is recommended to perform some non-weight-bearing or low-intensity exercises, such as swimming, Tai Chi, Ba Duan Jin, etc. For kyphosis, pull-ups, plank support, swimming, etc. can be performed appropriately. For kyphosis, extension training, Xiaoyan Fei, etc. can be performed appropriately.

[0220] In addition to the exercise type, the exercise plan can also include exercise intensity, exercise frequency, etc. For example: in the first training cycle (such as the 1st to 2nd week), the exercise intensity can be set lower and the exercise frequency can be set higher; in the second training cycle (the 3rd to 4th week), the exercise intensity can be appropriately increased.

[0221] After the exercise plan is generated, the exercise plan is pushed to the terminal used by the teenage user.

[0222] The embodiment of the present application subdivides spinal problems and recommends exercise plans suitable for adolescent users based on the specific types of spinal problems, thereby alleviating spinal problems in a targeted manner.

[0223] Based on the above embodiment, when recommending an exercise plan suitable for teenage users based on the type of spinal problems, the physical measurement data and exercise preferences of the teenage users can be obtained; then, an exercise plan suitable for the teenage users can be recommended based on the type of spinal problems, physical measurement data and exercise preferences.

[0224] Since different young users have different physical fitness and sports preferences, when recommending sports plans, physical test data of young users can be obtained. The physical test data is used to characterize the physical fitness of the young users, so that when specifying a sports plan, a suitable exercise intensity can be arranged. The physical test data can be obtained through physical tests conducted by schools or professional physical test institutions. It is also possible to obtain the sports preferences of young users, that is, what sports the young users like to do, so that when formulating sports plans, the sports that young users like can be recommended from a variety of suitable sports, which can increase the enthusiasm of young users for sports.

[0225] Based on the above embodiment, the method further includes:

[0226] Based on the monitoring data, the development of spinal problems in adolescent users is predicted after a preset period of time in the future if the adolescent users do not implement the recommended exercise plan.

[0227] In the specific implementation process, if the adolescent user has spinal problems according to the monitoring data analysis, it can also be determined based on the monitoring data that if the adolescent user continues to live according to the monitored data without performing targeted exercises, then what will happen to the adolescent user's spinal problems after a preset period of time (such as three months, six months or one year, etc.). For example: the current adolescent user's scoliosis is mild scoliosis. If he continues to live according to his current habits, then after one year, it may develop into moderate scoliosis. Specifically, prediction can be performed through a prediction model, which can be composed of an LSTM network.

[0228] The embodiments of the present application can predict the development of spinal problems of adolescent users after a preset period of time in the future if they do not exercise, thereby motivating adolescent users to persist in exercise and improve spinal problems.

[0229] Based on the above embodiment, after recommending the exercise plan to the young user, the method further includes:

[0230] Record historical monitoring data of young users according to preset time intervals;

[0231] Generate changing trends of spinal problems of adolescent users based on historical monitoring data.

[0232] During the specific implementation process, with respect to the situation where adolescent users exercise according to the recommended exercise plan, the electronic device can regularly record the adolescents' historical monitoring data. For example, the recorded historical monitoring data can be analyzed every other week to obtain analysis results, and the changing trend of the adolescent's spinal problems can be generated based on the results of each analysis. For example, for adolescents with moderate scoliosis, after one year of exercise, the scoliosis may gradually turn into mild scoliosis or no scoliosis.

[0233] The embodiment of the present application records the changing trends of spinal problems of adolescent users after they persist in exercising. The changing trends of spinal problems can let adolescent users know whether exercise can improve spinal problems, and if it can improve spinal problems, it can play a positive motivational role.

[0234] Based on the above embodiment, after recommending the exercise plan to the young user, the method further includes:

[0235] acquiring the monitoring data of the adolescent user again at a preset time interval, and evaluating the spinal problems of the adolescent user using the monitoring data;

[0236] Adjust your exercise plan based on the assessment results.

[0237] In the specific implementation process, in order to improve the effect of exercise, the monitoring data of the young user can be obtained again on a regular basis, and the monitoring data can be analyzed again using the spine analysis model to obtain the latest current situation of the young user, and then the exercise calculation can be adjusted according to the latest situation of the spine of the young user, so that the exercise plan is more reasonable. For example: for moderate scoliosis, the exercise type in the exercise plan can be playing basketball. When it is detected that the scoliosis of the young user becomes mild, it can be changed to jogging.

