Chronic disease risk management system based on big data analysis

Abstract

The invention discloses a chronic disease risk management system based on big data analysis, and particularly relates to the technical field of chronic disease risk management. Hardware level information of intelligent wearable equipment, sensor quality information and equipment wearing behavior information of a user are acquired; according to the hardware level differentiation coefficient, the sensor quality decline index and the user equipment wearing behavior coefficient, an error influence evaluation model is constructed, an error influence evaluation index is generated, and potential error risks of low-cost equipment caused by low hardware level, sensor quality decline and user wearing behavior nonlinear influence factors are intelligently identified. The method combines a personalized hardware level differentiation coefficient, a sensor quality decline index and a user equipment wearing behavior coefficient, dynamically adjusts the compensation intensity and method, breaks through the limitation of a traditional linear or simple compensation strategy, flexibly optimizes the compensation strategy, guarantees the data quality of a low-cost equipment user, and expands the applicable population of a chronic disease management system.

Description

Technical Field

[0001] The present invention relates to the technical field of chronic disease risk management, and more specifically, to a chronic disease risk management system based on big data analysis. Background Art

[0002] The chronic disease risk management system based on big data analysis is a system that uses big data technology to conduct early identification, risk assessment, early warning and intervention of chronic diseases (such as hypertension, diabetes, cardiovascular disease, etc.). The system integrates the patient's health data, including physiological indicators, living habits, genetic information, etc., to build a comprehensive risk assessment model, thereby achieving accurate prediction and management of chronic disease risks; remote monitoring and diagnosis have become one of the important directions of chronic disease management. Through remote smart wearable devices, the system can obtain the physical status data of the monitor anytime and anywhere, thereby achieving the management of chronic disease risks. However, compared with professional medical equipment, the measurement error of smart wearable devices is relatively high. The existing chronic disease risk management system based on big data analysis usually adopts a unified data self-compensation strategy for data with measurement errors. Its unified data self-compensation strategy usually assumes that the device error is linear or simple mode, while the actual device error may be nonlinear, change over time or be affected by external factors. Such a simplified model cannot effectively deal with complex errors, especially when different patients purchase low-cost smart wearable devices to reduce consumption expenditures. The hardware level of low-cost smart wearable devices, differences in sensor quality, and data errors caused by different patients' wearing habits affect the existing chronic disease risk management system. The data compensation strategy cannot be adjusted in time according to its impact, resulting in measurement errors not being effectively compensated, which may lead to false alarms (false warnings) or missed reports (actual risks not detected). The system may underestimate or overestimate the risk of chronic diseases, affecting the effect of early identification and intervention. Incorrect risk assessment may cause patients to lose trust in the system, thereby reducing the frequency of use and compliance, and weakening the effect of chronic disease management. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a chronic disease risk management system based on big data analysis to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A chronic disease risk management system based on big data analysis, including a hardware differentiation module, a sensor quality module, a wear behavior monitoring module, an error impact assessment module, and a compensation intervention module;

[0006] The hardware differentiation module is used to obtain the hardware level information of the smart wearable device and obtain the hardware level differentiation coefficient according to the hardware level information;

[0007] The sensor quality module is used to obtain the sensor quality information of the smart wearable device and obtain the sensor quality degradation index based on the sensor quality information;

[0008] The wearing behavior monitoring module is used to obtain the user's device wearing behavior information and obtain the user's device wearing behavior coefficient based on the device wearing behavior information;

[0009] The error impact assessment module is used to build an error impact assessment model and generate an error impact assessment index, which is used to evaluate the impact of the hardware level of smart wearable devices, sensor quality, and user's device wearing behavior on data errors;

[0010] The compensation intervention module is used to obtain the transmission data of the smart wearable device and optimize the data self-compensation strategy based on the hardware level of the smart wearable device, the quality of the sensor and the influence of the user's device wearing behavior on the data error.

