A recovery degree detection method and a wearable device

Abstract

This application provides a method and wearable device for detecting recovery degree, relating to the field of smart wearable device technology. The method, after acquiring sleep monitoring data and reference data, determines recovery evaluation indicators from the sleep monitoring data, then calculates core indicator scores based on the reference data and core indicators, and calculates cardiopulmonary function scores based on the reference data and cardiopulmonary function indicators. Finally, it calculates a recovery degree score based on the core indicator scores and cardiopulmonary function scores. The method uses physiological parameters collected by the wearable device to determine recovery evaluation indicators, including core indicators and cardiopulmonary function indicators, and calculates the core indicator scores and cardiopulmonary function scores respectively. Since the monitoring process only uses data directly collected by the wearable device or data fitted based on the collected data, the method can reduce the number of indicators required for recovery degree detection and improve the efficiency of recovery degree detection.

Description

Technical Field

[0001] This application relates to the field of smart wearable device technology, and in particular to a method for detecting resilience and a wearable device. Background Technology

[0002] Recovery rate, also known as physical recovery rate or exercise recovery rate, is a parameter of physical condition. Recovery rate refers to the degree to which various physiological functions and functional states of the body return to baseline levels after exercise training, physical labor, or other physical exertion. Recovery rate can be used to quantitatively assess the extent to which a user's body recovers from an exercise load or fatigue state to a normal state, providing a reference for users to rationally arrange rest and training processes.

[0003] To assess recovery progress, smart wearable devices or medical equipment can be used to measure data such as the user's daily exercise intensity, multiple physiological parameters, and subjective user evaluations. A comprehensive recovery score can then be calculated based on this data. For example, blood biochemical indicators such as creatine kinase, blood urea nitrogen, and cortisol ratio can be measured, combined with neuromuscular function tests such as vertical jump tests, grip strength tests, and isokinetic muscle strength tests, along with the user's subjective evaluation of their recovery, to comprehensively assess the recovery score.

[0004] However, in the aforementioned recovery rate testing process, the physiological parameters involved in the assessment are difficult to obtain directly and require medical testing, resulting in low efficiency and poor real-time performance. Furthermore, the large number of data indicators required to assess the recovery rate score, and the asynchronous testing processes for some indicators, limit the applicability of the recovery rate testing. Summary of the Invention

[0005] In view of this, embodiments of this application provide a method and wearable device for restoring strength detection to solve the problem of too many data indicators required for restoring strength detection.

[0006] According to a first aspect of this application, a method for detecting resilience is provided, the method comprising: Acquire sleep monitoring data and reference data; the sleep monitoring data includes physiological parameters collected during the user's sleep. Recovery evaluation indicators are determined from the sleep monitoring data. These indicators include core indicators and cardiopulmonary function indicators. The core indicators are used to characterize the degree of recovery and sleep quality during sleep, while the cardiopulmonary function indicators are used to characterize the degree of cardiopulmonary capacity recovery after sleep. The core indicator score is calculated based on the reference data and the core indicator, and the cardiopulmonary function score is calculated based on the reference data and the cardiopulmonary function indicator. The recovery score is calculated based on the core indicator score and the cardiopulmonary function score.

[0007] In some embodiments, acquiring sleep monitoring data and reference data includes: The latest data monitored during sleep behavior is used as the sleep monitoring data; If historical sleep data is stored, then the historical sleep data is extracted as the reference data; If historical sleep data is not stored, then the preset general data corresponding to the custom settings data is obtained as the reference data.

[0008] In some embodiments, the core metrics include sleep heart rate variability parameters and sleep scores; The cardiopulmonary function indicators include resting heart rate and / or respiratory rate.

[0009] In some embodiments, the sleep heart rate variability parameter is obtained as follows: Heart rate parameters are extracted from the sleep monitoring data, including the intervals between heartbeats monitored during the user's sleep. The variability evaluation value is calculated based on the heart rate parameters. The variability evaluation value includes at least one of the following: standard deviation of heartbeat interval, mean squared deviation of heartbeat interval difference, and proportion of heartbeat interval difference exceeding a preset difference threshold. The sleep heart rate variability parameter is determined based on the variability evaluation value.

[0010] In some embodiments, the sleep score is obtained as follows: Sleep evaluation indicators are extracted from the sleep monitoring data, including sleep duration and sleep structure indicators. Obtain the sleep score calculation weights, which include duration weights and sleep structure weights; The sleep score is calculated based on the sleep score calculation weights and the sleep evaluation indicators. The sleep score is a numerical value obtained by weighting the sleep evaluation indicators using the score calculation weights.

[0011] In some embodiments, calculating the sleep score based on the sleep score calculation weight and the sleep evaluation index includes: The recommended sleep duration is obtained and is correlated with the user's daily stress level and / or sleep debt value. A sleep duration score is calculated based on the sleep duration, the recommended sleep duration, and the duration weight. The sleep duration score is equal to the ratio of the sleep duration to the recommended sleep duration multiplied by the duration weight. The product of the quantitative value of the sleep structure index and the sleep structure weight is calculated to obtain the sleep structure score; The sleep score is obtained by summing the sleep duration score and the sleep structure score.

[0012] In some embodiments, the resting heart rate is obtained as follows: Extract body movement parameters from the sleep monitoring data; The resting state period is determined based on the body motion parameters, and the resting state period includes the start time of the resting state and the end time of the resting state; Obtain the heart rate parameters during the resting state period, and calculate the resting heart rate based on the heart rate parameters during the resting state period.

[0013] In some embodiments, the respiratory rate is obtained as follows: Respiratory correlation parameters are extracted from the sleep monitoring data, including photoplethysmography parameters and body movement parameters. The respiratory correlation parameters are input into a respiratory algorithm model to perform multimodal fusion of the photoplethysmography parameters and the body motion parameters, and to perform classification calculations based on the multimodal fusion results; the respiratory algorithm model is a deep learning model trained using sample respiratory data; the sample respiratory data are respiratory correlation parameters labeled with respiratory rate. Obtain the respiratory rate output by the respiratory algorithm model.

[0014] In some embodiments, the core metric score is obtained in the following manner: A heart rate variability baseline is determined based on the reference data, wherein when the reference data is historical sleep data, the heart rate variability baseline is calculated based on the heart rate parameters in the historical sleep data; when the reference data is the preset general data, the heart rate variability baseline is a general baseline parameter. Obtain the deviation of the sleep heart rate variability parameter from the heart rate variability baseline, and calculate the heart rate variability score based on the deviation; The core indicator score is calculated based on the heart rate variability score and the sleep score.

[0015] In some embodiments, calculating the cardiopulmonary function score based on the reference data and the cardiopulmonary function indicators includes: The cardiopulmonary function baseline is determined based on the reference data, wherein when the reference data is historical sleep data, the cardiopulmonary function baseline is obtained statistically based on the resting heart rate and / or respiratory rate in the historical sleep data; when the reference data is the preset general data, the cardiopulmonary function baseline is a general baseline parameter. Cardiopulmonary function scores are calculated based on the baseline cardiopulmonary function indicators and the cardiopulmonary function indicators themselves. The cardiopulmonary function scores include resting heart rate score and / or respiratory rate score.

[0016] In some embodiments, the baseline cardiopulmonary function indicators are obtained as follows: The set of cardiopulmonary parameters located within the analysis period in the reference data is traversed, and the set of cardiopulmonary parameters includes at least one of resting heart rate parameters and respiratory rate parameters; Calculate the average parameter value of the cardiopulmonary parameter set, wherein the average parameter value includes at least one of average resting heart rate and average respiratory rate; A baseline for cardiopulmonary function indicators is set based on the average parameter values, and the baseline for cardiopulmonary function indicators includes at least one of a resting heart rate baseline and a respiratory rate baseline.

