A method and system for predicting health in the elderly

Abstract

The application relates to the technical field of medical care information science, and discloses an old-age health prediction method and system, which comprises an information collection module, a phase-locked analysis module and a trend determination module, the information collection module synchronously acquires a behavior perturbation sequence and a physiological response sequence; the phase-locked analysis module calculates dynamic phase offset characteristics between the behavior perturbation sequence and the physiological response sequence, and determines a logical deviation residual of a measured phase offset relative to a reference mapping track by using a logic transfer check unit; and the trend determination module monitors a second-order cumulative change rate of the logical deviation residual in a sliding window, and outputs a warning signal when the second-order cumulative change rate exceeds a stability threshold value; the application can perceive the deterioration of body feedback regulation flexibility before a physiological parameter appears abnormal, and avoids prediction failure caused by a physiological compensation mechanism of an old individual by quantifying the information transmission efficiency of a feedback loop.

Description

Technical Field

[0001] This invention belongs to the field of healthcare informatics technology, and in particular relates to a method and system for predicting health in the elderly. Background Technology

[0002] Current health prediction systems for the elderly based on physiological parameter monitoring are an important means of ensuring home safety. Existing technologies collect heart rate, blood pressure and blood oxygen indicators, and use preset thresholds or statistical models to assess health status. This monitoring method plays a role in identifying sudden abnormalities and assessing basic health conditions.

[0003] The elderly possess physiological compensatory abilities. In the early stages of functional impairment, the body consumes reserve resources to maintain normal physiological parameter values. Physiological compensation masks the degeneration process of underlying regulatory mechanisms, making physiological indicators appear to have a normal steady-state distribution. When a single physiological parameter exceeds the compensatory boundary and produces numerical deviations, the body has often entered the decompensation stage, and the prediction system degenerates into an alarm tool, losing its predictive function. For example, Chinese invention patent CN119007995A discloses a method and device for predicting elderly health. This method integrates multimodal data from questionnaires, oral reports, and physiological parameters, and uses a weighted fusion of BERT and random forest models for prediction. This technology incorporates natural language processing to enhance prediction. In terms of information acquisition, the core logic for addressing the problem of unclear symptom descriptions in the elderly is based on feature extraction and statistical correlation of multi-source heterogeneous data. However, such static feature mapping models ignore the causal feedback mechanism behind physiological parameters and are unable to peel away the deteriorating regulatory elasticity features hidden under normal fluctuations during the compensatory stage before physiological values ​​deviate. Conventional improvement paths capture weak fluctuations by increasing sensor dimensions or lowering the judgment threshold. Lowering the threshold introduces environmental noise interference and increases the false alarm rate of the system. Relying on model mining of static values ​​cannot peel away the compensatory mechanism's effect of masking pathological features. There is a fundamental conflict between the dynamic response logic of the body to random behavioral stimuli and the numerical stability of the physiological results.

[0004] Therefore, the technical problem to be solved by this invention is to start from the dynamic coupling mechanism of the body in handling random behavioral perturbations, quantify the temporal phase stability between excitation and response, and detect the damage to the transmission efficiency of the feedback loop before the physiological index values ​​shift. Summary of the Invention

[0005] This invention provides an elderly health prediction system, including an information acquisition module, a phase-locked loop analysis module, and a trend determination module: The information acquisition module is used to synchronously acquire the behavioral perturbation sequence of the object to be monitored. and physiological response sequences Behavioral perturbation sequence The spontaneous micro-motion intensity pulse flow extracted by a triaxial accelerometer; The phase-locked loop analysis module integrates a logic transfer verification unit to process behavioral perturbation sequences. Defined as system stimulus input, and the physiological response sequence Defined as the system response output, the behavioral perturbation sequence is calculated using the sliding cross-correlation function. With physiological response sequence Dynamic phase shift characteristics between ; The logic transfer verification unit is used to determine the behavioral perturbation sequence. Instantaneous energy envelope And based on a predefined linear correlation operator characterizing the energy-phase coupling relationship under healthy conditions. Establishing an instantaneous energy envelope With dynamic phase shift characteristics The reference mapping trajectory between the two phases is determined, and the logical deviation residual of the measured phase offset relative to the reference mapping trajectory is determined. Logical deviation residual The calculation rules are as follows: ; The trend determination module is used to monitor logical deviations from residuals. The system calculates the second-order cumulative rate of change within a preset sliding window, and when the second-order cumulative rate of change exceeds a preset stability threshold, it determines that the transmission efficiency of the physiological feedback loop of the monitored object is impaired, and outputs a health decline warning signal.

[0006] Preferably, the system further includes a rhythm reference extraction module; the rhythm reference extraction module is used to extract rhythm references from behavioral perturbation sequences. When the intensity is within a preset resting range, analyze the physiological response sequence. The change in the autocorrelation function envelope is used to determine the endogenous rhythm phase of the current monitoring period. Phase-locked loop analysis module, utilizing intrinsic rhythm phase Logical deviation residuals Perform time-domain logical alignment correction to generate a corrected residual sequence for the trend determination module.

[0007] Preferably, the phase-locked loop analysis module further includes a response hysteresis analysis unit; the response hysteresis analysis unit is used to analyze behavioral perturbation sequences. Determine the dynamic phase shift characteristics when multiple consecutive action pulses are included. The slope of the change in the behavior pulse sequence number serves as the response hysteresis factor. The trend determination module, based on the response hysteresis factor... The degree of deviation from the preset linear range determines the physiological compensation reserve status of the monitored object.

