Methods, devices, computer program products, and electronic devices for assessing the degree and rate of aging

Abstract

The present disclosure relates to a method for evaluating the degree and speed of aging, comprising determining a chronic aging state of a target user at a current observation time point based on physiological indicator monitoring data of the target user at the current observation time point, the target user having one or more event objects, the event objects having corresponding event states; determining an age event acute impact state of the target user at the current observation time point based on the event objects of the target user, the age event acute impact state being updated retroactively in response to changes in the event states; generating a calibrated age difference of the target user at the current observation time point based on the chronic aging state of the target user at the current observation time point and the age event acute impact; determining a calibrated aging speed of the target user at the current observation time point based on a change trend of the chronic aging state of the target user within an observation window constructed with the current observation time point as the end point. The accuracy, stability and interpretability of the evaluation results can be improved.

Description

Technical Field

[0001] The embodiments disclosed herein generally relate to the fields of artificial intelligence and health monitoring technology, and more specifically to methods, computer program products, and electronic devices for assessing the degree and speed of aging. Background Technology

[0002] Existing digital physiological age and sleep age assessment technologies mainly predict biological age based on physiological characteristics such as heart rate, HRV, and sleep structure, and quantify the degree of aging by the difference between the predicted age and the actual age. At the same time, survival analysis models are used to verify the correlation between this difference and clinical outcomes.

[0003] However, due to the varying sensitivities of different diseases to physiological characteristics, using only clinical outcomes as a post-hoc validation means can lead to a lack of targeted risk calibration in aging assessments, making it difficult for age differences to accurately reflect the true physiological losses oriented by specific diseases. At the same time, existing methods fail to effectively reflect the dynamic evolution of health event states and their differentiated impact on the aging process. Acute events can easily cause instantaneous changes or unreasonable reverse-age fluctuations in aging rate indicators, resulting in long-term aging trends being masked by short-term disturbances, which significantly reduces the stability and interpretability of assessment results. Summary of the Invention

[0004] Embodiments of this disclosure provide methods, computer program products, and electronic devices for assessing the degree and rate of aging, designed to address one or more of the problems described above and other potential problems.

[0005] According to a first aspect of this disclosure, a method for assessing the degree and rate of aging is provided. The method includes determining the chronic aging state of a target user at the current observation time based on physiological indicator monitoring data of the target user at the current observation time. The target user has one or more event objects, each with a corresponding event state. The method also includes determining the target user's acute age-related event impact state at the current observation time based on the target user's event objects. The acute age-related event impact state is retrospectively updated in response to changes in the event state. The method further includes generating a calibrated age difference for the target user at the current observation time based on the target user's chronic aging state and the acute age-related event impact. Furthermore, the method includes determining the calibrated aging rate of the target user at the current observation time based on the changing trend of the target user's chronic aging state within an observation window constructed with the current observation time as the endpoint.

[0006] According to a second aspect of this disclosure, an apparatus for assessing the degree and rate of aging is provided. The apparatus includes a chronic state determination unit configured to determine the chronic aging state of a target user at the current observation time based on physiological indicator monitoring data of the target user at the current observation time. The target user has one or more event objects, each with a corresponding event state. The apparatus also includes an acute state determination unit configured to determine the target user's acute age-related event impact state at the current observation time based on the target user's event objects. The acute age-related event impact state is retrospectively updated in response to changes in the event state. The apparatus further includes an age difference generation unit configured to generate a calibrated age difference for the target user at the current observation time based on the target user's chronic aging state and the acute age-related event impact. Furthermore, the apparatus includes a rate generation unit configured to determine the calibrated aging rate of the target user at the current observation time based on the changing trend of the target user's chronic aging state within an observation window constructed with the current observation time as the endpoint.

[0007] According to a third aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method according to the first aspect.

[0008] According to a fourth aspect of this disclosure, an electronic device is provided, including one or more processors and a memory associated with the one or more processors. The memory is used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method of the first aspect. Attached Figure Description

[0009] The above and other objects, features, and advantages of embodiments of the present disclosure will become readily apparent from the following detailed description taken in conjunction with the accompanying drawings. Several embodiments of the present disclosure are illustrated in the drawings by way of example and not limitation.

[0010] Figure 1 This illustrates the application environment of a method for assessing the degree and rate of aging according to embodiments of this disclosure.

[0011] Figure 2 A method for assessing the degree and rate of aging according to embodiments of the present disclosure is shown.

[0012] Figure 3 The process for determining a chronic aging state according to an embodiment of the present disclosure is illustrated.

[0013] Figure 4 The process for determining the equivalent age for multiple disease risks according to embodiments of this disclosure is illustrated.

[0014] Figure 5 The process for determining the acute impact state of an age-related event according to an embodiment of the present disclosure is illustrated.

[0015] Figure 6 The state changes of an exemplary state direction function according to an embodiment of this disclosure are shown.

[0016] Figure 7 The process for determining the calibrated aging rate according to an embodiment of the present disclosure is illustrated.

[0017] Figure 8 The process for determining the calibrated aging rate according to an embodiment of the present disclosure is illustrated.

[0018] Figure 9 A schematic block diagram of an example apparatus according to some embodiments of the present disclosure is shown.

[0019] Figure 10 A block diagram of an electronic device according to several embodiments of the present disclosure is shown.

[0020] In the various figures, the same or corresponding reference numerals indicate the same or corresponding parts. Detailed Implementation

[0021] Preferred embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0022] The term "comprising" and its variations as used herein signify an open-ended inclusion, i.e., "including but not limited to". Unless otherwise stated, the term "or" means "and / or". The term "based on" means "at least partially based on". The terms "one example embodiment" and "one embodiment" mean "at least one example embodiment". The term "another embodiment" means "at least one additional embodiment". Terms such as "upper", "lower", "front", and "rear", indicating placement or positional relationship, are based on the orientation or positional relationship shown in the accompanying drawings and are used only for the purpose of describing the principles of this disclosure, and are not intended to indicate or imply that the elements referred to must have a specific orientation, be constructed or operated in a specific orientation, and therefore should not be construed as limiting this disclosure.

[0023] As mentioned earlier, existing digital physiological age and sleep age assessment technologies mainly predict biological age based on physiological characteristics such as heart rate, HRV, and sleep structure, and quantify the degree of aging by the difference between the predicted age and the actual age. At the same time, survival analysis models are used to verify the correlation between this difference and clinical outcomes.

