A health management device data analysis system and method based on the Internet of Things

Abstract

The application relates to the field of health equipment management, in particular to a health management equipment data analysis system and method based on the Internet of Things, which comprises a motion detection module, a cluster association module, a state recognition module, a heat monitoring module and an intelligent scheduling module; the motion detection module is used for detecting a motion state, collecting resting data and motion data; the cluster association module is used for generating a motion cluster; the state recognition module is used for judging a current motion type of a user; the heat monitoring module is used for estimating real-time heat consumption; and the intelligent scheduling module is used for detecting a recovery state of the user and matching an appropriate motion cluster. The application can perform personalized analysis on health data of a user, reasonably arrange a daily exercise plan, guarantee a physical state of the user, achieve better exercise effect, help the user record and analyze health data, improve detection accuracy of the health data and improve health management efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of health device management, in particular to a health management device data analysis system and method based on Internet of Things. BACKGROUND

[0002] The health management device is a kind of technical device for monitoring and managing personal health status, including smart watch, smart bracelet, sphygmomanometer, blood glucose meter, heart rate monitor and the like, to help users understand their own health status in real time. Commonly used sports health management device can judge the user's exercise state by detecting its own shaking, and present it in the form of steps or calories.

[0003] However, the user's exercise form is various, often not just walking, and the health management device uses steps as the measurement standard, which may not reflect the actual exercise state of the user, so as to give accurate calorie consumption evaluation, and the body states of users are different, so it is difficult to accurately measure user health data with a unified standard, affecting the reliability of data management.

[0004] In addition, the recovery time after exercise is different for different users, in order to avoid abnormal health data caused by excessive exercise, real-time health data is needed as a reference, and the traditional health management device cannot assist users in formulating a reasonable exercise plan according to the actual health data, so as to improve the user's health management experience. SUMMARY

[0005] The present application aims to provide a health management device data analysis system and method based on Internet of Things to solve the problems raised in the background.

[0006] In order to solve the above technical problems, the present application provides the following technical scheme: a health management device data analysis system based on Internet of Things, comprising: a motion detection module, a cluster association module, a state recognition module, a calorie monitoring module and an intelligent scheduling module.

[0007] The motion detection module is used to detect the user's exercise state from the sensor of the health management device, and collect health data every fixed time interval when the user is not in the exercise state as the user's resting data, and when in the exercise state, fit the exercise function of each detection item according to the last collected resting data, and store all exercise functions in the database in the form of linked list until the user stops exercising.

[0008] The cluster association module is used to obtain the correlation of each detection item by operating the correlation coefficient of the exercise function, store a plurality of exercise functions with correlation coefficient higher than the threshold value between each other in a plurality of exercise periods as an exercise function cluster, and mark the detection item corresponding to each exercise function, and generate an exercise cluster and give a name after packing the storage structure.

[0009] The state recognition module is configured to acquire various health data of the user in real time in a current exercise period, generate a detection function of detection data and time for each detection item, form a verification cluster with all the detection functions, calculate a consistency rate between the verification cluster and each exercise cluster, arrange all the exercise clusters according to the consistency rate, and display the exercise cluster at the top on the health device;

[0010] The heat monitoring module is configured to calculate the heat consumption of all the stored exercise clusters according to a MET heat consumption formula, accumulate the heat consumed by the user in real time according to the product of the consistency rate and the heat consumption of the exercise cluster, and display the exercise function of all the detection items and the real-time heat consumption on the health device.

[0011] The intelligent scheduling module is configured to obtain a recovery function of the detection item according to the detection results of the health data after the user finishes exercising, input a predicted exercise duration, and use the health data after the user finishes exercising as a starting coefficient to substitute into each exercise cluster, so that all the health data in the preset duration is lower than a warning value, output the exercise cluster meeting the condition, and if there is no exercise cluster meeting the condition, calculate the exercise cluster corresponding to the shortest recovery duration according to the recovery function of the health data.

[0012] Further, the exercise detection module includes a resting acquisition unit, an exercise acquisition unit and a function storage unit.

[0013] The resting acquisition unit is configured to acquire the health data of the user in a resting state and keep real-time updating in a fixed period.