[0017] In some embodiments, calculating the cardiopulmonary function score based on the baseline cardiopulmonary function index and the cardiopulmonary function index includes: Calculate the first deviation of the resting heart rate relative to the resting heart rate baseline; If the resting heart rate is lower than the resting heart rate baseline, and the first deviation value is less than or equal to a preset resting heart rate deviation judgment threshold, the resting heart rate score is calculated according to the first deviation value within a first value range; the minimum value of the first value range is greater than or equal to 0. If the resting heart rate is higher than the resting heart rate baseline, or the first deviation value is greater than the preset resting heart rate deviation judgment threshold, the resting heart rate score is calculated according to the first deviation value within a second value range; the maximum value of the second value range is less than 0; And / or, Calculate the second deviation of the respiratory rate relative to the respiratory rate baseline; If the respiratory rate is lower than the respiratory rate baseline, and the second deviation value is less than or equal to a preset respiratory rate deviation judgment threshold, the respiratory rate score is calculated according to the two deviation values ​​within a first value range; the minimum value of the first value range is greater than or equal to 0. If the respiratory rate is higher than the respiratory rate baseline, or the two deviation values ​​are greater than the preset respiratory rate deviation judgment threshold, the respiratory rate score is calculated according to the two deviation values ​​within a second value range; the maximum value of the second value range is less than 0.

[0018] In some embodiments, the method further includes: In response to a preset operation to display the recovery score, a target interface corresponding to the recovery score is displayed, the target interface including preset elements corresponding to the recovery score; The display strategy of the preset element is set according to the recovery score; Based on the display strategy, the preset elements are displayed in the target interface.

[0019] According to a second aspect of this application, a wearable device is provided, comprising: Biosensors are configured to collect physiological parameters; A controller is connected to the biosensor; the controller is configured to: Acquire sleep monitoring data and reference data; the sleep monitoring data includes physiological parameters collected during the user's sleep. Based on the sleep monitoring data, recovery evaluation indicators are determined. These indicators include core indicators and cardiopulmonary function indicators. The core indicators include sleep heart rate variability parameters and sleep scores. The cardiopulmonary function indicators include resting heart rate and / or respiratory rate. The core indicator score is calculated based on the reference data and the core indicator, and the cardiopulmonary function indicator score is calculated based on the reference data and the cardiopulmonary function indicator. The recovery score is calculated based on the scores of the core indicators and the cardiopulmonary function indicators.

[0020] By employing the above technical solutions, embodiments of this application provide a method and wearable device for detecting recovery degree. The method, after acquiring sleep monitoring data and reference data, determines recovery evaluation indicators from the sleep monitoring data, then calculates core indicator scores based on the reference data and core indicators, and calculates cardiopulmonary function scores based on the reference data and cardiopulmonary function indicators. Finally, it calculates a recovery degree score based on the core indicator scores and cardiopulmonary function scores. The method uses physiological parameters collected by the wearable device to determine recovery evaluation indicators, including core indicators and cardiopulmonary function indicators, and calculates the core indicator scores and cardiopulmonary function scores respectively. Since the monitoring process only uses data directly collected by the wearable device or data fitted based on the collected data, the method can reduce the number of indicators required for the recovery degree detection process and improve the detection efficiency of recovery degree.

[0021] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0022] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1This is a schematic diagram of the wearable device structure provided in an embodiment of this application; Figure 2 This is a schematic diagram of the recovery detection method provided in the embodiments of this application; Figure 3 This is a schematic diagram of the entire recovery detection process provided in the embodiments of this application; Figure 4 This is a schematic diagram of the recovery score display process provided in the embodiments of this application; Figure 5 This is a schematic diagram illustrating the display effect of the recovery score provided in an embodiment of this application. Detailed Implementation

[0023] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0024] In this application embodiment, recovery degree, also known as physical recovery degree or exercise recovery degree, is a physical state parameter. Recovery degree refers to the degree to which various physiological functions and functional states of the body return to baseline levels after experiencing exercise training, physical labor, or other physical exertion. Recovery degree can be used to quantitatively assess the extent to which a user's body recovers from an exercise load or fatigue state to a normal state, providing a reference for users to reasonably arrange rest and training processes.

[0025] Recovery rate detection is a data processing and data generation process that involves detecting a user's physiological parameters and calculating quantitative indicators of recovery based on the detection results. In some embodiments, to detect recovery rate, smart wearable devices or medical devices can be used to detect data such as the user's daily exercise intensity, multiple physiological parameters, and the user's subjective evaluation, and a comprehensive recovery rate score can be calculated based on the detection data.

[0026] For example, blood biochemical indicators such as creatine kinase, blood urea nitrogen, and cortisol ratio can be tested on users, and neuromuscular function tests such as vertical jump test, grip strength test, and isokinetic muscle strength test can be combined with the user's subjective evaluation of their own recovery to comprehensively assess the recovery score.

[0027] Wearable devices, also known as smart wearable devices, are portable devices that can be worn directly on a user or other wearer, or integrated into the wearer's clothing or accessories. Examples of wearable devices include, but are not limited to, smartwatches, smart bracelets, smart rings, smart glasses, and smart head-mounted devices. Wearable devices can not only detect physiological data during exercise, but also provide users with references to their physical condition, such as vitality scores, exercise readiness scores, or recovery scores.

[0028] like Figure 1 As shown, wearable devices can possess data monitoring, data processing, and communication capabilities. To this end, wearable devices can integrate sensors, controllers, and communicators internally. Sensors are used to collect users' health and behavioral data. Depending on the type and function of the wearable device, multiple sensors of various types can be integrated within it, forming a sensor suite. For example, a sensor suite may include accelerometers, gyroscopes, barometers, altimeters, heart rate sensors, temperature sensors, and electroactivity sensors.

[0029] The controller built into wearable devices can be used to process data collected by various sensors, control the operating status of the wearable device, and execute user commands. For example, the controller of a wearable device can be a microcontroller unit (MCU). The data processed by the controller can include internal data of the wearable device as well as data transmitted and received by the wearable device through communication connections.

[0030] The controller of a wearable device can perform data processing on the terminal side to convert the raw signals collected by sensors into monitoring data for specific purposes. For example, the controller can receive photoelectric signals collected by a photoplethysmography (PPG) sensor, extract pulse features from the photoelectric signals by running a heart rate detection algorithm, and then calculate heart rate variability (HRV) based on the pulse features to measure the fluctuation of heart rate intervals. As another example, after acquiring photoelectric signals, the controller can also extract subtle fluctuations in blood flow from the photoelectric signals by running a respiratory detection algorithm, and then infer the user's current respiratory rate (RR) based on the functional relationship between these subtle fluctuations and respiration to determine the number of breaths per unit time.

[0031] The controller of wearable devices can also fuse data collected by multiple sensors and make comprehensive judgments to obtain more diverse or more accurate physiological parameter detection results. For example, the controller can simultaneously receive body motion signals collected by an accelerometer and photoelectric signals collected by a PPG sensor, and determine whether the user is in a resting state through the body motion signals and photoelectric signals, and extract pulse characteristics from the photoelectric signals in the resting state to calculate the resting heart rate (RHR) to measure the number of heartbeats per unit time in a completely relaxed and undisturbed state.

[0032] The communicator in wearable devices enables data communication between wearable devices and between wearable devices and other devices. The communicator can include various types of communication modules depending on the supported communication methods. For example, communicators may include Bluetooth modules, GPS modules, and NFC modules. The Bluetooth module is used to connect with smartphones and other mobile terminals to achieve data synchronization, receive notifications, and control mobile terminal functions. The GPS module is used independently for location positioning and motion tracking. The NFC module is used for near-field wireless communication functions, such as payments and access cards.

[0033] Because the physiological parameters involved in the recovery assessment process are difficult to obtain directly and require medical testing, the detection efficiency and real-time performance are low. Furthermore, the large number of data indicators required to assess the recovery score, and the asynchronous testing processes for some indicators, limit the applicability of the recovery assessment.