[0024] In existing technologies, the association between age differences and outcomes such as death, cardiovascular disease, neurological diseases, and chronic respiratory diseases is validated using Cox models, survival analysis models, or other outcome prediction models. However, due to the varying sensitivities of different diseases to physiological characteristics, using only clinical outcomes as a posterior validation method leads to a lack of targeted risk calibration in aging assessments, making it difficult for age differences to accurately reflect the true physiological losses oriented by specific diseases. Some existing technologies also adjust age estimates based on health records documenting actual abnormal events that occurred to the user (e.g., diagnosis of atrial fibrillation, stroke, TIA, cardiovascular hospitalization, acute exacerbation of chronic respiratory disease, exclusion of test results, or equipment malfunctions). This adjustment method is prone to instantaneous abrupt changes or unreasonable reverse-age fluctuations in aging rate indicators due to acute events, and it also causes long-term aging trends to be masked by short-term disturbances, significantly reducing the stability and interpretability of the assessment results.

[0025] To address this, according to embodiments of this disclosure, a method for assessing the degree and rate of aging is provided. The method includes determining the chronic aging state of a target user at the current observation time based on physiological indicator monitoring data of the target user at the current observation time. The target user has one or more event objects, and each event object has a corresponding event state. The method also includes determining the acute impact state of age-related events of the target user at the current observation time based on the target user's event objects. The acute impact state of age-related events is updated retrospectively in response to changes in the event state. The method further includes generating a calibrated age difference of the target user at the current observation time based on the chronic aging state and the acute impact of age-related events of the target user at the current observation time. Furthermore, the method includes determining the calibrated aging rate of the target user at the current observation time based on the changing trend of the target user's chronic aging state within an observation window constructed with the current observation time as the endpoint. Using the method according to embodiments of this disclosure can improve the accuracy, stability, and interpretability of the assessment results.

[0026] The embodiments of this disclosure will now be described in further detail with reference to the accompanying drawings. Figure 1 An application environment 100 is shown for a method for assessing the degree and rate of aging according to an embodiment of this disclosure. For example... Figure 1 As shown, environment 100 includes computing unit 101, which can be any device with computing or processing capabilities. For example, computing unit 101 can be a server, desktop computer, laptop computer, tablet computer, and mobile phone, etc. Figure 1 As shown, environment 100 includes physiological indicator monitoring data of target user i at the current observation time point t. In the embodiments of this disclosure, the target user refers to an individual undergoing aging status monitoring and assessment, who may be a healthy person, a sub-healthy person, or a patient diagnosed with a specific chronic disease. The user's identity can be associated with an anonymized user identifier to ensure privacy compliance, while also supporting longitudinal tracking of the same user's physiological state changes at different observation time points. Physiological indicator monitoring data is typically acquired through one or more sensing acquisition devices associated with the target user. These devices include, but are not limited to, wearable smartwatches, wristbands, patch biosensors, portable electrocardiogram recorders, sleep monitoring pads, or home medical-grade testing terminals. The sensing acquisition devices can continuously or intermittently collect raw physiological data such as heart rate, heart rate variability (HRV), respiratory rate, body movement, blood oxygen saturation, skin temperature, and sleep stage-related signals using various modal sensing elements such as photoplethysmography (PPG), electrocardiogram (ECG), accelerometers, gyroscopes, blood oxygen probes, temperature sensors, and microphones. The collected raw data can undergo preliminary signal quality assessment, noise filtering, artifact removal, and feature pre-extraction at the sensing device to reduce subsequent transmission bandwidth requirements and improve data validity. Data can be transmitted between the sensing device and the computing unit wirelessly (e.g., via Bluetooth, Wi-Fi, cellular network) or via wired connections, or synchronized through a third-party health platform interface.

[0027] Each target user is associated with one or more event objects. Each event object is used to structurally record abnormal health events that actually occur within the monitoring period and have clear clinical or physiological significance. Examples include confirmed atrial fibrillation, stroke, TIA, cardiovascular hospitalization, acute exacerbation of chronic respiratory disease, exclusion of test results, or equipment false alarms. These events serve as key evidence for assessing the degree and rate of aging. Each event object has a corresponding event status, including "suspected" (preliminary detection pending verification), "confirmed" (clinically verified as a real event), "excluded" (confirmed as non-pathological or a false alarm), "under treatment" (currently receiving intervention), "stable" (condition well controlled), and "relapse" (recurrence of a previous event). Methods for assessing the degree and rate of aging can be executed on the computing unit 101. This is based on the target user's physiological indicator monitoring data. and its associated event object The calibrated age difference of the target user at the current observation time t is obtained. and calibrating the rate of aging .

[0028] The following is combined with Figure 2A method 200 for assessing the degree and rate of aging according to embodiments of the present disclosure is described. It should be understood that the numbers in the flowchart of method 200 do not indicate the order in which these steps are performed; some or all of these steps may be performed in parallel, or the order of execution may be interchanged, and the present disclosure does not limit this. Furthermore, Figure 2 The methods described may also include additional steps not shown and / or the steps shown may be omitted, and the scope of this disclosure is not limited in this respect.

[0029] like Figure 2 As shown in box 202, method 200 may include monitoring data of physiological indicators of target user i at the current observation time point t. Determine the chronic aging status of the target user at the current observation time point. In one or more embodiments of this disclosure, chronic aging status primarily reflects the overall degree of aging exhibited by the user at the current observation point t, mainly reflecting the current level of their sleep cardiopulmonary function and long-term health outcome risk. Chronic aging status can be based on physiological indicator monitoring data. The calculated physiological age offset or homeostasis index can directly quantify the degree of aging deviation relative to the same age group at the current observation point. Therefore, chronic aging status provides a "baseline level" and homeostasis characterization of a user's health.

[0030] In box 204, method 200 may include an event object based on the target user. Determine the age of target user i at the current observation time t and the acute shock state of the event. Acute shock state of age-related events It is based on the event state of event object E. The calculated acute impact of age-related events reflects the cumulative effect and impact load of sudden or confirmed abnormal health events (such as atrial fibrillation, stroke, cardiovascular hospitalization, etc.) on a user's current aging status. Acute impact of age-related events can be expressed as "equivalent aged years" or "standardized aging enhancement." The acute impact of age-related events changes as the event status changes. For example, when an abnormal indicator changes from "suspected" to "excluded" by clinical evidence, its impact should decrease or disappear; when the event is "confirmed" or experiences a "relapse," the acute impact of age-related events becomes more significant, leading to a steep deterioration in the assessed aging status in the short term. Furthermore, the acute impact should gradually decay over time. In the early stages of an event's occurrence or confirmation, the negative impact on the aging status is most severe; as time progresses, if the patient enters "treatment" or the condition reaches a "stable" state without new abnormalities, the body compensates and recovers, and the additional aging effects of the acute impact will decay (e.g., exponential or linear decay), gradually converging towards the user's true chronic aging baseline. Through this combined "state-driven + time-decay" modeling, acute event shocks can accurately quantify the actual interference of pathological stress on the aging process at different stages. The computational unit can record the event state changes of each event object (including event state type and change time), enabling changes in event state to not only affect the calculation of the acute shock state of age-related events after the change, but also to retrospectively update the acute shock states of one or more age-related events affected by the event state based on the evolution of the event state. This allows the impact of health events on the degree of aging to be updated and corrected in a timely manner. Therefore, the acute shock state of age-related events quantifies the instantaneous disturbance and short-term load imposed by sudden illness on the user's baseline health level.