[0014] The exercise acquisition unit is configured to detect the exercise state of the user according to the exercise amplitude of the user, and determine that the user is in an exercise state when the exercise amplitude in a unit duration is higher than a threshold.

[0015] The function storage unit is configured to acquire the parameter change of each health detection item in real time in an exercise state, and generate an exercise function.

[0016] Further, the cluster association module includes a correlation verification unit and a pattern extraction unit.

[0017] The correlation verification unit is configured to verify the correlation coefficient between each detection item in a historical period, and obtain the correlation of each detection item.

[0018] The pattern extraction unit is configured to cluster the detection items having correlation, and form an exercise cluster.

[0019] Further, the state recognition module includes a pattern verification unit and a consistency comparison unit.

[0020] The mode verification unit is configured to acquire motion functions of each health data in real time, and motion functions of all detection items constitute a verification cluster;

[0021] The consistency comparison unit is configured to compare consistency between the verification cluster and each motion cluster, and arrange each motion cluster according to the consistency ratio.

[0022] Further, the heat monitoring module comprises a heat accumulation unit and a device digital display unit.

[0023] The heat accumulation unit is configured to estimate heat consumption of each motion cluster, and then accumulate the heat consumption according to the consistency ratio of the current verification cluster and the motion cluster.

[0024] The device digital display unit is configured to output the accumulated real-time heat consumption, and display the health data and the heat consumption in the device.

[0025] Further, the intelligent scheduling module comprises a motion recovery unit, a cluster matching unit and a scheduling output unit.

[0026] The motion recovery unit is configured to detect recovery functions of each type of health data of the user after the user finishes the motion.

[0027] The cluster matching unit is configured to input the current health data of the user into each motion cluster, so that all health data are lower than the warning value.

[0028] The scheduling output unit is configured to set the rest duration and the motion cluster according to the recovery function, and output the motion arrangement.

[0029] A health management device data analysis method based on the Internet of Things, comprising the following steps:

[0030] Step S1. Detect the motion state of the user, collect the resting data of the user every fixed time interval when the user is not in the motion state, and use the resting data collected last time as the starting data when the user is in the motion state. For each type of health data, fit the time domain variation function, and store the fitting result as the motion function;

[0031] Step S2. Perform correlation analysis on each motion function in the historical period, extract the motion functions that are in the continuous motion period and have a correlation coefficient higher than a threshold value, store the function group in a set form, constitute a motion cluster and name it;

[0032] Step S3. Acquire each motion function in the current motion period, constitute a time domain verification cluster, calculate the consistency rate between the time domain verification cluster and each motion cluster, arrange all motion clusters in descending order of the consistency rate, and display the first motion cluster on the device.

[0033] Step S4. Calculate the consumed heat of all stored exercise clusters according to the MET heat consumption model, estimate the heat consumption of each exercise cluster, and accumulate the consumed heat in real time according to the product of the consistency rate and the heat consumption of the exercise cluster, and display the consumed heat in the device;

[0034] Step S5. After the user's exercise is completed, obtain the recovery function of each type of health data according to the detection result of the health data, input the expected exercise duration, and match in the stored exercise cluster to make all health data within the preset duration lower than the warning value. If there is no matching result, perform linear programming with the shortest recovery duration as the constraint condition, and output the programming result.

[0035] Further, step S1 includes:

[0036] Step S11. Place a motion sensor in the health management device. If the swing amplitude of the motion sensor exceeds the preset amplitude within a preset period, it is determined that the user is in an exercise state, and step S13 is performed. Otherwise, it is determined that the user is in a resting state, and step S12 is performed.

[0037] Step S12. In the user's resting state, collect the user's health data as resting data every preset duration. The health data includes heart rate, blood pressure, blood sugar, blood oxygen, respiratory rate, and body movement amplitude.

[0038] Step S13. Use the exponential decay model to fit the change function of each type of health data during exercise using the latest resting data as the starting data:

[0039] ;

[0040] Where Q(t) is the change function of the health data, Qv is the starting health data, k is the exercise coefficient, t is the time, and e is the base of the natural logarithm.

[0041] Step S14. For each type of health data, perform step S13 once to output the change function Q(t), and store the change function of each type of health data as the exercise function.