[0034] To address the issue of excessive data metrics required for recovery assessment, some embodiments of this application provide a recovery assessment method. This method uses heart rate variability (HRV) values ​​during sleep monitoring as the highest-weighted recovery evaluation indicator, comparing them with a long-term HRV baseline to determine the recovery level. Simultaneously, it combines baseline deviations from sleep scores, resting heart rate (RHR), and respiratory rate (RR) to comprehensively calculate the user's recovery level. This method can determine the user's recovery level using fewer data metrics, can be used with more wearable devices, and has a wider range of applications.

[0035] The method can be applied to wearable devices or electronic devices connected to wearable devices and possessing data processing capabilities. These electronic devices include, but are not limited to, computers, servers, mobile terminals, and smart wearable devices. For ease of description, this application embodiment uses a wearable device as the execution subject of the method. It should be understood that the method can also be applied to other types of execution subjects, which are not illustrated in this application embodiment. Figure 2 As shown, the method includes: S101. Obtain sleep monitoring data and reference data.

[0036] During recovery testing, sleep monitoring data and reference data can be acquired. The sleep monitoring data includes physiological parameters collected during the user's sleep. When a user wears a smartwatch or other wearable device while sleeping, the device can identify the user's sleep state through a set of physiological parameters collected by its built-in sensors. In this case, the set of physiological parameters collected by the biosensors can be used as sleep monitoring data.

[0037] Taking wearable devices such as smartwatches as an example, the sensors built into these devices can include inertial sensors, optical sensors, and potential sensors. Inertial sensors can include devices such as accelerometers and gyroscopes, forming a motion monitoring module to monitor the user's movement, posture, and other physiological parameters. Optical sensors can include photoplethysmography (PPG) sensors to monitor physiological parameters such as heart rate and blood oxygenation during wear. Potential sensors can include electrocardiogram (ECG) electrodes to monitor skin conductance signals and other physiological parameters during wear. Furthermore, wearable devices can also incorporate environmental sensors, such as temperature sensors, barometric pressure sensors, and humidity sensors, to detect information about the user's environment.

[0038] In practical implementation, acquiring sleep monitoring data can refer to wearable devices acquiring a set of physiological data of the user during sleep through sensors, or it can refer to a mobile terminal connected to the wearable device acquiring an existing set of physiological data from the wearable device. Here, sleep monitoring data includes a set of physiological parameters collected during the user's sleep. When the user wears the wearable device, the device can identify the user's activity status in real time, and then, upon determining that the user has entered sleep, compile the collected physiological data set into sleep monitoring data. In actual implementation, the wearable device can identify the set of physiological data collected by the sensors as sleep monitoring data each time it detects that the user has entered sleep, for use in sleep quality monitoring. Alternatively, the set of physiological data collected by the sensors can be identified as sleep monitoring data and sent to the linked mobile terminal for sleep quality monitoring.

[0039] Sleep monitoring data can be a set of physiological parameters collected in real time by wearable devices, or a set of physiological parameters accumulated by wearable devices over a specific period. For physiological parameters accumulated over a specific period, a data acquisition request can be sent to the wearable device or the cloud server connected to the wearable device when acquiring sleep monitoring data. Upon receiving the data acquisition request, the wearable device or cloud server can extract the set of physiological parameters for the corresponding monitoring period from local storage or cloud storage services, and feed the extracted set of physiological parameters back to the executing device as sleep monitoring data.

[0040] It should be noted that when a user wears a wearable device, the device can collect the user's physiological data in real time or periodically. The wearable device can identify whether the user is in a state of motion, rest, or sleep based on the collected physiological data. In all embodiments of this application, sleep monitoring data refers to data collected by the wearable device during the user's sleep process. That is, when the wearable device determines that the user is asleep based on the collected physiological data, the data collected during sleep can be used as sleep monitoring data.

[0041] It's easy to understand that data collected during movement and data collected at rest can be considered non-sleep monitoring data. Unlike non-sleep monitoring data, sleep monitoring data exhibits a certain temporal sequence and fluctuations in sleep states. Specifically, during sleep, users typically switch between REM sleep, light sleep, deep sleep, and brief awakenings. Based on this, in practical implementation, biological data characteristics can be used to identify whether a user is asleep or not, and then the physiological data collected during sleep can be combined into sleep monitoring data.

[0042] Reference data can be general data corresponding to historical monitoring data or custom settings. Historical monitoring data refers to physiological parameters collected by the wearable device within a preset monitoring period; that is, historical monitoring data is a set of physiological parameters collected by the wearable device over a relatively long monitoring period. Obviously, the monitoring period corresponding to historical monitoring data should be longer than the user's sleep duration, and its monitoring period should cover at least one complete cycle of sleep and non-sleep periods. Therefore, historical monitoring data includes not only physiological parameters collected by the wearable device during the user's sleep period over a past period, but also physiological parameters collected by the wearable device during the user's non-sleep period over a longer period.

[0043] For example, if the monitoring period corresponding to the historical monitoring data is 30 days, the historical monitoring data can include physiological parameters such as heart rate, body movement, and body surface temperature collected by the wearable device during the user's daily activities during the day; and the historical monitoring data can also include physiological parameters such as heart rate, body movement, and body surface temperature collected by the wearable device during the user's sleep at night.

[0044] Since the monitoring period corresponding to historical monitoring data is relatively long, the amount of raw data corresponding to historical monitoring data is huge. In order to reduce the amount of data processing in the recovery detection process, in some embodiments, historical monitoring data may only contain the necessary key raw data and the monitoring result data obtained by data transformation based on the raw data.

[0045] For example, historical monitoring data may not include all raw electrical signals collected by the smartwatch's built-in sensors over the past 30 days. Instead, it may include monitoring results obtained through parameter analysis and calculation based on the raw electrical signals, such as historical heart rate variability, historical sleep scores, historical resting heart rate, and historical respiratory rate. Furthermore, this monitoring result data can be sampled and stored in association according to specific monitoring time points, thereby reducing the overall data volume of historical monitoring data.

[0046] In order to obtain historical monitoring data, the execution device of the sleep quality monitoring method can respond to the recovery degree detection command, generate a historical data acquisition request, and send the historical data acquisition request to the storage medium of historical monitoring data, such as the memory of wearable devices, mobile terminals, and cloud servers, so that the storage medium can query the historical monitoring data within the corresponding analysis period according to the historical data acquisition request, and feed back the historical monitoring data to the execution device, so that the execution device can acquire the historical monitoring data.

[0047] Custom settings data are data pre-set by users based on their personalized needs. For example, custom settings data may include information representing a user's personalized needs, such as gender and age. Custom settings data can be recorded in a settings file and stored in a user database. When retrieving reference data, a data retrieval request can be sent to the user database to extract the settings file and read the custom settings items from the settings file, thereby obtaining the custom settings data.

[0048] Custom settings data can be used to obtain preset general data. This preset general data is monitoring data extracted by the wearable device operator from multiple wearable device monitoring data sets based on the user characteristics of the target users of the relevant products. This data is correlated with settings such as user gender and age in the custom settings data. General data can be used as reference data for recovery degree detection in the early stages of user use when insufficient historical monitoring data has been collected. Therefore, in some embodiments, when obtaining sleep monitoring data and reference data, the latest data monitored during sleep behavior can be obtained as sleep monitoring data. If historical sleep data is stored, it is extracted as the reference data; if no historical sleep data is stored, the preset general data corresponding to the custom settings data is obtained as the reference data.

[0049] For example, after obtaining user A's sleep monitoring data, the user ID (000A) can be retrieved from the sleep monitoring data. Based on the user ID, historical monitoring data can be queried in the user database. If the user database contains relevant historical monitoring data for user A, this historical monitoring data can be extracted as reference data. However, if the user database does not contain historical monitoring data for user A, or if the amount of historical monitoring data for user A is insufficient to meet the recovery rate detection requirements (e.g., the monitoring period corresponding to user A's historical monitoring data is less than 7 days), then user A's custom settings data can be queried from the user database to extract general data. This means obtaining monitoring data from other users with similar characteristics to user A, such as age, gender, occupation, and body type, as reference data.

