Physical health monitoring method and system based on big data

Abstract

The invention discloses a physical health monitoring method and system based on big data, and relates to the technical field of intelligent sensors, and the method comprises the steps: obtaining a posture data sequence related to physiological postures in daily activities of a user, and extracting the distribution characteristics of abnormal postures based on the posture data sequence; determining a quantized value of accumulated pressure based on the distribution characteristics of the abnormal postures; if the quantized value of the accumulated pressure exceeds a preset accumulated pressure quantized threshold value, judging a risk level of physiological health of the user based on the quantized value of the accumulated pressure, and generating a trend index reflecting a pressure accumulation rule based on the risk level; and based on the trend index, carrying out weighted adjustment on the distribution characteristics of the abnormal postures, and generating a health monitoring feedback sequence based on the adjusted distribution characteristics of the abnormal postures, so that the physical health monitoring accuracy can be improved.

Description

Technical Field

[0001] This application relates to the field of intelligent sensor technology, and in particular to a method and system for monitoring physical health based on big data. Background Technology

[0002] In modern society, physical health monitoring has become an important means of ensuring people's quality of life and preventing chronic diseases. With the accelerating pace of life and the influence of unhealthy habits such as prolonged sitting, physical health problems are becoming increasingly prominent, especially in areas such as spinal health and joint stress. There is an urgent need for scientific methods to identify risks and provide a basis for intervention.

[0003] Current methods for monitoring physical health often lack in-depth exploration of dynamic changes and long-term effects when faced with the complex relationship between human posture and health. They focus more on surface symptoms and ignore the deep connections hidden behind daily behaviors. In particular, they fail to effectively capture the regular characteristics of individual posture changes over different time periods, resulting in a lack of precision.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main purpose of this application is to provide a method and system for monitoring physical health based on big data, which aims to solve the technical problem that current methods for monitoring physical health focus more on surface symptoms and lack accuracy.

[0006] To achieve the above objectives, this application proposes a method for monitoring physical health based on big data, the method comprising:

[0007] Acquire posture data sequences related to physiological postures in users' daily activities, and extract distribution features of abnormal postures based on the posture data sequences;

[0008] The quantized value of the cumulative pressure is determined based on the distribution characteristics of the abnormal posture;

[0009] If the quantified value of the accumulated pressure exceeds the preset quantified threshold of accumulated pressure, the risk level of the user's physiological health is determined based on the quantified value of the accumulated pressure, and a trend indicator reflecting the pattern of pressure accumulation is generated based on the risk level.

[0010] Based on the trend indicators, the distribution characteristics of the abnormal postures are weighted and adjusted, and the triggering conditions for risk warnings are determined based on the adjusted distribution characteristics of the abnormal postures.

[0011] If the triggering conditions for the risk warning are met, a health monitoring feedback sequence is generated based on the triggering conditions and the trend indicators reflecting the cumulative stress pattern.

[0012] In one embodiment, the step of acquiring posture data sequences related to physiological postures in a user's daily activities and extracting distribution features of abnormal postures based on the posture data sequences includes:

[0013] The attitude data sequence is acquired through sensor devices;

[0014] The attitude data sequence is processed by recording in segments along the time axis;

[0015] The distribution features of the abnormal postures are extracted based on the processed posture data sequence.

[0016] In one embodiment, the step of determining the quantized value of the cumulative pressure based on the distribution characteristics of the abnormal posture includes:

[0017] Based on the distribution characteristics of the abnormal postures, posture segments related to physiological load are selected from a benchmark database constructed based on users' historical normal time period data.

[0018] The attitude segment is subjected to data smoothing processing;

[0019] Fluctuation parameters and abnormal amplitudes are calculated based on attitude segments after data smoothing.

[0020] The quantified value of the cumulative pressure is determined based on the fluctuation parameters and the abnormal amplitude.

[0021] In one embodiment, determining the risk level of a user's physiological health based on the quantified value of the accumulated pressure includes:

[0022] Analyze the rate of change of the quantified value of the accumulated pressure over time.

[0023] Periodic features are extracted based on the posture data sequence;

[0024] Based on the rate of change of the pattern, combined with the distribution characteristics of the abnormal posture and the periodic characteristics, the risk level of the user's physiological health is determined by a preset risk level judgment rule.

[0025] In one embodiment, generating a trend indicator reflecting the cumulative stress pattern based on the risk level includes:

[0026] Based on the risk level, historical pressure distribution and load cycle data related to biomechanical mechanisms are obtained from a preset database;

[0027] By integrating and analyzing the historical pressure distribution and the load cycle data using data sequence fusion technology, the trend index reflecting the pressure accumulation pattern is obtained.

[0028] In one embodiment, the step of weighting and adjusting the distribution characteristics of the abnormal postures based on the trend index, and determining the triggering conditions for risk warning based on the adjusted abnormal posture distribution characteristics includes:

[0029] Based on the load cycle data, the characteristics of abnormal load locations are obtained.

[0030] The distribution characteristics of the abnormal posture are weighted and adjusted by combining the trend indicators with the characteristics of the historical pressure distribution and the abnormal load points.

[0031] The triggering conditions for the risk warning are determined based on the distribution characteristics of the abnormal postures after weighted adjustment.

[0032] In one embodiment, the step of generating a health monitoring feedback sequence based on the triggering condition and the trend indicator reflecting the cumulative stress pattern if the risk warning triggering condition is met includes:

[0033] If the triggering conditions for the risk warning are met, a sequential list of intervention measures for physiological structures is generated based on the triggering conditions and the trend indicators reflecting the cumulative stress patterns.