[0011] In a preferred embodiment, by acquiring hardware level information of the smart wearable device, analyzing the hardware level of the smart wearable device, and obtaining a hardware level differentiation coefficient, the influence of the hardware level of the smart wearable device on the data error is measured;

[0012] The logic for obtaining the hardware level differentiation coefficient is as follows:

[0013] Step A1, within a preset time period T, obtain the hardware performance data of the smart wearable device at different times, take the initial time, middle time, and end time of the time period T as the initial clustering center, and construct an initial clustering cluster;

[0014] Step A2: for each moment of hardware performance data, calculate the Euclidean distance between it and each cluster center, and assign the moment to the cluster corresponding to the cluster center with the smallest Euclidean distance;

[0015] Step A3, calculating the average value of the hardware performance data at all times in each cluster, and taking the average value of the hardware performance data as the latest cluster center;

[0016] Step A4, repeating steps A2 and A3 until the cluster center no longer changes, and obtaining the final cluster;

[0017] Step A5, calculate the time difference between any two moments in the final cluster, compare the time difference between the two moments with the preset time difference threshold, if the time difference between the two moments is greater than the time difference threshold, calculate the hardware level differentiation coefficient Yjcy according to the time difference greater than the time difference threshold, the expression is as follows Among them skc ijIndicates the jth time difference within the ith final cluster that is greater than the time difference threshold.

[0018] In a preferred embodiment, by acquiring sensor quality information of the smart wearable device, analyzing the performance degradation degree of the sensor of the smart wearable device during actual use, and obtaining a sensor quality degradation index, the performance degradation degree of the sensor of the smart wearable device during actual use is measured;

[0019] The logic for obtaining the sensor quality degradation index is as follows:

[0020] Acquire sensor quality information data at different times within a preset time period T, wherein the sensor quality information data includes but is not limited to signal output strength, response time, battery voltage, and signal-to-noise ratio;

[0021] Calculate the probability distribution of each sensor quality information data, the expression is as follows Where P tn represents the probability distribution of the nth sensor quality information data at time t, x tn represents the specific value of the nth sensor quality information data obtained at the tth time, t={1,2,...,T}, T is a positive integer;

[0022] Calculate the entropy value of each sensor quality information data, the expression is as follows Among them SZ n Represents the entropy value of the nth sensor quality information data;

[0023] Calculate the impact factor of each sensor quality information data, the expression is as follows Among them, YX n represents the influencing factor of the nth sensor quality information data, n = {1, 2, 3, 4};

[0024] Calculate the sensor quality degradation index Cgqz, the expression is as follows

[0025] In a preferred embodiment, by acquiring the user's device wearing behavior information, analyzing the abnormal wearing behavior of the user's smart wearable device, and acquiring the user's device wearing behavior coefficient, the abnormal degree of the user's smart wearable device wearing behavior is measured;

[0026] The logic for obtaining the user device wearing behavior coefficient is as follows:

[0027] Acquire device wearing behavior information data within a preset time period T, wherein the device wearing behavior information data includes wearing position deviation, wearing tightness, and wearing time;

[0028] The accelerometer and gyroscope sensors are used to measure the coordinates of the user's wearing position each time, and the wearing position deviation is calculated. The expression is as follows Where wez represents the wearing position deviation, (Xs, Ys, Zs) represents the measured actual wearing position coordinates, and (Xq, Yq, Zq) represents the expected wearing position coordinates;

[0029] The contact area between the smart wearable device and the skin is measured by the pressure sensor, and the wearing tightness is calculated. The expression is as follows Where pdj represents the wearing tightness, M1 represents the actual pressure measured, and M2 represents the contact area;

[0030] Record the start and end time of each wearing of the device by calculating the difference between the start and end time of each wearing of the device and the wearing time.

[0031] The user device wearing behavior coefficient Pdxw is calculated based on the wearing position deviation, wearing tightness, and wearing time. The expression is as follows Among them wez g represents the wearing position deviation calculated by the g-th user wearing behavior, pdj g pds represents the wearing tightness calculated by the g-th user wearing behavior g represents the wearing time calculated for the g-th user wearing behavior, g = {1, 2, ..., G}, where G is a positive integer.

[0032] In a preferred embodiment, an error impact assessment model is constructed according to the hardware level differentiation coefficient, the sensor quality degradation index, and the user device wearing behavior coefficient to generate an error impact assessment index wcpg. The model is based on the following formula: wcpg=ln(w1*Yjcy+1)+sin(w2*Cgqz+w3*Pdxw), where w1, w2, and w3 represent preset proportional coefficients of the hardware level differentiation coefficient, the sensor quality degradation index, and the user device wearing behavior coefficient, respectively, and w1, w2, and w3 are all greater than 0.