[0050] It should be noted that, to ensure user privacy and security, the general data obtained when acquiring reference data is security monitoring data that has undergone privacy and security processing. That is, when wearable devices generate general data, they only upload monitoring results to the database, without uploading user names, user IDs, or other identifying information, and user authorization is required before uploading relevant monitoring data.

[0051] S102. Determine recovery evaluation indicators from sleep monitoring data.

[0052] like Figure 3 As shown, after acquiring sleep monitoring data, recovery evaluation indicators can be determined from the sleep monitoring data. These recovery evaluation indicators include core indicators and cardiopulmonary function indicators. Core indicators characterize the degree of recovery and sleep quality during sleep; that is, core indicators are those that have a major impact on the recovery score when calculating the recovery score. In some embodiments, core indicators include sleep heart rate variability (HRV) parameters. As a core judgment criterion, sleep HRV parameters have a veto power in the recovery detection process; if this sleep HRV parameter is poor, the recovery is definitely poor. For example, if the HRV value measured in this sleep study has not recovered to the HRV baseline range, it is determined that the recovery is not good, i.e., the recovery score is low.

[0053] In some embodiments, sleep heart rate variability parameters are obtained as follows: heart rate parameters are extracted from sleep monitoring data, including heartbeat intervals monitored during the user's sleep. A variability evaluation value is then calculated based on the heart rate parameters, wherein the variability evaluation value includes at least one of the following: standard deviation of heartbeat intervals, root mean square deviation of heartbeat interval differences, and the proportion of heartbeat interval differences exceeding a preset difference threshold. Sleep heart rate variability parameters are then generated based on the variability evaluation value.

[0054] For example, the fluctuation signal of the sinus interval can be measured using a PPG photoplethysmography sensor built into wearable devices such as smartwatches. A preprocessing threshold can be set to preprocess the fluctuation signal to remove outliers. This preprocessing threshold is used to define an interval outside ±30% of the mean as abnormal. Then, based on the time-domain signal characteristics, a variability evaluation value is calculated. For example, the standard deviation of normal-to-normal intervals (SDNN) can be calculated using the following formula:

[0055] Where SDNN represents the standard deviation of the heartbeat interval; RR i This represents the i-th heartbeat interval; This represents the average heartbeat interval; N represents the number of heartbeat intervals monitored.

[0056] Similarly, variability assessment values ​​can also be the root mean square of the differences between successful RR intervals (RMSSD) and the percentage of adjacent NN intervals differing by more than 50 milliseconds (pNN50). By collecting the user's normal heart rate intervals during sleep and calculating the variability assessment values, sleep heart rate variability parameters can be generated, reflecting the user's overall autonomic nervous system activity, parasympathetic (vagus nerve) activity, etc.

[0057] The sleep score is calculated by weighting sleep-related physiological parameters from sleep monitoring data. In some embodiments, when determining recovery evaluation indicators from sleep monitoring data, sleep evaluation indicators can be extracted from the sleep monitoring data first. These sleep evaluation indicators include sleep duration and sleep structure indicators. Sleep duration refers to the length of time a user sleeps, including total sleep time and the duration of various sleep types during sleep, such as deep sleep, light sleep, and rapid eye movement (REM) sleep.

[0058] To obtain sleep duration, physiological parameters, including heart rate and body movement parameters, can be extracted from sleep monitoring data. Then, sleep type determination criteria are obtained to identify the sleep type at each stage of the entire sleep process. These criteria include physiological parameter ranges for different sleep types.

[0059] For example, wearable devices can determine the sleep type corresponding to a user's sleep process based on heart rate and body movement monitoring results. To this end, wearable devices can use accelerometers to sense the amplitude and frequency of movements such as turning over and arm movements, use photoelectric sensors (PPG) to monitor heart rate changes, and determine the user's sleep state based on the physiological parameter range to which the monitoring results belong. For instance, in deep sleep, the user's body is almost still, and the heart rate drops to its lowest and remains stable; while in REM sleep, although the body is also still, the heart rate becomes irregular; the amplitude and frequency of body movements, as well as the heart rate, in light sleep fall between those of deep sleep and REM sleep. When both the heart rate and body movement parameters in the monitoring results fall within the heart rate and body movement ranges corresponding to deep sleep, it can be determined that the user is currently in deep sleep.

[0060] While determining deep sleep, wearable devices can also record the start and end times of heart rate and body movement parameters that meet the physiological parameter ranges for the corresponding sleep type, thus determining the duration and key moments of each sleep state. For example, in sleep monitoring data monitored at time T1, if the heart rate parameter is within the heart rate range of deep sleep, and the amplitude and frequency of body movements are within the body movement range of deep sleep, then the start time of deep sleep can be recorded as T1. Similarly, if the heart rate parameter is not within the heart rate range of deep sleep, or the amplitude and frequency of body movements are not within the body movement range of deep sleep, then the end time of deep sleep can be recorded as T2, and the corresponding duration of the current deep sleep state is (T2-T1).

[0061] Because the sleep process involves multiple cycles between deep sleep, light sleep, and REM sleep, the entire sleep process can be divided into multiple sleep periods based on the physiological parameter ranges to which the physiological parameters belong. Each sleep period is labeled with a corresponding sleep type, such as deep sleep, light sleep, and REM sleep. The cumulative duration of each sleep period is then calculated based on the sleep type to obtain the total sleep duration.

[0062] For example, a complete sleep cycle can consist of 4-6 cycles of approximately 90 minutes each. Within each 90-minute cycle, light sleep, deep sleep, and REM sleep occur in a fixed order. Therefore, when calculating sleep duration, it's necessary to calculate the duration of each sleep state (light sleep, deep sleep, and REM sleep) within each cycle, and then sum the durations corresponding to the same sleep type to obtain the sleep duration for each type: deep sleep duration, light sleep duration, and REM sleep duration. Finally, the sleep durations for each sleep type are summed to obtain the total sleep duration.

[0063] Sleep structure indicators can include sleep regularity indicators and sleep effectiveness indicators. Sleep regularity indicators reflect the regularity of a user's sleep process and can characterize a user's long-term sleep habits through the consistency of sleep data over a long period. To extract sleep regularity indicators, in some embodiments, sleep time information can first be extracted from sleep monitoring data based on physiological parameters. This sleep time information includes key sleep moments and sleep periods. Key sleep moments include the time of falling asleep and the time of waking up. Sleep periods are the time between falling asleep and waking up.

[0064] Next, historical monitoring data is acquired. This historical monitoring data includes historical time information recorded within a preset monitoring period, corresponding to the user's actual sleep time information. The historical time information includes historical key moments and average sleep duration. Historical key moments include average sleep onset time and average wake-up time. Both average sleep onset time and average wake-up time can be calculated by averaging the sleep onset time and wake-up time over multiple monitoring periods. The average sleep duration is the period between the average sleep onset time and the average wake-up time.

[0065] Then, regularity indicators are calculated based on sleep time information and historical monitoring data. These regularity indicators include indicators of the dispersion of key sleep moments relative to historical monitoring data. For example, regularity indicators may include the time difference between the time of falling asleep and the average time of falling asleep, and the variance of the time of falling asleep relative to multiple times of falling asleep in historical monitoring data, to reflect the degree of dispersion of key sleep moments relative to historical monitoring data.

[0066] Sleep effectiveness indicators are used to characterize the continuity and quality of sleep. Since deep sleep and REM sleep are most conducive to physical and mental recovery, sleep effectiveness indicators can reflect the continuity and quality of sleep by the proportion of deep sleep and REM sleep duration and the number of sleep interruptions.