[0008] Preferably, the trend determination module further includes a response phase verification unit; the response phase verification unit is used to perform phase verification on the behavioral perturbation sequence. Determine the physiological response sequence within the response time window after disappearance. The recovery phase-locked feature; the trend determination module, through the recovery phase-locked feature and the dynamic phase shift feature. The temporal symmetry between them allows for a consistent correction of health migration trends.

[0009] Preferably, the system further includes a dynamic zero-point drift hedging module; the dynamic zero-point drift hedging module is used to adjust the behavior perturbation sequence. When the intensity remains below a preset threshold, determine the physiological response sequence. The inherent response delay in a static environment serves as a reference zero-point characteristic; the phase-locked loop analysis module utilizes the reference zero-point characteristic to analyze the real-time dynamic phase shift characteristic. Perform offset compensation.

[0010] Preferably, the system also includes a consistency arbitration module for extracting behavioral perturbation sequences. With physiological response sequence The high-frequency components in the image are analyzed, and the coupling coherence between these components is determined. A trend determination module analyzes the dynamic phase shift characteristics based on the coupling coherence. The validity of the judgment is dynamically weighted.

[0011] Preferably, the response hysteresis analysis unit calculates the response hysteresis factor. The rule is: ,in, The corresponding serial number is The dynamic phase shift characteristics of the behavior pulse. The sequence index value of continuous behavioral pulses within the preset observation period; the trend determination module, in response hysteresis factor When a monotonically increasing trend is observed, it is determined that the body's compensatory reserves have been depleted.

[0012] Preferably, the system further includes an output validity auditing unit; the output validity auditing unit is used to determine the physiological response sequence. The endogenous coupling characteristics between different physiological frequency components within the body; the trend determination module, when the endogenous coupling characteristics deviate from the preset normal range, suppresses the warning output of the trend determination results.

[0013] Preferably, the phase-locked loop analysis module further includes an environmental coupling verification unit; the environmental coupling verification unit is used to detect behavioral perturbation sequences. The intensity is determined by suspending the parsing logic of the phase-locked loop analysis module when it falls below a preset noise threshold. The information acquisition module includes a triaxial accelerometer and a photoplethysmography (PPG) sensor; the PPG sensor is used to acquire physiological response sequences. Instantaneous heart rate characteristics; a triaxial accelerometer, used to acquire spontaneous behavioral pulse streams of the monitored object at a sampling frequency of not less than 25Hz.

[0014] A method for predicting health in older adults includes the following steps: Step 1101: Synchronously acquire the behavioral perturbation sequence of the object to be monitored. and physiological response sequences ; Step 1102, calculate the behavioral perturbation sequence With physiological response sequence Dynamic phase shift characteristics between ; Step 1103, determine the behavioral perturbation sequence Instantaneous energy envelope And based on the preset linear correlation operator Establishing an instantaneous energy envelope With dynamic phase shift characteristics The baseline mapping trajectory between; Step 1104: Determine the logical deviation residual of the measured phase offset relative to the reference mapping trajectory. ; Step 1105: Monitor logic deviation from residuals The second-order cumulative rate of change within a preset sliding window is calculated, and a healthy downward warning signal is output when the second-order cumulative rate of change exceeds a preset stability threshold.

[0015] Compared with existing technologies, the elderly health prediction system of this invention has the following advantages: 1. In predicting health in the elderly, establish a logical time delay analysis path between behavioral perturbation sequences and physiological response sequences. The health status determination is anchored to the body's autonomic nervous system's ability to synchronize responses to random stimuli. It is independent of physiological parameter value deviations and perceives changes in the elasticity of the feedback path regulation before organic abnormalities appear in physiological indicators. This avoids early prediction failure caused by elderly individuals forcibly maintaining normal physiological parameter values ​​due to physiological compensation mechanisms.

[0016] 2. The phase shift characteristics of the triggering phase and the phase-locked characteristics of the recovery phase are verified by time-domain mirror symmetry. Combined with the response hysteresis slope analysis caused by continuous behavioral pulses, a feedback closed-loop audit chain covering the initiation, duration and recovery process of physiological stress is constructed. The logical mutual verification of the evolution law of multiple characteristics is used to suppress false alarms caused by occasional environmental stress, and improve the confidence of the judgment results of the depletion of the body's systemic physiological reserves under complex home working conditions. The intrinsic rhythm phase and inherent response delay are extracted from the micro-static gap of the monitored object in the natural living state, so as to realize the dynamic logical alignment of the prediction benchmark and zero-point offset compensation. There are no additional external detection links to offset the background interference introduced by the diurnal biological rhythm changes and the tightness of the sensor, ensuring the consistency of the monitoring logic and the reference value of the data in long-term service.