[0034] Based on the proposed list of intervention measures, combined with the load cycle data and the trend indicators reflecting the cumulative pressure pattern, a personalized posture adjustment optimization path is obtained.

[0035] The health monitoring feedback sequence is generated based on the personalized posture adjustment optimization path.

[0036] In one embodiment, the distribution characteristics of the abnormal posture include the duration interval of the abnormal posture, and the step of generating the health monitoring feedback sequence based on the personalized posture adjustment optimization path includes:

[0037] Based on the personalized posture adjustment optimization path, the health monitoring feedback sequence is generated by integrating the fluctuation parameters and the duration range of the abnormal posture.

[0038] In one embodiment, the trend indicator reflecting the cumulative stress pattern includes long-term cumulative effect and stress change slope, and the generation of a sequential list of intervention recommendations targeting physiological structures based on the triggering condition and the trend indicator reflecting the cumulative stress pattern includes:

[0039] Based on the triggering conditions and the long-term cumulative effects, a sequential list of intervention recommendations targeting physiological structures is generated;

[0040] The process of obtaining a personalized posture adjustment optimization path based on the sequenced suggestion list of intervention measures, combined with the load cycle data and the trend indicators reflecting the cumulative stress pattern, includes:

[0041] Based on the proposed list of intervention measures, combined with the load cycle data and the pressure change slope, the personalized attitude adjustment optimization path is obtained.

[0042] Furthermore, to achieve the above objectives, this application also proposes a physical health monitoring system based on big data, wherein the physical health monitoring system based on big data includes:

[0043] The feature acquisition module is used to acquire posture data sequences related to physiological postures in the user's daily activities, and extract distribution features of abnormal postures based on the posture data sequences.

[0044] The quantization value determination module is used to determine the quantization value of the cumulative pressure based on the distribution characteristics of the abnormal posture;

[0045] The risk level judgment module is used to determine the risk level of the user's physiological health based on the quantitative value of the accumulated pressure if the quantitative value of the accumulated pressure exceeds the preset quantitative threshold of the accumulated pressure, and to generate a trend indicator reflecting the pattern of pressure accumulation based on the risk level.

[0046] The adjustment module is used to perform weighted adjustment of the distribution characteristics of the abnormal posture based on the trend index, and to determine the triggering conditions for risk warning based on the adjusted abnormal posture distribution characteristics.

[0047] The feedback sequence generation module is used to generate a health monitoring feedback sequence based on the triggering conditions and the trend indicators reflecting the cumulative stress pattern if the risk warning triggering conditions are met.

[0048] One or more technical solutions proposed in this application have at least the following technical effects:

[0049] This application acquires posture data sequences related to physiological postures during users' daily activities, extracts distribution characteristics of abnormal postures based on these data sequences, determines a quantified value of accumulated pressure based on these distribution characteristics, and assesses the user's physiological health risk level based on the quantified value of accumulated pressure if the accumulated pressure exceeds a preset threshold. A trend indicator reflecting the pattern of accumulated pressure is then generated based on this risk level. This process identifies the distribution patterns of abnormal postures from massive amounts of data and further reveals their profound impact on accumulated pressure in key body parts, thus providing a more accurate reflection of user health risks rather than relying on general standards or focusing on surface symptoms. Furthermore, this application adjusts the distribution characteristics of abnormal postures through trend indicators and dynamically updates risk warning trigger conditions, allowing for continuous iteration and optimization of risk warning conditions, thereby improving the accuracy of health monitoring. Attached Figure Description

[0050] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0051] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart illustrating an embodiment of a big data-based physical health monitoring method according to this application.

[0053] Figure 2 This is a flowchart illustrating a second embodiment of a big data-based physical health monitoring method according to this application.

[0054] Figure 3 This is a flowchart illustrating a third embodiment of a big data-based physical health monitoring method according to this application.

[0055] Figure 4 This is a schematic diagram of the module structure of a physical health monitoring system based on big data, according to an embodiment of this application.

[0056] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0057] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0058] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0059] The main solution of this application embodiment is as follows: Acquire posture data sequences related to physiological posture during the user's daily activities, and extract distribution features of abnormal postures based on the posture data sequences; determine the quantified value of accumulated pressure based on the distribution features of the abnormal postures; if the quantified value of accumulated pressure exceeds a preset accumulated pressure quantification threshold, determine the risk level of the user's physiological health based on the quantified value of accumulated pressure, and generate a trend indicator reflecting the pattern of pressure accumulation based on the risk level; perform weighted adjustment on the distribution features of the abnormal postures based on the trend indicator, and determine the triggering condition for risk warning based on the adjusted abnormal posture distribution features; if the triggering condition for risk warning is met, generate a health monitoring feedback sequence based on the triggering condition and the trend indicator reflecting the pattern of pressure accumulation.

[0060] In this embodiment, for ease of description, the physical health monitoring system will be used as the subject of the description.

