Sleep age assessment method and system

Abstract

The invention discloses a sleep age assessment method and system, and the method comprises the steps: obtaining user data and sleep monitoring data of a target user, the user data comprising personal basic data and inherent sleep data; based on the personal basic data and the inherent sleep data, determining a sleep age basic value of the target user; determining a sleep age deviation value of a target user based on sleep monitoring data and the inherent sleep data; and determining the actual sleep age of the target user based on the sleep age basic value and the sleep age deviation value. Through the inherent sleep data of the user, subjective evaluation deviation is avoided, complex equipment is not needed, and long-term convenient monitoring is achieved; the matching degree between the current sleep state of the user and the long-term inherent mode is accurately reflected by combining personal data and the difference between the long-term sleep data and the short-term sleep data according to the logic of'basic value + deviation value ', the long-term mode and the short-term fluctuation are considered, practical reference is provided for health of the user, and the defects of an existing method are overcome.

Description

Technical Field

[0001] This application relates to the field of smart home technology, and in particular to a method and system for assessing sleep age. Background Technology

[0002] In contemporary scientific research, the relationship between sleep and chronological age has become an important window into revealing the mechanisms of human aging. Sleep status not only reflects an individual's health status but also directly participates in regulating the physiological aging process. Therefore, assessing sleep age is of great significance for improving sleep and promoting health.

[0003] Current methods for assessing sleep age can be broadly categorized into two types: subjective assessment methods and objective assessment methods.

[0004] Subjective assessment methods include standardized scale assessments and sleep diaries. A sleep diary involves a user recording their sleep onset time, number and duration of awakenings, total sleep duration, and daytime functional status. Based on these parameters, the user's inherent sleep pattern can be obtained. This inherent sleep pattern is then used to assess the user's sleep age using a pre-defined algorithm model. This method is simple and easy to implement, and long-term recording can reflect long-term sleep patterns. However, it relies on subjective recall and may be subject to memory bias. Commonly used standardized scales include the Pittsburgh Sleep Quality Index (PSQI) and the Self-Rating Sleep Status Scale (SRSS), whose implementation principle is the same as that of sleep diaries. Sleep age is assessed based on the user's inherent sleep pattern obtained through these scales.

[0005] Objective assessment methods are mainly based on polysomnography (PSG), primarily used for acquiring physiological signals and determining subsequent sleep stages. By monitoring long-term stage results, sleep characteristics highly correlated with age (such as the proportion of deep sleep and REM latency) are extracted. Pre-set models are used to obtain sleep age. Commonly used pre-set models include linear / nonlinear regression, random forest, support vector machine, and end-to-end deep models. This method is considered the gold standard in sleep stage determination. However, the equipment is relatively complex to operate, requires a high level of professional expertise, and has low comfort, making it basically infeasible for long-term sleep monitoring. Summary of the Invention

[0006] Therefore, it is necessary to provide a sleep age assessment method and system to address the above-mentioned technical problems and solve at least one of the problems existing in the prior art.

[0007] Firstly, a method for assessing sleep age is provided, including: Acquire user data and sleep monitoring data of the target user. The user data includes personal basic data and inherent sleep data. Personal basic data refers to data that reflects the user's basic physiological characteristics, and inherent sleep data refers to stable sleep-related characteristic data that the user exhibits under normal conditions. Based on the aforementioned basic personal data and inherent sleep data, the baseline sleep age value of the target user is determined; Based on the sleep monitoring data and the inherent sleep data, the sleep age deviation value of the target user is determined; The actual sleep age of the target user is determined based on the baseline sleep age value and the sleep age deviation value.

[0008] In one possible implementation, determining the target user's baseline sleep age based on the individual's basic data and inherent sleep data includes: Based on the aforementioned basic personal data and inherent sleep data, a feature array is constructed; The feature array is input into a preset sleep age prediction model for prediction processing to generate the target user's basic sleep age value.

[0009] In one possible implementation, the intrinsic sleep data includes at least one of the following: an intrinsic sleep state array, intrinsic sleep onset time, intrinsic wakefulness time, intrinsic sleep midpoint, intrinsic sleep structure data, and intrinsic sleep quality data. The intrinsic sleep state array is obtained in the following manner: Obtain the target user's historical sleep monitoring data; Based on the historical sleep monitoring data, an array of sleep states corresponding to multiple sleep cycles is generated; The sleep state array corresponding to each sleep cycle is filtered to generate the inherent sleep state array.

[0010] In one possible implementation, determining the sleep age deviation value of the target user based on the sleep monitoring data and the inherent sleep data includes: Based on the sleep monitoring data and the inherent sleep data, the sleep assessment result of the target user is determined. The sleep assessment result includes at least one of the sleep state assessment result, sleep structure assessment result, and sleep quality assessment result. The sleep confidence factor is determined based on the proportion of valid sleep data that meets the preset duration threshold within a preset time window. Based on the sleep assessment results and the sleep confidence factor, the sleep age deviation value is determined.

[0011] In one possible implementation, the sleep state assessment results are obtained in the following manner: Based on the sleep monitoring data, the current key sleep milestones are obtained; Based on the inherent sleep data, key nodes of inherent sleep are obtained; Based on the current sleep key nodes and the inherent sleep key nodes, determine the relative differences corresponding to each node; The relative differences corresponding to each node are processed by attenuation calculation; After weighted summation of the relative differences corresponding to each node after the decay operation, normalization is performed to obtain the sleep state consistency score, which is used as the sleep state evaluation result.

[0012] In one possible implementation, the sleep structure assessment results are obtained in the following manner: Based on the sleep monitoring data, the duration of the current sleep stage and the percentage of the current sleep stage are obtained; Based on the intrinsic sleep data, the duration of intrinsic sleep stages and the percentage of intrinsic sleep stages are obtained. Based on the current sleep stage duration, current sleep stage percentage, inherent sleep stage duration, and inherent sleep stage percentage, a sleep structure consistency score is calculated using a preset weighted decay rule, which serves as the sleep structure evaluation result.

[0013] In one possible implementation, the sleep structure consistency score is calculated based on the current sleep stage duration, the current sleep stage percentage, the inherent sleep stage duration, and the inherent sleep stage percentage, using a preset weighted decay rule, including: Based on the current sleep stage duration and the inherent sleep stage duration, the relative difference in sleep stage duration is determined; Based on the current sleep stage percentage and the inherent sleep stage percentage, the relative difference in sleep stage percentage is determined; The relative differences in the duration of the sleep stages and the relative differences in the proportion of the sleep stages are respectively processed by attenuation calculation; After weighted summation of the relative differences in sleep stage duration and the relative differences in sleep stage proportion after attenuation calculation, normalization is performed to obtain the sleep structure consistency score.

[0014] In one possible implementation, the sleep quality assessment results are obtained in the following manner: Based on the sleep monitoring data, the current sleep efficiency and current sleep continuity are obtained; Based on the intrinsic sleep data, intrinsic sleep efficiency and intrinsic sleep continuity are obtained; Determine the relative difference in sleep efficiency between the current sleep efficiency and the inherent sleep efficiency, and the relative difference in sleep continuity between the current sleep continuity and the inherent sleep continuity; After weighted summation of the relative differences in sleep efficiency and sleep continuity, normalization is performed to obtain a sleep quality consistency score, which is used as the sleep quality assessment result.

