Cross-device health data fusion method

Abstract

The present invention relates to the field of health data processing technology, specifically to a cross-device health data fusion method, including the following steps: monitoring device connection status, identifying data loss and reissuing it, calibrating multi-device time and performing data interpolation, calculating device stability score to adjust confidence fusion data, extracting periodic health parameters and analyzing change trends, detecting factor activation status and combining sensitivity to derive risk scores, and outputting health processing results. In the present invention, by combining device connection status identification with data cache progress calculation, the accuracy of breakpoint data recovery is enhanced, timestamp alignment and interpolation correction mechanism are adopted to achieve continuity compensation of multi-source data, and stability score is used to adjust data weights, thereby improving the credibility of fused data. By combining behavioral feature trend extraction with risk factor sensitivity determination, the accuracy of chronic disease risk identification and individual differentiation capabilities are optimized, and the ability of personalized health management and disease prevention is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of health data processing, and in particular to a cross-device health data fusion method. Background Art

[0002] The field of health data processing technology includes methods and technologies for dynamic collection, standardized integration and systematic management of multi-dimensional health information generated by individuals in their daily lives and medical management. The core content of this technology field is to build a comprehensive health data system through real-time monitoring and processing of physiological parameters, behavioral characteristics and environmental data, and enhance the ability to identify and manage health risks. The overall technical system covers cross-device data collection standards, heterogeneous data format conversion, time series synchronization, key health indicator extraction and data security management mechanisms, supports continuous tracking of health status and chronic disease management, and is widely used in disease early warning, health intervention and telemedicine services, playing an important role in chronic disease prevention and control and personalized health management.

[0003] Among them, the cross-device health data fusion method refers to the multi-source heterogeneous health data obtained from smart wearable devices, mobile health terminals and auxiliary monitoring devices, relying on the device identification protocol to complete the data source confirmation, and through the unified data format conversion rules, the heart rate, blood pressure, blood sugar, blood oxygen and exercise behavior and other physiological and behavioral parameters output by different devices are standardized, and the time alignment mechanism is used to achieve the temporal consistency of cross-device data. Combined with the preset data integration process, chronic disease-related risk factors and stroke warning indicators are multi-dimensionally associated, and the fused health data is stored in a centralized management platform to realize the implementation and management of stroke risk monitoring, medication management, rehabilitation guidance and various health intervention measures.

[0004] Traditional cross-device health data fusion technology lacks efficient data loss compensation mechanisms and time offset calibration measures when processing multi-source data, resulting in time series inconsistencies and data breakpoints during the data integration process, limiting the depth and breadth of data analysis, and causing delays or misjudgments in health risk warnings. During data fusion, it lacks the ability to dynamically adjust to differences in signal quality of individual devices, limiting data accuracy and the personalization of health management. When faced with a diverse and large-scale user group, health intervention measures cannot be accurately matched to users' actual needs, affecting health management effectiveness and user satisfaction. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a cross-device health data fusion method.

[0006] In order to achieve the above objectives, the present invention adopts the following technical solution: a cross-device health data fusion method, comprising the following steps:

[0007] S1: Monitor the device connection status parameters, detect the device reconnection request, call the data capture progress parameters and cache pointer position parameters in the device data cache, calculate the data missing interval quantity based on the current data sequence number, and send a data retransmission request to generate the breakpoint resumable transmission compensation value;

[0008] S2: Calling the breakpoint resumable transmission compensation value, obtaining the collected data sequences of multiple devices, comparing the local timestamp of each device with the reference timestamp, calculating the time offset and performing calibration, identifying missing values in the data sequence and performing data interpolation, and generating an interpolation compensation result;

[0009] S3: Calling the interpolation compensation result, extracting the number of signal interruptions, the number of abnormal signal fluctuations, and the sampling interval offset parameters of each device, calculating the stability score of the device, adjusting the data confidence weight corresponding to each device, and fusing similar data to obtain a fused data set;

[0010] S4: Call the fused data set, extract walking volume parameters, sitting time parameters, sleep time parameters and heart rate fluctuation parameters, calculate the change rate characteristics of each parameter in multiple cycles, and analyze the time series change trend of the rate characteristics to generate a health data feature set.

[0011] As a further solution of the present invention, the breakpoint resumption compensation value is specifically the data missing interval amount, the reissue request instruction, and the data capture progress positioning information. The interpolation compensation result includes the time offset calibration value, the data sequence missing segment identifier, and the interpolation generated data point. The fusion data set is specifically the weighted fusion parameter value, the device confidence mapping table, and the unified time series data structure. The health data feature set includes the periodic change rate matrix, the trend change direction feature, and the continuity fluctuation range value.