[0033] In a preferred embodiment, the error impact assessment index is compared with a preset error impact assessment index threshold to evaluate the impact of the hardware level of the smart wearable device, the sensor quality, and the user's device wearing behavior on the data error, as follows:

[0034] If the error impact assessment index is greater than the error impact assessment index threshold, a first-level error impact signal is generated;

[0035] If the error impact assessment index is less than or equal to the error impact assessment index threshold, a secondary error impact signal is generated.

[0036] In a preferred embodiment, when a primary error impact signal is generated, the error impact evaluation index generated during the time period T is obtained, and the error impact evaluation index is subtracted from the error impact evaluation index threshold to obtain the error impact deviation. The transmission data of the smart wearable device within the time period T is compensated and corrected according to the error impact deviation. The compensation correction function is as follows: Where Dcorrect represents the transmission data after compensation correction, Doriginal represents the transmission data before compensation correction, Δwcpg represents the error influence deviation, wcpg yz represents the preset error impact assessment index threshold, and α represents the correction intensity factor.

[0037] Technical effects and advantages of the present invention:

[0038] 1. The present invention obtains the hardware level information, sensor quality information and user's device wearing behavior information of the smart wearable device, constructs an error impact assessment model according to the hardware level differentiation coefficient, sensor quality degradation index and user device wearing behavior coefficient, generates an error impact assessment index, and intelligently identifies the potential error risks caused by low hardware level, sensor quality degradation and nonlinear factors of user wearing behavior of low-cost devices. Combined with personalized hardware level differentiation coefficient, sensor quality degradation index and user device wearing behavior coefficient, the compensation intensity and method are dynamically adjusted to break through the limitations of traditional linear or simple compensation strategies, flexibly optimize compensation strategies, avoid false alarms or missed reports caused by error accumulation, improve the accuracy of risk assessment, ensure the data quality of low-cost device users, expand the applicable population of chronic disease management system, and improve the social inclusiveness of the system. Accurate error compensation strategies make chronic disease risk assessment models more credible, reduce the probability of false warnings or missed reports, and ensure the scientific nature of early identification and intervention of chronic diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0040] Figure 1 Flowchart of a system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] Embodiment: The present invention provides as Figure 1A chronic disease risk management system based on big data analysis is shown, including a hardware differentiation module, a sensor quality module, a wear behavior monitoring module, an error impact assessment module, and a compensation intervention module;

[0043] The hardware differentiation module is used to obtain the hardware level information of the smart wearable device and obtain the hardware level differentiation coefficient according to the hardware level information;

[0044] The sensor quality module is used to obtain the sensor quality information of the smart wearable device and obtain the sensor quality degradation index based on the sensor quality information;

[0045] The wearing behavior monitoring module is used to obtain the user's device wearing behavior information and obtain the user's device wearing behavior coefficient based on the device wearing behavior information;

[0046] The error impact assessment module is used to build an error impact assessment model and generate an error impact assessment index, which is used to evaluate the impact of the hardware level of smart wearable devices, sensor quality, and user's device wearing behavior on data errors;

[0047] The compensation intervention module is used to obtain the transmission data of the smart wearable device and optimize the data self-compensation strategy based on the hardware level of the smart wearable device, the quality of the sensor, and the influence of the user's device wearing behavior on the data error;

[0048] The hardware differentiation module is used to obtain the hardware level information of the smart wearable device and obtain the hardware level differentiation coefficient according to the hardware level information;

[0049] The hardware level differentiation coefficient is used to measure the degree of influence of the hardware performance of low-cost smart wearable devices on data quality, and is used to quantify the differences in the hardware of low-cost smart wearable devices. By calculating the hardware level differentiation coefficient, the level of low-cost smart wearable devices can be effectively identified, providing a clear basis for the optimization of data compensation strategies, so that the chronic disease risk management system can dynamically adjust the compensation method according to the actual performance of the device hardware, better adapt to the performance limitations of low-cost devices, and improve the accuracy of data analysis, the reliability of the system, and the user's trust and compliance. A larger hardware level differentiation coefficient indicates that the worse the hardware level of the low-cost smart wearable device, the deeper the impact on the data error. Conversely, a smaller hardware level differentiation coefficient indicates that the hardware level of the low-cost smart wearable device is less different from the hardware performance of the normal-level smart wearable device, and the impact on the data error is limited;

[0050] Therefore, by obtaining the hardware level information of the smart wearable device, analyzing the hardware level of the smart wearable device, and obtaining the hardware level differentiation coefficient, the influence of the hardware level of the smart wearable device on the data error is measured;