[0061] In modern society, physical health monitoring has become an important means of ensuring people's quality of life and preventing chronic diseases. With the accelerated pace of life and the influence of unhealthy habits such as prolonged sitting, physical health problems are becoming increasingly prominent, especially in areas such as spinal health and joint stress. Scientific methods are urgently needed to identify risks and provide a basis for intervention. Research in this field is not only related to individual health but also crucial for optimizing the public health system. However, current research and methods often lack in-depth exploration of dynamic changes and long-term effects when facing the complex relationship between human posture and health. Many programs focus more on recording surface symptoms, neglecting the deep connections hidden behind daily behaviors, especially failing to effectively capture the regular characteristics of postural changes in individuals over different time periods. This limitation makes health risk warnings and interventions lack precision and foresight. A deeper technical challenge lies in how to accurately quantify the impact of the duration distribution of abnormal postures on the body. The duration of abnormal postures directly determines the cumulative pressure on certain parts of the body; for example, prolonged bending or stooping can lead to chronic strain on the cervical or lumbar spine. This cumulative stress brings another challenge: understanding the long-term effects of specific postures on key areas like joints and the spine. These effects are often not linear but exhibit complex fluctuations over time and with individual differences. These two issues are intertwined, making it difficult to accurately determine when, where, and how a particular bad habit poses a substantial threat to health in actual monitoring. Therefore, identifying the duration distribution patterns of abnormal postures from massive amounts of data and further revealing their deep-seated mechanisms of cumulative stress on key body parts has become a crucial issue in the field of physical health monitoring. This application provides a solution that acquires posture data sequences related to physiological postures during users' daily activities, extracts distribution characteristics of abnormal postures based on these data sequences, determines a quantified value of accumulated pressure based on these distribution characteristics, and assesses the user's physiological health risk level based on the quantified value of accumulated pressure if the accumulated pressure exceeds a preset threshold. A trend indicator reflecting the pattern of pressure accumulation is then generated based on this risk level. This approach identifies the distribution patterns of abnormal postures from massive amounts of data and further reveals their profound impact on accumulated pressure in key body parts, thus providing a more accurate reflection of user health risks rather than relying on general standards or focusing on surface symptoms. Furthermore, this application adjusts the abnormal posture distribution characteristics through trend indicators and dynamically updates risk warning trigger conditions, allowing for continuous iteration and optimization of risk warning conditions, thereby improving the accuracy of health monitoring.

[0062] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone; or an electronic device capable of performing the above functions, a physical health monitoring device based on big data; or an electronic system capable of performing the above functions, a physical health monitoring system based on big data. The following description uses a physical health monitoring system as an example to illustrate this embodiment and the subsequent embodiments.

[0063] Based on this, the embodiments of this application provide a method for monitoring physical health based on big data, referring to... Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the physical health monitoring method based on big data in this application.

[0064] In this embodiment, the physical health monitoring method based on big data includes steps S1 to S5:

[0065] Step S1: Obtain posture data sequences related to physiological postures in the user's daily activities, and extract distribution features of abnormal postures based on the posture data sequences;

[0066] This can be achieved by using smart sensors to collect physiological posture data from users' daily activities in real time, forming a posture data sequence. For example, it can record the duration, angle, and frequency of actions such as standing, sitting, bending over, and lying down.

[0067] In this process, postures that do not conform to the normal physiological range (such as prolonged head-down posture, excessive bending angle, unilateral weight-bearing, etc.) can be screened from the posture data sequence, and their distribution characteristics (such as frequency of occurrence, duration, intensity, and occurrence scenario) can be statistically analyzed as the distribution characteristics of abnormal postures.

[0068] Step S2: Determine the quantized value of the cumulative pressure based on the distribution characteristics of the abnormal posture;

[0069] Based on the distribution characteristics of abnormal postures, combined with biomechanical models or empirical formulas, the cumulative pressure on tissues such as muscles, joints, and intervertebral discs is calculated. For example, for every 10° increase in the bending angle, the pressure on the lumbar intervertebral discs increases by approximately 15%.

[0070] Then, the calculation results of the cumulative pressure are converted into a unified quantitative value (such as 0~100 points) to facilitate subsequent risk level judgment.

[0071] Step S3: If the quantified value of the accumulated pressure exceeds the preset quantified threshold of the accumulated pressure, the risk level of the user's physiological health is determined based on the quantified value of the accumulated pressure, and a trend indicator reflecting the pattern of pressure accumulation is generated based on the risk level.

[0072] Among them, trend indicators are parameters that reflect the pattern of pressure accumulation and are used to predict future risks.

[0073] Specifically, the cumulative stress quantification value can be compared with a preset cumulative stress quantification threshold (e.g., 80 points). If it exceeds the cumulative stress quantification threshold, the risk level of the user's physiological health is determined based on the cumulative stress quantification value, and a trend indicator reflecting the stress accumulation pattern is generated based on the risk level.

[0074] As one implementation method, the risk level of a user's physiological health can be divided into low risk, medium risk, and high risk. Taking a risk value of 0-100 points and a preset cumulative stress quantification threshold of 60 points as an example, 60-70 points is low risk, 70-85 points is medium risk, and 85-100 points is high risk.

[0075] Step S4: Based on the trend index, the distribution characteristics of the abnormal posture are weighted and adjusted, and the triggering conditions for risk warning are determined based on the adjusted abnormal posture distribution characteristics.

[0076] Among them, the distribution characteristics of abnormal postures can be adjusted according to trend indicators (such as a rapid upward trend in pressure), and the triggering conditions for risk warnings can be determined based on the adjusted distribution characteristics of abnormal postures.

[0077] For example, if a trend indicator shows that users are bending over more frequently, the weight of the frequency of occurrence in the distribution characteristics of abnormal postures can be adjusted, and the trigger condition for risk warning can be determined as bending over more than twice per minute.

[0078] For example, if trend indicators show that pressure is accumulating too quickly, the weight of duration in the distribution characteristics is increased.

[0079] Step S5: If the triggering condition for the risk warning is met, a health monitoring feedback sequence is generated based on the triggering condition and the trend indicator reflecting the cumulative stress pattern.