[0031] In box 206, method 200 may include the chronic aging state of target user i at the current observation time t. and the acute impact of age-related events Generate the calibrated age difference of the target user at the current observation time point. Chronic aging represents the long-term depletion of a user's physiological reserves and the rate of aging, while acute age-related events represent the additional aging burden caused by disease stress at the current moment. The calibrated age difference obtained based on these two parameters includes both irreversible cumulative aging and reversible or time-mitigated pathological accelerated aging, thus providing users with a real-time, dynamic, and clinically relevant measure of physiological age shift.

[0032] In box 206, method 200 may include a chronic aging state of the target user within an observation window constructed with the current observation time point as the endpoint. The changing trend is used to determine the calibrated aging rate of the target user at the current observation time point t. The calibrated aging rate aims to capture intrinsic aging rates driven by long-term factors such as lifestyle, progression of chronic diseases, or depletion of organ reserves that cannot be explained by acute events. Therefore, the calibrated aging rate is derived only from the changing trend of chronic aging states within the observation window and does not include the instantaneous contribution of acute shock states of age-related events. This ensures that the impact of abnormal health events is not directly interpreted as abnormally rapid aging occurring in the user in a short period of time.

[0033] In this way, the calibrated age difference obtained by the method of this embodiment can decouple the acute impact state of age events from the chronic aging state, and the calibrated aging rate is mainly estimated by the long-term trajectory of the chronic aging state, thereby improving the stability and interpretability of the evaluation results.

[0034] The following is combined with Figure 3 The process for determining a chronic aging state according to embodiments of this disclosure is described. For example... Figure 3 As shown, in one or more embodiments of this disclosure, the physiological indicator monitoring data may include monitoring data of one or more physiological indicators such as heart rate, HRV (heart rate variability), respiration, sleep structure, activity, and rhythm. For example, the physiological indicator monitoring data may be based on nighttime as the basic observation unit, monitoring the target user's heart rate, HRV, respiratory stability, sleep structure, body movement, rhythm, and bed-leaning status, forming a feature vector of the target user i corresponding to the nighttime (i.e., the current observation time). Where K represents the total number of physiological indicator types included in the physiological indicator monitoring data. In one or more embodiments of this disclosure, the calculation unit can obtain the physiological age of target user i at the current observation time t through the basal sleep physiological age assessment model. The basal sleep physiological age model uses feature vectors... As input, through a pre-defined mapping function Output the physiological age of target user i at observation time t:

[0035]

[0036] Subsequently, the difference between the target user's chronological age and their chronological age is calculated:

[0037]

[0038] in, This represents the target user's actual age at the current observation point t. Actual age refers to age in the traditional sense, reflecting how long the user has lived. Actual age is irreversible; it increases by one year every 365 days. In one or more embodiments of this disclosure, actual age may also be updated in years or months. Physiological age reflects the user's health; it is a biological clock that characterizes the body's true aging state and may be larger or smaller than the actual age. If the physiological age is smaller than the actual age, it indicates better health and longer lifespan. In one or more embodiments of this disclosure, the mapping function... This can be implemented using machine learning or deep learning models, including but not limited to linear regression, support vector regression, random forest, gradient boosting tree, fully connected neural networks, recurrent neural networks, or Transformer architectures. The model is trained on a labeled dataset containing a large amount of historical physiological monitoring data from users and their corresponding reference physiological age labels. Model parameters are optimized by minimizing the loss function between the predicted values ​​and the labels. The basic sleep physiological model can be deployed on computing servers, edge computing devices, or terminal devices. The computing unit calls the trained basic sleep physiological model to... Input the basic physiological model of sleep and obtain the model's inference output. It should be noted that this disclosure does not limit the specific algorithm selection, network structure, training strategy, or deployment form of the basic sleep physiological age model. Any model that can achieve the mapping from physiological indicator monitoring data to physiological age can be applied to this embodiment. In this disclosure, the model serves as an upstream feature extraction module, and its core implementation lies in providing a stable and reproducible representation of sleep physiological age, rather than the innovation of the model itself.

[0039] In one or more embodiments of this disclosure, the computing unit may also be based on physiological indicator monitoring data. Generate the multi-disease risk equivalent age of target user i at the current observation time t. The multi-disease risk equivalent age is a comprehensive value that converts a user's sleep-related risks across multiple disease outcomes into an equivalent number of years older (or younger). In one or more embodiments of this disclosure, the multi-disease risk equivalent age can be measured in "years," providing a clearer picture of the overall health burden caused by sleep problems. Disease outcomes refer to specific health endpoints that are statistically significantly correlated with the user's baseline age, as verified by Cox proportional hazards models, survival analysis models, or other outcome prediction models. These endpoints include, but are not limited to, death, cardiovascular disease, neurological diseases, and chronic respiratory diseases. These outcomes not only have clear clinical significance but also possess quantifiable risk-age relationships, making them reliable anchors for calculating and validating risk equivalent age. For each disease outcome, the user's sleep-related cardiopulmonary risk can be converted into an age offset corresponding to that outcome, and these offsets can be aggregated into the multi-disease risk equivalent age.

[0040] In one or more embodiments of this disclosure, the calculation unit can calculate based on the baseline age difference. Equivalent age for multiple disease risks Generate chronic aging observations for target user i at the current observation time point:

[0041]

[0042] in, This is a coefficient used to control the proportion of people entering a state of chronic aging at the equivalent age for multiple disease risk.

[0043] In one or more embodiments of this disclosure, the chronic aging state can be updated using a quality-weighted smoothing method. Specifically, the computing unit can update the chronic aging observations of target user i at the current observation time t. The chronic aging status of the target users at the previous observation time point and based on physiological indicator monitoring data Data quality is used to determine the chronic aging status of the target user at the current observation time point:

[0044]

[0045] in, To be consistent with physiological indicator test data The quality coefficient is related to data quality. When data quality is high... Larger; when signal saturation, physical activity contamination, excessive time spent out of bed, or insufficient sleep duration, The decrease. The chronic aging state obtained in this way mainly reflects changes in sleep cardiopulmonary status and the risk of long-term disease outcomes.