[0042] Further, step S2 includes:

[0043] Step S21. Perform correlation analysis on the exercise functions of each type of health data to calculate the correlation between the exercise functions:

[0044] ;

[0045] Where W represents the correlation between the exercise functions, T is the preset period, Qr(t) and Qa(t) represent the exercise functions of two types of health data, respectively.

[0046] Step S22. Extract motion functions from the stored data. The extracted motion functions satisfy the first condition: the correlation coefficient between the motion functions is greater than the correlation threshold in k consecutive historical periods, where k is a preset parameter.

[0047] All extracted motion functions are stored as a set to form a motion cluster A, where A = {Q1(t), Q2(t), ..., Q...} n Let n be the number of motion functions in the cluster, Q1(t), Q2(t), ..., Qt. n (t) represents the motion function corresponding to the 1st to nth health data points;

[0048] Step S23. Generate and store all motion clusters that meet the first condition, and name each motion cluster.

[0049] Furthermore, step S3 includes:

[0050] Step S31. Summarize the motion functions collected by the user in the current motion state to obtain the verification cluster B, wherein B = {B1(t), B2(t), ..., B m (t)}, m is the number of health data types, B1(t), B2(t), ..., B m (t) represents the motion function corresponding to the 1st to mth health data points respectively;

[0051] Step S32. Calculate the consistency rate between the verification cluster and each motion cluster using the following formula:

[0052] ;

[0053] Where Y is the consistency rate, MAX is the maximum value function, t1 represents the time at the bottom edge of the sliding window, t2 represents the window duration, t1∈[0,T], and t2 is determined by the user's movement time, Q i (t) and B i (t) represents the health data corresponding to the i-th element in the motion cluster and the validation cluster, respectively, i∈{1,2,…,n};

[0054] Step S33. Sort all motion clusters in descending order of consistency rate, and display the motion cluster with the highest consistency rate on the device.

[0055] Furthermore, step S4 includes:

[0056] Step S41. Input the motion function of each motion cluster into the MET calorie consumption model and calculate the calorie relationship function D(t) between the total calorie consumption of the exercise and time.

[0057] Step S42. Accumulate the user's calorie consumption function U(t), where U(t) = p1·D1(t) + p2·D2(t) + ... p r ·D r (t), where r is the number of motion clusters, p1 to p r D1(t) represents the proportion of the consensus rate of the first to rth motion clusters in the total consensus rate, and D2(t) represents the proportion of the consensus rate of the first to rth motion clusters in the total consensus rate. r (t) represents the calorie relationship function for the first to rth exercise clusters, respectively, and displays the consumed calories in the health management device and keeps it updated in real time.

[0058] Furthermore, step S5 includes:

[0059] Step S51. After the user finishes exercising, re-detect the user's health data at fixed intervals and fit the recovery function of each health data using an exponential decay model;

[0060] Step S52. The user inputs the expected exercise duration, matches the exercise cluster, and ensures that all health data within the exercise duration are below the warning value. If a matching result exists, the matching result is output; otherwise, proceed to step S53.

[0061] Step S53. Perform linear programming to maximize heat consumption, with the shortest recovery time as the constraint:

[0062] ;

[0063] Where Tc represents recovery time, TE represents expected exercise time, MIN is the minimum function, and E(t) is the recovery function of user health data. The linear programming result is output on the device.

[0064] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0065] 1. This invention can collect health data at fixed intervals, calculate the resting coefficient of each health data point, and when user movement is detected, fit a motion function based on the resting coefficient to identify the user's motion clusters, perform personalized analysis of the user's health data, provide customized exercise suggestions, and offer users better health management services.

[0066] 2. This invention can extract motion clusters from historical periods, calculate the consistency rate between the extracted motion clusters and each preset motion cluster function group, and predict exercise calorie consumption based on the consistency rate, thereby helping users record and analyze health data, improving the accuracy of health data detection, and improving the efficiency of health management.