[0015] In one possible implementation, the baseline sleep age values ​​include a first baseline sleep age value and a second baseline sleep age value. Determining the target user's actual sleep age based on the baseline sleep age value and the sleep age deviation value includes: Based on the aforementioned basic personal data, the true age of the target user is determined; Determine the user usage frequency factor within a preset time range, wherein the user usage frequency factor includes the ratio of the number of effective sleeps to a preset threshold number of sleeps; Based on the first baseline sleep age value, the second baseline sleep age value, the actual age, and the sleep age deviation value, the actual sleep age is obtained by weighted fusion using the user usage frequency factor.

[0016] Secondly, a sleep age assessment system is provided, including: The sleep-related data acquisition unit is used to acquire user data and sleep monitoring data of the target user. The user data includes personal basic data and inherent sleep data. Personal basic data refers to data that reflects the user's own basic physiological characteristics, and inherent sleep data refers to stable sleep-related characteristic data that the user exhibits under normal conditions. The sleep age baseline value determination unit is used to determine the sleep age baseline value of the target user based on the personal baseline data and inherent sleep data. The sleep age deviation value determination unit is used to determine the sleep age deviation value of the target user based on the sleep monitoring data and the inherent sleep data. The actual sleep age determination unit is used to determine the actual sleep age of the target user based on the baseline sleep age value and the sleep age deviation value.

[0017] The aforementioned sleep age assessment method and system include the following steps: acquiring user data and sleep monitoring data of a target user, wherein the user data includes personal basic data and inherent sleep data; determining a baseline sleep age value for the target user based on the personal basic data and inherent sleep data; determining a sleep age deviation value for the target user based on the sleep monitoring data and inherent sleep data; and determining the actual sleep age of the target user based on the baseline sleep age value and the sleep age deviation value. In this embodiment, by acquiring the inherent sleep data of the target user (generated by filtering historical sleep monitoring data), the memory bias of subjective assessment is avoided, and there is no need to rely on complex PSG devices, achieving convenient long-term objective monitoring. Simultaneously, by using a two-layer logic of baseline sleep age value and deviation value, combined with the multi-dimensional differences between personal basic data, inherent sleep data, and short-term sleep monitoring data, the system accurately reflects the degree of matching between the target user's sleep state and physiological age, reflecting both long-term sleep patterns and short-term sleep fluctuations, providing a more practical reference for users to improve sleep and maintain health, effectively compensating for the shortcomings of existing assessment methods. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of an application environment for a sleep age assessment method according to an embodiment of this application; Figure 2 This is a flowchart illustrating a sleep age assessment method in one embodiment of this application; Figure 3 This is a schematic diagram of the distribution of historical sleep data in one embodiment of this application; Figure 4 This is a schematic diagram of historical sleep-wake distribution in one embodiment of this application; Figure 5 This is a schematic diagram of the historical sleep-wake distribution after filtering in one embodiment of this application; Figure 6 This is a schematic diagram of the sleep age assessment system in one embodiment of this application; Figure 7 This is a schematic diagram of a computer device according to one embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] The sleep age assessment method provided in this embodiment can be applied to, for example... Figure 1 In the application environment, it includes server S1, client S2 and device S3, which can communicate and connect with each other.

[0022] Specifically, the server S1 can be implemented using a standalone server or a server cluster consisting of multiple servers. It can be a local server or a cloud server. It is used for user data management, specifically including user data storage and information distribution. User data includes, but is not limited to, user personal information, historical sleep data, historical physiological parameter information, and user's inherent sleep pattern parameter data. User personal information includes, but is not limited to, height, weight, age, and gender; historical sleep data includes, but is not limited to, user's sleep state information at various points in time, including bedtime, bedtime, sleep onset time, and wakefulness time; sleep state information can be categorized into bed-out, awake, light sleep, deep sleep, and REM sleep; or multiple classification formats such as bed-out, awake, N1, N2, N3, and REM sleep; historical physiological parameter information mainly includes resting heart rate and resting respiratory rate, which refer to the user's average heart rate and average respiratory rate in a resting state, respectively. The resting state refers to the time segment in which the user remains motionless for more than the threshold TH7 (e.g., 5 minutes) and has not entered sleep. Resting heart rate and respiratory rate are updated once daily at a fixed time; user's inherent sleep pattern parameter data... This includes: a user's inherent sleep-wake state array, the duration of each inherent sleep phase, the percentage of each inherent phase, the user's inherent sleep efficiency, and the user's inherent sleep continuity. The sleep efficiency is calculated as sleep duration / time spent in bed × 100%. Sleep continuity is determined by combining sleep phase characteristics and micro-awake characteristics during sleep, and is evaluated using a preset sleep continuity calculation model to obtain a continuity score. The sleep phase characteristics include the stage of each sleep segment, the duration of that stage, and the sleep cycle number of that stage. The micro-awake characteristics include micro-awake intensity, micro-awake duration, the sleep stage of the micro-awake, the sleep cycle of the micro-awake, and the sleep duration before the micro-awake. The preset sleep continuity calculation model can be a simple linear model or a deep learning model. Client S2 includes, but is not limited to, various personal computers, laptops, smartphones, tablets and portable wearable devices, which are pre-installed with APPs or mini-programs for displaying sleep age results, entering user personal information, and uploading relevant data to the server. The main carrier of the S3 device can be smart home devices such as smart bed frames or smart mattresses, which can include five parts: signal acquisition module, signal processing module, sleep stage module, inherent sleep pattern acquisition module, and sleep age calculation module. The responsibilities of each module are as follows: The signal acquisition module is used to collect sensor data. The type of sensor data can be a piezoelectric sensor signal placed directly below the shoulder, or a photoplethysmography signal collected from the fingers, earlobe, or wrist. The signal processing module is used to perform noise filtering and feature extraction on the acquired sensor signals. The noise filtering can be a simple infinite impulse corresponding Butterworth filter or a combination of filters, or an autoencoder filter based on a deep learning model. The features extracted in the feature extraction include the heartbeat interval array (the heartbeat interval is the time difference between adjacent heartbeat peaks), the standard deviation of the heartbeat interval (SDNN), the respiratory interval array (the expiratory interval is the time difference between adjacent respiratory wave peaks), the standard deviation of the respiratory interval, and body movement intensity. The sleep staging module is used to classify the current sleep state based on the feature array composed of features extracted by the signal processing module and a preset sleep staging model. The feature array includes, in addition to the heart interval array, respiratory interval array, heart interval standard deviation, respiratory interval standard deviation, and body movement intensity within the current window time, the current state and the duration of the current state. The preset sleep staging model can be a deep learning model or a traditional machine learning model such as random forest. The inherent sleep pattern acquisition module is used to extract and generate user-specific inherent sleep pattern parameter data.

[0023] Sleep consistency calculation module: Based on inherent sleep pattern parameters, it performs consistency calculations on sleep state, sleep structure, and sleep quality, and outputs preliminary assessment results.

[0024] The specific implementation process is as follows: First, users can enter their basic personal data (height, weight, age, gender, etc.) through the client S2. The entered information is uploaded to the server S1 for storage via the communication link. At the same time, the signal acquisition module of the device S3 continuously collects the user's physiological sensor signals (such as piezoelectric sensor signals and photoplethysmography pulse wave signals) and transmits the raw sensor signals to the signal processing module for noise filtering and feature extraction to obtain core feature data such as heart rate interval, respiratory interval, and body movement intensity. Then, the sleep staging module of the device S3 can complete real-time sleep staging based on the extracted feature data to obtain the current sleep state information. The inherent sleep pattern acquisition module combines historical sleep staging data (which can be supplemented by historical sleep data retrieved from the server S1) to generate the user's inherent sleep pattern parameter data. The device S3 synchronously uploads the real-time sleep data, extracted feature data, and inherent sleep pattern parameter data to the server S1. The server S1 classifies and stores this data and, combined with the stored user personal information, provides complete data support for subsequent sleep age assessment. Next, the sleep age calculation module on server S1 or device S3 calls a preset model. Based on the user's basic personal data and inherent sleep pattern parameters provided by server S1, as well as the real-time sleep monitoring data uploaded by device S3, it calculates the baseline and deviation values ​​of sleep age, and finally merges them to obtain the user's actual sleep age. Finally, the calculated actual sleep age result is sent from server S1 to client S2, which displays the assessment result to the user in a visual format (such as numerical values, charts, and text interpretations). At the same time, server S1 supplements the user's historical data with the current sleep monitoring data and sleep age assessment result, completing the data update and iteration, and providing data accumulation for more accurate assessments in the future.