[0046] The following is combined with Figure 4The process for determining the risk-equivalent age for multiple diseases according to embodiments of this disclosure is described. Figure 4 As shown, in one or more embodiments of this disclosure, physiological monitoring index data can be processed. The monitoring data of various physiological indicators k were standardized and processed to obtain the risk characteristics of each physiological indicator. Standardization aims to eliminate the incomparability between different physiological indicators caused by differences in dimensions, orders of magnitude, and individual baselines, making them more comparable. It not only preserves the relative change information of the original features, but also has comparability across individuals, time periods, and feature types, providing a consistent data foundation for calculating the equivalent age of risk for multiple diseases.

[0047] Subsequently, the calculation unit can determine the target user's linear risk prediction value for each disease outcome based on the risk weights of each physiological indicator and the risk characteristics of each physiological indicator. Where D represents the total number of disease outcome types. Specifically, for the disease outcome d of target user i at the current observation time t, a linear risk prediction value excluding age can be calculated based on a pre-defined disease-specific risk parameter library:

[0048]

[0049] in, These are the risk weights corresponding to disease outcome d recorded in the disease-specific risk parameter library. The risk weights recorded for each disease outcome d in the disease-specific risk parameter library can be quantitative coefficients that reflect the independent contribution of various sleep cardiopulmonary characteristics to a specific disease, determined through statistical modeling and external validation using large-scale longitudinal cohort data.

[0050] Furthermore, the computing unit can determine the additional outcome risk for each disease outcome d for the target user based on the target user's linear risk prediction values ​​for each disease outcome:

[0051]

[0052] Additional outcome risk is used to characterize the additional outcome risk of target user i relative to the conditionally matched reference population, when When, it indicates that the target user's current sleep cardiopulmonary risk regarding disease outcome d is higher than that of a reference population with similar or identical physiological age, chronological age, gender, etc.; when When, it indicates that the target user's current risk of disease outcome d is lower than that of the reference population; when When the value approaches 0, it indicates that the target user's current risk regarding disease outcome d is largely explained by physiological age and baseline demographic factors. Through the aforementioned residualization, it becomes... It can retain the original disease risk model weights and their explanatory power for outcomes while removing the risk components that have already been explained by physiological age, actual age, and gender.

[0053] In one or more embodiments of this disclosure, For conditional risk reference models established for disease outcome d, based on a training cohort or an independent calibration cohort:

[0054]

[0055] in, Let be the physiological age of subject i in the cohort. The actual age of subject i For subject i, the gender or a pre-defined population stratification variable, This is used to represent the linear predictive value of the risk of disease outcome d typically possessed by a reference population under conditions of the same or similar physiological age, chronological age, and population stratification variables. In one or more embodiments of this disclosure, the conditional reference risk model can be established using a machine learning regression model (such as random survival forest, gradient boosting decision tree, support vector regression, or deep learning-based survival neural network, etc.).

[0056] Subsequently, risk effect indices for each disease outcome were calculated based on the target users' actual age. and the calculated risks of each additional outcome. Generate the risk-equivalent age of the target user for each disease outcome. Wherein, the risk-equivalent age of target user i for disease outcome d is:

[0057]

[0058] The risk weights of physiological indicators recorded in the pre-defined disease-specific risk parameter library for disease outcome d are used to characterize the logarithmic risk change of disease outcome d for each year of increase in actual age. It is the risk age for a single disease dimension, which can be understood as "how many years older are you, if we only consider the risk of disease d?" Only the risk of unexplained additional outcomes As input, the additional sleep-related cardiopulmonary risk of target user i beyond the conditionally matched reference population is calculated, equivalent to the number of years the reference risk level of disease outcome d increases or decreases with actual age. Therefore, it is possible to reduce the repeated inclusion of homologous sleep-related cardiopulmonary information in the chronic aging state. The risks.

[0059] Finally, based on the importance weights of each disease outcome and the risk-equivalent age of the target users for each disease outcome. Generate the multi-disease risk equivalent age of target user i at the current observation time t:

[0060]

[0061] in, This is a numeric truncation function used to limit the input value to a specified minimum value. and maximum value between, For risk transformation function, Additional disease outcome risk for target users relative to a conditionally matched reference population. The risk age for multiple diseases was considered to characterize the combined effects of these diseases.

[0062] In one or more embodiments of this disclosure, and The values ​​can be -20 and 20 respectively. Preferably, the risk transformation function can be... For example, a user has three age-equivalent disease risk levels:

[0063] stroke +5 0.4 Atrial fibrillation +2 0.3 Chronic respiratory diseases −3 0.3

[0064] The user's multi-disease risk equivalent age is Although the user had a higher risk of stroke and atrial fibrillation, the risk of chronic respiratory disease was lower than that of the reference population, which offset some of the positive risk, resulting in a final multi-disease risk equivalent age of 1.7 years.

[0065] In one or more embodiments of this disclosure, risk models can be established for disease outcomes in cohort data with follow-up outcomes, and a disease-specific risk parameter library can be constructed based on these models. Disease outcomes may include stroke or TIA, cardiovascular events, atrial fibrillation or arrhythmias, chronic respiratory diseases, all-cause mortality, hospitalization risk, or other target outcomes. Survival analysis can be performed using Cox proportional hazards models, competing hazard models, discrete-time survival models, or other models capable of outputting outcome risk parameters. For example, survival analysis can be performed using a Cox proportional hazards model, with physiological characteristics used as covariates in the Cox proportional hazards model.

[0066]

[0067] in, This represents the risk function of time t under the condition of covariate Z (i.e., physiological index characteristics), which serves as the conditional risk function of disease d. The baseline risk function for disease d is independent of physiological characteristics. Based on this, the Cox proportional hazards model outputs the risk effect index of actual age a on disease outcome d. and the initial risk weight of physiological indicator k on disease outcome d :

[0068]

[0069] in, This is an intermediate result of the model output, indicating how many times the instantaneous risk of disease d increases after feature k increases by 1 unit.

[0070] To avoid unstable features from being included in the aging rate calibration, the system generates risk weights for physiological index k on disease outcome d based on significance, resampling stability, and directional consistency.

[0071]

[0072] in, The significance index is the result of multiple test correction. For Bootstrap or cross-validation stability coefficients, This is the directional consistency coefficient. When a feature's direction is unstable, its significance is insufficient, or its interpretability is insufficient in different data segments, its weight is reduced or set to zero.