[0238] And you can also take the physical fitness of young users into comprehensive consideration. The amount of exercise can be set smaller in the early stage, and then you can increase the amount of exercise after exercising for a period of time.

[0239] The embodiment of the present application enables adolescent users to use a more reasonable exercise plan for exercise by regularly adjusting the exercise plan, thereby improving the efficiency of improving spinal problems.

[0240] On the basis of the above embodiments, the embodiments of the present application can also provide a real-time reminder function. The monitoring data of the adolescent users obtained includes continuous data and instantaneous data. The continuous data includes sitting time, standing time, sleeping time, etc., and the instantaneous data includes the spinal curvature, cervical angle, etc. at a certain moment.

[0241] You can pre-set spinal health reminder rules, for example: when you sit for more than half an hour continuously, an alarm will be triggered; or when the current posture is incorrect, such as the spinal curvature or cervical angle reaches a certain degree, an alarm will be triggered.

[0242] The acquired monitoring data is matched with the spinal health reminder rules. If the match is successful, an alarm prompt is triggered, thereby promptly reminding young users to get up and move around, or adjust their posture.

[0243] In addition, you can also provide reminders for sleep problems, nutritional intake problems, etc.

[0244] The embodiment of the present application sends out an alarm in real time when it is found that the posture of the adolescent user does not meet the spinal health requirements, so as to remind the adolescent user to adjust the posture in time and reduce the probability of spinal problems.

[0245] Based on the above embodiment, after obtaining the monitoring data of the adolescent user in daily life, the method further includes:

[0246] Get the age of teenage users;

[0247] Determining the weight of the monitoring data according to age;

[0248] The monitoring data is analyzed using the spine analysis model to obtain analysis results, including:

[0249] The monitoring data and weights are input into the spinal analysis model to obtain the analysis results.

[0250] In the specific implementation process, teenagers of different age groups have different characteristics, such as: teenagers aged 7-10, their academic workload is not so heavy, and their growth and development is slower than that after the age of 10, mainly due to the posture of carrying a schoolbag, playing with mobile phones, sleeping posture, exercise, etc. The main factors. Teenagers after the age of 10 have heavier academic workload, which greatly increases the length of time sitting to study, and teenagers after the age of 10 grow and develop faster, so nutritional intake is also more important. Therefore, the weights of monitoring parameters can be set differently according to different age stages. The weights of the monitoring parameters that account for the main factors are set to be relatively large, and the weights of the monitoring parameters that account for the secondary factors are set to be relatively small. The specific weight setting can be determined according to the actual situation, and the embodiments of the present application do not specifically limit this.

[0251] The monitoring parameters and weights are input into the spinal analysis model to obtain analysis results.

[0252] Since adolescent users of different age groups have different characteristics in their daily study and life, the weight of the monitoring data can be determined according to the age of the adolescent users, and analysis can be performed based on the weight to improve the accuracy of the analysis.

[0253] Figure 3 This is a schematic diagram of the structure of a spinal monitoring device for adolescents provided in an embodiment of the present application. The device may be a module, program segment or code on an electronic device. It should be understood that the device is similar to the above-mentioned Figure 1 The method embodiment corresponds to and can be executed Figure 1 The various steps involved in the method embodiment and the specific functions of the device can be found in the above description. To avoid repetition, detailed description is appropriately omitted here. The device includes: a data acquisition module 301 and an analysis module 302, wherein:

[0254] The data acquisition module 301 is used to acquire monitoring data of the adolescent user in daily life; the monitoring data includes at least one of sitting posture data, standing posture data, lying posture data and backpack data;

[0255] The analysis module 302 is used to analyze the monitoring data using a spinal analysis model to obtain an analysis result; the analysis result is used to characterize whether the adolescent user has a spinal health risk in the current state.

[0256] Based on the above embodiment, the sitting posture data includes a first spinal curvature and a first cervical vertebra angle in a sitting state;

[0257] The first spinal curvature and the first cervical vertebra angle are acquired through a posture sensor arranged on the adolescent user.

[0258] Based on the above embodiment, the sitting posture data includes pressure data collected by a pressure sensor provided on the seat back and / or the seat cushion.

[0259] Based on the above embodiment, the sitting posture data further includes sitting time and sitting frequency;

[0260] The sedentary time and sedentary frequency are monitored and obtained through a smart bracelet or a smart seat cushion.