[0025] In one embodiment, such as Figure 2 As shown, a sleep age assessment method is provided, which can be applied to... Figure 1 Taking the device side as an example, the explanation includes the following steps: In step S110, user data and sleep monitoring data of the target user are acquired. The user data includes personal basic data and inherent sleep data. Personal basic data refers to data that reflects the user's basic physiological characteristics, and inherent sleep data refers to stable sleep-related characteristic data that the user exhibits under normal conditions. The personal basic data includes, but is not limited to, age, weight, height, and gender; the user's inherent sleep data includes, but is not limited to, the user's inherent sleep onset time, inherent wakefulness time, inherent sleep midpoint, inherent sleep duration, inherent percentage of each sleep stage, inherent duration of each sleep stage, inherent sleep efficiency, and inherent sleep continuity; sleep monitoring data refers to the target user's sleep monitoring data from the previous night, which includes, but is not limited to, sleep onset time, wakefulness time, sleep midpoint, sleep duration, duration and percentage of each sleep stage (deep sleep, light sleep, REM sleep, wakefulness), sleep efficiency, sleep continuity, and sleep state sequence.

[0026] It should be noted that sleep monitoring data can be continuously collected by the signal acquisition module of the device (such as smart bed frames, smart mattresses, and other smart home devices) to collect physiological sensor signals of the target user during sleep (including piezoelectric sensor signals placed directly under the shoulder or photoplethysmography signals collected from the fingers, earlobes, and wrists). After noise filtering (using Infinite Impulse Response (IIR) Butterworth filters, filter combinations, or autoencoder filters based on deep learning models, etc.) and feature extraction (extracting features such as heart interval arrays, heart interval standard deviation (SDNN), respiratory interval arrays, respiratory interval standard deviation, and body movement intensity), the sleep staging module uses the extracted features to form a feature array (which may include heart interval arrays, respiratory interval arrays, heart interval standard deviations, respiratory interval standard deviations, body movement intensity, current state, and duration of the current state within the current window). The current sleep state is then classified using a preset sleep staging model (such as deep learning models, random forests, and other traditional machine learning models), ultimately generating sleep monitoring data.

[0027] User data can be manually entered by users through pre-installed apps or mini-programs on client devices (such as smartphones, tablets, portable wearable devices, etc.). This data includes age, weight, height, gender, etc. The entered information is uploaded to the server for storage via communication links (such as mobile networks, Wi-Fi, etc.). The server can then distribute the data to the device or use it directly for data processing.

[0028] In addition, the inherent sleep data is obtained by acquiring historical sleep monitoring data of the target user within a preset time window (such as 30 consecutive days) that is consistent with the above sleep monitoring data. After cleaning and removing invalid sleep data (such as sleep duration less than the preset threshold TH5, for example, 3 hours) and abnormal data (such as sleep efficiency exceeding 100%), the cleaned valid historical sleep data is subjected to feature statistical analysis. The statistical values ​​of each sleep parameter are calculated respectively (where the inherent sleep onset time, inherent wake-up time, inherent sleep midpoint, and inherent sleep duration are the average, median, or mode of the corresponding parameters; the inherent percentage of each stage, inherent duration of each stage, and inherent sleep efficiency are the average values ​​of the corresponding parameters; and inherent sleep continuity is the average value of the sleep continuity score in the valid historical data or the continuity benchmark value obtained by analyzing the sleep state sequence in the valid historical data through a preset model). Finally, the parameters obtained above are integrated to generate the inherent sleep data of the target user.

[0029] In step S120, based on the personal basic data and inherent sleep data, the basic value of the target user's sleep age is determined; Optionally, personal baseline data and inherent sleep data can be preprocessed to generate a structured feature data sequence that meets the input requirements of a preset sleep age prediction model. This sequence is then input into a pre-trained preset sleep age prediction model (such as an Extreme Gradient Boosting Tree (XGBoost) model with 20 trees and a depth of 6, or a deep learning model composed of fully connected layers) for prediction. This model has been trained on a dataset containing a large amount of personal information, long-term sleep parameters, and their actual age labels. It can learn the correlation between sleep characteristics and physiological age. The model calculates and analyzes the input personalized feature array and finally outputs a value representing the user's long-term sleep pattern health status, i.e., the baseline sleep age value.

[0030] In step S130, the sleep age deviation value of the target user is determined based on the sleep monitoring data and the inherent sleep data; Optionally, the current sleep monitoring data (such as the previous night's sleep onset time, sleep structure, and sleep efficiency) and the inherent sleep data (such as inherent sleep onset time, inherent proportion of each sleep stage, and inherent sleep efficiency) obtained from long-term historical data statistical analysis of the target user are first acquired. Then, for at least one dimension of sleep state (such as key time points such as sleep onset and wakefulness), sleep structure (such as the duration and proportion of each sleep stage), and sleep quality (such as sleep efficiency and sleep continuity), the relative difference between the sleep monitoring data and the inherent sleep data is calculated. By processing these relative differences, a consistency score for at least one dimension (such as at least one of sleep state consistency score, sleep structure consistency score, and sleep quality consistency score) is obtained, which is used as the sleep age deviation value. This deviation value can reflect the degree of deviation of the user's current sleep state from its long-term inherent pattern, thereby providing a dynamic adjustment basis for the final sleep age assessment.

[0031] In step S140, the actual sleep age of the target user is determined based on the baseline sleep age value and the sleep age deviation value.

[0032] Optionally, a user usage frequency factor can be calculated from relevant data within a preset statistical window (e.g., 30 days, 60 days, or 120 days). This user usage frequency factor can be determined by the ratio of the number of effective sleep sessions within the window to a preset number of sessions. Effective sleep refers to records where the total sleep duration on a given day exceeds a set threshold (e.g., 3 hours). Then, based on the pre-obtained sleep age deviation value, the baseline sleep age values ​​(including the two baseline sleep age values ​​corresponding to the current moment and the previous day's inherent sleep data) are weighted and fused. This fusion result is then weighted and integrated again with the user's current real age according to the usage frequency factor and its corresponding complementary ratio, ultimately yielding the target user's actual sleep age. This actual sleep age simultaneously reflects the correlation between the user's long-term sleep pattern, recent sleep consistency, and real age.