[0051] The logic for obtaining the hardware level differentiation coefficient is as follows:

[0052] Step A1, within a preset time period T, obtain the hardware performance data of the smart wearable device at different times, take the initial time, middle time, and end time of the time period T as the initial clustering center, and construct an initial clustering cluster;

[0053] It should be noted that hardware performance data includes but is not limited to processor clock frequency, number of cores, RAM size, storage speed, battery capacity, Bluetooth connection rate, and data transmission speed;

[0054] Step A2: for each moment of hardware performance data, calculate the Euclidean distance between it and each cluster center, and assign the moment to the cluster corresponding to the cluster center with the smallest Euclidean distance;

[0055] Step A3, calculating the average value of the hardware performance data at all times in each cluster, and taking the average value of the hardware performance data as the latest cluster center;

[0056] Step A4, repeating steps A2 and A3 until the cluster center no longer changes, and obtaining the final cluster;

[0057] Step A5, calculate the time difference between any two moments in the final cluster, compare the time difference between the two moments with the preset time difference threshold, if the time difference between the two moments is greater than the time difference threshold, calculate the hardware level differentiation coefficient Yjcy according to the time difference greater than the time difference threshold, the expression is as follows Among them skc ij Indicates the jth time difference within the ith final cluster that is greater than the time difference threshold;

[0058] The sensor quality module is used to obtain the sensor quality information of the smart wearable device and obtain the sensor quality degradation index based on the sensor quality information;

[0059] The sensor quality degradation index is used to measure the degree of performance degradation of the sensors of smart wearable devices during actual use. The sensor quality degradation index is calculated to quantify the key performance changes of the sensor from the initial state to the current state, evaluate the impact of sensor quality degradation on data accuracy and system stability, and provide support for data compensation and system optimization. In the chronic disease risk management system, the sensors of smart wearable devices are mainly used to collect key physiological data of the wearer in real time, providing a basis for chronic disease risk assessment. A larger sensor quality degradation index indicates that the sensor is severely degraded and has a significant impact on data quality. The deeper the impact on data errors, the more likely it is that the uncertainty of system analysis results will increase, leading to erroneous health risk assessment and intervention. A smaller sensor quality degradation index indicates that the sensor still maintains a high performance, has a limited impact on data errors, and has a high credibility of the measured data.

[0060] Therefore, by obtaining the sensor quality information of the smart wearable device, analyzing the performance degradation degree of the sensor of the smart wearable device during actual use, and obtaining the sensor quality degradation index, the performance degradation degree of the sensor of the smart wearable device during actual use is measured;

[0061] The logic for obtaining the sensor quality degradation index is as follows:

[0062] Acquire sensor quality information data at different times within a preset time period T, wherein the sensor quality information data includes but is not limited to signal output strength, response time, battery voltage, and signal-to-noise ratio;

[0063] Calculate the probability distribution of each sensor quality information data, the expression is as follows Where P tn represents the probability distribution of the nth sensor quality information data at time t, x tn represents the specific value of the nth sensor quality information data obtained at the tth time, t={1,2,...,T}, T is a positive integer;

[0064] Calculate the entropy value of each sensor quality information data, the expression is as follows Among them SZ n Represents the entropy value of the nth sensor quality information data;

[0065] Calculate the impact factor of each sensor quality information data, the expression is as follows Among them, YX n represents the influencing factor of the nth sensor quality information data, n = {1, 2, 3, 4};

[0066] Calculate the sensor quality degradation index Cgqz, the expression is as follows

[0067] The wearing behavior monitoring module is used to obtain the user's device wearing behavior information and obtain the user's device wearing behavior coefficient based on the device wearing behavior information;

[0068] The user device wearing behavior coefficient is a key indicator used to measure the degree of influence of the user's behavior of wearing smart wearable devices on the accuracy of device data measurement. The user device wearing behavior coefficient is calculated based on the user's device wearing behavior information (such as wearing position, contact tightness, wearing time, etc.), reflecting the comprehensive influence of the device wearing state on sensor performance and data collection quality. By calculating the user device wearing behavior coefficient, the influence of the user's device wearing behavior on data error can be effectively evaluated, and potential risks to data quality can be identified in time. A larger user device wearing behavior coefficient indicates that the user's wearing behavior is improper, which may increase the influence on data error. A smaller user device wearing behavior coefficient indicates that factors such as wearing position, tightness, and wearing time are all good, and the influence on data error is small.