[0080] Among them, the health monitoring feedback sequence is a layered and time-based health intervention suggestion generated based on the user's status. It can form a closed loop through the user's status and user feedback, guiding the user to adaptively adjust their behavior.

[0081] When the triggering conditions for a risk warning are met, a health monitoring feedback sequence can be generated by combining the triggering conditions and trend indicators (such as an upward trend in stress). This sequence includes immediate suggestions, long-term plans, and medical alerts. For example, an immediate suggestion could be "Stop bending over immediately and perform spinal extension exercises." A long-term plan could be "Set three rest reminders daily and stretch for five minutes each time." A medical alert could be "High-risk warnings for three consecutive days; it is recommended to schedule an appointment with an orthopedic doctor."

[0082] As one implementation method, it can be presented via terminal devices (such as mobile phones, smartwatches, and health management platforms) in the form of text, charts, voice, or vibration.

[0083] This embodiment provides a big data-based method for monitoring physical health. This application acquires posture data sequences related to physiological postures during users' daily activities, extracts distribution characteristics of abnormal postures based on these data sequences, determines a quantitative value of accumulated stress based on these distribution characteristics, and if the quantitative value of accumulated stress exceeds a preset accumulated stress threshold, then the user's physiological health risk level is judged based on the quantitative value of accumulated stress. A trend indicator reflecting the pattern of stress accumulation is generated based on the risk level. This allows for the identification of abnormal posture distribution patterns from massive amounts of data and further reveals their profound impact on accumulated stress in key body parts, thus more accurately reflecting the user's health risk rather than relying on general standards or focusing on surface symptoms. Furthermore, this application adjusts the abnormal posture distribution characteristics through trend indicators and dynamically updates risk warning trigger conditions, allowing for continuous iterative optimization of risk warning conditions, thereby improving the accuracy of health monitoring.

[0084] Based on Embodiment 1 of this application, in Embodiment 2 of this application, the content that is the same as or similar to that in Embodiment 1 can be referred to the above description, and will not be repeated hereafter. Based on this, refer to... Figure 2 Step S1, acquiring posture data sequences related to physiological postures during the user's daily activities, and extracting distribution features of abnormal postures based on the posture data sequences, includes steps S11 to S13:

[0085] Step S11: Acquire the attitude data sequence through sensor devices;

[0086] Among them, intelligent sensor devices can be inertial sensors used to measure the acceleration and angular velocity of various parts of the body and reflect the state of motion; pressure sensors embedded in seats or wearable devices to monitor the pressure distribution between the body and the contact surface; and cameras that capture the angles of human joints through image recognition, etc.

[0087] Step S12: Process the attitude data sequence by recording it in segments along the time axis;

[0088] The attitude data sequence is segmented according to the time axis to obtain the processed attitude data sequence.

[0089] For example, the data can be segmented into 5-minute intervals.

[0090] Step S13: Extract the distribution features of the abnormal posture based on the processed posture data sequence.

[0091] Among these methods, abnormal pose identification can be performed on the processed pose data sequence through threshold determination or a pre-trained machine learning model, marking the data sequence with abnormal poses, and extracting the distribution features of abnormal poses based on the data sequence with abnormal poses.

[0092] The distribution characteristics of abnormal postures can include time and space dimensions. The time dimension includes abnormal frequency, peak period, duration, etc., while the space dimension includes high-risk areas, etc.

[0093] This application embodiment acquires the posture data sequence through sensor devices, processes the posture data sequence by recording it in segments along the time axis, and extracts the distribution characteristics of the abnormal postures based on the processed posture data sequence. This allows for the identification of distribution patterns of abnormal postures from massive amounts of data, thereby more accurately reflecting user health risks.

[0094] Based on the above embodiments of this application, a third embodiment of this application is proposed. In this third embodiment, content that is the same as or similar to the above embodiments can be referred to the above description, and will not be repeated hereafter. Based on this, refer to... Figure 3 Step S2, determining the quantified value of the cumulative pressure based on the distribution characteristics of the abnormal posture, includes steps S21 to S24:

[0095] Step S21: Based on the distribution characteristics of the abnormal postures, select posture segments related to physiological load from a benchmark database constructed based on the user's historical normal time period data.

[0096] Among these methods, a baseline database can be built in the cloud based on the user's historical normal time period data, and then posture segments related to physiological load can be selected from the baseline database according to the distribution characteristics of abnormal postures.

[0097] Among them, posture segments related to physiological load are posture data segments that are directly related to human physiological stress. They are obtained by filtering from the database based on the distribution characteristics of abnormal postures (such as location and intensity). For example, prolonged sitting with a spinal forward tilt angle >30° and lifting movements with shoulder pressure >60N are high-load postures, while upright sitting postures and natural standing postures, which have less physiological stress, are low-load postures.

[0098] Step S22: Perform data smoothing processing on the attitude segment;

[0099] Among these methods, filtering algorithms can be used to smooth the attitude segments, thereby removing noise or short-term fluctuations in the attitude data and improving the accuracy of subsequent parameter calculations.

[0100] For example, a 5-point moving average filtering algorithm is used to smooth the acceleration value sequence [2.6, 2.8, 3.0, 2.7, 2.5]. The smoothed value is calculated as [(2.6+2.8+3.0+2.7+2.5) / 5=2.72], thereby reducing the influence of noise and ensuring the accuracy of subsequent analysis.

[0101] Step S23: Calculate the fluctuation parameters and abnormal amplitude based on the attitude segment after data smoothing.