[0033] This application provides a method for assessing sleep age, comprising: acquiring user data and sleep monitoring data of a target user, wherein the user data includes personal basic data and inherent sleep data; determining a baseline sleep age value for the target user based on the personal basic data and inherent sleep data; determining a sleep age deviation value for the target user based on the sleep monitoring data and inherent sleep data; and determining the actual sleep age of the target user based on the baseline sleep age value and the sleep age deviation value. This application, by acquiring the inherent sleep data of the target user, avoids the memory bias of subjective assessment and eliminates the need for complex PSG devices, achieving convenient long-term objective monitoring. Simultaneously, by employing a two-layer logic of baseline and deviation values ​​for sleep age, combined with the multi-dimensional differences between personal basic data, inherent sleep data, and short-term sleep monitoring data, it accurately reflects the degree of matching between the target user's sleep state and physiological age, reflecting both long-term sleep patterns and short-term sleep fluctuations, providing a more practical reference for users to improve sleep and maintain health, effectively compensating for the shortcomings of existing assessment methods.

[0034] In one embodiment of this application, determining the baseline sleep age value of the target user based on the personal basic data and inherent sleep data includes: Based on the aforementioned basic personal data and inherent sleep data, a feature array is constructed; The feature array is input into a preset sleep age prediction model for prediction processing to generate the target user's basic sleep age value.

[0035] Optionally, personal basic data (such as the user's age, weight, height, gender, etc.) and inherent sleep data (including inherent sleep onset / wake-up / sleep midpoint time, inherent sleep duration, the proportion and duration of each sleep stage, inherent sleep efficiency, and inherent sleep continuity) are first integrated into an input feature array, which is then fed into a preset sleep age prediction model (an Extreme Gradient Boosting Tree (XGBoost) model configured with 20 trees and a tree depth of 6 can be used; a deep learning model composed of fully connected layers can also be used). The model outputs the corresponding baseline sleep age value. This baseline sleep age value may include the baseline sleep age value corresponding to the inherent sleep data at the current moment and the baseline sleep age value corresponding to the inherent sleep data of the previous day.

[0036] It should be noted that this preset sleep age prediction model can be trained on a specific dataset: this dataset covers people aged 15-80, with a body mass index (BMI: weight / height). The weight (in kilograms, height in meters) ranges from 10 to 40. Each data point includes a feature set and an age label (the subject's actual age). The feature set requires continuous sleep data collection for a preset duration (TH2, e.g., 30 days), ensuring no invalid sleep data or extreme abnormal sleep events are recorded during the collection period. The feature content must be consistent with the input feature array. The age label represents the subject's current age. During training, the feature data is input into a preset model structure, and the weights are iteratively optimized using the age label as the target value. Taking XGBoost as an example, the training parameters are configured as follows: learning rate 0.1, maximum tree depth 6, number of trees 30, and the root mean square error (RMSE) is used as the loss function evaluation metric.

[0037] Invalid sleep data refers to sleep data with a duration of less than TH5 (e.g., 3 hours).

[0038] In one embodiment of this application, the inherent sleep data includes at least one of the following: an inherent sleep state array, inherent sleep onset time, inherent wakefulness time, inherent sleep midpoint, inherent sleep structure data, and inherent sleep quality data. The inherent sleep state array is obtained in the following manner: Obtain the target user's historical sleep monitoring data; Based on the historical sleep monitoring data, an array of sleep states corresponding to multiple sleep cycles is generated; The sleep state array corresponding to each sleep cycle is filtered to generate the inherent sleep state array.

[0039] Optionally, firstly, historical sleep monitoring data of the target user within a preset time window TH2 (e.g., 30 days) can be obtained. This historical sleep monitoring data may specifically include an array of sleep state information at various times, including bedtime, bedtime, sleep onset time, and wake-up time. An example of the distribution of the sleep state information array at each time point can be found here. Figure 3 , Figure 3 The graph displays the results of 50 days of continuous sleep monitoring for a single user. The horizontal axis represents the time sequence, with 1 indicating day 1. The vertical axis represents the time points, from 18:00 the previous day to 18:00 the current day. The sleep states shown in the graph include out of bed, awake, light sleep, deep sleep, rapid eye movement (REM) sleep, and nap. The criteria for determining a nap are: if the time interval between the sleep segment and the main sleep segment exceeds 3 hours, and the occurrence time of the sleep segment is not within a specified time segment, then the sleep segment is determined to be a nap. The sleep segment states include light sleep, deep sleep, and REM. The main sleep segment refers to the time segment with the longest sleep duration in the sleep stage data of the day. The specified time segment can be 0:00-8:00.

[0040] Then, at least one of the following can be extracted from the above historical sleep monitoring data: the intrinsic sleep state array, the intrinsic sleep onset time, the intrinsic wake-up time, the intrinsic sleep midpoint, the intrinsic sleep structure data, and the intrinsic sleep quality data.

[0041] Specifically, the process of obtaining the inherent sleep state array can be as follows: extract daily sleep state data (such as light sleep, deep sleep, REM sleep, nap, etc.) and wakefulness data from historical sleep monitoring data, and construct a daily inherent sleep state array, that is, a binary array representing sleep (1) and wakefulness (0) states. The times corresponding to getting out of bed and wakefulness can be set to 0, and the times corresponding to light sleep, deep sleep, REM sleep, and nap can be set to 1. The time resolution of the sleep-wake binary array can be 30 seconds, that is, one state value every 30 seconds. The length of the daily sleep-wake binary array can be 2880. An example of the distribution of processed sleep-wake data is shown below. Figure 4 .

[0042] Then, for the sleep-wake binary arrays at the same time each day, a low-pass filter is applied along the time dimension (i.e., the same time across multiple days of data) based on a preset sleep data filtering template. The resulting filtered sleep-wake binary array is as follows: Figure 5 As shown, the filtered sleep-wake binary array (which contains the sleep probability value at each moment) is truncated by the following steps: values ​​in the filtered sleep-wake binary array that are greater than a preset threshold TH3 (e.g., 0.5) are set to 1, indicating a sleep state, and values ​​that are less than the preset threshold TH3 are set to 0, indicating a wake state. The resulting binary array is the user's inherent sleep state array for that day.

[0043] It should be noted that during the filtering process, if the sleep data for a certain day is invalid, the original sleep data for that day needs to be replaced with the user's existing inherent sleep information before participating in the filtering process. The replacement includes replacing the user's sleep-wake array for that day with the user's inherent sleep-wake state array for that day, and replacing the user's sleep onset time, wake-up time, sleep midpoint, percentage of each phase, duration of each phase, sleep efficiency, and sleep continuity for that day with the user's inherent sleep onset time, inherent wake-up time, inherent sleep midpoint, inherent percentage of each phase, inherent duration of each phase, inherent sleep efficiency, and inherent sleep continuity, respectively. Invalid sleep data refers to sleep data with a sleep duration less than TH5 (e.g., 3 hours).

[0044] The sleep data filtering template can be a mean filter template, an infinite impulse response (IIR) Butterworth low-pass filter template, or a Gaussian filter template. Taking Gaussian filtering as an example, the Gaussian filter template can be as follows: ; Where i corresponds to the index of the filter template, that is, the i-th value of the template; Pi; The standard deviation is denoted as σ, which can be 20; e is the natural constant; N is the mean of the template length, taking a template length of 60 as an example, where N is 30; the obtained Gaussian filter template is then normalized, and the normalized Gaussian coefficients are... The acquisition process is as follows: ; Where M is the length of the Gaussian template, i.e., 60; the first half of the Gaussian filter template is truncated, and all coefficients are multiplied by 2 to ensure that the sum of the filter coefficients is 1. The coefficient array processed above is used as the sleep-wake filter array. The sleep-wake state distribution after filtering is as follows: Figure 5 As shown in the figure, the brighter the light, the greater the likelihood of being asleep, and the darker the light, the greater the likelihood of being awake. At the very beginning of the figure, due to insufficient data, the daily sleep patterns have a greater impact on the filtered results, so the changes in the initial time period are more volatile than in later time periods.