[0067] 3.The application can use each health coefficient after the user's movement as the starting coefficient to replace each movement cluster, make all health coefficients within the preset time length lower than the warning value, and calculate the shortest recovery time length corresponding to the movement cluster, output the movement time arrangement, help the user reasonably arrange the daily exercise plan, ensure the user's physical state, and achieve better movement effect. BRIEF DESCRIPTION OF DRAWINGS

[0068] The accompanying drawings are included to provide a further understanding of the application, and constitute a part of the specification, illustrate the application, and are used to explain the application together with the embodiments of the application, and do not constitute a limitation on the application. In the drawings:

[0069] Figure 1 is a structural schematic diagram of a health management equipment data analysis system based on the Internet of Things according to the application;

[0070] Figure 2 is a step schematic diagram of a health management equipment data analysis method based on the Internet of Things according to the application. DETAILED DESCRIPTION

[0071] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0072] Please refer to Figure 1 The application provides a technical solution: a health management equipment data analysis system based on the Internet of Things, comprising: a movement detection module, a cluster association module, a state recognition module, a heat monitoring module, and an intelligent scheduling module.

[0073] The movement detection module is used to detect the movement state of the user from the sensor of the health management equipment, collect health data every fixed time length as the resting data of the user when the user is not in the movement state, fit the movement function of each detection item according to the resting data collected last time when the user is in the movement state, store all movement functions in the database in the form of a linked list, and stop until the user stops moving.

[0074] The movement detection module comprises a resting collection unit, a movement collection unit, and a function storage unit.

[0075] The resting collection unit is used to collect health data in the resting state of the user and keep real-time update within a fixed period.

[0076] The motion acquisition unit is configured to detect the motion state of the user according to the motion amplitude of the user, and determine that the user is in the motion state when the motion amplitude in a unit time period is higher than a threshold value.

[0077] The function storage unit is configured to acquire the parameter change of each health detection item in real time in the motion state, and generate a motion function.

[0078] The cluster association module is configured to obtain the correlation of each detection item by operation of the correlation coefficient of the motion function, store a plurality of motion functions in which the correlation coefficient between each other in a plurality of continuous motion periods is higher than a threshold value as a motion function cluster, mark the detection item corresponding to each motion function, package the storage structure, generate a motion cluster, and name the motion cluster.

[0079] The cluster association module comprises a correlation verification unit and a mode extraction unit.

[0080] The correlation verification unit is configured to verify the correlation coefficient between each detection item in a historical period, and obtain the correlation of each detection item.

[0081] The mode extraction unit is configured to cluster the detection items having the correlation, and form a motion cluster.

[0082] The state recognition module is configured to acquire the health data of the user in real time in a current motion period, generate a detection function of the detection data and time for each detection item, form a verification cluster by using all the detection functions, calculate the consistency rate between the verification cluster and each motion cluster, arrange all the motion clusters according to the consistency rate, and display the motion cluster arranged at the first position on the health device.

[0083] The state recognition module comprises a mode verification unit and a consistency comparison unit.

[0084] The mode verification unit is configured to acquire the motion function of each health data in real time, and form a verification cluster by using the motion function of all the detection items.

[0085] The consistency comparison unit is configured to compare the consistency between the verification cluster and each motion cluster, and arrange each motion cluster according to the consistency.

[0086] The heat monitoring module is configured to calculate the heat consumption of all the stored motion clusters according to a MET heat consumption formula, accumulate the heat consumed by the user in real time according to the product of the consistency rate and the heat consumption of the motion cluster, and display the motion function of all the detection items and the real-time heat consumption on the health device.

[0087] The heat monitoring module comprises a heat accumulation unit and a device display unit.

[0088] The heat accumulation unit is used to estimate the heat consumption of each exercise cluster, and then accumulate the consistency rate proportion of the current verification cluster and the exercise cluster;

[0089] The device digital display unit is used to output the accumulated real-time heat consumption, and display the health data and the heat consumption in the device.

[0090] The intelligent scheduling module is used to obtain the recovery function of the detection item after the user's exercise ends according to the detection results of each health data, input the expected exercise duration by the user, and use each health data after the user's exercise as the starting coefficient to substitute into each exercise cluster, so that all health data within the preset duration is below the warning value, output the exercise cluster that meets the condition, and if there is no exercise cluster that meets the condition, calculate the exercise cluster corresponding to the shortest recovery duration according to the recovery function of the health data.