[0067] To extract sleep effectiveness indicators, in some embodiments, the sleep duration percentage can be calculated based on the total sleep duration. This sleep duration percentage includes the percentage of deep sleep and the percentage of REM sleep. Specifically, the deep sleep percentage is the ratio of deep sleep duration to total sleep duration; the REM sleep percentage is the ratio of REM sleep duration to total sleep duration. Simultaneously with calculating the sleep duration percentage, the number of sleep interruptions can be counted according to the sleep type corresponding to each sleep period, and sleep effectiveness indicators can be generated based on the sleep duration percentage and the number of sleep interruptions.

[0068] For example, after obtaining the sleep duration for each sleep type through statistical analysis, the deep sleep duration ΔTd, REM sleep duration ΔTr, and total sleep duration ΔTs can be extracted from the sleep duration data. Then, by calculating the ratio of deep sleep duration ΔTd to total sleep duration ΔTs, the proportion of deep sleep, Rd, can be obtained, i.e., Rd = ΔTd / ΔTs. Similarly, the proportion of rapid eye movement (REM) sleep, Rr, can be calculated based on the REM sleep duration ΔTr and total sleep duration ΔTs, i.e., Rr = ΔTr / ΔTs.

[0069] While calculating the percentage of sleep duration, the number of sleep interruptions (Ni) can also be obtained by iterating through the number of transitions from sleep to wakefulness throughout the entire sleep process. For example, wearable devices can monitor a user's body movement and heart rate parameters during sleep, and then use algorithms to estimate the number and duration of sleep interruptions. If, at a certain moment, the body movement and heart rate parameters are within the range of those in the wakefulness state, and significant movements such as large turning over or changing position, or arm waving are detected, and the heart rate rises by 10-20 bpm within 3-10 seconds, then a sleep interruption can be identified. At this point, the number of sleep interruptions and their corresponding times can be recorded, and the number of interruptions can be accumulated to obtain the number of sleep interruptions (Ni).

[0070] After extracting sleep evaluation indicators from sleep monitoring data, sleep score calculation weights are obtained. These weights include duration weights and sleep structure weights. For example, among the sleep evaluation indicators, sleep duration best reflects a user's sleep state. By comparing historical scores, if a user's sleep duration score deviates significantly from their ideal sleep duration score, then the sleep duration score can be assigned the highest weight, such as 0.5. Sleep regularity and sleep effectiveness have relatively weaker impacts on sleep quality; therefore, the sum of the weights for sleep regularity and sleep effectiveness can be set to 0.5. For example, a regularity weight of 0.25 means the regularity score accounts for 25%, and a sleep effectiveness weight of 0.25 means the effectiveness score accounts for 25%.

[0071] Then, a sleep score is calculated based on the sleep score calculation weights and sleep evaluation indicators. That is, the sleep score is a numerical value obtained by weighting the sleep evaluation indicators using the score calculation weights. For example, the sleep score can be the weighted sum of the score calculation weights and the sleep evaluation indicators, i.e.: S S = α 1 S d + α 2 S r + α 3 S e ; in, S S Indicates sleep score; α 1 indicates the weight of the sleep duration score calculation; S d This represents the score corresponding to sleep duration; α 2 indicates the weighting of the sleep regularity score calculation; S r This represents the score corresponding to the sleep regularity index;α 3 indicates the weighting of the sleep effectiveness score calculation; S e This represents the score corresponding to the sleep effectiveness index.

[0072] To calculate a sleep score, in some embodiments, when calculating the sleep score based on sleep score calculation weights and sleep evaluation indicators, a recommended sleep duration can be obtained first. This recommended sleep duration is correlated with the user's daily stress level and / or sleep debt. When obtaining the recommended sleep duration, daily monitoring data can be acquired first. This daily monitoring data includes daily physiological parameters and historical sleep data. The daily physiological parameters are a set of physiological parameters collected by the user through wearable devices during non-sleep periods; the historical sleep data includes sleep data from multiple monitoring periods and a baseline sleep demand.

[0073] Then, based on daily monitoring data, the user's actual physiological data is generated. This actual physiological data includes at least one of daily stress level and sleep debt. The daily stress level is calculated based on daily physiological parameters; the sleep debt is calculated based on historical sleep data.

[0074] Then, a recommended sleep duration is calculated based on the user's actual physiological data. This recommended sleep duration falls within a preset sleep duration range and is positively correlated with the cumulative amount of daily stress levels and / or sleep debt. To calculate the recommended sleep duration, statistical sample data can be obtained first. This statistical sample data includes sample physiological parameters labeled with the recommended sleep duration. A duration calculation model is then constructed based on the statistical sample. This duration calculation model can be a duration parameter fitting function obtained through linear fitting; or it can be a neural network model trained using the statistical sample data. Finally, the user's actual physiological data is input into the duration calculation model to obtain the recommended sleep duration output by the model.

[0075] After obtaining the recommended sleep duration, a sleep duration score is calculated based on the total sleep duration, the recommended sleep duration, and the duration weight. The sleep duration score is equal to the ratio of the total sleep duration to the recommended sleep duration multiplied by the duration weight. Therefore, the ratio of the total sleep duration to the recommended sleep duration can be calculated to obtain the sleep achievement level. Then, the sleep duration score is calculated based on the sleep achievement level; that is, the sleep duration score is the product of the sleep achievement level and the duration weight.

[0076] For example, analysis of a user's daily or long-term physiological data can determine their stress levels and / or sleep debt. Since stress levels and sleep debt have a certain long-term cumulative effect, assigning the highest weight to sleep duration can account for the impact of individual sleep duration on sleep quality, as well as the cumulative impact of long-term sleep insufficiency. If a user is chronically under stress or experiencing accumulating sleep debt, the recommended sleep duration may increase. Therefore, even with the actual sleep duration remaining constant, a higher recommended sleep duration can lower the corresponding sleep duration score, thus representing the impact of long-term sleep insufficiency on sleep quality.

[0077] Understandably, since the recommended sleep duration is derived from fitting the user's actual physiological data and is related to the user's daily stress level and sleep debt, the degree to which the recommended sleep duration is met can, to some extent, represent the user's degree of stress relief and / or sleep sufficiency over a period of time. For example, if the user's actual sleep duration is comparable to the recommended sleep duration, then the actual sleep duration and recommended sleep duration are close to 1, and the sleep duration score is close to 50.

[0078] Similarly, while calculating sleep duration scores, sleep structure scores can also be calculated based on sleep regularity and sleep effectiveness indicators. That is, the sleep structure score includes both a sleep regularity score and a sleep effectiveness score. For example, the sleep regularity or consistency score is primarily determined based on the degree of deviation between single sleep data and long-term sleep habit data. Indicators for the sleep regularity score can include at least one of the following: sleep onset time, wake-up time, and core sleep period. Therefore, the sleep onset time range and wake-up time range can be determined based on long-term sleep habit data. The corresponding basic score is determined by using the degree of deviation between the current sleep onset time and the current sleep onset time range, as well as the degree of deviation between the current wake-up time range and the current sleep onset time range in the single sleep data. The product of this basic score and the weight of sleep regularity is the sleep regularity score.

[0079] Sleep effectiveness score can be determined based on at least one of the following: the percentage of total deep sleep time, the percentage of total REM (Recurrent Eye Movement) sleep time, and the number of sleep interruptions. For example, if the percentage of total deep sleep time and / or the percentage of total REM sleep time are outside the preset range during a sleep cycle, and / or there are many sleep interruptions and / or high stress levels during sleep, a low sleep effectiveness score will result. Conversely, if the percentage of total deep sleep time and / or the percentage of total REM sleep time are both within the preset range during a sleep cycle, and / or there are few sleep interruptions and / or low stress levels during sleep, a high sleep effectiveness score will result.