[0045] Based on historical sleep monitoring data, the inherent sleep onset time, inherent wakefulness time, and inherent sleep midpoint information of the target user can be obtained. The inherent sleep onset time refers to the time point when the first state of the main sleep segment is sleep, the inherent wakefulness time refers to the time point when the last state of the main sleep segment is sleep, and the inherent sleep midpoint is the midpoint (i.e., the average value) between the inherent sleep onset time and the inherent wakefulness time.

[0046] Simultaneously, based on historical sleep monitoring data, the user's inherent sleep structure and sleep quality information can also be obtained. The inherent sleep structure information includes: the inherent percentage of each sleep stage and the inherent duration of each sleep stage. The percentage of each sleep stage includes: inherent deep sleep percentage, inherent light sleep percentage, inherent REM sleep percentage, and inherent wakefulness percentage. The method for obtaining the percentage of each sleep stage is to perform low-pass filtering along the time dimension on the percentage of each sleep stage in the user's historical sleep data, and the specific acquisition method is consistent with the aforementioned sleep-wake binary array processing method. The statistical range for the percentages of deep sleep, light sleep, REM sleep, and wakefulness is from the time the user falls asleep to the time the user wakes up. The percentages of each stage after low-pass filtering are normalized by dividing the percentage of each stage by the sum of the percentages of all stages after filtering, to ensure that the sum of the percentages of each stage is 100%. The inherent duration of each stage includes: inherent deep sleep duration, inherent light sleep duration, inherent REM sleep duration, and inherent wakefulness duration. The inherent duration of each stage is obtained by calculating the time interval between the user's inherent fall-off time and inherent wakefulness time, i.e., the user's inherent sleep duration, and then multiplying the inherent sleep duration by the inherent percentage of each stage to obtain the inherent duration of each stage. Inherent sleep quality information includes: inherent sleep efficiency and inherent sleep continuity. Inherent sleep efficiency can be obtained by performing low-pass filtering on the user's historical sleep data along the time dimension, and the method is the same as above; inherent sleep continuity can be obtained by performing low-pass filtering on the user's historical sleep data along the time dimension, and the method is the same as above.

[0047] In one embodiment of this application, determining the sleep age deviation value of the target user based on the sleep monitoring data and the inherent sleep data includes: Based on the sleep monitoring data and the inherent sleep data, the sleep assessment result of the target user is determined. The sleep assessment result includes at least one of the sleep state assessment result, sleep structure assessment result, and sleep quality assessment result. The sleep confidence factor is determined based on the proportion of valid sleep data that meets the preset duration threshold within a preset time window. Based on the sleep assessment results and the sleep confidence factor, the sleep age deviation value is determined.

[0048] Optionally, based on the user's current sleep monitoring data and personal inherent sleep data, at least one dimension is selected from sleep state (such as consistency of sleep onset / wake time), sleep structure (such as consistency of the proportion of each sleep stage), and sleep quality (such as sleep efficiency and continuity consistency) to determine the sleep assessment result; then, based on the proportion of valid sleep data that meets the duration threshold (such as 3 hours) within a preset time window, a sleep confidence factor reflecting the reliability of the data is calculated; finally, the multi-dimensional sleep assessment results are combined with the sleep confidence factor, and the final sleep age deviation value is calculated through a preset weighted fusion rule.

[0049] For example, taking a sleep assessment result that includes three components—sleep state assessment result, sleep structure assessment result, and sleep quality assessment result—where the sleep state assessment result includes a sleep state consistency score, the sleep structure assessment result includes a sleep structure consistency score, and the sleep quality assessment result includes a sleep quality consistency score, the sleep assessment result (i.e., the sleep assessment consistency score) can be calculated using the following formula: ; in, For sleep confidence factors, , , The preset weighting coefficients are set based on empirical values, and can all be 0.333. The sleep state consistency score, Assess sleep structure consistency score, The consistency score for sleep quality.

[0050] In one embodiment of this application, the sleep state assessment result is obtained in the following manner: Based on the sleep monitoring data, the current key sleep milestones are obtained; Based on the inherent sleep data, key nodes of inherent sleep are obtained; Based on the current sleep key nodes and the inherent sleep key nodes, determine the relative differences corresponding to each node; The relative differences corresponding to each node are processed by attenuation calculation; After weighted summation of the relative differences corresponding to each node after the decay operation, normalization is performed to obtain the sleep state consistency score, which is used as the sleep state evaluation result.

[0051] Optionally, key sleep nodes such as sleep onset time, wake-up time, and sleep midpoint are extracted from the current sleep monitoring data, and corresponding key sleep nodes are extracted from the inherent sleep data. Then, the relative differences between the current and inherent key sleep nodes (such as the duration of sleep onset time advancement / delay, wake-up time deviation, etc.) are calculated, and the differences of each node are processed by nonlinear decay operation according to preset rules. Finally, the decayed differences of each node are weighted and summed according to preset weights (such as sleep onset time weight 0.4, wake-up time weight 0.4, and sleep midpoint weight 0.2), and the total difference value of the weighted sum is normalized (such as converting the result to the 0-1 interval through the maximum-minimum value mapping method) to obtain the sleep state consistency score, which is used as the sleep state assessment result.

[0052] The sleep consistency score can be calculated using the following formula: ; Where i is the array index, and For preset weight parameters, This is the time to fall asleep; It is a fixed time to fall asleep; Midpoint of sleep; This is the midpoint of intrinsic sleep; The point in time when one is awake; This refers to the inherent point in time when the person is awake.

[0053] It should be noted that, and It can be obtained by parameter fitting on a specific dataset 1. The fitting process refers to using the inherent sleep onset time, inherent sleep midpoint, inherent wake-up time, sleep onset time, sleep midpoint, and wake-up time of the day as input features, and using the sleep state consistency score as the target value based on a preset weight distribution template, to perform fitting or hyperparameter search.

[0054] Specific dataset 1 refers to a sleep dataset in which user sleep data is continuously collected for a duration of TH2, during which there is no invalid sleep data, and which carries a sleep state consistency score label. The invalid sleep data refers to the sleep data with a duration of less than TH5 (e.g., 3 hours). The sleep state consistency score label refers to the sleep state consistency score given by multiple (e.g., 3) human sleep state experts to the user sleep data with a window duration of TH2. The scoring is based on the sleep consistency score scale of the last day within the window time TH2, the inherent sleep onset time of the last day, the inherent sleep midpoint of the last day, the inherent wake-up time of the last day, the sleep onset time of the last day, the sleep midpoint of the last day, and the wake-up time of the last day. The average of the multiple scoring results is calculated to obtain the final sleep state consistency score label.

[0055] In one embodiment of this application, the sleep structure assessment results are obtained in the following manner: Based on the sleep monitoring data, the duration of the current sleep stage and the percentage of the current sleep stage are obtained; Based on the intrinsic sleep data, the duration of intrinsic sleep stages and the percentage of intrinsic sleep stages are obtained. Based on the current sleep stage duration, current sleep stage percentage, inherent sleep stage duration, and inherent sleep stage percentage, a sleep structure consistency score is calculated using a preset weighted decay rule, which serves as the sleep structure evaluation result.