[0091] The intelligent scheduling module includes an exercise recovery unit, a cluster matching unit and a scheduling output unit;

[0092] The exercise recovery unit is used to detect the recovery function of each type of health data of the user after the user's exercise ends;

[0093] The cluster matching unit is used to input the current health data of the user into each exercise cluster, so that all health data is below the warning value;

[0094] The scheduling output unit is used to set the rest duration and the exercise cluster according to the recovery function, and output the exercise arrangement.

[0095] As shown in Figure 2 A health management device data analysis method based on the Internet of Things, comprising the following steps:

[0096] Step S1. Detect the exercise state of the user, collect the resting data of the user every fixed duration when the user is not in the exercise state, and use the resting data collected last time as the starting data when the user is in the exercise state. For each type of health data, fit the time domain variation function, and store the fitting result as the exercise function;

[0097] Step S1 includes:

[0098] Step S11. Place a motion sensor in the health management device, and when the swing amplitude of the motion sensor exceeds the preset amplitude within a preset period, it is judged that the user is in the exercise state, and the process goes to step S13, otherwise it is judged that the user is in the resting state, and the process goes to step S12;

[0099] Step S12. In the resting state of the user, collect the health data of the user every fixed duration as resting data, the health data includes: heart rate, blood pressure, blood sugar, blood oxygen, respiratory rate and body movement amplitude;

[0100] Step S13. Using the latest resting data as the starting data, the exponential decay model is used to fit the change function of each type of health data during exercise:

[0101] ;

[0102] Wherein Q(t) is the change function of health data, Qv is the starting health data, k is the exercise coefficient, t is the time, and e is the base of natural logarithm;

[0103] Step S14. For each type of health data, step S13 is performed once, and the change function Q(t) is output. The change functions of each type of health data are stored as exercise functions.

[0104] Step S2. Correlation analysis is performed on the exercise functions in the historical period. The exercise functions in the continuous exercise period and with a correlation coefficient higher than the threshold value are extracted, the function group is stored in a set form, the exercise cluster is constructed and named;

[0105] Step S2 includes:

[0106] Step S21. Correlation analysis is performed on the exercise functions of each type of health data, and the correlation between the exercise functions is calculated:

[0107] ;

[0108] Wherein W represents the correlation between the exercise functions, T is a preset period, and Qr(t) and Qa(t) represent the exercise functions of two types of health data, respectively;

[0109] Step S22. The exercise functions are extracted from the stored data, and the extracted exercise functions satisfy the first condition: the correlation coefficient between the exercise functions is greater than the correlation threshold value in the continuous k historical periods, and k is a preset parameter;

[0110] All extracted exercise functions are stored as a set to form an exercise cluster A, and the A={Q1(t),Q2(t),…,Q n (t)}, n is the number of exercise functions in the cluster, Q1(t), Q2(t), …, Q n (t) represent the exercise functions corresponding to the first to n types of health data, respectively;

[0111] Step S23. All exercise clusters meeting the first condition are generated and stored, and each exercise cluster is named.

[0112] Step S3. In the current exercise period, each exercise function is obtained to form a time domain verification cluster, the consistency rate between the time domain verification cluster and each exercise cluster is calculated, all exercise clusters are arranged in descending order of the consistency rate, and the first exercise cluster is displayed on the device;

[0113] Step S3 includes:

[0114] Step S31. Summarize the motion functions collected by the user in the current motion state to obtain the verification cluster B, wherein B = {B1(t), B2(t), ..., B m (t)}, m is the number of health data types, B1(t), B2(t), ..., B m (t) represents the motion function corresponding to the 1st to mth health data points respectively;

[0115] Step S32. Calculate the consistency rate between the verification cluster and each motion cluster using the following formula:

[0116] ;

[0117] Where Y is the consistency rate, MAX is the maximum value function, t1 represents the time at the bottom edge of the sliding window, t2 represents the window duration, t1∈[0,T], and t2 is determined by the user's movement time, Q i (t) and B i (t) represents the health data corresponding to the i-th element in the motion cluster and the validation cluster, respectively, i∈{1,2,…,n};

[0118] Step S33. Sort all motion clusters in descending order of consistency rate, and display the motion cluster with the highest consistency rate on the device.

[0119] Step S4. Calculate the calorie consumption of all stored exercise clusters according to the MET calorie consumption model, estimate the calorie consumption of each exercise cluster, and accumulate the consumed calorie in real time according to the product of the consistency rate ratio and the calorie consumption of the exercise cluster, and display it in the device.