[0012] As a further solution of the present invention, the step of obtaining the breakpoint resume compensation value is specifically as follows:

[0013] S111: Monitor device connection status parameters, identify the device connection status in real time, including connection and disconnection, detect device reconnection requests in real time, obtain a reconnection request identifier and a current device connection status code, and generate a device reconnection status identifier value;

[0014] S112: calling the device reconnection status identification value, extracting the data capture progress parameter and the cache pointer position parameter in the device data cache, calculating the offset between the data capture progress parameter and the cache pointer position parameter, and generating a cache offset value;

[0015] S113: Based on the cache offset value, call the current data sequence number and the last valid data sequence number, calculate the difference between the current data sequence number and the last valid data sequence number, obtain the data missing interval, send a data retransmission request, and generate a breakpoint resume compensation value.

[0016] As a further solution of the present invention, the step of obtaining the interpolation compensation result is specifically as follows:

[0017] S211: Calling the breakpoint resume compensation value, obtaining the acquisition data sequences of multiple devices, extracting the device number and timestamp corresponding to each device, and generating a device acquisition sequence set value;

[0018] S212: Based on the device acquisition sequence set value, calculate the difference between the local timestamp of each device and the reference timestamp to obtain a time offset and correct the acquisition data sequence of each device to generate a time calibration sequence value;

[0019] S213: calling the time calibration sequence value, identifying missing values in the continuous data time axis, extracting adjacent data points of the missing values, calculating the values of the interpolation points, performing data interpolation on the collected data sequence, and generating an interpolation compensation result.

[0020] As a further solution of the present invention, the steps of obtaining the fused data set are specifically as follows:

[0021] S311: Calling the interpolation compensation result, extracting the signal interruption number parameter, signal abnormal fluctuation number parameter, and sampling interval offset parameter of each device, and generating a basic parameter set for device stability;

[0022] S312: Extract the number of signal interruptions and abnormal fluctuations for each device in each data segment based on the device stability basic parameter set, obtain the sampling interval offset and signal duration parameters, and call the device data segment number parameter to calculate the device stability score;

[0023] S313: Call the device stability score value, adjust the corresponding data confidence weight according to the score value of each device, perform data fusion on similar data, and obtain a fused data set.

[0024] As a further solution of the present invention, the steps of obtaining the health data feature set are specifically as follows:

[0025] S411: Calling the fused data set, extracting walking volume parameters, sitting time parameters, sleeping time parameters, and heart rate fluctuation parameters within multiple time periods, and generating a periodic health parameter set;

[0026] S412: Based on the set of periodic health parameters, extract the numerical changes and periodic time points of each type of health parameter in continuous periods, construct a period number sequence and the total number of periods, calculate the average change rate of multiple health parameters in the period sequence, and obtain a period rate characteristic sequence value;

[0027] S413: Call the periodic rate characteristic sequence value, arrange the rate characteristic values of each health parameter in periodic order, extract the time trend change pattern and determine the trend range, and obtain the health data feature set.

[0028] As a further embodiment of the present invention, the method further comprises:

[0029] S5: Calling the health data feature set to obtain the periodic change rate characteristic values and time series fluctuation characteristic values of multiple parameters, comparing them with the preset chronic disease risk factor characteristic thresholds, identifying the activation status of the risk factors, combining the sensitivity parameters of multiple risk factors, calculating the user's chronic disease risk score, and obtaining the health data processing results;

[0030] The health data processing results specifically refer to chronic disease risk score values, activated risk factor types, and sensitivity impact parameter combinations.

[0031] As a further solution of the present invention, the steps for obtaining the health data processing results are specifically as follows:

[0032] S511: Calling the health data feature set, extracting the periodic change rate characteristic values and time series fluctuation characteristic values of multiple health parameters, and comparing them with the preset chronic disease risk factor characteristic thresholds, identifying and marking the activation status of each risk factor, and obtaining the activation status comparison results;

[0033] S512: Based on the activation status comparison result, by identifying the activation status of the risk factor of each health parameter, extracting the sensitivity parameter of each activated risk factor, evaluating the activation intensity of each parameter, and generating activation intensity data;

[0034] S513: Call the activation intensity data, calculate the risk score of each activation factor, obtain the user's chronic disease risk score, and output the health data processing result.

[0035] Compared with the prior art, the advantages and positive effects of the present invention are:

[0036] In the present invention, by combining device connection status identification and data cache progress calculation, the accuracy of breakpoint data recovery is enhanced, timestamp alignment and interpolation correction mechanism are adopted to achieve continuity compensation of multi-source data, and stability score is used to adjust data weight to improve the credibility of fused data. Combined with behavioral feature trend extraction and risk factor sensitivity determination, the accuracy of chronic disease risk identification and individual differentiation ability are optimized, thereby enhancing the ability of personalized health management and disease prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the main steps of the present invention;

[0038] Figure 2 A flow chart for obtaining the compensation value for breakpoint resume transmission according to the present invention;

[0039] Figure 3 A flow chart for obtaining interpolation compensation results of the present invention;