[0102] The fluctuation parameter describes the degree of change of attitude data over time. The fluctuation parameter can include standard deviation and coefficient of variation. The standard deviation reflects the dispersion of the data, and the coefficient of variation is the ratio of the standard deviation to the mean, which is used to compare the volatility of data with different dimensions.

[0103] Abnormal amplitude refers to the degree to which attitude data deviates from the normal range. Abnormal amplitude can include absolute amplitude and relative amplitude. Absolute amplitude is the difference between the abnormal value and the normal threshold, while relative amplitude is the ratio of the abnormal value to the normal mean.

[0104] Step S24: Determine the quantified value of the cumulative pressure based on the fluctuation parameters and the abnormal amplitude.

[0105] Among them, by combining fluctuation parameters and abnormal amplitude, the total physiological stress that the human body suffers due to abnormal posture over a period of time can be quantified.

[0106] As one implementation method, the quantized value of the cumulative pressure can be determined based on the fluctuation parameter and the abnormal amplitude as follows: Quantized value of cumulative pressure = fluctuation parameter * abnormal amplitude * duration.

[0107] As another implementation method, the cumulative pressure can be obtained by weighted summation of the fluctuation parameters and the abnormal amplitude.

[0108] This application embodiment, based on the distribution characteristics of the abnormal postures, filters posture segments related to physiological load from a benchmark database constructed based on users' historical normal time period data, performs data smoothing processing on the posture segments, calculates fluctuation parameters and abnormal amplitude based on the smoothed posture segments, and determines the quantitative value of the cumulative pressure based on the fluctuation parameters and the abnormal amplitude. Thus, by revealing the deep impact of the abnormal posture distribution pattern on the cumulative pressure of key parts of the body, it can more accurately reflect the user's health risk.

[0109] Based on the above embodiments of this application, Embodiment Four of this application is proposed. In Embodiment Four, content that is the same as or similar to the above embodiments can be referred to the above description and will not be repeated hereafter. Based on this, the step of determining the risk level of the user's physiological health based on the quantified value of the accumulated pressure includes steps S31~S33:

[0110] Step S31: Analyze the rate of change of the quantified value of the accumulated pressure on the time axis.

[0111] The pattern change rate is the dynamic change of the cumulative stress quantification value over time, reflecting the accumulation or relief trend of physiological load. An increased pattern change rate indicates that the user is experiencing frequent high-load actions in a short period of time, while a decreased pattern change rate indicates that the user is in a long-term high-load state without effective relief.

[0112] As one implementation method, the rate of change of the quantified value of cumulative pressure on the time axis can be obtained through slope analysis or differential calculation.

[0113] Step S32: Extract periodic features based on the attitude data sequence;

[0114] Among them, periodic features extracted from posture data sequences can reveal regular patterns in user behavior.

[0115] As one implementation method, Fourier transform or autocorrelation analysis can be used to extract periodic features.

[0116] The Fourier transform converts time series data from the time domain to the frequency domain. By analyzing the peak frequency in the frequency domain, it identifies the main periodic components in the data and extracts the main period frequency, period intensity, and multi-period combinations.

[0117] Autocorrelation analysis, in particular, detects recurring patterns in time series data by calculating the correlation between the data at different time lags. If a significant correlation exists at a lag of τ, it indicates that the data has a periodicity τ.

[0118] Step S33: Based on the rate of change of the pattern, combined with the distribution characteristics of the abnormal posture and the periodic characteristics, the risk level of the user's physiological health is determined by a preset risk level judgment rule.

[0119] Among them, the normalized pattern change rate, abnormal posture distribution and periodic features can be input into the rule engine, and the risk level of the user's physiological health can be determined by the preset risk level judgment rules.

[0120] As one implementation method, the risk level judgment rule can be set as follows: when the pressure increase is greater than a preset threshold, and abnormal posture occurs frequently and periodic characteristics disappear, the risk level is judged as high risk level; when the pressure fluctuation increases, occasional abnormal posture occurs, and periodic characteristics deviate slightly, the risk level is judged as medium risk level; when the pressure is stable, there are no abnormal posture and periodic characteristics are normal, the risk level is judged as low risk level.

[0121] This application embodiment analyzes the rate of change of the quantified value of the accumulated pressure over time, extracts periodic features based on the posture data sequence, and, based on the rate of change of the pattern, combines the distribution characteristics of the abnormal posture with the periodic features, and judges the risk level of the user's physiological health through preset risk level judgment rules, thus achieving accurate assessment from posture data to physiological health risk.

[0122] Based on the above embodiments of this application, Embodiment Five of this application is proposed. In Embodiment Five, content that is the same as or similar to the above embodiments can be referred to the above description and will not be repeated hereafter. Based on this, the step of generating a trend indicator reflecting the cumulative pressure pattern based on the risk level includes steps S34~S35:

[0123] Step S34: Based on the risk level, obtain historical pressure distribution and load cycle data related to biomechanical mechanisms from a preset database;

[0124] Biomechanical mechanisms are a discipline that studies the structural, kinematic, and functional changes of organisms under mechanical action, with a focus on the effects of pressure distribution, load cycles, and other factors on tissues (such as muscles and bones).

[0125] Historical pressure distribution refers to data on the changes in pressure values ​​of specific body parts or equipment components over time or space, recorded over a period of time.

[0126] Based on the current risk level, historical stress distribution and load cycle data related to the biomechanical mechanisms of similar risks are filtered from the database. For example, if the current risk level is high (excessive lordosis angle), the spinal stress distribution (such as hourly stress value) and load cycle (such as continuous bending work every 4 hours) of all high-risk users in the past are extracted from the database.

[0127] Among them, load cycle data is the regular pattern of load recurrence over time, including cycle length and intensity.