[0261] Based on the above embodiment, the sitting posture data further includes a back image in a sitting state;

[0262] The back image is acquired by an image acquisition device.

[0263] Based on the above embodiment, the standing posture data includes a second spinal curvature and a second cervical vertebra angle in a standing posture;

[0264] The second spinal curvature and the second cervical angle are acquired by collecting a posture sensor disposed on the adolescent user.

[0265] Based on the above embodiment, the standing posture data further includes a back image in a standing posture;

[0266] The back image is obtained by capturing an image of the young user in a standing state through an image capturing device.

[0267] Based on the above embodiment, the lying posture data includes sleeping posture, sleeping duration and sleeping quality;

[0268] The sleeping posture, sleeping duration and sleeping quality are collected and obtained through a smart mattress and / or a smart bracelet.

[0269] Based on the above embodiment, the backpack data includes the weight of the backpack, the pressure difference between the two shoulders and the backpack time;

[0270] The backpack weight, the pressure difference between the two shoulders and the time of carrying the backpack are collected and obtained through sensors arranged on the shoulder straps of the backpack.

[0271] Based on the above embodiment, the monitoring data further includes exercise data, and the exercise data includes exercise type and exercise duration;

[0272] The exercise type and the exercise time are collected and obtained through a smart bracelet.

[0273] Based on the above embodiment, the spinal column analysis model includes a sub-analysis model constructed for each type of monitoring data; the analysis module 302 is specifically used for:

[0274] Input the monitoring data into the corresponding sub-model, use the corresponding sub-model to perform analysis, and obtain the intermediate analysis results output by each sub-model;

[0275] The intermediate analysis result is analyzed again based on a preset decision strategy to obtain the analysis result; wherein the preset decision strategy includes a softmax gating function or a preset probability distribution function.

[0276] Based on the above embodiment, the analysis module 302 is specifically used for:

[0277] Extracting features from the time series data in the monitoring data to obtain time series features; wherein the time series data includes at least one of the sitting time and sitting frequency in the sitting data, the standing time and standing frequency in the standing data, and the sleeping time in the sleeping data; extracting features from the spatial data in the monitoring data to obtain spatial features; wherein the spatial data includes at least one of the spinal curvature, cervical angle, and pressure data corresponding to the sitting state and the standing state, respectively;

[0278] Fusing the temporal feature with the spatial feature to obtain a temporal-spatial feature;

[0279] The time-space features are input into the spine analysis model to obtain the analysis results.

[0280] Based on the above embodiment, the monitoring data includes the back image of the adolescent user, and the analysis module 302 is further specifically used for:

[0281] Performing feature extraction on the back image to obtain the image features;

[0282] Performing feature fusion on the image feature and the temporal-spatial feature to obtain fused features;

[0283] The fused features are input into the spine analysis model to obtain the analysis results.

[0284] Based on the above embodiment, the device further includes a recommendation module, which is used to:

[0285] In the case where the analysis result indicates that the adolescent user has a spinal health risk, using a spine type recognition model to recognize the back image of the adolescent user to determine the type of spinal problem;

[0286] An exercise program suitable for the teenage user is recommended according to the type of spinal problem.

[0287] Based on the above embodiment, the recommendation module is specifically used for:

[0288] Acquiring the physical test data and sports preferences of the adolescent user;

[0289] An exercise plan suitable for the teenage user is recommended based on the spinal problem type, the physical measurement data and the exercise preference.

[0290] Based on the above embodiment, the device further includes a prediction module, which is used to:

[0291] Based on the monitoring data, it is predicted that if the adolescent user does not perform the recommended exercise plan, the development of the adolescent user's spinal problems will occur after a preset period of time in the future.

[0292] Based on the above embodiment, the device further includes a trend analysis module, which is used to:

[0293] Recording historical monitoring data of the youth user according to a preset time interval;

[0294] A change trend of the spinal problem of the adolescent user is generated based on the historical monitoring data.

[0295] Based on the above embodiment, the device further includes a plan adjustment module, which is used to:

[0296] acquiring the monitoring data of the adolescent user again according to a preset time interval, and evaluating the spinal problems of the adolescent user using the monitoring data;

[0297] The exercise plan is adjusted according to the evaluation results.

[0298] Based on the above embodiment, the device further includes an alarm module, which is used to:

[0299] According to the monitoring data, if it is detected that the current posture of the adolescent user does not meet the spinal health requirements, an alarm prompt is issued.