[0040] Figure 4 A flowchart for obtaining a fused data set according to the present invention;

[0041] Figure 5 A flow chart for obtaining a health data feature set of the present invention;

[0042] Figure 6 The flowchart for obtaining the health data processing results of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0044] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0045] See also Figure 1 The present invention provides a technical solution: a cross-device health data fusion method, comprising the following steps:

[0046] S1: Monitor the device connection status parameters, detect the device reconnection request, call the data capture progress parameters and cache pointer position parameters in the device data cache, calculate the data missing interval quantity based on the current data sequence number, and send a data retransmission request to generate the breakpoint resumable transmission compensation value;

[0047] S2: Call the breakpoint resumable transmission compensation value to obtain the collected data series of multiple devices. By comparing the local timestamp of each device with the reference timestamp, the time offset is calculated and calibrated. Missing values in the data series are identified and data interpolation is performed to generate the interpolation compensation result.

[0048] S3: Call the interpolation compensation result, extract the signal interruption times, signal abnormal fluctuation times, and sampling interval offset parameters of each device, calculate the stability score of the device, adjust the data confidence weight corresponding to each device, and fuse similar data to obtain a fused data set;

[0049] S4: Call the fusion data set to extract walking volume parameters, sitting time parameters, sleep time parameters, and heart rate fluctuation parameters, calculate the change rate characteristics of each parameter in multiple cycles, and analyze the time series change trend of the rate characteristics to generate a health data feature set;

[0050] S5: Call the health data feature set to obtain the periodic change rate characteristic values and time series fluctuation characteristic values of multiple parameters, compare them with the preset chronic disease risk factor characteristic thresholds, identify the activation status of the risk factors, combine the sensitivity parameters of multiple risk factors, calculate the user's chronic disease risk score, and obtain the health data processing results.

[0051] The compensation value for breakpoint resumption specifically includes the data missing interval, reissue request instruction, and data capture progress positioning information. The interpolation compensation result includes the time offset calibration value, the missing segment identifier of the data sequence, and the interpolation generated data point. The fusion data set specifically includes the weighted fusion parameter value, the device confidence mapping table, and the unified time series data structure. The health data feature set includes the periodic change rate matrix, the trend change direction feature, and the continuous fluctuation range value. The health data processing result specifically refers to the chronic disease risk score value, the activation risk factor type, and the sensitivity influence parameter combination.

[0052] See also Figure 2 The specific steps for obtaining the compensation value for resuming breakpoint transmission are as follows:

[0053] S111: Monitor device connection status parameters, identify the device connection status in real time, including connection and disconnection, detect device reconnection requests in real time, obtain a reconnection request identifier and a current device connection status code, and generate a device reconnection status identifier value;

[0054] Each monitoring device connected to the network is numbered, for example, device A, device B, and device C are numbered 1, 2, and 3 respectively. When the system background detects that a new device is connected or a disconnection event occurs, the connection status detection module is called to read the real-time identification bit information of the device status register to obtain the current connection status code of the device. The status code is 0 for disconnected, 1 for connected, 2 for stable connection, and 3 for disconnected. If the device re-reports a heartbeat packet within 30 seconds, the system recognizes it as a "device reconnection request" and the device unique identifier ( The system log records the device's network connection information (e.g., MAC address or device number), reconnection timestamp, and reconnection status flag (for example, a change from 0 to 1). The disconnection duration is calculated by taking the time difference between the reconnection timestamp and the disconnection timestamp. In this example, device A is disconnected at 09:15:03 on April 24, 2025, and reconnected at 09:15:17. The disconnection duration is 14 seconds, which is greater than the system's minimum fault tolerance interval of 5 seconds. This is considered a valid disconnection event, and the generated reconnection status flag value is "R1-14," where "R1" represents the reconnection event number 1 and "14" represents the disconnection duration in seconds.

[0055] S112: calling the device reconnection status identification value, extracting the data capture progress parameter and the cache pointer position parameter in the device data cache, calculating the offset between the data capture progress parameter and the cache pointer position parameter, and generating a cache offset value;

[0056] Read the data capture progress pointer position and the current cache write pointer position stored in the circular array structure in the device cache. The device collects 1 record per second, and the cache space retains a maximum of 1800 records (i.e. 30 minutes of data). For example, in a reconnection event, the cache write pointer position is , the data crawl progress pointer is , then the calculated offset is ,in, is the cache offset, Write the pointer position for the current cache, The data capture progress pointer position indicates that the device has cached 26 uncaptured data records during the disconnection period. The system uses the offset value The index buffer area is read one by one from the 998th to the 1024th item. Each data is saved in a fixed structure, including timestamp, heart rate value, step number, and status flag. During the reading process, the system marks the currently processed cache segment and updates the crawl progress pointer, and finally generates a cache offset value of 26.