[0128] Step S35: The historical pressure distribution and the load cycle data are integrated and analyzed using data sequence fusion technology to obtain the trend index reflecting the pressure accumulation pattern.

[0129] Among them, data sequence fusion technology is a technique that integrates historical pressure distribution and load cycle data into a single feature or indicator through statistical methods or machine learning models.

[0130] As one implementation method, the covariance and correlation between the pressure values ​​of historical pressure distribution and the load cycle can be calculated, the rate of change of pressure with the load cycle can be extracted, and the rate of change of pressure with the load cycle can be used as a trend indicator reflecting the pressure accumulation law.

[0131] As another implementation method, machine learning models such as LSTM and Transformer can be used to learn the joint pattern of pressure distribution and load cycle, and predict trend indicators that reflect the cumulative pressure pattern.

[0132] The embodiments of this application obtain trend indicators that reflect the pressure accumulation pattern by integrating historical pressure distribution and load cycle data. This can more accurately reflect the pressure accumulation pattern, rather than a single pressure value or cycle length, thereby improving the accuracy of risk prediction.

[0133] Based on the above embodiments of this application, Embodiment Six of this application is proposed. In Embodiment Six, content that is the same as or similar to the above embodiments can be referred to the above description and will not be repeated hereafter. Based on this, step S4, based on the trend indicator, weights the distribution characteristics of the abnormal posture and determines the triggering conditions for risk warning based on the adjusted abnormal posture distribution characteristics, including steps S41~S43:

[0134] Step S41: Obtain the characteristics of abnormal load locations based on the load cycle data;

[0135] Among them, abnormal load points are the time points or body parts where the pressure value deviates significantly from the normal range during the load cycle. The characteristics of abnormal load points describe the quantitative attributes of abnormal load points, including pressure value, time of occurrence, duration, and degree of deviation from the normal value.

[0136] Among them, load cycle data can be scanned to mark points where the pressure value deviates significantly from the normal range, which are then identified as abnormal load points. For abnormal load points, features such as pressure value, occurrence time, duration, and degree of deviation from the normal value can be extracted.

[0137] Step S42: Using the trend index, combined with the characteristics of the historical pressure distribution and the abnormal load points, the distribution characteristics of the abnormal posture are weighted and adjusted.

[0138] It should be noted that the distribution characteristics of abnormal postures include frequency of occurrence, duration, spatial distribution, periodicity, etc. Each characteristic in the distribution characteristics of abnormal postures has a weight, and the weight of each characteristic is affected by trend indicators, historical pressure distribution, and the characteristics of abnormal load points.

[0139] For example, if the rate of pressure accumulation in the trend indicator increases, the weight of duration in the abnormal posture distribution feature should be increased because prolonged high pressure will accelerate damage.

[0140] The weighted adjustment of the distribution characteristics of abnormal postures is to assign higher weights to features that are strongly correlated with injury, and lower weights to features that are weakly correlated or interfering. Furthermore, the weights are dynamically adjusted based on the user's stress tolerance.

[0141] Step S43: Determine the triggering conditions for the risk warning based on the distribution characteristics of the weighted and adjusted abnormal postures.

[0142] One implementation method is to trigger a risk warning if any adjusted feature exceeds a threshold.

[0143] As another implementation method, the trigger condition for risk warning can also be that the sum of the weighted adjusted distribution characteristics of abnormal postures exceeds a preset threshold.

[0144] This application uses the trend indicators, combined with the characteristics of the historical pressure distribution and the abnormal load points, to weight and adjust the distribution characteristics of the abnormal posture. This allows for higher weighting of features strongly correlated with damage and lower weighting of weakly correlated or interfering features, thereby improving the accuracy of risk identification.

[0145] Based on the above embodiments of this application, Embodiment Seven of this application is proposed. In Embodiment Seven, content that is the same as or similar to the above embodiments can be referred to the above description and will not be repeated hereafter. Based on this, step S5, if the triggering condition of the risk warning is met, then generating a health monitoring feedback sequence according to the triggering condition and the trend indicator reflecting the cumulative stress pattern includes steps S51~S54:

[0146] Step S51: If the triggering condition for the risk warning is met, then a sequential list of intervention measures for physiological structures is generated based on the triggering condition and the trend indicator reflecting the cumulative stress pattern.

[0147] Among them, trend indicators are parameters that reflect the pattern of pressure accumulation and are used to predict future risks.

[0148] Among them, the Sequential Recommendation List of Interventions for Physiological Structures is a standardized, step-by-step prevention or correction plan developed for risks or functional impairments in specific physiological structures (such as the spine, joints, muscles, etc.). The Sequential Recommendation List of Interventions for Physiological Structures is a static solution library that provides standardized and actionable steps.

[0149] As one implementation method, if the triggering conditions for risk warning are met, a standardized, step-by-step prevention or correction plan for physiological structures is generated based on the triggering conditions and trend indicators reflecting the cumulative stress pattern. This plan is then used as a list of intervention measures to sequence recommendations. This list serves as a static solution library, providing users with standardized operational guidelines to reduce the risk of physiological structural damage or improve functional impairments by clearly defining action specifications (such as adjusting posture frequency, using assistive tools, and performing specific training).

[0150] The trend indicator reflecting the cumulative effect of pressure includes the long-term cumulative effect, and step S51 includes step S511:

[0151] Step S511: Based on the triggering conditions and the long-term cumulative effects, generate a list of sequential recommendations for intervention measures targeting physiological structures.