[0120] Step S4 includes:

[0121] Step S41. Input the motion function of each motion cluster into the MET calorie consumption model and calculate the calorie relationship function D(t) between the total calorie consumption of the exercise and time.

[0122] Step S42. Accumulate the user's calorie consumption function U(t), where U(t) = p1·D1(t) + p2·D2(t) + ... p r ·D r (t), where r is the number of motion clusters, p1 to p r D1(t) represents the proportion of the consensus rate of the first to rth motion clusters in the total consensus rate, and D2(t) represents the proportion of the consensus rate of the first to rth motion clusters in the total consensus rate. r (t) represents the calorie relationship function for the first to rth exercise clusters, respectively, and displays the consumed calories in the health management device and keeps it updated in real time.

[0123] Step S5. After the user's exercise ends, a recovery function of each type of health data is obtained according to a detection result of the health data, a predicted exercise duration is input, and matching is performed in the stored exercise cluster, so that all health data within the preset duration is below an alarm value. If there is no matching result, linear programming is performed with the shortest recovery duration being the shortest as a constraint condition, and a programming result is output.

[0124] Step S5 includes:

[0125] Step S51. After the user's exercise ends, the health data of the user is detected again every fixed duration, and an exponential decay model is used to fit a recovery function of the health data.

[0126] Step S52. The user inputs a predicted exercise duration, and matching is performed in the exercise cluster, so that all health data within the exercise duration is below an alarm value. If there is a matching result, the matching result is output. If there is no matching result, step S53 is performed.

[0127] Step S53. Linear programming with the maximum heat consumption is performed with the shortest recovery duration being the shortest as a constraint condition:

[0128]

[0129] wherein Tc represents a recovery duration, TE represents a predicted exercise duration, MIN is a minimum function, and E(t) is a recovery function of the health data of the user. The linear programming result is output on the device.

[0130] Embodiment: Analysis of the historical record of the user shows that the user has three exercise modes, each of which contains a corresponding exercise function of heart rate and blood pressure. At a certain moment, the user is in an exercise state, and the current exercise function of the user is sent to the exercise mode for matching, and the consistency rates are 0.5, 0.2 and 0.3, respectively. The heat consumptions of the three exercise modes are 100 Kcal, 50 Kcal and 40 Kcal, respectively. Therefore, the heat consumption output of the user is 72 Kcal.

[0131] It should be noted that the relational terms such as first and second and the like are used merely to distinguish one entity or action from another, without necessarily requiring or implying that there is any such actual relationship or order between such entities or actions. In addition, the terms "comprises", "comprising", or any other variant thereof are intended to cover non-exclusive inclusions, so that a process, method, article or apparatus that includes a list of elements does not only include those elements, but also includes other elements not expressly listed or inherent to such process, method, article or apparatus.