[0057] S113: Based on the cache offset value, the current data sequence number and the last valid data sequence number are called, the difference between the current data sequence number and the last valid data sequence number is calculated, the data missing interval is obtained, and a data retransmission request is sent to generate a breakpoint resumable transmission compensation value;

[0058] After the device is disconnected and restored, the system compares its current data sequence number The last valid data sequence number , which are 14276 and 14247 respectively, so the amount of missing data is ,in, is the data missing interval, The data sequence number reported by the current device. is the valid data sequence number captured last time. If the cache offset is 26, it means that there are still 3 data not stored in the cache, that is, 3 data are lost during the actual disconnection. The system will send a retransmission request to the device based on the cache offset, with the request segment sequence number interval [14248,14273] and the data type mask. After receiving the request, the device retrieves the local cache and retransmits the data in the interval. Finally, the receiving end merges the original data sequence and the retransmitted data to form a complete data stream, generating a breakpoint resumable transmission compensation value of 29, and the compensation rate is ,where the numerator 26 is the number of records successfully reissued, and the denominator 29 is the theoretical total number of missing records, indicating that the system successfully recovered most of the data. Table 1 lists the device cache status and compensation result examples.

[0059] Table 1 Device cache offset and reissue compensation table

[0060]

[0061] As shown in Table 1, the compensation request content and actual success rate are calculated by the offset and sequence number difference.

[0062] See also Figure 3 , the steps for obtaining the interpolation compensation results are as follows:

[0063] S211: Calling the breakpoint resumable transmission compensation value, obtaining the acquisition data sequence of multiple devices, extracting the device number and timestamp corresponding to each device, and generating a device acquisition sequence set value;

[0064] Call the breakpoint resume compensation value, and the system will call the previously recorded breakpoint resume compensation value array in sequence , locate the device number list corresponding to each compensation interval through the internal index mechanism, where the device numbers are D1, D2 and D3 respectively. The system verifies the legitimacy of each device through the device identity authentication module and obtains its original collected data within the compensation period in the form of time periods. The collected data field structure is {timestamp, heart rate, number of steps, signal flag}. For each data, the system extracts its timestamp value and pairs it with the device number to which the data belongs, and constructs multiple groups of time series data sets indexed by the device number. The time series data is automatically sorted from small to large by timestamp to form a set value. For example, the D1 device obtains a total of 30 records from 09:15:00 to 09:15:30, and the timestamp interval is 1 second. The set form is ,The system finally completes the construction of the set structure and generates the ,device acquisition sequence set value.

[0065] S212: Based on the device collection sequence set value, calculate the difference between the local timestamp of each device and the reference timestamp to obtain the time offset and correct the collection data sequence of each device to generate a time calibration sequence value;

[0066] Based on the device collection sequence set value, the system extracts the local timestamp for each data item in the set And the benchmark timestamp set uniformly with the server Perform difference operation, the calculation method is ,in, is the time offset value, Report timestamp for the device. The reference timestamp corresponding to the current receiving time of the server. If the device D1 reports the time as 09:15:03 and the server receives the time as 09:15:00, the offset is 3 seconds. The system performs the same offset correction on all time series data of the D1 device, that is, all timestamps are uniformly subtracted by 3 seconds to correct the offset, thereby completing the standardization operation of the collected time series. The system records the corrected time series structure as follows: , complete the data time standardization processing of all devices and generate time calibration sequence values.

[0067] S213: Call the time calibration series value, identify the missing values in the continuous data time axis, and extract the adjacent data points of the missing values using the formula:

[0068] ;

[0069] Calculate the value of the interpolation point, perform data interpolation on the collected data sequence, and generate the interpolation compensation result;

[0070] in, is the value corresponding to the interpolation point, is the value of the previous valid data point in the missing interval, is the value of the next valid data point after the missing interval, is the timestamp of the previous valid data point in the missing interval, is the timestamp of the next valid data point after the missing interval, is the timestamp of the current interpolation point, is the fine-tuning factor value;

[0071] Call the time calibration sequence value, the system scans all standardized time series and marks the gaps in the continuous time points where the interval exceeds the preset sampling interval threshold (such as 1 second). If it is found that the interval between the timestamps 09:15:11 and 09:15:13 of the D2 device is 2 seconds, it is confirmed that there is a missing point 09:15:12 in the middle. The system locates the valid data points before and after the missing point. and , and define the interpolation points as , the system sets the fine-tuning factor value to , substitute the following formula:

[0072] ;

[0073] ;

[0074] The final interpolated value is 79.4. The system inserts this value into the original time series to construct continuous interpolation compensation data. The same process is repeated for other devices, compensating for missing data segments to form a complete series and generating the interpolation compensation result. Table 2 lists the actual data and results of the interpolation calculations for multiple devices. The formula automatically repairs missing data in the time series due to disconnection or sampling failure. First, the values provided by the preceding and following valid points are used to determine the reasonable position of the current gap in the state evolution process. For example, if the heart rate jumps from 78 to 80, and the missing point is located at the midpoint between the two, the system tends to interpret this point as a transition state in the linear trend. The interpolation process simulates this linear transition process. Furthermore, in reality, the sampling process of devices is affected by environmental factors or fluctuations in device status, which may cause the sampling trend to be slightly asymmetric or fluctuate. Therefore, a fine-tuning factor is introduced to offset the trend based on the amplitude of the jump. This allows the interpolation result to be dynamically adjusted without disrupting the preceding and following trends, enhancing the smoothness and stability of the overall data series.