[0017] 3. Phase mapping nonlinear transfer residual monitoring quantifies the deviation residual of the measured phase from the baseline mapping trajectory and its second-order cumulative rate of change, realizing a leap from statistical distribution characteristics to an essential measure of information transmission efficiency. It identifies the damage to the integrity of the feedback link hidden behind fluctuations in normal physiological values. Since phase-locked analysis focuses on temporal causal relationships and non-signal amplitudes, it enhances the system's stability against sensor aging and signal distortion. Auditing the endogenous coupling characteristics between different frequency components within the physiological response sequence, it uses the phase-locked relationship between the body's cardiac and respiratory systems to provide self-perceived quality endorsement for the main prediction logic. Before outputting health trend judgment results, it excludes the deviation of conclusions caused by physical contact instability, ensuring the engineering rigor and clinical reference significance of the warning signal output. Attached Figure Description

[0018] Figure 1 This is a flowchart of the predictive early warning logic of phase-locked loop and residual monitoring in this invention; Figure 2 This is a time-domain waveform diagram of the behavioral perturbation excitation and physiological response following characteristics of the present invention; Figure 3 This is a timing diagram of the multi-module interaction of the system for introducing phase verification in this invention. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0020] It should be noted that all directional and positional terms used in this invention, such as: up, down, left, right, front, back, vertical, horizontal, inner, outer, top, low, lateral, longitudinal, center, etc., are only used to explain the relative positional relationship and connection between components in a specific state (as shown in the accompanying drawings). They are only for the convenience of describing this invention and do not require that this invention be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention. In addition, the descriptions of "first," "second," etc., in this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly indicating the number of technical features indicated.

[0021] In the description of this invention, unless otherwise explicitly specified and limited, the terms installation, connection, and linking should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections; they can refer to direct connections or indirect connections through an intermediate medium; they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0022] In the description of this specification, references to the terms "an embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example, and the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0023] The following disclosure is intended to describe the present invention in detail through specific embodiments, so that those skilled in the art can understand and implement the present invention; it should be noted that the following embodiments are only used to explain the present invention, and are not intended to limit the scope of protection of the present invention.

[0024] This invention discloses a method and system for predicting health in the elderly, comprising an information acquisition module, a phase-locked loop analysis module, and a trend determination module; the information acquisition module is used to simultaneously acquire the behavioral perturbation sequence of the subject to be monitored. and physiological response sequences Behavioral perturbation sequence Extracted from a triaxial accelerometer in a wearable device, the triaxial accelerometer at a speed of not less than The sampling frequency is used to collect acceleration pulses of the limbs of the monitored object in three-dimensional space. After filtering out the gravitational acceleration component, the spontaneous micro-motion intensity pulse stream is obtained, forming a behavioral perturbation sequence. Physiological response sequence The pulse-time (PPG) sensor extracts pulse-time characteristics from the photoplethysmography (PPG) sensor. This PPG sensor collects changes in the intensity of reflected light from the skin surface of the subject being monitored, extracts instantaneous heart rate features, and forms a physiological response sequence. The phase-locked loop (PLL) analysis module is used to process the synchronously acquired data stream; this PLL analysis module will analyze the behavioral perturbation sequence. Defined as system stimulus input, and the physiological response sequence Defined as the system response output; the phase-locked loop analysis module calculates the behavioral perturbation sequence by executing the sliding cross-correlation function. Physiological response sequence Dynamic phase shift characteristics between The specific procedure involves setting a fixed-length sliding time window and searching within this window for the time shift that maximizes the correlation between the excitation sequence and the response sequence; this time shift is the dynamic phase shift characteristic at the current moment. .

[0025] The phase-locked loop analysis module integrates a logic transfer verification unit, which is used to determine the behavioral perturbation sequence. Instantaneous energy envelope ; By analyzing behavioral perturbation sequences Obtained by performing a Hilbert transform; the logic transfer verification unit is based on a preset linear correlation operator. Establishing an instantaneous energy envelope With dynamic phase shift characteristics The reference mapping trajectory between them is determined, and the logical deviation residual of the measured phase offset relative to the reference mapping trajectory is calculated. Logical deviation residual The calculation rules are shown in the following formula: ,in, For logical deviation residuals, To measure the dynamic phase shift characteristics, The instantaneous energy envelope of the behavioral perturbation sequence. For the preset linear correlation operator, this linear correlation operator Used to characterize the coupling relationship between energy and phase in a healthy state.

[0026] When establishing the baseline mapping trajectory, the information acquisition module performs individualized calibration during the first day-night cycle after the monitored object is accessed. The calibration includes a synchronous acquisition frequency of not less than [missing information]. Behavioral perturbation sequence With physiological response sequence The phase-locked loop analysis module extracts no less than The group contains instantaneous energy envelopes With dynamic phase shift characteristics The feature vector pairs are subjected to linear regression fitting by the logic transfer verification unit using the least squares method to determine the preset linear correlation operator. slope term With intercept term ,satisfy The mapping relationship represents the inherent excitation response function of the body's regulatory loop under healthy homeostasis for a specific individual, and the real-time calculated logical deviation residual. The instantaneous deviation of measured phase data from the steady-state function is characterized by an individualized calibration process that eliminates baseline errors introduced by the heterogeneity of autonomic nervous system fundamental frequency regulation among different elderly individuals; the trend determination module is used to monitor the aforementioned logical deviation residuals. The second-order cumulative rate of change within a preset sliding window; when the second-order cumulative rate of change exceeds a preset stability threshold, the trend determination module outputs a health decline warning signal; this determination logic is used to identify the impaired transmission efficiency of the feedback link hidden behind the fluctuation of normal physiological values, and can detect the decline in the body's feedback regulation elasticity before the physiological parameters become abnormal, thus avoiding prediction failure caused by the physiological compensation mechanism of elderly individuals maintaining normal surface parameters.