[0080] Corresponding to the core indicators, cardiopulmonary function indicators are used to characterize the degree of cardiopulmonary recovery after sleep. Cardiopulmonary function indicators refer to other indicators that can affect the degree of recovery, such as resting heart rate and / or respiratory rate. In some embodiments, when resting heart rate is included in the cardiopulmonary function indicators, body movement parameters can be extracted from sleep monitoring data first, and the start and end times of the resting state can be determined based on the body movement parameters. The start time of the resting state is the time recorded when the body movement parameters begin to meet the resting body movement parameter range; the end time of the resting state is the time recorded when the body movement parameters no longer meet the resting body movement parameter range.

[0081] For example, a user's sleep period is from 11:00 PM to 7:00 AM, a total of 8 hours. Wearable devices such as smartwatches record body movement and heart rate parameters in 1-minute increments, and determine the start and end times of the resting state based on the collected body movement and heart rate parameters. If at 11:00 PM, the user has just gotten into bed and is turning over, the body movement parameters will fluctuate significantly, not meeting the resting body movement parameter range. At 11:05 PM, the body movement parameters tend to stabilize, and the resting state begins. At 11:40 PM, a larger turning over is detected, indicating the end of the resting state. At 11:45 PM, the body movement parameters stabilize again, confirming that the user has re-entered rest. Therefore, the start time of the first resting state can be determined as 11:05 PM, and the end time as 11:40 PM.

[0082] Then, based on the start and end times of the resting state, at least one resting state period is divided from the entire sleep period. Heart rate parameters within the resting state period are then obtained, and the resting heart rate is calculated based on these parameters.

[0083] For example, given the start time of the resting state at 23:05 and the end time of the resting state at 23:40, the first resting state period can be determined to be from 23:05 to 23:40. At this time, the heart rate parameters within this period can be obtained, that is, all valid heart rate data within this period are extracted, possible abnormal values ​​of the sensor are excluded, and then the resting heart rate is calculated by taking the lowest value, the average value, a certain percentile, and the median.

[0084] In some embodiments, when the cardiopulmonary function indicators include respiratory rate, respiratory-related parameters can be extracted from sleep monitoring data first, wherein the respiratory-related parameters include photoplethysmography parameters and body movement parameters.

[0085] The respiratory correlation parameters are then input into the respiratory algorithm model to perform multimodal fusion of photoplethysmography parameters and body motion parameters. Based on the multimodal fusion result, classification calculations are performed to obtain the respiratory rate output by the respiratory algorithm model. The respiratory algorithm model is a deep learning model trained using sample respiratory data; the sample respiratory data consists of respiratory correlation parameters labeled with the respiratory rate.

[0086] For example, based on artificial intelligence algorithms such as deep learning, a massive amount of sample data can be used to train an AI model, such as data collected from professional medical devices and watches labeled with respiratory rate. The trained deep learning model is called a respiratory algorithm model. This model can ultimately learn to accurately identify weak breathing-related patterns from noisy PPG and accelerometer signals and calculate the number of breaths a user takes per unit of time. During calculations, the respiratory algorithm model can perform multimodal index fusion, comprehensively analyzing data from multiple sensors such as PPG and accelerometers to improve the accuracy and stability of the calculation results.

[0087] Therefore, during recovery testing, multimodal respiratory parameters, such as subtle fluctuations in vascular blood flow, motion acceleration signals, and posture signals, can be collected using sensors like PPG sensors, accelerometers, and gyroscopes in wearable devices such as smartwatches during the user's sleep. These respiratory parameters are then input into a respiratory algorithm model, which performs multimodal fusion and classification calculations to obtain the respiratory rate output by the model.

[0088] S103. Calculate the core indicator score based on reference data and core indicators, and calculate the cardiopulmonary function score based on reference data and cardiopulmonary function indicators.

[0089] After obtaining the reference data, scores can be calculated. Specifically, core indicator scores and cardiopulmonary function scores can be calculated based on the reference data and core indicators. Core indicator scores characterize the degree of influence of core indicators on recovery and can be calculated using the reference data and core indicators.

[0090] In some embodiments, when calculating the core indicator score, the heart rate variability baseline can be determined first based on reference data. When the reference data is historical sleep data, the heart rate variability baseline is calculated based on the heart rate parameters in the historical sleep data; when the reference data is the preset general data, the heart rate variability baseline is a general baseline parameter.

[0091] Next, the deviation of the sleep heart rate variability parameters from the heart rate variability baseline is obtained, and the heart rate variability score is calculated based on the deviation. Then, the core indicator score is calculated based on the heart rate variability score and the sleep score.

[0092] When calculating heart rate variability scores, the set of heart rate parameters to be analyzed can be obtained by traversing the heart rate parameters in the reference data within the analysis period, and the average heart rate variability parameter of the heart rate parameter set to be analyzed can be calculated. Then, a preset variability interval is obtained, and a heart rate variability baseline is set based on the variability interval and the average heart rate variability parameter.

[0093] For example, after acquiring historical monitoring data, multiple analysis periods can be determined based on the corresponding monitoring cycles: the most recent 7 days, 14 days, 21 days, and 30 days. Then, based on each analysis period, the heart rate parameter HRV is extracted from the historical monitoring data. i It also calculates the average value of the heart rate parameters for each analysis cycle, i.e., the mean heart rate variability parameter. .

[0094] Next, obtain the preset variability interval, such as ±5%, and then set the heart rate variability baseline based on the variability interval and the mean heart rate variability parameter, i.e., the heart rate variability baseline is... ±5%×( ).

[0095] Next, the deviation of the sleep heart rate variability (HRV) parameter from the HRV baseline is obtained, and the HRV score is calculated based on the deviation. This score is then combined with the sleep score to calculate the core indicator score. For example, since both the HRV score and sleep score are positively correlated with the user's recovery level and are both bonus indicators, the recovery score is equal to the weighted sum of the HRV score and sleep score calculated according to the recovery level weights. This can be expressed as: S r = β 1 S HRV + β 2 S S ; in, S r Indicates the recovery score; β 1 represents the core weight, which is the weight used to calculate the recovery rate corresponding to the HRV score; S HRV Indicates HRV score; β 2 represents the relevant weight, that is, the weight for calculating the recovery rate of the sleep score; S S This represents the sleep score. The HRV score can be determined based on the distance between the current sleep monitoring data and the HRV baseline. The sleep score is at least based on the score obtained by fitting sleep structure and sleep duration.

[0096] It's easy to understand that a user's physical recovery is related to their sleep quality; the higher the sleep quality, the greater the recovery. Therefore, in addition to considering the deviation between HRV heart rate and HRV baseline, sleep scores can also be taken into account, which can improve the scientific rigor, accuracy, and stability of determining recovery levels.

[0097] Cardiopulmonary function score is used to characterize the degree of influence of cardiopulmonary function indicators on recovery. It can be calculated based on reference data, using resting heart rate and / or respiratory rate. In some embodiments, when calculating the cardiopulmonary function score based on reference data and cardiopulmonary function indicators, a baseline of cardiopulmonary function indicators can be determined based on the reference data. Specifically, when the reference data is historical sleep data, the baseline is obtained statistically based on the resting heart rate and / or respiratory rate from the historical sleep data; when the reference data is preset general data, the baseline is a general baseline parameter. Then, the cardiopulmonary function score, including resting heart rate score and / or respiratory rate score, is calculated based on the baseline and the cardiopulmonary function indicators.

[0098] The cardiopulmonary parameter set within the analysis period is traversed from the reference data, and the average parameter value of the cardiopulmonary parameter set is calculated. A baseline for cardiopulmonary function indicators is then set based on the average parameter value. The cardiopulmonary parameter set includes at least one of resting heart rate and respiratory rate parameters. Correspondingly, the average parameter value includes at least one of average resting heart rate and average respiratory rate; the baseline for cardiopulmonary function indicators includes at least one of resting heart rate baseline and respiratory rate baseline.