[0073] Finally, a disease-specific risk parameter library can be constructed based on the risk effect index of each actual age on each disease outcome and the risk weight of each physiological indicator on each disease outcome.

[0074] The following is combined with Figure 5 This document describes the process for determining an acute shock state of an age-related event according to embodiments of the present disclosure. In one or more embodiments of the present disclosure, an event object is generated when an abnormal health event triggering condition is met. Each event object includes at least an event identifier, event type, event occurrence time, source of evidence, current evidence status, event credibility, event severity, disease association matrix, valid event interval, and update log. Figure 5 As shown, we can first determine the contribution index of each event object e to each disease outcome d. The contribution index of an event object indicates the degree of influence of the event on the disease outcome. For example, the contribution index of the target user's event e to disease outcome d at observation time point t is:

[0075]

[0076] in, The correlation strength between disease outcome d and event e recorded in the disease association matrix; The credibility of event e is used to reflect whether the relevant evidence for the event is reliable. This represents the severity of event e, reflecting the nature of the event and how serious the anomaly is. The event state of event e at the current observation time t; Let be the state direction function. Both are phase functions, and their determination is based on the event state. The decay period is determined based on the event type, event severity, and current stage, and is dynamically updated as the event state changes. This refers to the time when the event occurred. In this way, the impact of an event can be expressed in different dimensions and aspects through severity, stage functions, or decay periods. For example, it can be reduced by... , or shorten To express "recovery effect after treatment".

[0077] Figure 6 The state changes of an exemplary state direction function according to an embodiment of this disclosure are shown. For example... Figure 6 As shown, in one or more embodiments of this disclosure, the value of the directional function is 1 when the event status is suspected, confirmed, or recurrent; the value of the directional function is 0 when the event status is ruled out; and the value of the directional function is between 0 and 1 when the event status is under treatment or stable. Through the state directional function, it is possible to recalculate the historical observation time points affected by the event when it is ruled out. This updates the contribution index of the event at the historical observation point to 0, instead of adding any new negative event.

[0078] In one or more embodiments of this disclosure, the calculation unit can determine the event impact equivalent age of the target user based on the importance weight of each disease outcome, the event impact ageization coefficient of each disease outcome, and the contribution index of each event object of the target user to each disease outcome:

[0079]

[0080] in, is the ageization coefficient of the event impact on disease outcome d, which is obtained periodically by expert rules; and These are the upper and lower limits of the event's equivalent age, which can be -20 and 20 respectively.

[0081] Finally, based on the equivalent age of the event impact and the age of the target user at the previous observation point, the acute impact state of the event... It can calculate the target user's age at the current observation point in time and the state of acute impact of the event:

[0082]

[0083] in, For state-dependent shock retention factors, This sets the proportion of the pre-defined event impact that enters the calibration age difference. The impact retention coefficient can be adjusted according to the credibility of the event. For example, a major event with a high degree of confirmed severity can have a higher impact retention coefficient, a suspected event can have a lower impact retention coefficient, and excluded events can trigger a retrospective update and make the impact retention coefficient approach 0.

[0084] In one or more embodiments of this disclosure, the event state set can be defined as:

[0085]

[0086] In this system, 0 indicates suspected, 1 indicates confirmed, 2 indicates ruled out, 3 indicates under treatment, 4 indicates stable, and 5 indicates relapse. The following are typical triggering conditions and corresponding actions for event state transitions:

[0087] Suspected → Confirmed Upload examination materials, doctor's confirmation, or structured medical record confirmation. Improve event credibility and severity, correct event occurrence time, and extend decay period. Suspected → Excluded Exclusion by re-examination, false alarms by manual review, or denial by multimodal evidence. Reduce the credibility of the original event or set the contribution index of the original event to zero, and recalculate it retrospectively. Confirmed → Under Treatment Users begin treatment, medication, or hospitalization management. Update stage coefficients and enter the treatment observation period. During treatment → Stable No recurrence for N consecutive days or N months, and stable follow-up indicators Reduce the severity, stage factor, or decay period of the event to reduce its impact. Stable → Relapse Recurrence of similar abnormalities or diagnosis of relapse Restart or enhance the event impact retention factor and record the number of recurrences.

[0088] Ultimately, the computational unit can use the sum of chronic aging states and acute shock states from age-related events as the calibration age difference, i.e.:

[0089]

[0090] In this way, this disclosure can decouple the calibrated age difference into two parts: chronic and acute. It primarily reflects changes in sleep cardiopulmonary status and long-term outcome risk, particularly the acute portion. It primarily reflects abnormal health events and subsequent changes in evidence, thus avoiding false jumps in the rate of aging caused by these events.

[0091] In one or more embodiments of this disclosure, the calculation unit may further calculate the calibrated physiological age of the target user based on the calibrated age difference:

[0092]

[0093] The following is combined with Figure 7 The process for determining the calibrated aging rate according to embodiments of this disclosure is described. For example... Figure 7As shown, in one or more embodiments of this disclosure, the calibrated aging rate is not directly calculated as the slope of the calibrated age difference containing acute impacts of age-related events, but rather as the calibrated aging rate by performing a quality-weighted robust slope estimation on the chronic aging state. Specifically, to ensure the reliability and longitudinal comparability of sleep cardiopulmonary risk features extracted from the original physiological indicator monitoring signals, a multi-dimensional quality assessment can be performed on the signals and states of each observation night (or each observation window), and a set of quality parameters with values ​​preferably between 0 and 1 can be generated: the closer the value is to 1, the more ideal the quality of that dimension is and it does not constitute a penalty; the lower the value, the more severe the defect of that dimension. The quality parameters may include the effective data ratio, body movement contamination weight, signal saturation weight, sleep duration or in-bed effectiveness weight, and outlier weight. The effective data ratio can be the proportion of effective signals that can be used for feature calculation within the monitoring period, reflecting the completeness and statistical power of the data; if it is below a preset lower limit, the data for that night is generally unusable. The body movement contamination weight quantifies the degree of contamination of the signal by body movement based on accelerometer or artifact detection algorithms; the less body movement, the closer it is to 1, used to characterize the level of residual motion interference. Signal saturation weight characterizes the degree to which the waveform is clipped when the signal reaches the upper or lower limits of the sensor's measurement range. The lower the proportion of saturated samples, the closer it is to 1, and it is used to measure the reliability of waveform morphology information. Sleep duration or bedtime effectiveness weight is based on sleep stage or bedtime detection results, measuring whether the actual sleep duration is sufficient and whether the bedtime status is continuous and effective, avoiding misuse of data from awake or short-term wear. Outlier weight is based on the user's own historical baseline or the distribution of peers, using robust statistical methods to detect the degree of outlier characteristics of the night, in order to suppress the contamination of measurement abnormal nights and avoid misinterpreting single-night failures as health mutations.