[0075] Table 2 Interpolation calculation results

[0076]

[0077] As shown in Table 2, the results show that the values of the missing data points are restored after interpolation calculation and can be used in the subsequent fusion and fluctuation analysis steps.

[0078] See also Figure 4, the specific steps for obtaining the fusion dataset are:

[0079] S311: Call the interpolation compensation result, extract the signal interruption number parameter, signal abnormal fluctuation number parameter, and sampling interval offset parameter of each device, and generate a basic parameter set for device stability;

[0080] Call the interpolation compensation result, the system reads the continuous sampling sequence of each device after the interpolation is completed, and performs a signal analysis process on each record. First, the signal connection status flag of each data segment is identified. If the flag changes from "valid" to "disconnected" and then to "valid", it is counted as a signal interruption in the segment. The number of interruptions in all data segments is counted as the signal interruption number parameter , then analyze the amplitude change of each sampling data segment. If the difference between any two consecutive sampling points in the same segment is greater than the device set threshold , it is recorded as an abnormal signal fluctuation, and the cumulative number of abnormalities is used as the abnormal fluctuation number parameter , again extract the theoretical sampling interval from the sampling time series Seconds, compare the actual time interval of each pair of adjacent sampling points and calculate the mean square error, the result is recorded as the sampling interval offset Finally, the length of time for each continuous sampling is counted in seconds and recorded as the signal duration , combine the above parameters and add the number of data segments divided by the device in this round of collection , and finally build a complete basic parameter set of equipment stability.

[0081] S312: Based on the device stability basic parameter set, extract the number of signal interruptions and abnormal fluctuations of each device in each data segment, obtain the sampling interval offset and signal duration parameters, and call the device data segment number parameter using the formula:

[0082] ;

[0083] Calculate the device stability score;

[0084] in, For the The stability score of each device, For the Device No. Number of signal interruptions in the segment, For the Device No. The number of abnormal signal fluctuations in the segment, For the The sampling interval offset of each device, For the The signal duration of each device, For the The number of continuous data segments of each device, is the sampling frequency, is the normalized benchmark value of abnormal intensity, is the normalized reference value of the timing disturbance, Indicates the total number of data segments corresponding to each device. Indicates the device number index, Indicates the device data segment number index;

[0085] According to the basic parameter set of device stability, the system traverses each device number , extract the number of signal interruptions of each segment in turn and the number of abnormal signal fluctuations , multiply the corresponding elements and sum them to form the term , if the number of interruptions recorded by device D1 in the three segments of data is , the number of abnormal fluctuations is , then the sum of products is , assuming the normalized baseline value of anomaly intensity , then the normalized term is , continue to extract the sampling interval offset of device D1 With signal duration , sampling frequency Hz, and calculate its disturbance modulus as , assuming the disturbance normalized reference value , then the normalized value is The absolute value of the difference between the two is , the number of device data segments is , substitute into the formula:

[0086] ;

[0087] ;

[0088] The device stability score is a dimensionless value calculated based on four indicators: signal interruption frequency, abnormal fluctuation level, sampling time stability, and data continuity structure. It represents the operational stability and data reliability of a device during the acquisition process. Higher scores indicate frequent interruptions, frequent fluctuations, or unstable sampling times, indicating poor stability. Lower scores indicate stable operation, complete data, and good time synchronization. This score serves as the basis for adjusting confidence weights in the subsequent data fusion phase, directly influencing the fusion ratio and credibility assessment of data from different devices. It is an important basis for controlling data weight allocation and selecting high-quality fusion data. The formula aims to quantitatively characterize device data quality from multiple dimensions, enabling identification of device behavioral stability. This score is then projected into a standardized score, facilitating cross-device ranking and subsequent weighting. The design uses product multiplication to reflect coupling strength, square root processing time impact, normalization criteria to introduce dimensionless control, absolute value to prevent directional errors, and continuous segment correction to enhance score credibility. The results of practical applications for other devices are shown in Table 3, ultimately resulting in a device stability score.

[0089] Table 3 Equipment stability score calculation table

[0090]

[0091] As shown in Table 3, different devices have different stability scores according to their abnormality levels and time stability differences.