[0099] For example, for resting heart rate (RHR), 7-day, 14-day, 21-day, or 30-day RHR baselines can be established. These baselines are used to compare the RHR monitored during the user's sleep with the baseline. If the RHR is slightly lower than the baseline, it indicates good recovery; if it is too high, it indicates poor recovery. The impact of this parameter on the final recovery level is assessed using a deduction system; the greater the deviation, the more points are deducted.

[0100] Similarly, for respiratory rate (RR), RR baselines can be established for 7 days, 14 days, 21 days, or 30 days. The RR during nighttime sleep is compared to the RR baseline. Generally, an RR slightly lower than the baseline indicates good recovery, while a RR that is too high indicates poor recovery. The impact of this parameter on the final recovery rate is assessed using a deduction system; the greater the deviation, the more points are deducted.

[0101] In some embodiments, when calculating cardiopulmonary function index scores, a first deviation value of the resting heart rate relative to the resting heart rate baseline can be calculated first. Then, the resting heart rate is compared with the resting heart rate baseline, and the first deviation value is compared with a preset resting heart rate deviation judgment threshold. If the resting heart rate is lower than the resting heart rate baseline, and the first deviation value is less than or equal to the preset resting heart rate deviation judgment threshold, a resting heart rate score is calculated based on the first deviation value within a first value range. Wherein, the minimum value within the first value range is greater than or equal to 0, indicating that the resting heart rate at this point has a positive impact on recovery and is considered a bonus.

[0102] Similarly, if the resting heart rate is higher than the resting heart rate baseline, or the first deviation value is greater than the preset resting heart rate deviation judgment threshold, the resting heart rate score is calculated according to the first deviation value within the second value range. Where the maximum value within the second value range is less than 0, it indicates that the resting heart rate at this time has a negative impact on the recovery rate, and is considered a deduction item.

[0103] For example, when calculating cardiopulmonary function scores, resting heart rate (RHR) is an indicator that indirectly affects the recovery score. When users consistently engage in aerobic exercise, such as running, cycling, or swimming, their resting heart rate gradually decreases. Conversely, when users experience sleep deprivation, high stress, fatigue, or a cold, their resting heart rate increases. For the same user, a stable and low resting heart rate indicates stronger physical fitness and recovery ability. In other words, a lower resting heart rate indicates better physical condition and a less strained heart.

[0104] Therefore, by establishing RHR baselines for 7 days, 14 days, 21 days, and 30 days, the monitored RHR can be compared with the baseline each time the recovery is measured. If the RHR is slightly lower than the baseline, it indicates good recovery, and the RHR score is equal to or greater than 0. If the RHR is equal to or greater than the baseline, it indicates poor recovery, and the RHR score is less than 0.

[0105] Similar to RHR, when calculating the respiratory rate score, a second deviation value of the respiratory rate relative to the respiratory rate baseline can be calculated. If the respiratory rate is lower than the respiratory rate baseline and the second deviation value is less than or equal to a preset respiratory rate deviation judgment threshold, the respiratory rate score is calculated according to the second deviation value within a first value range; if the respiratory rate is higher than the respiratory rate baseline, or the second deviation value is greater than the preset respiratory rate deviation judgment threshold, the respiratory rate score is calculated according to the second deviation value within a second value range.

[0106] For example, by establishing 7-day, 14-day, 21-day, or 30-day RR baselines, the RR score can be determined by comparing the current night's sleep RR with the RR baseline each time recovery is measured. It's easy to understand that when a user is at rest, a lower respiratory rate indicates better cardiopulmonary function and stronger physical fitness. Slower breathing is due to stronger lung function, allowing for more oxygen exchange per breath. High cardiopulmonary efficiency makes oxygen delivery easier. A more stable parasympathetic nervous system leads to greater relaxation and stronger recovery ability. Therefore, in recovery measurement, when a user's body has recovered well, the respiratory rate is low and stable, with a uniform rhythm and not shallow or rapid. This indicates sufficient fatigue clearance, good sleep quality, and a non-excited sympathetic nervous system. If the body has not recovered, is fatigued, under stress, has stayed up late, or is sick, the respiratory rate increases significantly, and breathing becomes shallow and rapid. Based on this, when calculating recovery, the RR from the current sleep monitoring data is compared with the RR baseline; generally, a slightly lower RR is better, while a higher RR indicates poor recovery. The impact of this parameter on the final recovery rate is assessed using a deduction system; the greater the deviation, the more points are deducted.

[0107] S104. Calculate the recovery score based on the core indicator score and the cardiopulmonary function score.

[0108] After calculating the core indicator score and cardiopulmonary function score, the recovery score can be calculated based on these scores. The recovery score is the sum of the core indicator score and the cardiopulmonary function score; that is, it is calculated by adding the core indicator score and the cardiopulmonary function score together.

[0109] In some embodiments, the recovery score can also be a weighted sum of the core indicator score and the cardiopulmonary function score. When calculating the recovery score, recovery calculation weights can be obtained. These recovery calculation weights include a core weight and at least one cardiopulmonary function weight, and the core weight is greater than or equal to the sum of the cardiopulmonary function weights. Then, a weighted summation calculation is performed on the core indicator score and related indicator scores based on the recovery calculation weights to obtain the recovery score.

[0110] For example, if the core score is 0.5, and the cardiopulmonary function weights include resting heart rate and respiratory rate weights, each 0.25, then the recovery score can be calculated by weighting the obtained score using the core score, resting heart rate weight, and respiratory rate weight. That is, the recovery score is 0.5 × core score + 0.25 × resting heart rate score + 0.25 × respiratory rate score.

[0111] By applying the technical solutions of the above embodiments, the recovery degree detection method described in the above embodiments can calculate the user's recovery degree using relatively little physiological data, such as resting heart rate and sleep score. Since the sleep score is determined based on sleep structure and sleep duration, and sleep structure and sleep duration can be fitted based on the user's heart rate, it is not necessary to use wearable devices to obtain multiple and different dimensions of user data to fit the user's recovery degree score. Therefore, the method has a wider range of applications.

[0112] In some embodiments, as a refinement and extension of the specific implementation of the above embodiments, and to fully illustrate the specific implementation process of this embodiment, some embodiments of this application also provide a method for restoring the degree of recovery. The difference between this method and the above embodiments is that the restoring score can be visualized, such as... Figure 4 As shown, the method includes: S201. In response to the preset operation of displaying the recovery score, display the target interface corresponding to the recovery score; S202. Set the display strategy for preset elements based on the recovery score; S203. Based on the display strategy, display preset elements in the target interface.

[0113] To visualize the recovery score, after its generation, user interactions can be monitored. Upon detecting a preset user input indicating the recovery score, a target interface corresponding to that score is displayed. This target interface includes preset elements corresponding to the recovery score. A display strategy for these preset elements is then set based on the recovery score. This strategy can include displaying at least one of the following: a corresponding color, icon, or animation. Finally, based on this display strategy, the preset elements are displayed on the target interface.

[0114] For example, the wearable device could be a smartwatch with a built-in screen, in which case the recovery score can be visualized through the smartwatch. Specifically, the smartwatch obtains the recovery score every 24 hours by performing the aforementioned recovery detection method, and then displays it on the smartwatch's app interface. During the display process, the recovery score can display corresponding preset elements according to a specific display strategy; that is, based on the specific recovery score value, at least one of the following can be displayed: a color, an icon, and an animation.

[0115] Figure 5 A schematic diagram illustrating the display effect of the recovery score provided in an embodiment of this application is shown. Figure 5As shown, the preset elements in the target interface displayed by the wearable device can be a fraction ring and / or a fraction. When setting the display strategy of the preset elements based on the recovery score, the target color can be determined according to the preset range in which the score is located, and this color is the display color of the preset element.

[0116] For example, you can set the default element color to green when the recovery score is in the range of 67-99%; the default element color to yellow when the recovery score is in the range of 34-66%; and the default element color to red when the recovery score is in the range of 1-33%.