[0027] The elderly health prediction system of this invention also includes a rhythm benchmark extraction module; used for extracting rhythm benchmarks from behavioral perturbation sequences. When the intensity is within a preset resting range, analyze the physiological response sequence. The change in the autocorrelation function envelope is used to determine the endogenous rhythm phase of the current monitoring period. The phase-locked loop analysis module utilizes this intrinsic rhythm phase. Logical deviation residuals A time-domain logical alignment correction is performed to generate a corrected residual sequence for use by the trend determination module; this step is used to eliminate background bias caused by biological rhythms; the phase-locked loop analysis module also includes a response hysteresis analysis unit; used in behavioral perturbation sequences Determine the dynamic phase shift characteristics when multiple consecutive action pulses are included. Follows the pulse sequence number The slope of the change, as a response hysteresis factor Response hysteresis factor The calculation rule is shown in Formula 2 below: ,in, In response to the hysteresis factor, The corresponding serial number is The dynamic phase shift characteristics of the behavior pulse. The trend determination module uses the response hysteresis factor to determine the sequence index value of continuous behavioral pulses within a preset observation period. The degree of deviation from the preset linear range determines the physiological compensation reserve state; if the response lag factor The trend shows a monotonically increasing trend, indicating depletion of the body's compensatory reserves; the trend determination module also includes a recovery phase verification unit, used in behavioral perturbation sequences. Determine the physiological response sequence within the response time window after disappearance. The trend determination module uses the phase-locked loop recovery characteristics and dynamic phase shift characteristics to determine the phase-locked loop recovery characteristics. The temporal symmetry relationship between them is used to make consistent corrections to the health migration trend. This mirror symmetry verification is used to suppress false alarms caused by occasional environmental stress.

[0028] The system also includes a dynamic zero-point drift hedging module; used for behavioral perturbation sequences. When the intensity remains below a preset threshold, determine the physiological response sequence. The inherent response delay under static conditions serves as a reference zero-point characteristic; the phase-locked loop analysis module utilizes this reference zero-point characteristic to analyze the real-time dynamic phase shift characteristics. Offset compensation is performed to counteract background interference introduced by fluctuations in sensor wearing tightness; the consistency arbitration module is used to extract behavioral perturbation sequences. With physiological response sequence The high-frequency components are analyzed, and the coupling coherence between these components is determined. The trend determination module then uses this coupling coherence to analyze the dynamic phase shift characteristics. The validity of the judgment is dynamically weighted; if the coherence decreases, the corresponding weight is reduced to ensure logical consistency in complex scenarios; the output validity audit unit is used to determine the physiological response sequence. The intrinsic coupling characteristics between different physiological frequency components within the body; when the intrinsic coupling characteristics deviate from the preset normal range, the trend determination module suppresses the warning output of the trend determination results; this process utilizes the intrinsic phase-locked relationship between the body's cardiac and respiratory systems to eliminate deviations in conclusions caused by physical contact instability; the phase-locked analysis module also includes an environmental coupling verification unit; this unit detects behavioral perturbation sequences. When the intensity is lower than the preset noise threshold, the parsing logic of the phase-locked loop analysis module is suspended.

[0029] The elderly health prediction method of the present invention is implemented through the above system and includes the following steps: Step 1101, synchronously acquiring the behavioral perturbation sequence of the subject to be monitored through a triaxial accelerometer and a photoplethysmography-pulse sensor. and physiological response sequences Step 1102: Calculate the behavioral perturbation sequence using the sliding cross-correlation function. With physiological response sequence Dynamic phase shift characteristics between Step 1103: Determine the behavioral perturbation sequence through Hilbert transform. Instantaneous energy envelope And based on the preset linear correlation operator Establishing an instantaneous energy envelope With dynamic phase shift characteristics The reference mapping trajectory between; Step 1104, determine the logical deviation residual of the measured phase offset relative to the reference mapping trajectory. Step 1105: Monitor logic deviation from residuals. The second-order cumulative rate of change within a preset sliding window is used to output a healthy downward warning signal when the second-order cumulative rate of change exceeds a preset stability threshold.

[0030] To address the systematic phase interference caused by asynchronous clock acquisition from cross-modal sensors, the phase-locked loop analysis module employs a preset delay compensation matrix before calculation. Timing alignment of data streams across physical channels, delay compensation matrix This includes the fixed circuit transmission delay difference between the triaxial accelerometer and the photoplethysmography (PPG) sensor measured during the manufacturing process. The phase-locked loop analysis module uses a cubic spline interpolation algorithm to resample the original sampling sequence into a uniform value. Interval time axis to synchronize and align residuals Maintain at Below, in Perform cross-correlation calculations within a sliding window of each sampling point, obtain the displacement corresponding to the maximum value of the correlation coefficient, and define the displacement as the dynamic phase shift feature. When the intensity of behavioral perturbation is below a preset threshold in a resting state, the trend determination module's statistical logic deviates from the residual. Background fluctuation standard deviation The preset stability threshold is set to times The measured values ​​enable the generation of early warning signals based on the physical degradation of the machine's feedback loop transmission efficiency, eliminating environmental noise interference; in engineering implementation, the above logic deviates from the residual. The monitoring is through a length of The process is executed within a sliding window of hours, with the sampling step size set to [value]. The second-order cumulative rate of change calculated by the trend determination module is shown in the following formula: (Hourly value) ,in, It is the second-order cumulative rate of change. The logical deviation residual at the current moment, Let be the sampling step size, when Three consecutive sampling points exceeded the preset stability threshold When this happens, an alert is triggered.