[0092] S313: Calling the device stability score value, adjusting the corresponding data confidence weight according to the score value of each device, performing data fusion on similar data, and obtaining a fused data set;

[0093] Call the device stability score value, the system sets the initial confidence level of each device data to 1.0, and reads its corresponding score value , set the confidence attenuation threshold table according to the scoring level, when The corresponding confidence weight is 0.9. When set to 0.8, When 0.7, When the score is 0.6, the corresponding score of device D1 is 0.4255, which is between 0.4 and 0.5. Then its confidence is adjusted to 0.7. The system traverses all devices and redistributes the weighted weights in data fusion. The data of each device is weighted averaged according to the adjusted weights based on each type of data (such as heart rate and number of steps). Finally, a single data value stream at the same time point is synthesized to obtain a fused data set.

[0094] See also Figure 5 ,The steps for obtaining the health data feature set are as follows:

[0095] S411: Calling the fusion data set, extracting walking volume parameters, sitting time parameters, sleep time parameters, and heart rate fluctuation parameters within multiple time periods, and generating a periodic health parameter set;

[0096] The fused data set is called, and the system traverses all data points in the fused data, dividing the data into segments by day or hour, with each segment defined as a complete cycle. , extract four types of health data dimensions in each period, including walking volume, sitting time, sleep time and heart rate fluctuation. Each dimension extracts the corresponding field from the original record. For example, the walking volume is counted according to the cumulative step field of the device. The sitting time is marked as "stationary state" based on the posture sensor or action recognition module and the time is summed. The sleep time is recorded based on the low movement at night + the lower limit of heart rate fluctuation. The heart rate fluctuation is extracted by the difference between the maximum and minimum heart rate in this period. The system records the periodic data structure of each type of parameter in turn, forming a data structure like (walking volume), (sleep duration) and other periodic parameter values to establish a periodic health parameter set of multiple health parameters.

[0097] S412: Based on the periodic health parameter set, extract the value changes and period time points of each type of health parameter in continuous periods, construct a period number sequence and the total number of periods, using the formula:

[0098] ;

[0099] Calculate the average change rate of multiple health parameters in the periodic sequence to obtain the periodic rate characteristic sequence value;

[0100] in, Indicates the The average change rate characteristic value of the class health parameter, Indicates in In the cycle The value of the class health parameter, Indicates the value of the parameter in the next adjacent cycle. Indicates the The time point or sequence number of each cycle, Indicates the time points or sequence numbers of adjacent cycles, Indicates the total number of cycles, Represents the cycle number index, Indicates the health parameter type number;

[0101] Based on the periodic health parameter set, the system performs health parameter Construct its sequence in all periods , and define the serial time points for all cycle numbers , taking walking volume as an example, the three cycle values are , the cycle number is , then perform the following calculation:

[0102] ;

[0103] Substituting the walking amount parameters into the equation:

[0104] ;

[0105] The corresponding sleep duration data is , calculated as:

[0106] ;

[0107] Heart rate fluctuations , calculated as:

[0108] ;

[0109] The final result, the average rate of change characteristic value, is a rate of change indicator formed by the numerical changes and time differences between the adjacent cycles of various health parameters in a continuous cycle. After cycle-level averaging, the output is a numerical feature representing the activity of each type of health behavior change. The higher the value, the more frequent and rapid the fluctuations of the health behavior throughout the cycle sequence, reflecting a high degree of behavioral instability or dynamism; the lower the value, the more stable and stable the behavior. The parameter is used as an important quantitative basis for further identifying individual health status trends, constructing behavioral stability profiles, and screening high-frequency abnormal behaviors. The formula is calculated continuously between cycles, systematically revealing the frequency and speed of changes in behavioral patterns such as walking volume, sitting time, sleep time, and heart rate fluctuations. The absolute value is used to eliminate directional interference, making it more suitable for measuring stability trends, monitoring abnormal activities, or establishing behavioral fluctuation thresholds. By averaging all cycles, it characterizes the full-cycle change activity characteristics of each type of health behavior, facilitating subsequent feature comparison, trend extraction, or behavior clustering.

[0110] Table 4 Health parameter period rate characteristic value table

[0111]

[0112] As shown in Table 4, different health parameters have different fluctuation amplitudes within the cycle, and the rate characteristic values intuitively reflect their change intensity and stability.

[0113] S413: Calling the periodic rate characteristic sequence value, arranging the rate characteristic values of each health parameter in periodic order, extracting its time trend change pattern and determining the trend range, and obtaining the health data feature set;

[0114] Call the period rate characteristic sequence value, the system will The values are arranged in sequence, and a rate characteristic trend trajectory diagram is constructed according to the periodic sequence number. The change values at each time point are connected to form a trend line, and the direction of change of the trend line is judged. If three consecutive periodic rate values form a monotonically increasing sequence, the trend is marked as an "increasing interval", if it is decreasing, it is a "decreasing interval", and if the fluctuation amplitude is below 0.1, it is recorded as a "stable interval". The upper and lower fluctuation interval boundaries are set respectively. The system divides the trend status of each type of parameter trend sequence into sections. For example, the walking volume trend line is 300-300, which is stable; the sleep duration trend is 0.2, rising to 0.5 and then falling back to 0.35, marked as oscillation; the heart rate fluctuation trend is 1.7, falling to 0.4, which is an obvious downward trend. Finally, the trend label of each type of parameter and the original value are integrated to form a health data feature set.