[0094] When calculating the calibrated aging rate, one can first base it on the effective data ratio of physiological indicator monitoring data at each observation time point within the observation window. Physical pollution weight Signal saturation weights Weighting of sleep duration (or bed effectiveness) and outlier weights One or more of the following factors determine the overall weight of data quality for the corresponding observation time point:

[0095]

[0096] Subsequently, the comprehensive data quality weight can be used as a gating condition to perform quality-weighted robust slope estimation of chronic aging states, ensuring the repeatability and reliability of each chronic aging state data point entering the calibration aging rate calculation. Specifically, based on the chronic aging state and comprehensive data quality weight of target user i at each observation time point within the observation window, the calibration aging rate of target user i at the current observation time point t can be determined:

[0097]

[0098] Where w is the width of the sliding observation window, The slope of the trend line fitted by weighted least squares estimation using data quality weights.

[0099] In one or more embodiments of this disclosure, the change in event impact can also be calculated. and overall display speed And these are included as part of the method output to improve the interpretability of the evaluation results:

[0100]

[0101]

[0102] in, The weighting coefficient is used to control the contribution of the weighted least squares slope of the data quality comprehensive weighting to the overall display speed in response to event impacts.

[0103] The following is combined with Figure 8 The process for determining the calibrated aging rate according to embodiments of this disclosure is described. For example... Figure 8 As shown, in one or more embodiments of this disclosure, when the event state of an event object changes, the calculation unit not only updates the contribution index of the event object for the current night (i.e., the current observation time), but also retrospectively updates the contribution index of related historical observation time points based on the effective event interval of the event object. Specifically, when the event state changes, the calculation unit first updates the contribution index based on the effective event time interval of the event object. One or more observation time points that are within the valid time interval of the event are identified as historical observation time points that need to be backtracked and updated.

[0104] Subsequently, for each historical observation point, the contribution index of that event object to each disease outcome d will be recalculated:

[0105]

[0106] in, This is the set of parameters for the event object after the event state changes. These are historical observation points within the valid time zone of the event. This indicates that based on historical observation time points The updated parameter set storage management system recalculates the contribution index of event objects to each disease outcome d. Subsequently, it calculates the difference in contribution index between the updated event objects and the events before and after the event state change.

[0107]

[0108] in, It refers to the contribution index of the event object to disease outcome b before the event state changes.

[0109] Finally, the computing unit can update the target user i at that historical observation time point based on the calculated contribution difference. The impact of the event on the equivalent age Age-time shock state and calibrated age In this way, the confirmation, exclusion, treatment, stabilization, and relapse of events can all form traceable, explainable, and revocable health status updates.

[0110] Figure 9 A schematic block diagram of an example device 900 according to some embodiments of the present disclosure is shown. Device 900 can be implemented by software, hardware, or a combination of both. Figure 9 As shown, the device 900 may include a chronic state determination unit 901, configured to determine the chronic aging state of the target user at the current observation time based on physiological indicator monitoring data of the target user at the current observation time. The target user has one or more event objects, and each event object has a corresponding event state. The device also includes an acute state determination unit 902, configured to determine the acute impact state of age-related events of the target user at the current observation time based on the event objects of the target user. The acute impact state of age-related events is updated retrospectively in response to changes in the event state. The device also includes an age difference generation unit 903, configured to generate a calibrated age difference of the target user at the current observation time based on the chronic aging state of the target user at the current observation time and the acute impact of age-related events. In addition, the device also includes a velocity generation unit, configured to determine the calibrated aging velocity of the target user at the current observation time based on the changing trend of the chronic aging state of the target user within an observation window constructed with the current observation time as the endpoint.

[0111] The following examples illustrate the evaluation mechanisms for the calibration age difference and calibration aging rate in embodiments of this disclosure under different event state transition scenarios:

[0112] Example 1: State transition and dynamic calibration after a diagnosed atrial fibrillation event

[0113] A user continuously uses a smart bed for nighttime sleep monitoring. The system can first obtain the user's continuous sleep cardiopulmonary data, calculate physiological age, baseline age difference, and equivalent age for multiple disease risks, and update the user's chronic aging status accordingly.

[0114] During two consecutive nights of monitoring, the smart bed's device algorithm identified atrial fibrillation-like rhythm abnormalities. Upon the occurrence of the abnormality, an event object e1 was generated, set as an atrial fibrillation-related event, with the evidence source recorded as device algorithm identification, and the initial state set to "suspected." Based on the device algorithm's confidence level, the number of consecutive occurrences of the abnormality, and signal quality, the event's credibility, severity, stage coefficient, and decay period were initialized. Simultaneously, based on a pre-defined event-disease correlation matrix, disease association parameters between the event and atrial fibrillation, stroke, and cardiovascular disease outcomes were determined. The event's contribution index to disease outcomes was calculated based on these event parameters, and the event's impact-equivalent age E was aggregated to generate the age-related acute impact state H. At this point, the event was recorded as a suspected state, therefore its impact on H was limited by lower credibility and stage coefficient.

[0115] After a user uploads a hospital's dynamic electrocardiogram (ECG) report confirming atrial fibrillation, the status of the same event object e1 can be updated from "suspected" to "confirmed," instead of creating a new confirmed event. Furthermore, based on the newly added medical evidence, the event credibility of the event object is increased, the event severity is updated according to the atrial fibrillation load or clinical condition, the stage coefficient corresponding to the confirmed status is increased, the decay period is adjusted, and the event impact equivalent age E and the acute impact state H of the age event are recalculated, thereby updating the calibrated age difference and calibrated physiological age.

[0116] After the user begins medication, the event status further transitions from "Confirmed" to "Under Treatment." If no atrial fibrillation-like abnormalities are detected again for three consecutive months and the follow-up results are stable, the status transitions from "Under Treatment" to "Stable." The system response reduces the event severity, stage coefficient, or shortens the decay period, causing the age of the event participant and the acute impact state H of the event to gradually decline with the event.

[0117] Throughout the process, the acute effects of atrial fibrillation events primarily enter the acute impact state H of age-related events, and do not directly serve as a permanent increment of the chronic aging state. The calibrated aging rate is still calculated based on the long-term trajectory of the chronic aging state A. As can be seen from Example 1, the same real disease event can be continuously updated with the changes in evidence from "suspected—confirmed—treatment—stability," rather than establishing independent event records at each stage, thus achieving dynamic changes in the impact of the event with the evidence and treatment status.