[0031] Example 1: In a long-term home physiological monitoring scenario targeting elderly individuals in the subclinical compensatory stage, the monitored subjects' physiological parameters such as heart rate and blood pressure were all within the industry-recognized normal range at rest, i.e., heart rate was maintained at [value missing] per minute. Next Traditional monitoring systems, relying on physiological parameter deviations for judgment, fail to identify the functional decline of underlying feedback mechanisms during periods of excessive depletion of regulatory reserves to maintain parameter stability. This invention's elderly health prediction system acquires data streams from the monitored subject in their natural daily routine through an information acquisition module. The wearable device includes a triaxial accelerometer... Frequency output summation acceleration pulses form a behavioral perturbation sequence. The synchronously integrated photoplethysmography (PPG) sensor outputs a physiological response sequence. Phase-locked analysis module with Using a sliding window with sampling points, the behavioral perturbation sequence is calculated. With physiological response sequence The sliding cross-correlation function between them is used to determine the dynamic phase shift characteristics. When the monitored object performs subtle limb movements such as rolling over or sitting up, the phase-locked loop analysis module observes dynamic phase shift characteristics. Changes in behavioral incentives.

[0032] Because the elderly can mask feedback delays by increasing autonomic nervous system regulatory load during the compensatory period, a logic transmission verification unit intervenes and performs quantitative verification of the transmission efficiency of the regulatory path; this unit uses Hilbert transform to extract behavioral perturbation sequences. Instantaneous energy envelope The logic transfer verification unit verifies the logic based on the preset linear correlation operator. A baseline mapping trajectory for energy and phase is established, and the logical deviation residual of the measured data relative to the baseline trajectory is calculated using the formula. : ,in, For logical deviation residuals, This is a dynamic phase shift characteristic. The instantaneous energy envelope of the behavioral perturbation sequence. As a preset linear correlation operator, when the body's regulatory reserves are impaired, even if the dynamic phase shift characteristics are... The absolute value did not exceed the physiological alarm threshold, and its logical deviation from the residual was... It still shows an unstable expansion trend.

[0033] The trend determination module receives the above logical deviation residual. and with Hours is the sampling step size Perform stability assessment; this module calculates the residual sequence in the past according to Formula 3. Second-order cumulative rate of change within hours : ,in, It is the second-order cumulative rate of change. The logical deviation residual at the current moment, This is the sampling step size; in the monitoring cycle... The system detected at 1 hour. The calculated value is from Growth to This value exceeds the preset stability threshold. The trend determination module determines that the transmission efficiency of the physiological feedback loop is impaired and generates a health decline warning signal, which reflects the physical deterioration of the body's feedback regulation elasticity, rather than the pathological analysis result of a single indicator. During the processing, the system performs hash mapping desensitization processing on the original physiological signals and completes the core logic operation locally on the wearer, only transmitting the hash-anonymized technical indicators to the backend management terminal. The output validity audit unit confirms the physical connection stability of the signal acquisition process by calculating the endogenous coupling characteristics between the respiratory modulation component and the heart rate fundamental frequency component, and eliminates the interference of sensor wearing tightness fluctuations on the judgment results. The final state of the system is to achieve early perception of health risks of elderly individuals before organic deviations occur in physiological parameter values.

[0034] Example 2: This experiment was conducted in a controlled home-based physiological monitoring simulation environment. A physical experimental platform containing a triaxial accelerometer and a high-sensitivity photoplethysmography (PPG) sensor was used to collect raw data streams from the subjects. The triaxial accelerometer had a range of [missing information]. Nonlinearity less than The PPG sensor, used to capture spontaneous, minute movement intensity pulses at the extremities, has a quantization resolution of [missing information]. The sampling frequency is set to The sampling frequency is set based on limb micro-motion signals. to The main energy distribution range, in order to balance the real-time performance of data processing and avoid spectral aliasing, the sampling frequency must be determined according to the sampling theorem to be no less than the signal bandwidth. The final selection was made by combining the transition band characteristics of the front-end filter with the multiples of the filter. As a standard value, and superimposed on the signal-to-noise ratio in the sensor signal source, it is... Gaussian white noise was used to simulate electromagnetic interference and physical artifacts in a real home environment, thereby constructing a behavioral perturbation sequence of the object to be monitored. Physiological response sequence .

[0035] The experimental design includes a sample group based on the present invention and a control sample group. The sample group based on the present invention fully applies the phase-locked analysis and second-order cumulative rate of change monitoring logic described in the aforementioned specific embodiments, while the control sample group uses static threshold monitoring logic based on the absolute value deviation of heart rate. By adjusting the exercise load intensity and rest interval of the subjects, a gradient-like state of bodily functional reserve is simulated. The observation period is set as follows: hour, sampling step size Set as Hours, dynamic phase shift characteristics during the experiment Logical deviation residual and the second-order cumulative rate of change The specific evolution process is recorded in Table 1.

[0036] Table 1: Comparison of Health Prediction Indicators at Different Functional Reserve Stages Based on the analysis of the measurement values ​​recorded in Table 1, the body's functional reserves increased from... Descending to During the compensatory phase, the heart rate measurement of the control group remained at a constant value per minute. Next Within the normal fluctuation range, the alarm logic based on numerical deviation was not triggered, indicating a masking effect of physiological parameters on underlying functional deterioration; while the sample group of this invention, with functional reserves reduced to At that time, its logic deviates from the residual. Depend on The number of seconds suddenly increased to The calculated second-order cumulative rate of change is given in seconds. achieve Exceeding the preset stability threshold And output a warning signal; due to this stability threshold The determination was achieved through zero-position calibration under background noise conditions, that is, in Measurements of healthy individuals at noise levels The standard deviation of background fluctuation is Select Using multiple standard deviations as the decision boundary ensures the system's determinism in identifying impaired feedback link transmission performance, and this is further reinforced as functional reserves continue to decline. It exhibits a non-stationary accelerating trend, reflecting the loss of certainty in the coupling relationship between stimulus and response.