[0056] Optionally, from the current sleep monitoring data, extract the duration data (such as deep sleep duration and light sleep duration) and the corresponding percentage data (such as deep sleep percentage and light sleep percentage) corresponding to each sleep stage, including deep sleep, light sleep, REM sleep, and wakefulness. At the same time, from the intrinsic sleep data, retrieve the intrinsic sleep stage duration (such as intrinsic deep sleep duration and intrinsic light sleep duration) and the intrinsic sleep stage percentage (such as intrinsic deep sleep percentage and intrinsic light sleep percentage). Then, simultaneously substitute the duration and percentage of the current sleep stage and the duration and percentage of the intrinsic sleep stage into the preset weighted decay rule, and calculate the sleep structure consistency score. This score is the sleep structure assessment result.

[0057] In one embodiment of this application, the sleep structure consistency score is calculated based on the current sleep stage duration, the current sleep stage percentage, the inherent sleep stage duration, and the inherent sleep stage percentage, using a preset weighted decay rule, including: Based on the current sleep stage duration and the inherent sleep stage duration, the relative difference in sleep stage duration is determined; Based on the current sleep stage percentage and the inherent sleep stage percentage, the relative difference in sleep stage percentage is determined; The relative differences in the duration of the sleep stages and the relative differences in the proportion of the sleep stages are respectively processed by attenuation calculation; After weighted summation of the relative differences in sleep stage duration and the relative differences in sleep stage proportion after attenuation calculation, normalization is performed to obtain the sleep structure consistency score.

[0058] Optionally, for each sleep stage, such as deep sleep, light sleep, REM sleep, and wakefulness, the relative difference between the duration of the current sleep stage and the duration of the inherent sleep stage (e.g., the deviation ratio of deep sleep duration) and the relative difference between the proportion of the current sleep stage and the proportion of the inherent sleep stage (e.g., the deviation of the proportion of light sleep) are calculated. Then, attenuation operations are performed on these two types of differences to reduce the interference of extreme fluctuations on the evaluation results. Subsequently, according to a preset weighting rule (e.g., assigning higher weights to the differences in core stages such as deep sleep and REM sleep), the attenuated duration differences and proportion differences are weighted and summed. Finally, the summation result is normalized (e.g., mapped to a consistency score range of 0-1), and the final value is the sleep structure consistency score.

[0059] The sleep structure consistency score can be calculated using the following formula: ; Where i is the array index, and Preset weight parameters; For deep sleep duration, The duration of light sleep, REM duration Duration of wakefulness; For inherent deep sleep duration, Due to the inherent duration of light sleep, The inherent REM duration, The inherent duration of consciousness; For the percentage of deep sleep, The percentage of light sleep, REM percentage, The percentage of those who are sober; As the inherent percentage of deep sleep, The proportion of inherent light sleep, This represents the inherent REM ratio. The inherent percentage of sobriety.

[0060] It should be noted that, and It can be obtained by fitting a specific dataset 2. The fitting process refers to using the duration of each sleep stage, the proportion of each sleep stage, the inherent duration of each sleep stage, and the inherent proportion of each sleep stage as input features, based on the calculation form of sleep structure consistency score, using the sleep structure consistency score as the target value, and performing fitting or hyperparameter search.

[0061] Specifically, Specific Dataset 2 refers to a sleep dataset where the continuous collection duration of user sleep data reaches or exceeds TH2, there is no invalid sleep data during this period, and it carries a sleep structure consistency score label. The invalid sleep data refers to the corresponding sleep data with a sleep duration less than TH5. The sleep structure consistency score label is a sleep structure consistency score 3 manually labeled by multiple people (e.g., 3 people) on the sleep data within the window time TH2. The scoring basis is the sleep consistency score scale of the last day within the window time TH2, the duration of each sleep stage on the last day, the proportion of each sleep stage on the last day, the inherent duration of each sleep stage on the last day, and the inherent proportion of each sleep stage on the last day. The average of the multiple people's scoring results is calculated to obtain the final sleep structure consistency score label.

[0062] In one embodiment of this application, the sleep quality assessment result is obtained in the following manner: Based on the sleep monitoring data, the current sleep efficiency and current sleep continuity are obtained; Based on the intrinsic sleep data, intrinsic sleep efficiency and intrinsic sleep continuity are obtained; Determine the relative difference in sleep efficiency between the current sleep efficiency and the inherent sleep efficiency, and the relative difference in sleep continuity between the current sleep continuity and the inherent sleep continuity; After weighted summation of the relative differences in sleep efficiency and sleep continuity, normalization is performed to obtain a sleep quality consistency score, which is used as the sleep quality assessment result.

[0063] Optionally, core indicators reflecting sleep quality are extracted from current sleep monitoring data, including current sleep efficiency (e.g., the ratio of actual sleep time to bedtime) and current sleep continuity (e.g., the continuity of the number of awakenings and the total duration of awakenings during sleep). Simultaneously, the corresponding inherent sleep efficiency and inherent sleep continuity (i.e., the user's long-term stable sleep quality benchmark) are retrieved from the user's inherent sleep data. The relative difference between current sleep efficiency and inherent sleep efficiency (e.g., the deviation ratio of current efficiency from inherent efficiency) and the relative difference between current sleep continuity and inherent sleep continuity (e.g., the fluctuation range of current continuity from inherent continuity) are calculated respectively. Then, according to a preset weighting rule (e.g., assigning higher weight to sleep efficiency differences to match their core impact on sleep quality), these two types of relative differences are weighted and summed. Finally, the summation result is normalized to a scoring range of 0-1, and the resulting value is the sleep quality consistency score. This score can be directly used as the sleep quality assessment result; the closer the score is to 1, the higher the matching degree between the current sleep quality and the user's inherent sleep quality.

[0064] The sleep quality consistency score can be calculated using the following formula: ; in, , Preset weight parameters; For sleep efficiency; Inherent sleep efficiency; For sleep continuity; For inherent sleep continuity; It should be noted that, among them , It can be obtained by fitting a specific dataset 3. The fitting process refers to using the sleep efficiency, sleep continuity, inherent sleep efficiency, and inherent sleep continuity of the last day within the window time TH2 as input features, and using the sleep quality consistency score as the target value based on the calculation form of the sleep quality consistency score, to perform fitting or hyperparameter search.

[0065] Specifically, the specific dataset 3 refers to a sleep dataset in which user sleep data is continuously collected for a duration of TH2 or longer, during which there is no invalid sleep data, and which carries a sleep quality consistency score label. The invalid sleep data refers to the sleep data with a duration of less than TH5. The sleep quality consistency score label refers to the sleep quality consistency score manually labeled by multiple people (e.g., 3 people) on the sleep data within the window time TH2. The multiple people can be 3. The scoring criteria are the sleep quality consistency score scale of the last day within the window time TH2, the sleep efficiency of the last day, the sleep continuity of the last day, the inherent sleep efficiency of the last day, and the inherent sleep continuity of the last day. The average of the scores from multiple people is calculated to obtain the final sleep quality consistency score label.

[0066] In one embodiment of this application, the baseline sleep age value includes a first baseline sleep age value and a second baseline sleep age value. Determining the actual sleep age of the target user based on the baseline sleep age value and the sleep age deviation value includes: Based on the aforementioned basic personal data, the true age of the target user is determined; Determine the user usage frequency factor within a preset time range, wherein the user usage frequency factor includes the ratio of the number of effective sleeps to a preset threshold number of sleeps; Based on the first baseline sleep age value, the second baseline sleep age value, the actual age, and the sleep age deviation value, the actual sleep age is obtained by weighted fusion using the user usage frequency factor.

[0067] The first baseline sleep age value refers to the baseline sleep age value corresponding to the user's inherent sleep data at the current moment, calculated based on the user's inherent sleep data at the current moment (i.e., the user's long-term stable sleep characteristic data). The second baseline sleep age value refers to the baseline sleep age value corresponding to the user's inherent sleep data from the previous day, calculated based on the user's inherent sleep data from the previous day. Both reflect the reference level of sleep age corresponding to the user's inherent sleep state at different time points.