[0115] See also Figure 6 , the specific steps for obtaining health data processing results are:

[0116] S511: Calling the health data feature set, extracting the periodic change rate characteristic values and time series fluctuation characteristic values of multiple health parameters, and comparing them with the preset chronic disease risk factor characteristic thresholds, identifying and marking the activation status of each risk factor, and obtaining the activation status comparison results;

[0117] Call the health data feature set. The system presets multiple chronic disease risk factors. Each risk factor corresponds to a multi-parameter associated feature set. The system extracts the change rate characteristics of each type of parameter in multiple cycles from the health data feature set. and fluctuation range , and compare them item by item with the characteristic conditions in the risk factor definition to determine the activation status. For example, for the risk factor "metabolic decline risk", its definition characteristics are: 、 、 , when the system detects that the corresponding parameters of the current user are 、 、 If any two of the three indicator activation conditions are met, the system will determine that the risk factor "metabolic decline risk" is activated. The system will perform row-by-row judgment on multiple risk factors in accordance with the risk factor-parameter matching matrix, and finally generate a comparison result of the risk factor activation status.

[0118] S512: Based on the activation status comparison result, by identifying the activation status of the risk factor of each health parameter, extracting the sensitivity parameter of each activated risk factor, evaluating the activation intensity of each parameter, and generating activation intensity data;

[0119] Based on the activation status comparison results, the system traverses all risk factors marked as "activated" and extracts and defines the sensitivity coefficients of their associated parameters. ,in represents the risk factor number, Indicates the factor on which the For example, the corresponding parameters and sensitivities for "metabolic decline risk" are: walking volume 0.6, sitting time 0.8, and heart rate fluctuation 0.9. The system calculates the percentage of each parameter that exceeds the threshold and multiplies it by its sensitivity coefficient to form the single parameter activation intensity. , such as walking volume: , meditation duration: , heart rate fluctuations: .

[0120] Table 5 Composition of activation intensity of chronic disease risk factors

[0121]

[0122] As shown in Table 5, the system merges the activation strength of all relevant parameters of the current risk factor and generates the risk factor activation strength value using a weighted sum or average strategy. , the final activation intensity of the metabolic decline risk factor is , other risk factors are calculated in sequence, and finally a complete activation intensity matrix is constructed to form an activation intensity dataset.

[0123] S513: Calling the activation intensity data, calculating the risk score of each activation factor, obtaining the user's chronic disease risk score, and outputting the health data processing result;

[0124] Calling the activation intensity data, the system integrates the activation intensity corresponding to each risk factor , generates the user's chronic disease risk score. The score is calculated by summing the strengths of all activated risk factors and normalizing them to the total score upper limit of 1.0. In the current example, the activation strength of "metabolic decline risk" is 0.38, and the activation strength of "sleep quality disorder risk" is 0.28, so the total score is The system divides the scores into segments according to the score level definition. A score <0.3 is "low risk", 0.3-0.6 is "medium risk", and >0.6 is "high risk". The current score belongs to the "high risk" level. The system combines this level with the activation factor name, corresponding strength and other information to form a health risk score structure, and outputs the health data processing results, including: risk score value, classification label, activation factor list, activation parameters of each factor, sensitivity and score weight mapping table, full life cycle label mapping and other dimensional field information to form the final chronic disease risk assessment result.