[0300] Based on the above embodiment, the device further includes a weight determination module, which is used to:

[0301] Obtaining the age of the teenage user;

[0302] determining a weight of the monitoring data according to the age;

[0303] The analysis module 302 is specifically used for:

[0304] The monitoring data and the weights are input into the spinal column analysis model to obtain the analysis results.

[0305] Figure 4 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present application, such as Figure 4As shown, the electronic device includes: a processor (processor) 401, a memory (memory) 402 and a bus 403; wherein,

[0306] The processor 401 and the memory 402 communicate with each other via the bus 403;

[0307] The processor 401 is used to call the program instructions in the memory 402 to execute the methods provided by the above-mentioned method embodiments, for example, including: obtaining monitoring data of adolescent users in daily life; the monitoring data includes at least one of sitting data, standing data, lying data and backpack data; using a spinal analysis model to analyze the monitoring data to obtain analysis results; the analysis results are used to characterize whether the adolescent user has a spinal health risk in the current state.

[0308] The processor 401 may be an integrated circuit chip with signal processing capabilities. The processor 401 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. It may implement or execute various methods, steps and logic block diagrams disclosed in the embodiments of the present application. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0309] The memory 402 may include, but is not limited to, random access memory (RAM), read only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable read-only memory (EEPROM), etc.

[0310] The present embodiment discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided by the above-mentioned method embodiments, for example, including: obtaining monitoring data of adolescent users in daily life; the monitoring data includes at least one of sitting data, standing data, lying data and backpack data; using a spinal analysis model to analyze the monitoring data to obtain analysis results; the analysis results are used to characterize whether the adolescent user has spinal health risks in the current state.

[0311] The present embodiment provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable the computer to execute the methods provided by the above-mentioned method embodiments, for example, including: obtaining monitoring data of adolescent users in daily life; the monitoring data includes at least one of sitting data, standing data, lying data and backpack data; using a spinal analysis model to analyze the monitoring data to obtain analysis results; the analysis results are used to characterize whether the adolescent user has a spinal health risk in the current state.

[0312] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0313] In addition, the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0314] Furthermore, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0315] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0316] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for monitoring the spine of adolescents, characterized in that: include: Acquire monitoring data of the adolescent user in daily life; the monitoring data includes at least one of sitting posture data, standing posture data, lying posture data and backpack data; The monitoring data is analyzed using a spine analysis model to obtain analysis results; the analysis results are used to characterize whether the adolescent user has spinal health risks in the current state.

2. The method according to claim 1, characterized in that The sitting posture data includes a first spinal curvature and a first cervical vertebra angle in a sitting state; The first spinal curvature and the first cervical vertebra angle are acquired through a posture sensor arranged on the adolescent user.

3. The method according to claim 2, characterized in that The sitting posture data also includes pressure data collected by pressure sensors arranged on the seat back and / or the seat cushion.

4. The method according to claim 2, characterized in that: The sitting posture data also includes sitting time and sitting frequency; The sedentary duration and sedentary frequency are obtained through monitoring by a smart bracelet or a smart seat cushion.

5. The method according to claim 2, characterized in that: The sitting posture data also includes a back image in a sitting state; The back image is acquired by an image acquisition device.

6. The method according to claim 1, characterized in that The standing posture data includes a second spinal curvature and a second cervical vertebra angle in a standing posture; The second spinal curvature and the second cervical angle are acquired by collecting a posture sensor disposed on the adolescent user.

7. The method according to claim 6, characterized in that The standing posture data also includes a back image in a standing posture; The back image is obtained by capturing an image of the young user in a standing state through an image capturing device.

8. The method according to claim 1, characterized in that: The lying posture data includes sleeping posture, sleeping duration and sleeping quality; The sleeping posture, sleeping duration and sleeping quality are collected and obtained through a smart mattress and / or a smart bracelet.

9. The method according to claim 1, characterized in that: The backpack data includes the weight of the backpack, the pressure difference between the two shoulders and the time of carrying the backpack; The backpack weight, the pressure difference between the two shoulders and the time of carrying the backpack are collected and obtained through sensors arranged on the shoulder straps of the backpack.

10. The method according to claim 1, characterized in that The monitoring data also includes exercise data, and the exercise data includes exercise type and exercise duration; The exercise type and the exercise duration are collected and obtained through a smart bracelet.