[0118] Example 2: Retrospective Update of Historical Contribution Index after Suspected Atrial Fibrillation Has Been Ruled Out

[0119] A user continuously uses a smart bed for nighttime sleep monitoring. The system can first obtain the user's continuous sleep cardiopulmonary data, calculate physiological age, baseline age difference, and equivalent age for multiple disease risks, and update the user's chronic aging status accordingly.

[0120] During two consecutive nights of monitoring, the smart bed's algorithm detected a suspected atrial fibrillation-like rhythm abnormality. Upon the occurrence of the abnormality, an event object e2 was generated, recording the initial time of the abnormality, the abnormal segment, the algorithm's result, and the signal quality. The evidence status was set to "suspected." Since the current evidence originates from the algorithm and lacks medical diagnostic support, a relatively low event confidence level and event stage coefficient were assigned to this event. The event severity and initial decay period were determined based on the duration of the abnormality. Similar to Example 1, the event impact equivalent age E was calculated based on the event parameters and included in the acute impact state H of the age event. Therefore, the user's calibrated age difference and calibrated physiological age may show a certain degree of short-term increase, but this effect is limited by the "suspected" event status.

[0121] After a period of time, the user completes a dynamic electrocardiogram (ECG) examination, the doctor confirms no risk of atrial fibrillation, and the user uploads corresponding exclusion evidence. The event status of the original event object e2 changes from "suspected" to "excluded," instead of creating a new negative event to offset the original suspected event. Simultaneously, the effective time interval of event e2 is located. Based on the new event status "excluded," the directional state function values ​​of each observed event point within the effective event interval are updated to 0, and the contribution index of this event to each disease outcome at each historical observed event point is recalculated. Based on this, starting from the earliest affected historical observed event point, the event impact equivalent age E and the acute impact of age-related events H are recalculated sequentially, and the calibrated age difference, calibrated physiological age, and comprehensive display results of the historical observed event points are simultaneously corrected, thereby reversing the historical risk increase caused by the original suspected event.

[0122] Since the acute impact of the age-related event H is decoupled from the chronic aging state A, if excluding the event does not change the continuous sleep cardiopulmonary data and the equivalent age of multiple disease risk, the calibrated aging rate calculated from the chronic aging state A will also remain unchanged. Therefore, the calibrated aging rate does not change due to the exclusion of the event. In addition, event records can be kept for "device suspected, medical exclusion" samples, and abnormal device fragments, exclusion evidence, and event state transition logs can be saved for subsequent evaluation of device algorithm false alarm rates or model optimization.

[0123] As can be seen from Example 2, when an early device malfunction is refuted by subsequent medical evidence, this disclosure does not simply deduct a score at the current point in time, but modifies the original event object and retroactively cancels its contribution index at the historical observation point, thereby preventing erroneous events from remaining in the user's health trajectory for a long time.

[0124] Example 3: Decoupling the acute impact of a major stroke event from the rate of chronic aging.

[0125] A user continuously uses a smart bed for long-term sleep monitoring. The system calculates the user's physiological age, baseline age difference, and equivalent age for multiple disease risks based on their nighttime heart rate, HRV, breathing, and sleep structure, and updates the user's chronic aging status and historical baseline for calibrating the rate of aging accordingly.

[0126] If a user subsequently suffers a stroke and is hospitalized, a stroke event object e3 can be generated based on the user's uploaded hospitalization records or medical materials. Since this event already has clear medical diagnostic evidence, its initial event status can be set to "confirmed" without first going through a "suspected" state. Based on the source of the medical diagnosis, the event credibility is set to a high level. Based on the event type (i.e., stroke), clinical severity, and hospitalization status, the event severity is set to a high level, and the attenuation period for this type of major event is also set to be relatively long. In the event-disease association matrix, this event establishes a high association strength with stroke outcomes, cardiovascular outcomes, and all-cause risk outcomes. Based on each association strength, the contribution index of this event on each disease outcome dimension can be calculated. Based on this, a high event impact equivalent age E is obtained, which significantly increases the acute impact H of age-related events. Therefore, after a stroke, the user's calibrated age difference and calibrated physiological age show a significant structural increase.

[0127] Because this disclosure simultaneously preserves two independent trajectories—the acute impact of the age event H and the chronic aging state A—the equivalent age E resulting from the event impact on the day of the stroke is only included in the acute impact of the age event H and not directly written into the chronic aging state A. The calibrated aging rate is still calculated based on the long-term change slope of the chronic aging state within a preset sliding observation window. Therefore, even if an immediate stroke event causes a significant increase in the calibrated age difference in the short term, it will not be directly interpreted as the user experiencing abnormally rapid aging within a few days, but rather as an acute event impact that can be displayed independently.

[0128] After a user begins treatment, the event status transitions from "Confirmed" to "Under Treatment." If subsequent recovery is stable, it further transitions to "Stable." By reducing the stage coefficient or shortening the decay period, the acute impact H of the age-related event can be gradually reduced. Furthermore, this disclosure will continue to utilize subsequent physiological indicator monitoring data to calculate physiological age and multi-disease risk equivalent age. If, weeks or months after a stroke, there is a persistent decrease in HRV, abnormal heart rate, deterioration of sleep structure, or an increase in multi-disease risk equivalent age, these long-term changes will enter a chronic aging state A through the baseline age difference and multi-disease risk equivalent age. Only then will the calibrated aging rate show a sustained increase.

[0129] As can be seen from Example 3, the short-term structural shock caused by a major disease event is primarily represented by the acute shock H of the age event. Only the persistent changes in sleep cardiopulmonary function and long-term disease outcome risk after the event affect the calibrated aging rate through the chronic aging state A, thus avoiding the misinterpretation of a stroke event as "rapid aging" in a short period of time.

[0130] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this specification are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)).

[0131] Figure 10 A block diagram of an electronic device 1000 that can implement various embodiments of the present disclosure is shown. For example... Figure 10 As shown, the electronic device 1000 includes a processor 1010, a disk drive 1020, an input / output interface 1030, a network interface 1040, and a memory 1050. The processor 1010, disk drive 1020, input / output interface 1030, network interface 1040, and memory 1050 can communicate with each other via a communication bus 1060.

[0132] The processor 1010 can be implemented using a general-purpose CPU, microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits to execute relevant programs and implement the technical solution provided in this application.