[0068] Optionally, the user's true age is determined based on their personal basic data (such as date of birth); simultaneously, the number of effective sleep episodes (i.e., the number of sleep records that meet the sleep duration threshold) within a preset time range (e.g., the most recent 30 days) is counted, and the ratio of this number to the preset threshold is calculated to obtain a user usage frequency factor used to measure data reliability; finally, the first sleep age baseline value (the sleep age benchmark value calculated based on the current moment's inherent sleep data), the second sleep age baseline value (the sleep age benchmark value calculated based on the previous day's inherent sleep data), the true age, and the sleep age deviation value are weighted and fused. During the fusion process, the weight of each parameter is dynamically adjusted through the user usage frequency factor (for example, when the user's usage frequency is high, the first and second sleep age baseline values ​​and the sleep age deviation value are given higher weights to more accurately reflect the current sleep state; when the usage frequency is low, the weight of the true age is appropriately increased to ensure the stability and reliability of the results), ultimately obtaining the target user's actual sleep age.

[0069] The actual sleep age can be calculated using the following formula: ; in, The frequency factor for user usage within the window time TH4 (i.e., the frequency of user usage within the time range from TH4 days before the current time to the current time) is calculated as follows: TH4 can be 30 days, 60 days, or 120 days. The calculation method is to divide the number of effective sleeps within the window time TH4 by TH6. The method for counting the number of effective sleeps is to consider the sleep data of the day as effective sleep data if the total sleep duration of the day exceeds TH5, and to increment the number of effective sleeps by one. TH6 can be TH4 / 2, or other values. The baseline sleep age value is obtained from the existing sleep data at the current moment. The baseline sleep age value is obtained from the previous day's intrinsic sleep data. This is the user's current real age.

[0070] In this embodiment, by acquiring the target user's inherent sleep data (generated by filtering historical sleep monitoring data), the memory bias of subjective assessment is avoided, and there is no need to rely on complex PSG equipment, thus achieving long-term convenient objective monitoring. At the same time, by using a two-layer logic of baseline sleep age value and deviation value, combined with the multi-dimensional differences of personal basic data, inherent sleep data and short-term sleep monitoring data, the degree of matching between the target user's sleep state and physiological age is accurately reflected. It not only reflects long-term sleep patterns, but also covers short-term sleep fluctuations, providing users with a more practical reference for improving sleep and maintaining health, and effectively making up for the shortcomings of existing assessment methods.

[0071] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0072] In one embodiment, a sleep age assessment system is provided, which corresponds one-to-one with the sleep age assessment methods described in the above embodiments. For example... Figure 6 As shown, the sleep age assessment system includes a sleep-related data acquisition unit 10, a sleep age baseline value determination unit 20, a sleep age deviation value determination unit 30, and an actual sleep age determination unit 40. Detailed descriptions of each functional module are as follows: The sleep-related data acquisition unit 10 is used to acquire user data and sleep monitoring data of the target user. The user data includes personal basic data and inherent sleep data. Personal basic data refers to data that reflects the user's own basic physiological characteristics, and inherent sleep data refers to stable sleep-related characteristic data that the user exhibits under normal conditions. The sleep age baseline value determination unit 20 is used to determine the sleep age baseline value of the target user based on the personal baseline data and inherent sleep data. The sleep age deviation value determination unit 30 is used to determine the sleep age deviation value of the target user based on the sleep monitoring data and the inherent sleep data. The actual sleep age determination unit 40 is used to determine the actual sleep age of the target user based on the baseline sleep age value and the sleep age deviation value.

[0073] In one embodiment of this application, the sleep age baseline value determination unit 20 is further configured to: Based on the aforementioned basic personal data and inherent sleep data, a feature array is constructed; The feature array is input into a preset sleep age prediction model for prediction processing to generate the target user's basic sleep age value.

[0074] In one embodiment of this application, the inherent sleep data includes at least one of the following: an inherent sleep state array, inherent sleep onset time, inherent wakefulness time, inherent sleep midpoint, inherent sleep structure data, and inherent sleep quality data. The inherent sleep state array is obtained in the following manner: Obtain the target user's historical sleep monitoring data; Based on the historical sleep monitoring data, an array of sleep states corresponding to multiple sleep cycles is generated; The sleep state array corresponding to each sleep cycle is filtered to generate the inherent sleep state array.

[0075] In one embodiment of this application, the sleep age deviation value determination unit 30 is further configured to: Based on the sleep monitoring data and the inherent sleep data, the sleep assessment result of the target user is determined. The sleep assessment result includes at least one of the sleep state assessment result, sleep structure assessment result, and sleep quality assessment result. The sleep confidence factor is determined based on the proportion of valid sleep data that meets the preset duration threshold within a preset time window. Based on the sleep assessment results and the sleep confidence factor, the sleep age deviation value is determined.

[0076] In one embodiment of this application, the sleep state assessment result is obtained in the following manner: Based on the sleep monitoring data, the current key sleep milestones are obtained; Based on the inherent sleep data, key nodes of inherent sleep are obtained; Based on the current sleep key nodes and the inherent sleep key nodes, determine the relative differences corresponding to each node; The relative differences corresponding to each node are processed by attenuation calculation; After weighted summation of the relative differences corresponding to each node after the decay operation, normalization is performed to obtain the sleep state consistency score, which is used as the sleep state evaluation result.

[0077] In one embodiment of this application, the sleep structure assessment results are obtained in the following manner: Based on the sleep monitoring data, the duration of the current sleep stage and the percentage of the current sleep stage are obtained; Based on the intrinsic sleep data, the duration of intrinsic sleep stages and the percentage of intrinsic sleep stages are obtained. Based on the current sleep stage duration, current sleep stage percentage, inherent sleep stage duration, and inherent sleep stage percentage, a sleep structure consistency score is calculated using a preset weighted decay rule, which serves as the sleep structure evaluation result.

[0078] In one embodiment of this application, the step of calculating a sleep structure consistency score based on the current sleep stage duration, the current sleep stage percentage, the inherent sleep stage duration, and the inherent sleep stage percentage, using a preset weighted decay rule, includes: Based on the current sleep stage duration and the inherent sleep stage duration, the relative difference in sleep stage duration is determined; Based on the current sleep stage percentage and the inherent sleep stage percentage, the relative difference in sleep stage percentage is determined; The relative differences in the duration of the sleep stages and the relative differences in the proportion of the sleep stages are respectively processed by attenuation calculation; After weighted summation of the relative differences in sleep stage duration and the relative differences in sleep stage proportion after attenuation calculation, normalization is performed to obtain the sleep structure consistency score.

[0079] In one embodiment of this application, the sleep quality assessment result is obtained in the following manner: Based on the sleep monitoring data, the current sleep efficiency and current sleep continuity are obtained; Based on the intrinsic sleep data, intrinsic sleep efficiency and intrinsic sleep continuity are obtained; Determine the relative difference in sleep efficiency between the current sleep efficiency and the inherent sleep efficiency, and the relative difference in sleep continuity between the current sleep continuity and the inherent sleep continuity; After weighted summation of the relative differences in sleep efficiency and sleep continuity, normalization is performed to obtain a sleep quality consistency score, which is used as the sleep quality assessment result.