11. The method according to any one of claims 1 to 10, characterized in that: The spine analysis model includes a sub-analysis model constructed for each type of monitoring data; the monitoring data is analyzed using the spine analysis model to obtain analysis results, including: Input the monitoring data into the corresponding sub-model, use the corresponding sub-model to perform analysis, and obtain the intermediate analysis results output by each sub-model; The intermediate analysis result is analyzed again based on a preset decision strategy to obtain the analysis result; wherein the preset decision strategy includes a softmax gating function or a preset probability distribution function.

12. The method according to any one of claims 1 to 10, characterized in that: The step of analyzing the monitoring data using the spine analysis model to obtain analysis results includes: Extracting features from the time series data in the monitoring data to obtain time series features; wherein the time series data includes at least one of the sitting time and sitting frequency in the sitting data, the standing time and standing frequency in the standing data, and the sleeping time in the sleeping data; Extracting features from the spatial data in the monitoring data to obtain spatial features; wherein the spatial data includes at least one of the spinal curvature, cervical angle and pressure data corresponding to the sitting state and the standing state respectively; Fusing the temporal feature with the spatial feature to obtain a temporal-spatial feature; The time-space features are input into the spine analysis model to obtain the analysis results.

13. The method according to claim 12, characterized in that The monitoring data includes a back image of the adolescent user, and the inputting of the temporal-spatial features into the spine analysis model to obtain the analysis result includes: Performing feature extraction on the back image to obtain image features; Performing feature fusion on the image feature and the temporal-spatial feature to obtain fused features; The fused features are input into the spine analysis model to obtain the analysis results.

14. The method according to claim 1, characterized in that After obtaining the analysis result, the method further comprises: In the case where the analysis result indicates that the adolescent user has a spinal health risk, using a spine type recognition model to recognize the back image of the adolescent user to determine the type of spinal problem; An exercise program suitable for the teenage user is recommended according to the type of spinal problem.

15. The method according to claim 14, characterized in that The step of recommending an exercise plan suitable for the adolescent user according to the type of spinal problem comprises: Acquiring the physical test data and sports preferences of the adolescent user; An exercise plan suitable for the teenage user is recommended based on the spinal problem type, the physical measurement data and the exercise preference.

16. The method according to claim 14, characterized in that The method further comprises: Based on the monitoring data, it is predicted that if the adolescent user does not perform the recommended exercise plan, the development of the adolescent user's spinal problems will occur after a preset period of time in the future.

17. The method according to claim 14 or 15, characterized in that After recommending the exercise plan to the teenage user, the method further includes: Recording historical monitoring data of the youth user according to a preset time interval; A change trend of the spinal problem of the adolescent user is generated based on the historical monitoring data.

18. The method according to claim 14 or 15, characterized in that After recommending the exercise plan to the teenage user, the method further includes: acquiring the monitoring data of the adolescent user again according to a preset time interval, and evaluating the spinal problems of the adolescent user using the monitoring data; The exercise plan is adjusted according to the evaluation results.

19. The method according to claim 1, characterized in that After obtaining the monitoring data of the adolescent user in daily life, the method further includes: According to the monitoring data, if it is detected that the current posture of the adolescent user does not meet the spinal health requirements, an alarm prompt is issued.

20. The method according to claim 1, characterized in that After obtaining the monitoring data of the adolescent user in daily life, the method further includes: Obtaining the age of the teenage user; determining a weight of the monitoring data according to the age; The step of analyzing the monitoring data using the spine analysis model to obtain analysis results includes: The monitoring data and the weights are input into the spinal column analysis model to obtain the analysis results.

21. A spine monitoring device for adolescents, characterized in that: include: A data acquisition module, used to acquire monitoring data of the adolescent user in daily life; the monitoring data includes at least one of sitting posture data, standing posture data, lying posture data and backpack data; An analysis module, used to analyze the monitoring data using a spine analysis model to obtain analysis results; The analysis result is used to characterize whether the adolescent user has spinal health risks in the current state.

22. An electronic device, characterized in that: include: processor, memory and bus, wherein, The processor and the memory communicate with each other via the bus; The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the method according to any one of claims 1 to 20.

23. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, which, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 20.

24. A computer program product, characterized in that The method comprises computer program instructions, and when the computer program instructions are read and executed by a processor, the method according to any one of claims 1 to 20 is executed.

Images