[0125] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A cross-device health data fusion method, characterized in that: The following steps are involved: S1: Monitor the device connection status parameters, detect the device reconnection request, call the data capture progress parameters and cache pointer position parameters in the device data cache, calculate the data missing interval quantity based on the current data sequence number, and send a data retransmission request to generate the breakpoint resumable transmission compensation value; The compensation value for breakpoint resuming transmission is specifically the data missing interval, the reissue request instruction, and the data capture progress positioning information; The specific steps for obtaining the breakpoint resume compensation value are as follows: S111: Monitor device connection status parameters, identify the device connection status in real time, including connection and disconnection, detect device reconnection requests in real time, obtain a reconnection request identifier and a current device connection status code, and generate a device reconnection status identifier value; S112: calling the device reconnection status identification value, extracting the data capture progress parameter and the cache pointer position parameter in the device data cache, calculating the offset between the data capture progress parameter and the cache pointer position parameter, and generating a cache offset value; S113: Based on the cache offset value, the current data sequence number and the last valid data sequence number are called, the difference between the current data sequence number and the last valid data sequence number is calculated, the data missing interval is obtained, and a data retransmission request is sent to generate a breakpoint resume compensation value; S2: Calling the breakpoint resumable transmission compensation value, obtaining the collected data sequences of multiple devices, comparing the local timestamp of each device with the reference timestamp, calculating the time offset and performing calibration, identifying missing values in the data sequence and performing data interpolation, and generating an interpolation compensation result; S3: Calling the interpolation compensation result, extracting the number of signal interruptions, the number of abnormal signal fluctuations, and the sampling interval offset parameters of each device, calculating the stability score of the device, adjusting the data confidence weight corresponding to each device, and fusing similar data to obtain a fused data set; S4: Call the fused data set, extract walking volume parameters, sitting time parameters, sleep time parameters and heart rate fluctuation parameters, calculate the change rate characteristics of each parameter in multiple cycles, and analyze the time series change trend of the rate characteristics to generate a health data feature set.

2. The cross-device health data fusion method according to claim 1, characterized in that: The interpolation compensation result includes a time offset calibration value, a missing segment identifier of a data sequence, and interpolation-generated data points. The fusion data set specifically includes a weighted fusion parameter value, a device confidence mapping table, and a unified time series data structure. The health data feature set includes a periodic change rate matrix, a trend change direction feature, and a continuity fluctuation range value.

3. The cross-device health data fusion method according to claim 1, characterized in that: The steps for obtaining the interpolation compensation result are specifically as follows: S211: Calling the breakpoint resume compensation value, obtaining the acquisition data sequences of multiple devices, extracting the device number and timestamp corresponding to each device, and generating a device acquisition sequence set value; S212: Based on the device acquisition sequence set value, calculate the difference between the local timestamp of each device and the reference timestamp to obtain a time offset and correct the acquisition data sequence of each device to generate a time calibration sequence value; S213: calling the time calibration sequence value, identifying missing values in the continuous data time axis, extracting adjacent data points of the missing values, calculating the values of the interpolation points, performing data interpolation on the collected data sequence, and generating an interpolation compensation result.

4. The cross-device health data fusion method according to claim 3, characterized in that: The steps for obtaining the fused data set are specifically as follows: S311: Calling the interpolation compensation result, extracting the signal interruption number parameter, signal abnormal fluctuation number parameter, and sampling interval offset parameter of each device, and generating a basic parameter set for device stability; S312: Extract the number of signal interruptions and abnormal fluctuations for each device in each data segment based on the device stability basic parameter set, obtain the sampling interval offset and signal duration parameters, and call the device data segment number parameter to calculate the device stability score; S313: Call the device stability score value, adjust the corresponding data confidence weight according to the score value of each device, perform data fusion on similar data, and obtain a fused data set.

5. The cross-device health data fusion method according to claim 4, characterized in that: The steps for obtaining the health data feature set are specifically as follows: S411: Calling the fused data set, extracting walking volume parameters, sitting time parameters, sleeping time parameters, and heart rate fluctuation parameters within multiple time periods, and generating a periodic health parameter set; S412: Based on the set of periodic health parameters, extract the numerical changes and periodic time points of each type of health parameter in continuous periods, construct a period number sequence and the total number of periods, calculate the average change rate of multiple health parameters in the period sequence, and obtain a period rate characteristic sequence value; S413: Call the periodic rate characteristic sequence value, arrange the rate characteristic values of each health parameter in periodic order, extract the time trend change pattern and determine the trend range, and obtain the health data feature set.

6. The cross-device health data fusion method according to claim 1, characterized in that: The method further comprises: S5: Calling the health data feature set to obtain the periodic change rate characteristic values and time series fluctuation characteristic values of multiple parameters, comparing them with the preset chronic disease risk factor characteristic thresholds, identifying the activation status of the risk factors, combining the sensitivity parameters of multiple risk factors, calculating the user's chronic disease risk score, and obtaining the health data processing results; The health data processing results specifically refer to chronic disease risk score values, activated risk factor types, and sensitivity impact parameter combinations.

7. The cross-device health data fusion method according to claim 6, characterized in that: The steps for obtaining the health data processing results are specifically as follows: S511: Calling the health data feature set, extracting the periodic change rate characteristic values and time series fluctuation characteristic values of multiple health parameters, and comparing them with the preset chronic disease risk factor characteristic thresholds, identifying and marking the activation status of each risk factor, and obtaining the activation status comparison results; S512: Based on the activation status comparison result, by identifying the activation status of the risk factor of each health parameter, extracting the sensitivity parameter of each activated risk factor, evaluating the activation intensity of each parameter, and generating activation intensity data; S513: Call the activation intensity data, calculate the risk score of each activation factor, obtain the user's chronic disease risk score, and output the health data processing result.

Images