[0132] ​Finally, it should be noted that the above only describes the preferred embodiments of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art will appreciate that the technical solutions described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalent ones. 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 data analysis method for health management devices based on the Internet of Things, characterized in that, The method includes the following steps: Step S1. Detect the user's motion state. When the user is not in motion, collect the user's resting data at fixed intervals. When the user is in motion, use the most recently collected resting data as the starting data. For each type of health data, fit a time-domain variation function and store the fitting result as a motion function. Step S2. Perform correlation analysis on each motion function within the historical cycle, extract motion functions that are in continuous motion cycles and whose correlation coefficients are higher than the threshold, store the function groups in the form of sets, form motion clusters and name them. Step S3. Obtain each motion function in the current motion cycle to form a time-domain verification cluster, calculate the consistency rate between the time-domain verification cluster and each motion cluster, sort all motion clusters in descending order of consistency rate, and display the first motion cluster on the device. Step S4. Calculate the calorie consumption of all stored exercise clusters according to the MET calorie consumption model, estimate the calorie consumption of each exercise cluster, and accumulate the consumed calorie in real time according to the product of the consistency rate and the calorie consumption of the exercise cluster, and display it in the device. Step S5. After the user finishes exercising, based on the detection results of health data, obtain the recovery function of various health data, input the expected exercise duration, match it in the stored exercise cluster, so that all health data within the expected exercise duration are lower than the warning value. If there is no matching result, perform linear programming with the shortest recovery duration as the constraint condition, and output the planning result. Step S4 includes: Step S41. Input the motion function of each motion cluster into the MET calorie consumption model and calculate the calorie relationship function D(t) between the total calorie consumption of the exercise and time. Step S42. Accumulate the user's calorie consumption function U(t), where U(t) = p1·D1(t) + p2·D2(t) + ... p r ·D r (t), where r is the number of motion clusters, p1 to p r D1(t) represents the proportion of the consensus rate of the first to rth motion clusters in the total consensus rate, and D2(t) represents the proportion of the consensus rate of the first to rth motion clusters in the total consensus rate. r (t) represents the calorie relationship function of the first to rth exercise clusters, respectively, and displays the consumed calories in the health management device and keeps it updated in real time; Step S5 includes: Step S51. After the user finishes exercising, re-detect the user's health data at fixed intervals and fit the recovery function of each health data using an exponential decay model; Step S52. The user inputs the expected exercise duration, matches the exercise cluster, and ensures that all health data within the exercise duration are below the warning value. If a matching result exists, the matching result is output; otherwise, proceed to step S53. Step S53. Perform linear programming to maximize heat consumption while minimizing recovery time: ； Where Q(t) is the change function of health data, Tc represents the recovery time, TE represents the expected exercise time, MIN is the minimum value function, E(t) is the recovery function of user health data, and B1(t), B2(t), ..., Bm(t) represent the exercise functions corresponding to the 1st to mth health data, respectively. The linear programming result is output on the device.

2. The data analysis method for health management devices based on the Internet of Things according to claim 1, characterized in that: Step S1 includes: Step S11. Insert a motion sensor into the health management device. If the swing amplitude of the motion sensor exceeds the preset amplitude within a preset period, it is determined that the user is in motion and proceeds to step S13. Otherwise, it is determined that the user is in resting state and proceeds to step S12. Step S12. When the user is at rest, collect the user's health data as resting data at preset intervals. The health data includes: heart rate, blood pressure, blood sugar, blood oxygen, respiratory rate and body movement amplitude. Step S13. Using the most recent resting data as the starting data, fit the change function of various health data during exercise using an exponential decay model: ； Where Q(t) is the function of change of health data, Qv is the initial health data, k is the exercise coefficient, t is time, and e is the base of the natural logarithm; Step S14. For each type of health data, execute step S13 once, output the change function Q(t), and store the change functions of each type of health data as motion functions.

3. The data analysis method for health management devices based on the Internet of Things according to claim 2, characterized in that: Step S2 includes: Step S21. Perform correlation analysis on the motion functions of various health data, and calculate the correlation between the motion functions: ； Where W represents the correlation between motion functions, T is the preset period, and Qr(t) and Qa(t) represent the motion functions of the two types of health data, respectively; Step S22. Extract motion functions from the stored data. The extracted motion functions satisfy the first condition: the correlation coefficient between the motion functions is greater than the correlation threshold in k consecutive historical periods, where k is a preset parameter. All extracted motion functions are stored as a set to form a motion cluster A, where A = {Q1(t), Q2(t), ..., Q...} n Let n be the number of motion functions in the cluster, Q1(t), Q2(t), ..., Qt. n (t) represents the motion function corresponding to the 1st to nth health data points; Step S23. Generate and store all motion clusters that meet the first condition, and name each motion cluster.

4. The data analysis method for health management devices based on the Internet of Things according to claim 3, characterized in that: Step S3 includes: Step S31. Summarize the motion functions collected by the user in the current motion state to obtain the verification cluster B, wherein B = {B1(t), B2(t), ..., B m (t)}, m is the number of health data types, B1(t), B2(t), ..., B m (t) represents the motion function corresponding to the 1st to mth health data points respectively; Step S32. Calculate the consistency rate between the verification cluster and each motion cluster using the following formula: ； Where Y is the consistency rate, MAX is the maximum value function, t1 represents the time at the bottom edge of the sliding window, t2 represents the window duration, t1∈[0,T], and t2 is determined by the user's movement time, Q i (t) and B i (t) represents the i-th corresponding motion function in the motion cluster and the verification cluster, respectively, i∈{1,2,…,n}; Step S33. Sort all motion clusters in descending order of consistency rate and display the motion cluster with the highest consistency rate on the device.