[0037] During the test run, the behavioral perturbation sequence extracted by the phase-locked loop analysis module Instantaneous energy envelope With dynamic phase shift characteristics A physical temporal causal relationship is maintained between them. The logical transmission verification unit quantifies the deviation of the measured phase from the reference mapping trajectory to eliminate the artifact of parameter stability caused by normal physiological compensation; the output validity audit unit synchronously monitors the physiological response sequence. The internal breathing modulation frequency component ensures that the superposition... Even after noise interference, the intrinsic phase-locked characteristics of the data stream still conform to the physiological characteristic distribution. The final state of the system is to achieve advanced perception of health risks in the elderly by identifying the nonlinear growth of information transmission loss in the feedback loop before the absolute value of the heart rate exceeds the limit.

[0038] Example 3: This example combines Figures 1 to 3 A description of a method and system for predicting health in the elderly, such as... Figure 1 As shown, the information acquisition module receives behavioral perturbation sequences and physiological response sequences in parallel. This module is responsible for synchronously acquiring triaxial acceleration and pulse wave data and transmitting the data stream to the next level. The phase-locked loop analysis module receives the data stream to calculate the dynamic phase shift characteristics between sequences. Then, the logic transfer verification unit determines the logical deviation residual of the measured value relative to the baseline mapping trajectory based on the output of the upper level. Subsequently, the trend determination module monitors the second-order cumulative rate of change of the logical deviation residual within the sliding window in real time. Finally, the system enters the logic branch judgment stage. If the monitored rate of change exceeds the preset stability threshold, the system executes the output warning operation and generates a healthy downward warning signal. If the rate of change does not exceed the stability threshold, the judgment result is negative and the control flow returns to the information acquisition module to maintain the continuous monitoring state.

[0039] like Figure 2 As shown in the figure, a coordinate system is constructed with time as the horizontal axis and normalized amplitude as the vertical axis. The horizontal axis covers the entire monitoring period from 0 to 23 hours, and the vertical axis scale ranges from 0.1 to 0.7. The solid curve represents the behavioral perturbation sequence P(t), which exhibits a pulse fluctuation pattern with multiple peaks. The dashed curve represents the physiological response sequence R(t), whose waveform follows the solid line but exhibits a significant phase lag. Both curves are marked with circular markers to identify discrete sampling points. Figure 3 As shown, the object under monitoring emits a behavioral perturbation excitation. The information acquisition module receives the excitation and generates a behavioral perturbation sequence P(t), which is then sent to the phase-locked loop analysis module. The phase-locked loop analysis module then calculates the dynamic phase shift characteristics. The trigger phase characteristics are recorded in the response phase verification unit. When the perturbation of the monitored object's behavior stops, the information acquisition module notifies the response phase verification unit to enter the response time window. During this period, the response phase verification unit continuously monitors the physiological response sequence R(t) to determine the response phase-locked characteristics and compares the trigger phase with the response phase to verify the temporal symmetry relationship. The judgment logic is divided into two branches. If there is temporal mirror symmetry, the response phase verification unit sends a confirmation instruction to the trend judgment module to confirm that the feedback loop is normal. The trend judgment module performs health migration trend consistency correction and outputs the corrected judgment result to the monitored object. If the symmetry relationship is abnormal, the response phase verification unit marks possible occasional stress. The trend judgment module suppresses the current warning output accordingly and feeds back the instruction to the monitored object to continue observation and wait for confirmation.

[0040] Example 4: In the scenario of system initialization parameter calibration for newly connected users, due to the physiological heterogeneity of basal metabolic levels and autonomic nervous system regulatory fundamental frequencies among different individuals, a uniform judgment threshold is difficult to adapt to the time delay distribution characteristics of personalized feedback paths. To establish a monitoring benchmark that excludes interference from individual physiological differences, the system uses the raw data stream within the first 24-hour cycle of the monitored object to perform a deterministic calibration procedure; the information acquisition module simultaneously acquires the behavioral perturbation sequence within this cycle. With physiological response sequence And unify the data storage format to A floating-point array, and control the timestamp alignment error of the two sequences to be less than... The phase-locked loop analysis module analyzes behavioral perturbation sequences. With physiological response sequence Perform discrete cross-correlation calculations; at each span of Within a sliding observation window of each sampling point, the time delay displacement is iterated through using a formula. exist to Relevant values ​​within the second interval: ,in, The cross-correlation function value, For the first observation window The amplitude of the behavioral perturbation at each sampling point For physiological response sequences in time delay The corresponding value below, The index value of the sample points within the observation window. The time delay displacement is extracted by the phase-locked loop analysis module. When the maximum value is obtained As a dynamic phase shift characteristic The instantaneous energy envelope of the behavioral perturbation sequence within the window is extracted synchronously by the logic transfer verification unit. .