[0152] Among them, the long-term cumulative effect is the risk of chronic damage to tissues caused by stress or abnormal posture.

[0153] Based on triggering conditions and the risk of chronic damage to tissues caused by stress or abnormal posture, a sequential list of intervention recommendations for physiological structures can be generated, providing users with standardized operational guidelines to reduce the risk of chronic damage.

[0154] Step S52: Based on the intervention measure sequence suggestion list, and combined with the load cycle data and the trend indicators reflecting the pressure accumulation pattern, obtain a personalized posture adjustment optimization path;

[0155] Among them, the personalized posture adjustment optimization path is a sequence of optimal posture adjustment strategies for a specific user, generated based on individual physiological structure characteristics, load cycle data, and trend indicators of pressure accumulation patterns.

[0156] As one implementation method, high-risk behavior patterns of users can be identified based on load cycle data (such as sitting for 6 hours a day or bending over too often). Combined with trend indicators, the most effective adjustment actions can be selected from the intervention list (such as recommending dynamic sitting posture switching rather than simply straightening the back for sedentary individuals). Based on the user's physiological characteristics (such as height and spinal curvature) and behavioral habits (such as work scenario), the frequency, intensity, and timing of the actions can be dynamically adjusted (such as reminding once every 25 minutes and maintaining the correct posture for 5 minutes after adjustment).

[0157] The trend indicator reflecting the cumulative pressure pattern also includes the slope of pressure change, and step S52 includes step S521:

[0158] Step S521: Based on the intervention measure sequence suggestion list, and combined with the load cycle data and the pressure change slope, obtain the personalized posture adjustment optimization path.

[0159] Among them, the slope of the pressure change is the rate at which the pressure changes over time, reflecting the urgency of the pressure increase.

[0160] As one implementation method, high-risk behavior patterns of users can be identified based on load cycle data. Combined with the slope of stress change, the most effective adjustment actions can be selected from the intervention list. Based on the user's physiological characteristics and behavioral habits, the frequency, intensity and timing of the actions can be dynamically adjusted.

[0161] Step S53: Generate the health monitoring feedback sequence based on the personalized posture adjustment optimization path.

[0162] Among them, based on personalized posture adjustment and path optimization, instant suggestions, long-term plans, and medical tips are generated, and the generated instant suggestions, long-term plans, and medical tips are used as health monitoring feedback sequences.

[0163] Step S53 includes step S531:

[0164] Step S531: Based on the personalized posture adjustment optimization path, the fluctuation parameters and the duration range of the abnormal posture are fused to generate the health monitoring feedback sequence.

[0165] To generate a health monitoring feedback sequence, a personalized posture adjustment optimization path is required, which integrates fluctuation parameters and the duration range of abnormal postures. The correlation between the two is dynamically analyzed: when fluctuation parameters show a rapid increase in pressure or posture abnormalities, and the duration reaches or approaches the injury threshold, immediate suggestions (such as emergency vibration alerts + voice commands) are triggered first; if fluctuation parameters show a slow upward trend and the duration range is long, a long-term plan is generated (such as pushing an alert every 2 hours); if fluctuation parameters and duration range match the characteristics of high-risk injury, medical prompts are generated simultaneously (such as suggesting contacting a doctor for professional evaluation), thus generating a health monitoring feedback sequence.

[0166] This application embodiment provides a standardized action library through an intervention sequence suggestion list, and selects and combines the optimal action combination from the intervention sequence suggestion list through a personalized posture adjustment optimization path. Based on the personalized posture adjustment optimization path, the fluctuation parameters and the duration range of the abnormal posture are fused to generate a health monitoring feedback sequence, thereby realizing personalized health management.

[0167] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the big data-based physical health monitoring method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0168] This application also provides a physical health monitoring system based on big data; please refer to [reference needed]. Figure 4 The big data-based physical health monitoring system includes:

[0169] The feature acquisition module 10 is used to acquire posture data sequences related to physiological postures in the user's daily activities, and extract distribution features of abnormal postures based on the posture data sequences.

[0170] The quantization value determination module 20 is used to determine the quantization value of the cumulative pressure based on the distribution characteristics of the abnormal posture;

[0171] The risk level judgment module 30 is used to determine the risk level of the user's physiological health based on the quantitative value of the accumulated pressure if the quantitative value of the accumulated pressure exceeds the preset quantitative threshold of the accumulated pressure, and to generate a trend indicator reflecting the pattern of pressure accumulation based on the risk level.

[0172] The adjustment module 40 is used to perform weighted adjustment on the distribution characteristics of the abnormal posture based on the trend index, and to determine the triggering conditions for risk warning based on the adjusted abnormal posture distribution characteristics.

[0173] The feedback sequence generation module 50 is used to generate a health monitoring feedback sequence based on the triggering conditions and the trend indicators reflecting the cumulative stress pattern if the triggering conditions of the risk warning are met.

[0174] The physical health monitoring system based on big data provided in this application, employing the physical health monitoring method based on big data in the above embodiments, can solve the technical problem that current physical health monitoring methods focus more on surface symptoms and lack accuracy. Compared with the prior art, the beneficial effects of the physical health monitoring system based on big data provided in this application are the same as those of the physical health monitoring method based on big data provided in the above embodiments, and other technical features of the physical health monitoring system based on big data are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0175] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for monitoring physical health based on big data, characterized in that, The method includes: Acquire posture data sequences related to physiological postures in users' daily activities, and extract distribution features of abnormal postures based on the posture data sequences; The quantized value of the cumulative pressure is determined based on the distribution characteristics of the abnormal posture; If the quantified value of the accumulated pressure exceeds the preset quantified threshold of accumulated pressure, the risk level of the user's physiological health is determined based on the quantified value of the accumulated pressure, and a trend indicator reflecting the pattern of pressure accumulation is generated based on the risk level. Based on the trend indicators, the distribution characteristics of the abnormal postures are weighted and adjusted, and the triggering conditions for risk warnings are determined based on the adjusted distribution characteristics of the abnormal postures. If the triggering conditions for the risk warning are met, a health monitoring feedback sequence is generated based on the triggering conditions and the trend indicators reflecting the cumulative stress pattern.