[0133] The memory 1050 can be implemented in the form of ROM (Read Only Memory), RAM (Read Access Memory), static memory, dynamic storage devices, etc. The memory 1050 can store the operating system 1051 for controlling the operation of the electronic device 1000, and the basic input / output system (BIOS) 1052 for controlling the low-level operations of the electronic device 1000. Additionally, it can store a web browser 1053, a data storage management system 1054, etc. In summary, when implementing the technical solution provided in this application through software or firmware, the relevant program code is stored in the memory 1050 and is called and executed by the processor 1010.

[0134] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0135] The network interface 1040 is used to connect the communication module (not shown in the figure) to enable communication and interaction between the device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0136] Bus 1060 includes a pathway for transmitting information between various components of the device, such as processor 1010, disk drive 1020, input / output interface 1030, network interface 1040, and memory 1050.

[0137] It should be noted that although the above-described device only shows the processor 1010, disk drive 1020, input / output interface 1030, network interface 1040, memory 1050, bus 1060, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the method of this application, and does not necessarily include all the components shown in the figures.

[0138] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0139] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing. Furthermore, although operations are depicted in a specific order, this should be understood as requiring that such operations be performed in the specific order shown or in sequential order, or requiring that all illustrated operations be performed to achieve the desired result. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the foregoing discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented individually or in any suitable sub-combination in multiple implementations.

[0140] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method for assessing the degree and rate of aging, characterized in that, include: Based on the physiological indicator monitoring data of the target user at the current observation time, the chronic aging state of the target user at the current observation time is determined. The target user has one or more event objects, and the event objects have corresponding event states. Based on the event object of the target user, the acute impact state of the age event of the target user at the current observation time is determined, and the acute impact state of the age event is updated retrospectively in response to the change of the event state. Based on the target user’s chronic aging state and acute age-related events at the current observation time, a calibrated age difference for the target user at the current observation time is generated. and Based on the changing trend of the target user's chronic aging state within the observation window constructed with the current observation time point as the endpoint, the calibrated aging rate of the target user at the current observation time point is determined.

2. The method according to claim 1, characterized in that, The physiological indicator monitoring data includes monitoring data of one or more of the following physiological indicators: heart rate, HRV, respiration, sleep structure, activity, and rhythm. Furthermore, determining the chronic aging status of the target user at the current observation time point based on the physiological indicator monitoring data at the current observation time point includes: Based on the physiological indicator monitoring data, the baseline age difference between the target user's physiological age and their actual age, as well as the target user's multi-disease risk equivalent age, are determined. Based on the multi-disease risk equivalent age and the baseline age difference, the chronic aging observation of the target user at the current observation time point is determined; and Based on the chronic aging observations of the target user at the current observation time, the chronic aging status of the target user at the previous observation time, and the data quality of the physiological indicator monitoring data, the chronic aging status of the target user at the current observation time is determined.

3. The method according to claim 2, characterized in that, The determination of the equivalent age for multiple disease risk includes: Based on the physiological indicator monitoring data, physiological indicator risk characteristics of each physiological indicator of the target user are generated; Based on the risk weights of each of the aforementioned physiological indicators for each disease outcome and the risk characteristics of each of the aforementioned physiological indicators, the linear risk prediction values ​​of the target user for each of the aforementioned disease outcomes are determined. Based on the target user's linear risk prediction values ​​for each of the disease outcomes, determine the target user's additional outcome risk for each disease outcome; Based on the risk effect index of the actual age for each of the disease outcomes and the risk of each of the additional outcomes, determine the risk-equivalent age of the target user for each of the disease outcomes; and The risk-equivalent age for each disease outcome is determined based on the importance weight of each disease outcome and the risk-equivalent age of the target user for each disease outcome.

4. The method according to claim 1, characterized in that, The event object also has a corresponding event credibility, event severity, and disease correlation matrix, and the determination of the acute shock state of the age event includes: Based on the disease association matrix, event credibility, event severity, and event status, the contribution index of each event object to each disease outcome is determined. Based on the importance weights of each disease outcome, the event impact ageization coefficients of each disease outcome, and the contribution indexes of each event subject of the target user to each disease outcome, the event impact equivalent age of the target user is determined; and Based on the equivalent age of the event impact and the acute impact state of the age event of the target user at the previous observation time, the acute impact state of the age event of the target user at the current observation time is determined.

5. The method according to claim 1, characterized in that, The determination of the calibrated aging rate includes: Based on one or more of the following factors—the effective data ratio of physiological indicator monitoring data at each observation time point within the observation window, the weight of body movement contamination, the weight of signal saturation, the weight of sleep duration, and the weight of outliers—a comprehensive data quality weight is determined for the corresponding observation time point; and Based on the chronic aging status and data quality weight of the target user at each observation time point within the observation window, the calibration aging rate of the target user at the current observation time point is determined.

6. The method according to claim 1, characterized in that, The event status includes one or more of the following: suspected, confirmed, ruled out, under treatment, stable, and relapsed. The event object has a corresponding valid time interval, and the retrospective update of the acute impact of the age event includes: In response to a change in the event state, based on the effective time interval of the event object corresponding to the event state, one or more historical observation time points related to the event object are determined; and Determine the difference in contribution index of the event object to each disease outcome before and after the change in the event state at each of the aforementioned historical observation time points; and Based on the contribution differences at each of the historical observation time points, update the target user's acute impact status of age events and calibrated age at each of the historical observation time points.

7. The method according to any one of claims 1-6, characterized in that, Also includes: Based on the target user's acute impact state of age-related events at the current observation time and the acute impact state of age-related events at the previous observation time, the change in event impact of the target user at the current observation time is determined; and Based on the calibrated aging rate of the target user at the current observation event point and the acute impact state of the age event of the target user at the current observation time point, a comprehensive aging display rate of the target user at the current observation time point is generated.

8. A device for assessing the degree and rate of aging, characterized in that, include: The chronic state determination unit is configured to determine the chronic aging state of the target user at the current observation time based on the physiological index monitoring data of the target user at the current observation time. The target user has one or more event objects, and the event objects have corresponding event states. An acute state determination unit is configured to determine the acute impact state of an age event of the target user at the current observation time point based on the event object of the target user, and the acute impact state of the age event is updated retrospectively in response to changes in the event state. The age difference generation unit is configured to generate a calibrated age difference for the target user at the current observation time, based on the target user's chronic aging state and acute age-related events at the current observation time. The speed generation unit is configured to determine the calibrated aging speed of the target user at the current observation time point based on the changing trend of the target user's chronic aging state within an observation window constructed with the current observation time point as the endpoint.

9. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-7.

10. Electronic devices, including: One or more processors, and A memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method according to any one of claims 1-7.

Images