[0041] The logic transfer verification unit utilizes the accumulated data within this calibration period, which is not less than For each pair of valid data, the linear coefficients are determined using a first-order linear regression algorithm. With intercept The value is used to establish a baseline mapping trajectory for that specific user. It also calculates the logistic deviation residuals of each sample point in real time during the calibration phase. The statistical mean of the residual sequence during the statistical calibration phase of the trend determination module. with standard deviation And determine the user's stability threshold according to Formula 5. : ,in, As the stability threshold, This represents the statistical mean of the logical deviation residuals. To determine the standard deviation of the logical deviation residuals, this adaptive parameter calibration procedure based on background noise benchmarks enables the system to achieve logical alignment with personalized physiological backgrounds. Recognizing that sensor positions may shift slightly in a home environment, the system incorporates a dynamic zero-point drift offset module into the calibration procedure. It identifies behavioral perturbation sequences. Intensity consistently below The microstatic gap is determined, and the physiological response sequence is identified within this window. The inherent response delay under static conditions serves as a reference zero-point characteristic; the phase-locked loop analysis module utilizes this reference zero-point characteristic to analyze the real-time dynamic phase shift characteristics. Subtraction compensation is performed to eliminate logic zero-point drift caused by changes in wearing position; the output validity audit unit monitors the physiological response sequence. The endogenous coupling characteristics between different frequency components within the body suppress the current warning output when the endogenous coupling characteristics deviate from the preset normal range; ultimately, a highly reliable individual health prediction benchmark is established, so that the subsequent output warning signals accurately characterize the physical deterioration of the body's feedback regulation elasticity.

[0042] Example 5: In a distributed monitoring deployment, the triaxial accelerometers and photoplethysmography (PPG) sensors of different wearable devices have different gain characteristics. The system performs normalization processing before connecting to the phase-locked loop (PLL) analysis module; this includes reading the sensor's underlying registers to obtain the raw code value and using a preset sensitivity correction coefficient. Converting code values ​​to acceleration amplitude and pulse reflection intensity in physical units for behavioral perturbation sequences Physiological response sequence Perform moving average removal separately, that is, within a length of... After calculating the sequence mean within the calibration period, subtraction bias correction is performed. To calibrate the time length, in seconds, so that both processed data streams converge to a fluctuation pattern centered on zero, eliminating the phase deviation caused by DC drift of the analog front-end circuit.

[0043] When establishing the baseline trajectory for the first cycle, the logic transfer verification unit performs data quality auditing to eliminate baseline deviations caused by unstable physical contact of the sensors; in determining the instantaneous energy envelope... With dynamic phase shift characteristics linear coefficients Previously, the system calculated the behavioral perturbation sequence. With physiological response sequence Peak value of cross-correlation coefficient within the current sliding observation window and set no less than Quality audit boundaries; when When the data falls below this audit boundary, data pairs within the current sampling interval are removed from the linear regression sample set until the cumulative threshold is reached. The feature vector pairs and the logical transfer verification unit use first-order linear fitting to calculate the baseline mapping trajectory. The slope and intercept, and the logical deviation residuals generated subsequently. This process is used to characterize the performance fluctuations of physiological regulatory pathways. By eliminating sample points with cross-correlation below a threshold, the stability of the monitoring logic in heterogeneous hardware environments is improved.

[0044] Example 6: In a distributed monitoring scenario involving asynchronous clocks from multiple sensors, the triaxial accelerometer and the photoplethysmography (PPG) sensor use independent sampling clock sources. The system extracts a preset delay compensation matrix before performing phase calculations. The phase-locked loop (PLL) analysis module obtains the interpolation parameters by measuring the time interval of the fixed variance excitation in each channel during the factory manufacturing process. Based on the transmission bandwidth of the sampling channel, the LPL module adjusts the interpolation parameters and uses a cubic spline interpolation algorithm to interpolate the asynchronously sampled behavioral perturbation sequence. With physiological response sequence Mapped to a unified timeline grid, the processed synchronously aligned residuals Stable at Within, of which, To synchronize and align residuals, the unit is milliseconds. The dynamic phase offset characteristics are calculated using the delay compensation matrix. Eliminate phase distortion interference caused by hardware clock jitter.

[0045] When the system faces physical response benchmark drift caused by long-term sensor service, the logic transfer verification unit executes the benchmark self-healing procedure to identify the monitored object in a continuous natural state. The hourly micro-static gap serves as a recalibration window; the system calculates the logic deviation residual within this window. sliding variance and in continued The number of observation periods exceeded the average value obtained in the initial calibration phase. And physiological response sequence When the statistical characteristics are within the preset normal range, the linear correlation operator is executed. Online updates, including For the sliding variance of the logic deviation residual, the logic transfer check unit re-accumulates the variance within this window. For each pair of feature vectors, the linear coefficients are corrected using the weighted least squares method. With intercept This enables logical alignment between the baseline mapping trajectory and the current physical response characteristics of the hardware, maintaining consistency in the determination of healthy downlink warning signals throughout the sensor's lifecycle.

[0046] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit of this application and the scope of protection of this invention, and all of these forms are within the protection scope of this application.