2. The method as described in claim 1, characterized in that, The step of acquiring posture data sequences related to physiological postures in users' daily activities and extracting distribution features of abnormal postures based on the posture data sequences includes: The attitude data sequence is acquired through sensor devices; The attitude data sequence is processed by recording in segments along the time axis; The distribution features of the abnormal postures are extracted based on the processed posture data sequence.

3. The method as described in claim 1, characterized in that, The step of determining the quantized value of cumulative pressure based on the distribution characteristics of the abnormal posture includes: Based on the distribution characteristics of the abnormal postures, posture segments related to physiological load are selected from a benchmark database constructed based on users' historical normal time period data. The attitude segment is subjected to data smoothing processing; Fluctuation parameters and abnormal amplitudes are calculated based on attitude segments after data smoothing. The quantified value of the cumulative pressure is determined based on the fluctuation parameters and the abnormal amplitude.

4. The method as described in claim 1, characterized in that, The method of determining the risk level of a user's physiological health based on the quantified value of the accumulated pressure includes: Analyze the rate of change of the quantified value of the accumulated pressure over time. Periodic features are extracted based on the posture data sequence; Based on the rate of change of the pattern, combined with the distribution characteristics of the abnormal posture and the periodic characteristics, the risk level of the user's physiological health is determined by a preset risk level judgment rule.

5. The method as described in claim 3, characterized in that, The trend indicators generated based on the risk level to reflect the cumulative stress pattern include: Based on the risk level, historical pressure distribution and load cycle data related to biomechanical mechanisms are obtained from a preset database; By integrating and analyzing the historical pressure distribution and the load cycle data using data sequence fusion technology, the trend index reflecting the pressure accumulation pattern is obtained.

6. The method as described in claim 5, characterized in that, The step of weighting and adjusting the distribution characteristics of the abnormal postures based on the trend indicators, and determining the triggering conditions for risk warnings based on the adjusted abnormal posture distribution characteristics includes: Based on the load cycle data, the characteristics of abnormal load locations are obtained. The distribution characteristics of the abnormal posture are weighted and adjusted by combining the trend indicators with the characteristics of the historical pressure distribution and the abnormal load points. The triggering conditions for the risk warning are determined based on the distribution characteristics of the abnormal postures after weighted adjustment.

7. The method as described in claim 6, characterized in that, If the triggering condition for the risk warning is met, the step of generating a health monitoring feedback sequence based on the triggering condition and the trend indicator reflecting the cumulative stress pattern includes: If the triggering conditions for the risk warning are met, a sequential list of intervention measures for physiological structures is generated based on the triggering conditions and the trend indicators reflecting the cumulative stress patterns. Based on the proposed list of intervention measures, combined with the load cycle data and the trend indicators reflecting the cumulative pressure pattern, a personalized posture adjustment optimization path is obtained. The health monitoring feedback sequence is generated based on the personalized posture adjustment optimization path.

8. The method as described in claim 7, characterized in that, The distribution characteristics of the abnormal posture include the duration interval of the abnormal posture, and the step of generating the health monitoring feedback sequence based on the personalized posture adjustment optimization path includes: Based on the personalized posture adjustment optimization path, the health monitoring feedback sequence is generated by integrating the fluctuation parameters and the duration range of the abnormal posture.

9. The method as described in claim 7, characterized in that, The trend indicators reflecting the cumulative stress pattern include long-term cumulative effects and the slope of stress changes. The generation of a sequential list of intervention recommendations targeting physiological structures based on the triggering conditions and the trend indicators reflecting the cumulative stress pattern includes: Based on the triggering conditions and the long-term cumulative effects, a sequential list of intervention recommendations targeting physiological structures is generated; The process of obtaining a personalized posture adjustment optimization path based on the sequenced suggestion list of intervention measures, combined with the load cycle data and the trend indicators reflecting the cumulative stress pattern, includes: Based on the proposed list of intervention measures, combined with the load cycle data and the pressure change slope, the personalized attitude adjustment optimization path is obtained.

10. A physical health monitoring system based on big data, characterized in that, The system includes: The feature acquisition module is used to acquire posture data sequences related to physiological postures in the user's daily activities, and extract distribution features of abnormal postures based on the posture data sequences. The quantization value determination module is used to determine the quantization value of the cumulative pressure based on the distribution characteristics of the abnormal posture; The risk level judgment module is used to determine the risk level of the user's physiological health based on the quantitative value of the accumulated pressure if the quantitative value of the accumulated pressure exceeds the preset quantitative threshold of the accumulated pressure, and to generate a trend indicator reflecting the pattern of pressure accumulation based on the risk level. The adjustment module is used to perform weighted adjustment on the distribution characteristics of the abnormal posture based on the trend index, and to determine the triggering conditions for risk warning based on the adjusted abnormal posture distribution characteristics. The feedback sequence generation module is used to generate a health monitoring feedback sequence based on the triggering conditions and the trend indicators reflecting the cumulative stress pattern if the risk warning triggering conditions are met.

Images