5. A data analysis system for health management devices based on the Internet of Things (IoT), wherein the system executes the data analysis method for health management devices based on the IoT as described in claim 1, characterized in that, The system includes the following modules: motion detection module, cluster association module, status recognition module, calorie monitoring module, and intelligent scheduling module; The motion detection module is used to detect the user's motion status from the sensors of the health management device. When the user is not in motion, health data is collected at fixed intervals as the user's resting data. When the user is in motion, motion functions for each detection item are fitted based on the most recently collected resting data, and all motion functions are stored in the database in the form of a linked list until the user stops exercising. The cluster association module is used to obtain the correlation of each detection item by calculating the correlation coefficient of the motion function, store multiple motion functions with a correlation coefficient higher than the threshold in multiple consecutive motion cycles as a motion function cluster, label the detection items corresponding to each motion function, package the storage results, generate a motion cluster and name it. The status recognition module is used to acquire various health data of the user in real time during the current exercise cycle. For each detection item, a detection function of detection data and time is generated. All detection functions constitute a verification cluster. The consistency rate between the verification cluster and each exercise cluster is calculated. All exercise clusters are arranged in descending order of consistency rate. The exercise cluster ranked first is displayed on the health device. The calorie monitoring module is used to calculate the calorie consumption of all stored exercise clusters according to the MET calorie consumption formula, and to accumulate the user's calorie consumption in real time by multiplying the consistency rate ratio by the calorie consumption of the exercise clusters. The module also displays the exercise function and real-time calorie consumption of all detected items in the health device. The intelligent scheduling module is used to obtain the recovery function of the detection items based on the detection results of each health data after the user's exercise ends. The user inputs the expected exercise duration, and the health data after the user's exercise ends are used as the starting coefficients to be substituted into each exercise cluster. This ensures that all health data within the expected exercise duration are lower than the warning value, and outputs the exercise cluster that meets the conditions. If there is no exercise cluster that meets the conditions, the exercise cluster corresponding to the shortest recovery duration is calculated based on the recovery function of the health data.

6. The data analysis system for health management devices based on the Internet of Things according to claim 5, characterized in that: The motion detection module includes: a resting acquisition unit, a motion acquisition unit, and a function storage unit; The resting data acquisition unit is used to collect health data from the user at rest and maintains real-time updates within a fixed period. The motion acquisition unit is used to detect the user's motion state based on the user's motion amplitude. When the motion amplitude is higher than a threshold within a unit of time, it is judged as a motion state. The function storage unit is used to collect parameter changes of various health detection items in real time during motion and generate motion functions.

7. The IoT-based health management device data analysis system according to claim 6, characterized in that: The cluster association module includes: a related verification unit and a pattern extraction unit; The correlation verification unit is used to verify the correlation coefficient between each detection item within a historical period to obtain the correlation of each detection item. The pattern extraction unit is used to cluster related detection items to form a motion cluster. The state recognition module includes: a pattern verification unit and a consistency comparison unit; The pattern verification unit is used to acquire the motion functions of each health data in real time, and the motion functions of all detection items constitute a verification cluster. The consistency comparison unit is used to compare the consistency rate between the verification cluster and each motion cluster, and to arrange each motion cluster according to the consistency rate ratio.

8. The IoT-based health management device data analysis system according to claim 7, characterized in that: The heat monitoring module includes: a heat accumulation unit and a device digital display unit; The heat accumulation unit is used to estimate the heat consumption of each sports cluster, and then accumulate it according to the consistency ratio between the current verification cluster and the sports cluster. The device's digital display unit is used to output the accumulated real-time heat consumption and display the health data and heat consumption on the device.

9. A data analysis system for health management devices based on the Internet of Things according to claim 8, characterized in that: The intelligent scheduling module includes: a motion recovery unit, a cluster matching unit, and a scheduling output unit; The exercise recovery unit is a function used to detect various health data of the user after the user finishes exercising; The cluster matching unit is used to input the user's current health data into each sports cluster, so that all health data are below the warning value. The scheduling output unit is used to set the rest duration and exercise cluster according to the recovery function and output the exercise schedule.

Images