[0069] Therefore, by obtaining the user's device wearing behavior information, analyzing the abnormal wearing behavior of the user's smart wearable device, and obtaining the user's device wearing behavior coefficient, the abnormal degree of the user's smart wearable device wearing behavior is measured;

[0070] The logic for obtaining the user device wearing behavior coefficient is as follows:

[0071] Acquire device wearing behavior information data within a preset time period T, wherein the device wearing behavior information data includes wearing position deviation, wearing tightness, and wearing time;

[0072] The accelerometer and gyroscope sensors are used to measure the coordinates of the user's wearing position each time, and the wearing position deviation is calculated. The expression is as follows Where wez represents the wearing position deviation, (Xs, Ys, Zs) represents the measured actual wearing position coordinates, and (Xq, Yq, Zq) represents the expected wearing position coordinates;

[0073] The contact area between the smart wearable device and the skin is measured by the pressure sensor, and the wearing tightness is calculated. The expression is as follows Where pdj represents the wearing tightness, M1 represents the actual pressure measured, and M2 represents the contact area;

[0074] Record the start and end time of each wearing of the device by the user and calculate the difference to get the wearing time;

[0075] The user device wearing behavior coefficient Pdxw is calculated based on the wearing position deviation, wearing tightness, and wearing time. The expression is as follows Among them wezg represents the wearing position deviation calculated by the g-th user wearing behavior, pdj g pds represents the wearing tightness calculated by the g-th user wearing behavior g represents the wearing time calculated by the g-th user wearing behavior, g = {1, 2, ..., G}, G is a positive integer;

[0076] The error impact assessment module is used to build an error impact assessment model and generate an error impact assessment index, which is used to evaluate the impact of the hardware level of smart wearable devices, sensor quality, and user's device wearing behavior on data errors;

[0077] According to the hardware level differentiation coefficient, sensor quality degradation index, and user device wearing behavior coefficient, an error impact assessment model is constructed to generate an error impact assessment index wcpg. The model is based on the following formula wcpg=ln(w1*Yjcy+1)+sin(w2*Cgqz+w3*Pdxw), where w1, w2, and w3 represent the preset proportional coefficients of the hardware level differentiation coefficient, the sensor quality degradation index, and the user device wearing behavior coefficient, respectively, and w1, w2, and w3 are all greater than 0;

[0078] It should be noted that before constructing the error impact assessment model, it is necessary to ensure that the hardware level differentiation coefficient, sensor quality degradation index, and user device wearing behavior coefficient are all normalized. Common normalization methods include Min-Max normalization and Z-Score standardization. w1, w2, and w3 are set according to actual conditions. For example, the expert empowerment method is adopted, that is, experts in related fields are invited to determine the preset proportional coefficients of various indicators through professional opinion surveys and comprehensive evaluations.

[0079] It can be seen from the above calculation expression that the larger the hardware level differentiation coefficient, the larger the sensor quality degradation index, and the larger the user device wearing behavior coefficient, the larger the error impact assessment index, indicating that the hardware level of the smart wearable device, the sensor quality, and the user's device wearing behavior have a greater impact on the data error. Conversely, the smaller the hardware level differentiation coefficient, the smaller the sensor quality degradation index, and the smaller the user device wearing behavior coefficient, the smaller the error impact assessment index, indicating that the hardware level of the smart wearable device, the sensor quality, and the user's device wearing behavior have a smaller impact on the data error.

[0080] The error impact assessment index is compared with the preset error impact assessment index threshold to evaluate the impact of the hardware level of the smart wearable device, the sensor quality, and the user's device wearing behavior on the data error, as follows:

[0081] If the error impact assessment index is greater than the error impact assessment index threshold, it indicates that the hardware level of the smart wearable device, the sensor quality, and the user's device wearing behavior have a great impact on the data error. The existing data self-compensation strategy may not be able to fully offset the error impact, generating a first-level error impact signal;

[0082] If the error impact assessment index is less than or equal to the error impact assessment index threshold, it indicates that the hardware level of the smart wearable device, the sensor quality and the user's device wearing behavior have little impact on the data error. The existing data self-compensation strategy can complete the compensation of the data error and generate a secondary error impact signal.