[0080] In one embodiment of this application, the baseline sleep age value includes a first baseline sleep age value and a second baseline sleep age value, and the actual sleep age determination unit 40 is further configured to: Based on the aforementioned basic personal data, the true age of the target user is determined; Determine the user usage frequency factor within a preset time range, wherein the user usage frequency factor includes the ratio of the number of effective sleeps to a preset threshold number of sleeps; Based on the first baseline sleep age value, the second baseline sleep age value, the actual age, and the sleep age deviation value, the actual sleep age is obtained by weighted fusion using the user usage frequency factor.

[0081] In this embodiment, by acquiring the target user's inherent sleep data (generated by filtering historical sleep monitoring data), the memory bias of subjective assessment is avoided, and there is no need to rely on complex PSG equipment, thus achieving long-term convenient objective monitoring. At the same time, by using a two-layer logic of baseline sleep age value and deviation value, combined with the multi-dimensional differences of personal basic data, inherent sleep data and short-term sleep monitoring data, the degree of matching between the target user's sleep state and physiological age is accurately reflected. It not only reflects long-term sleep patterns, but also covers short-term sleep fluctuations, providing users with a more practical reference for improving sleep and maintaining health, and effectively making up for the shortcomings of existing assessment methods.

[0082] Specific limitations regarding the sleep age assessment system can be found in the limitations of the sleep age assessment method described above, and will not be repeated here. Each module in the aforementioned sleep age assessment system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0083] In one embodiment, a computer device is provided, which may be a terminal device, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a readable storage medium storing computer-readable instructions. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer-readable instructions implement a sleep age assessment method. The readable storage medium provided in this embodiment includes both non-volatile and volatile readable storage media.

[0084] In this application embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, it implements the steps of the sleep age assessment method described above.

[0085] In one embodiment of the application, a readable storage medium is provided, which stores computer-readable instructions. When the computer-readable instructions are executed by a processor, they implement the steps of the sleep age assessment method described above.

[0086] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware with computer-readable instructions. These computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When executed, these computer-readable instructions can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0087] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0088] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A sleep age assessment method, characterized by, The method comprises: obtaining user data and sleep monitoring data of a target user, the user data comprising personal basic data and inherent sleep data, wherein the personal basic data refers to data reflecting the basic physiological characteristics of the user, and the inherent sleep data refers to sleep-related characteristic data with stability exhibited by the user in a regular state; determining a sleep age base value of the target user based on the personal basic data and the inherent sleep data; determining a sleep age deviation value of the target user based on the sleep monitoring data and the inherent sleep data; determining an actual sleep age of the target user based on the sleep age base value and the sleep age deviation value.

2. The sleep age assessment method of claim 1, wherein, The determination of the sleep age base value of the target user based on the personal basic data and the inherent sleep data comprises: constructing a feature array based on the personal basic data and the inherent sleep data; inputting the feature array into a preset sleep age prediction model for prediction processing to generate the sleep age base value of the target user.

3. The sleep age assessment method of claim 1, wherein, The inherent sleep data comprises at least one of an inherent sleep state array, an inherent sleep time point, an inherent wake time point, an inherent sleep intermediate time point, inherent sleep structure data, and inherent sleep quality data, and the inherent sleep state array is obtained by: obtaining historical sleep monitoring data of the target user; generating sleep state arrays corresponding to multiple sleep cycles based on the historical sleep monitoring data; performing filtering processing on the sleep state arrays corresponding to each sleep cycle to generate the inherent sleep state array.

4. The sleep age assessment method of claim 1, wherein, The determination of the sleep age deviation value of the target user based on the sleep monitoring data and the inherent sleep data comprises: determining a sleep evaluation result of the target user based on the sleep monitoring data and the inherent sleep data, the sleep evaluation result comprising at least one of a sleep state evaluation result, a sleep structure evaluation result, and a sleep quality evaluation result; determining a sleep confidence factor based on a proportion of effective sleep data satisfying a preset time length threshold within a preset time window; determining the sleep age deviation value based on the sleep evaluation result and the sleep confidence factor.

5. The sleep age assessment method of claim 4, wherein, The sleep state evaluation result is obtained by: obtaining a current sleep key node based on the sleep monitoring data; obtaining an inherent sleep key node based on the inherent sleep data; determining relative differences corresponding to each node based on the current sleep key node and the inherent sleep key node; performing attenuation operation processing on the relative differences corresponding to each node; performing weighted summation on the relative differences corresponding to each node after the attenuation operation processing, and then performing normalization processing to obtain a sleep state consistency score as the sleep state evaluation result.

6. The sleep age assessment method of claim 4, wherein, The sleep structure evaluation result is obtained by: obtaining a current sleep stage duration and a current sleep stage proportion based on the sleep monitoring data; obtaining an inherent sleep stage duration and an inherent sleep stage proportion based on the inherent sleep data; The sleep structure consistency score is calculated based on the current sleep staging time length, the current sleep staging proportion, the inherent sleep staging time length, and the inherent sleep staging proportion through a preset weighted attenuation rule, and is taken as the sleep structure evaluation result.

7. The sleep age assessment method of claim 6, wherein, The sleep structure consistency score is calculated based on the current sleep staging time length, the current sleep staging proportion, the inherent sleep staging time length, and the inherent sleep staging proportion through a preset weighted attenuation rule, and includes: A sleep staging time length relative difference is determined based on the current sleep staging time length and the inherent sleep staging time length; A sleep staging proportion relative difference is determined based on the current sleep staging proportion and the inherent sleep staging proportion; The sleep staging time length relative difference and the sleep staging proportion relative difference are respectively subjected to attenuation operation processing; After the sleep staging time length relative difference and the sleep staging proportion relative difference subjected to the attenuation operation processing are weighted and summed, normalization processing is performed to obtain the sleep structure consistency score.

8. The sleep age assessment method of claim 4, wherein, The sleep quality evaluation result is obtained in the following manner: A current sleep efficiency and a current sleep continuity are obtained based on the sleep monitoring data; An inherent sleep efficiency and an inherent sleep continuity are obtained based on the inherent sleep data; A sleep efficiency relative difference of the current sleep efficiency and the inherent sleep efficiency, and a sleep continuity relative difference of the current sleep continuity and the inherent sleep continuity are determined; After the sleep efficiency relative difference and the sleep continuity relative difference are weighted and summed, normalization processing is performed to obtain a sleep quality consistency score as the sleep quality evaluation result.

9. The sleep age assessment method of any one of claims 1-8, wherein, The sleep age base value includes a first sleep age base value and a second sleep age base value, and the actual sleep age of the target user is determined based on the sleep age base value and the sleep age deviation value, including: A real age of the target user is determined based on the personal base data; A user usage frequency factor in a preset time range is determined, the user usage frequency factor including a ratio of an effective sleep number to a preset number threshold; The actual sleep age is obtained by weighted fusion of the first sleep age base value, the second sleep age base value, the real age, the sleep age deviation value, and the user usage frequency factor.

10. A sleep age assessment system, characterized by, The system includes: A sleep-related data acquisition unit is configured to acquire user data and sleep monitoring data of a target user, the user data including personal base data and inherent sleep data, wherein the personal base data refers to data reflecting basic physiological characteristics of a user, and the inherent sleep data refers to sleep-related characteristic data that is stable and exhibited by a user in a regular state; A sleep age base value determination unit is configured to determine a sleep age base value of the target user based on the personal base data and the inherent sleep data; A sleep age deviation value determination unit is configured to determine a sleep age deviation value of the target user based on the sleep monitoring data and the inherent sleep data; The actual sleep age determining unit is configured to determine an actual sleep age of the target user based on the sleep age base value and the sleep age deviation value.

Images