Claims

1. An elderly health prediction system, characterized in that, It includes an information acquisition module, a phase-locked loop analysis module, and a trend determination module: The information acquisition module is used to synchronously acquire the behavioral perturbation sequence of the object to be monitored. and physiological response sequences Behavioral perturbation sequence The spontaneous micro-motion intensity pulse flow extracted by a triaxial accelerometer; The phase-locked loop analysis module integrates a logic transfer verification unit to process behavioral perturbation sequences. Defined as system stimulus input, and the physiological response sequence Defined as the system response output, the behavioral perturbation sequence is calculated using the sliding cross-correlation function. With physiological response sequence Dynamic phase shift characteristics between ; The logic transfer verification unit is used to determine the behavioral perturbation sequence. Instantaneous energy envelope And based on a predefined linear correlation operator characterizing the energy-phase coupling relationship under healthy conditions. Establishing an instantaneous energy envelope With dynamic phase shift characteristics The reference mapping trajectory between the two phases is determined, and the logical deviation residual of the measured phase offset relative to the reference mapping trajectory is determined. Logical deviation residual The calculation rules are as follows: ; The trend determination module is used to monitor logical deviations from residuals. The system calculates the second-order cumulative rate of change within a preset sliding window, and when the second-order cumulative rate of change exceeds a preset stability threshold, it determines that the transmission efficiency of the physiological feedback loop of the monitored object is impaired, and outputs a health decline warning signal.

2. The elderly health prediction system according to claim 1, characterized in that, The system also includes a rhythm reference extraction module; the rhythm reference extraction module is used to extract rhythm references from behavioral perturbation sequences. When the intensity is within a preset resting range, analyze the physiological response sequence. The change in the autocorrelation function envelope is used to determine the endogenous rhythm phase of the current monitoring period. ; Phase-locked loop analysis module utilizes intrinsic rhythm phase Logical deviation residuals Perform time-domain logical alignment correction to generate a corrected residual sequence for the trend determination module.

3. The elderly health prediction system according to claim 1, characterized in that, The phase-locked loop analysis module also includes a response hysteresis analysis unit; the response hysteresis analysis unit is used to analyze behavioral perturbation sequences. Determine the dynamic phase shift characteristics when multiple consecutive action pulses are included. The slope of the change in the behavior pulse sequence number serves as the response hysteresis factor. ; The trend determination module, based on the response hysteresis factor The degree of deviation from the preset linear range determines the physiological compensation reserve status of the monitored object.

4. The elderly health prediction system according to claim 1, characterized in that, The trend determination module also includes a response phase verification unit; the response phase verification unit is used to perform phase verification on behavioral perturbation sequences. Determine the physiological response sequence within the response time window after disappearance. The response phase-locked feature; The trend determination module uses the recovery phase-locked loop characteristics and dynamic phase shift characteristics. The temporal symmetry between them allows for a consistent correction of health migration trends.

5. The elderly health prediction system according to claim 1, characterized in that, The system also includes a dynamic zero-drift hedging module; the dynamic zero-drift hedging module is used to adjust the behavior perturbation sequence. When the intensity remains below a preset threshold, determine the physiological response sequence. The inherent response delay under static conditions is used as a baseline zero-point characteristic; The phase-locked loop analysis module utilizes the reference zero-point characteristics to analyze the real-time dynamic phase shift characteristics. Perform offset compensation.

6. The elderly health prediction system according to claim 1, characterized in that, The system also includes a consistency arbitration module for extracting behavioral perturbation sequences. With physiological response sequence The high-frequency components in the data are identified, and the coupling coherence between the high-frequency components is determined. trend The determination module determines the dynamic phase shift characteristics based on the coupling coherence. The validity of the judgment is dynamically weighted.

7. The elderly health prediction system according to claim 3, characterized in that, Response hysteresis analysis unit calculates response hysteresis factor The rule is: ,in, For the corresponding serial number is The dynamic phase shift characteristics of the behavior pulse, The sequence index value of continuous behavioral pulses within the preset observation period; the trend determination module, in response hysteresis factor When a monotonically increasing trend is observed, it is determined that the body's compensatory reserves have been depleted.

8. The elderly health prediction system according to claim 1, characterized in that, The system also includes an output validity auditing unit; Output validity audit unit, used to determine physiological response sequence The endogenous coupling characteristics between different physiological frequency components within the body; The trend determination module suppresses the warning output of the trend determination results when the endogenous coupling characteristics deviate from the preset normal range.

9. The elderly health prediction system according to claim 1, characterized in that, The phase-locked loop analysis module also includes an environmental coupling verification unit; this unit is used to detect behavioral perturbation sequences. The intensity is determined by suspending the parsing logic of the phase-locked loop analysis module when it falls below a preset noise threshold. The information acquisition module includes a triaxial accelerometer and a photoplethysmography (PPG) sensor; the PPG sensor is used to acquire physiological response sequences. Instantaneous heart rate characteristics; A triaxial accelerometer is used to acquire spontaneous behavioral pulse streams of the object under monitoring at a sampling frequency of not less than 25 Hz.

10. A method for predicting health in the elderly, used to implement the elderly health prediction system of claim 1, characterized in that, Includes the following steps: Step 1101: Synchronously acquire the behavioral perturbation sequence of the object to be monitored. and physiological response sequences ; Step 1102, calculate the behavioral perturbation sequence With physiological response sequence Dynamic phase shift characteristics between ; Step 1103, determine the behavioral perturbation sequence Instantaneous energy envelope And based on the preset linear correlation operator Establishing an instantaneous energy envelope With dynamic phase shift characteristics The baseline mapping trajectory between; Step 1104: Determine the logical deviation residual of the measured phase offset relative to the reference mapping trajectory. ; Step 1105, monitor logic deviation residuals The second-order cumulative rate of change within a preset sliding window is calculated, and a healthy downward warning signal is output when the second-order cumulative rate of change exceeds a preset stability threshold.

Images