[0083] The compensation intervention module is used to obtain the transmission data of the smart wearable device and optimize the data self-compensation strategy based on the hardware level of the smart wearable device, the quality of the sensor, and the influence of the user's device wearing behavior on the data error;

[0084] When the first-level error impact signal is generated, the error impact evaluation index generated in the time period T is obtained, and the difference between it and the error impact evaluation index threshold is calculated to obtain the error impact deviation. According to the error impact deviation, the transmission data of the smart wearable device in the time period T is compensated and corrected. The compensation correction function is as follows Where Dcorrect represents the transmission data after compensation correction, Doriginal represents the transmission data before compensation correction, Δwcpg represents the error influence deviation, wcpg yz represents the preset error impact assessment index threshold, and α represents the correction intensity factor;

[0085] It should be noted that the correction intensity factor is set by technicians in this field according to the actual needs of the device type and usage scenario, and will not be described in detail here;

[0086] The present invention obtains the hardware level information, sensor quality information and device wearing behavior information of the smart wearable device, constructs an error impact assessment model according to the hardware level differentiation coefficient, the sensor quality degradation index and the user device wearing behavior coefficient, generates an error impact assessment index, and intelligently identifies the potential error risks of low-cost devices caused by low hardware level, sensor quality degradation and nonlinear influencing factors of user wearing behavior. Combined with the personalized hardware level differentiation coefficient, sensor quality degradation index and user device wearing behavior coefficient, the compensation intensity and method are dynamically adjusted to break through the limitations of traditional linear or simple compensation strategies, flexibly optimize the compensation strategy, avoid false alarms or missed alarms caused by error accumulation, improve the accuracy of risk assessment, ensure the data quality of low-cost device users, expand the applicable population of the chronic disease management system, and improve the social inclusiveness of the system. The accurate error compensation strategy makes the chronic disease risk assessment model more credible, reduces the probability of false warnings or missed alarms, and ensures the scientific nature of early identification and intervention of chronic diseases.

[0087] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0088] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0089] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0090] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0091] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A chronic disease risk management system based on big data analysis, characterized by: It includes hardware differentiation module, sensor quality module, wear behavior monitoring module, error impact assessment module, and compensation intervention module; The hardware differentiation module is used to obtain the hardware level information of the smart wearable device and obtain the hardware level differentiation coefficient according to the hardware level information; The sensor quality module is used to obtain the sensor quality information of the smart wearable device and obtain the sensor quality degradation index based on the sensor quality information; The wearing behavior monitoring module is used to obtain the user's device wearing behavior information and obtain the user's device wearing behavior coefficient based on the device wearing behavior information; The error impact assessment module is used to build an error impact assessment model and generate an error impact assessment index, which is used to evaluate the impact of the hardware level of smart wearable devices, sensor quality, and user's device wearing behavior on data errors; The compensation intervention module is used to obtain the transmission data of the smart wearable device and optimize the data self-compensation strategy based on the hardware level of the smart wearable device, the quality of the sensor and the influence of the user's device wearing behavior on the data error.

2. According to claim 1, a chronic disease risk management system based on big data analysis is characterized by: By obtaining the hardware level information of the smart wearable device, analyzing the hardware level of the smart wearable device, and obtaining the hardware level differentiation coefficient, the influence of the hardware level of the smart wearable device on the data error is measured; The logic for obtaining the hardware level differentiation coefficient is as follows: Step A1, within a preset time period T, obtain the hardware performance data of the smart wearable device at different times, take the initial time, middle time, and end time of the time period T as the initial clustering center, and construct an initial clustering cluster; Step A2: for each moment of hardware performance data, calculate the Euclidean distance between it and each cluster center, and assign the moment to the cluster corresponding to the cluster center with the smallest Euclidean distance; Step A3, calculating the average value of the hardware performance data at all times in each cluster, and taking the average value of the hardware performance data as the latest cluster center; Step A4, repeating steps A2 and A3 until the cluster center no longer changes, and obtaining the final cluster; Step A5, calculate the time difference between any two moments in the final cluster, compare the time difference between the two moments with the preset time difference threshold, if the time difference between the two moments is greater than the time difference threshold, calculate the hardware level differentiation coefficient Yjcy according to the time difference greater than the time difference threshold, the expression is as follows Among them skc ij Indicates the jth time difference within the ith final cluster that is greater than the time difference threshold.

3. According to claim 1, a chronic disease risk management system based on big data analysis is characterized by: By obtaining sensor quality information of smart wearable devices, analyzing the performance degradation of sensors of smart wearable devices during actual use, and obtaining a sensor quality degradation index, the performance degradation of sensors of smart wearable devices during actual use is measured; The logic for obtaining the sensor quality degradation index is as follows: Acquire sensor quality information data at different times within a preset time period T, wherein the sensor quality information data includes but is not limited to signal output strength, response time, battery voltage, and signal-to-noise ratio; Calculate the probability distribution of each sensor quality information data, the expression is as follows Where P tn represents the probability distribution of the nth sensor quality information data at time t, x tn represents the specific value of the nth sensor quality information data obtained at the tth time, t={1,2,...,T}, T is a positive integer; Calculate the entropy value of each sensor quality information data, the expression is as follows Among them SZ n Represents the entropy value of the nth sensor quality information data; Calculate the impact factor of each sensor quality information data, the expression is as follows Among them, YX n represents the influencing factor of the nth sensor quality information data, n = {1, 2, 3, 4}; Calculate the sensor quality degradation index Cgqz, the expression is as follows 4. According to claim 1, a chronic disease risk management system based on big data analysis is characterized by: By obtaining the user's device wearing behavior information, analyzing the abnormal wearing behavior of the user's smart wearable device, and obtaining the user's device wearing behavior coefficient, the abnormal degree of the user's smart wearable device wearing behavior is measured; The logic for obtaining the user device wearing behavior coefficient is as follows: Acquire device wearing behavior information data within a preset time period T, wherein the device wearing behavior information data includes wearing position deviation, wearing tightness, and wearing time; The accelerometer and gyroscope sensors are used to measure the coordinates of the user's wearing position each time, and the wearing position deviation is calculated. The expression is as follows Where wez represents the wearing position deviation, (Xs, Ys, Zs) represents the measured actual wearing position coordinates, and (Xq, Yq, Zq) represents the expected wearing position coordinates; The contact area between the smart wearable device and the skin is measured by the pressure sensor, and the wearing tightness is calculated. The expression is as follows Where pdj represents the wearing tightness, M1 represents the actual pressure measured, and M2 represents the contact area; Record the start and end time of each wearing of the device by calculating the difference between the start and end time of each wearing of the device and the wearing time. The user device wearing behavior coefficient Pdxw is calculated based on the wearing position deviation, wearing tightness, and wearing time. The expression is as follows Among them wez g represents the wearing position deviation calculated by the g-th user wearing behavior, pdj g pds represents the wearing tightness calculated by the g-th user wearing behavior g represents the wearing time calculated for the g-th user wearing behavior, g = {1, 2, ..., G}, where G is a positive integer.

5. According to claim 1, a chronic disease risk management system based on big data analysis is characterized in that: An error impact assessment model is constructed based on the hardware level differentiation coefficient, the sensor quality degradation index, and the user device wearing behavior coefficient to generate the error impact assessment index wcpg. The model is based on the following formula: wcpg=ln(w1*Yjcy+1)+sin(w2*Cgqz+w3*Pdxw), where w1, w2, and w3 represent the preset proportional coefficients of the hardware level differentiation coefficient, the sensor quality degradation index, and the user device wearing behavior coefficient, respectively, and w1, w2, and w3 are all greater than 0.

6. According to claim 5, a chronic disease risk management system based on big data analysis is characterized in that: The error impact assessment index is compared with the preset error impact assessment index threshold to evaluate the impact of the hardware level of the smart wearable device, the sensor quality, and the user's device wearing behavior on the data error, as follows: If the error impact assessment index is greater than the error impact assessment index threshold, a first-level error impact signal is generated; If the error impact assessment index is less than or equal to the error impact assessment index threshold, a secondary error impact signal is generated.

7. The chronic disease risk management system based on big data analysis according to claim 6 is characterized by: When the first-level error impact signal is generated, the error impact evaluation index generated in the time period T is obtained, and the difference between it and the error impact evaluation index threshold is calculated to obtain the error impact deviation. According to the error impact deviation, the transmission data of the smart wearable device in the time period T is compensated and corrected. The compensation correction function is as follows Where Dcorrect represents the transmission data after compensation correction, Doriginal represents the transmission data before compensation correction, Δwcpg represents the error influence deviation, wcpg yz represents the preset error impact assessment index threshold, and α represents the correction intensity factor.

Images