[0117] As one possible implementation, the preset element can also be a pre-set icon element. For example, the pre-set icon element may include: an "○" icon to represent a high recovery status, a "√" icon to represent a normal recovery status, and an "×" icon to represent a low recovery status.

[0118] For example, when the recovery score is in the range of 67% to 99%, a circular icon, or "○", is displayed; when the recovery score is in the range of 34% to 66%, a checkmark icon, or "√", is displayed; and when the recovery score is in the range of 1% to 33%, a cross icon, or "×", is displayed.

[0119] As another possible implementation, the preset element can also be any of the following: a score, an energy progress bar, or a battery level icon.

[0120] For example, the display strategy for the score value can be a scoring animation. For instance, it could rapidly calculate the score from 0 to the recovery score. The display strategy for the energy progress bar can be a simulated progress bar being dragged. For instance, it could drag the progress bar from its initial position to any intermediate position or its final position. The display strategy for the battery level icon can be to map the recovery score to the battery level, that is, to represent the recovery score through the battery level displayed in the battery level icon. For example, based on the recovery score value corresponding to a percentage of battery level, such as a recovery score of 90, the battery level icon could display 90% of the battery level to represent that recovery score.

[0121] It should be noted that the recovery score can also be visualized via mobile devices such as smartphones. Specifically, when the wearable device is a screenless smart bracelet or smart ring, data can be periodically synchronized to the mobile device. The mobile device can then use an app to execute this solution, obtain and display the aforementioned recovery score. Furthermore, the recovery score is displayed in the mobile app after executing the visualization steps described above.

[0122] In some embodiments, as a specific implementation of the recovery detection method described in the above embodiments, some embodiments of this application also provide a wearable device, including: a biosensor and a controller. The biosensor is configured to collect physiological parameters; the controller is connected to the biosensor. Furthermore, the controller is configured to: Acquire sleep monitoring data and reference data; the sleep monitoring data includes physiological parameters collected during the user's sleep. Based on the sleep monitoring data, recovery evaluation indicators are determined. These indicators include core indicators and cardiopulmonary function indicators. The core indicators include sleep heart rate variability parameters and sleep scores. The cardiopulmonary function indicators include resting heart rate and / or respiratory rate. The core indicator score is calculated based on the reference data and the core indicator, and the cardiopulmonary function indicator score is calculated based on the reference data and the cardiopulmonary function indicator. The recovery score is calculated based on the scores of the core indicators and the cardiopulmonary function indicators.

[0123] By applying the technical solutions of the above embodiments, the wearable device provided in this application can, after acquiring sleep monitoring data and reference data, determine recovery evaluation indicators from the sleep monitoring data, then calculate core indicator scores based on reference data and core indicators, and calculate cardiopulmonary function scores based on reference data and cardiopulmonary function indicators, and finally calculate the recovery degree score based on the core indicator scores and cardiopulmonary function scores. The method determines recovery evaluation indicators, including core indicators and cardiopulmonary function indicators, using physiological parameters collected by the wearable device, and calculates the core indicator scores and cardiopulmonary function scores respectively. Since the monitoring process only uses data directly collected by the wearable device or data fitted based on the collected data, the wearable device can reduce the number of indicators required for the recovery degree detection process and improve the detection efficiency of the recovery degree.

[0124] In one embodiment, a computer-readable storage medium is also provided, which may be non-volatile or volatile, and a computer program is stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0125] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0126] The embodiments described above are merely examples of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application.

Claims

1. A method for detecting resilience, characterized in that, The method includes: Acquire sleep monitoring data and reference data; the sleep monitoring data includes physiological parameters collected during the user's sleep. Recovery evaluation indicators are determined from the sleep monitoring data. These indicators include core indicators and cardiopulmonary function indicators. The core indicators are used to characterize the degree of recovery and sleep quality during sleep, while the cardiopulmonary function indicators are used to characterize the degree of cardiopulmonary capacity recovery after sleep. The core indicator score is calculated based on the reference data and the core indicator, and the cardiopulmonary function score is calculated based on the reference data and the cardiopulmonary function indicator. The recovery score is calculated based on the core indicator score and the cardiopulmonary function score.

2. The method according to claim 1, characterized in that, The acquisition of sleep monitoring data and reference data includes: The latest data monitored during sleep behavior is used as the sleep monitoring data; If historical sleep data is stored, then the historical sleep data is extracted as the reference data; If historical sleep data is not stored, then the preset general data corresponding to the custom settings data is obtained as the reference data.

3. The method according to claim 2, characterized in that, The core indicators include sleep heart rate variability parameters and sleep scores; The cardiopulmonary function indicators include resting heart rate and / or respiratory rate.

4. The method according to claim 3, characterized in that, The sleep heart rate variability parameters were obtained in the following manner: Heart rate parameters are extracted from the sleep monitoring data, including the intervals between heartbeats monitored during the user's sleep. The variability evaluation value is calculated based on the heart rate parameters. The variability evaluation value includes at least one of the following: standard deviation of heartbeat interval, mean squared deviation of heartbeat interval difference, and proportion of heartbeat interval difference exceeding a preset difference threshold. The sleep heart rate variability parameter is determined based on the variability evaluation value.

5. The method according to claim 3, characterized in that, The sleep score was obtained in the following manner: Sleep evaluation indicators are extracted from the sleep monitoring data, including sleep duration and sleep structure indicators. Obtain the sleep score calculation weights, which include duration weights and sleep structure weights; The sleep score is calculated based on the sleep score calculation weights and the sleep evaluation indicators. The sleep score is a numerical value obtained by weighting the sleep evaluation indicators using the score calculation weights.

6. The method according to claim 4, characterized in that, The calculation of the sleep score based on the sleep score calculation weight and the sleep evaluation index includes: The recommended sleep duration is obtained and is correlated with the user's daily stress level and / or sleep debt value. A sleep duration score is calculated based on the sleep duration, the recommended sleep duration, and the duration weight. The sleep duration score is equal to the ratio of the sleep duration to the recommended sleep duration multiplied by the duration weight. The product of the quantitative value of the sleep structure index and the sleep structure weight is calculated to obtain the sleep structure score; The sleep score is obtained by summing the sleep duration score and the sleep structure score.

7. The method according to claim 3, characterized in that, The resting heart rate was obtained in the following manner: Extract body movement parameters from the sleep monitoring data; The resting state period is determined based on the body motion parameters, and the resting state period includes the start time of the resting state and the end time of the resting state; Obtain the heart rate parameters during the resting state period, and calculate the resting heart rate based on the heart rate parameters during the resting state period.

8. The method according to claim 3, characterized in that, The respiratory rate is obtained as follows: Respiratory correlation parameters are extracted from the sleep monitoring data, including photoplethysmography parameters and body movement parameters. The respiratory correlation parameters are input into a respiratory algorithm model to perform multimodal fusion of the photoplethysmography parameters and the body motion parameters, and to perform classification calculations based on the multimodal fusion results; the respiratory algorithm model is a deep learning model trained using sample respiratory data; the sample respiratory data are respiratory correlation parameters labeled with respiratory rate. Obtain the respiratory rate output by the respiratory algorithm model.

9. The method according to claim 3, characterized in that, The core indicator scores are obtained in the following way: A heart rate variability baseline is determined based on the reference data, wherein when the reference data is historical sleep data, the heart rate variability baseline is calculated based on the heart rate parameters in the historical sleep data; when the reference data is the preset general data, the heart rate variability baseline is a general baseline parameter. Obtain the deviation of the sleep heart rate variability parameter from the heart rate variability baseline, and calculate the heart rate variability score based on the deviation; The core indicator score is calculated based on the heart rate variability score and the sleep score.

10. The method according to claim 3, characterized in that, The calculation of the cardiopulmonary function score based on the reference data and the cardiopulmonary function indicators includes: The cardiopulmonary function baseline is determined based on the reference data, wherein when the reference data is historical sleep data, the cardiopulmonary function baseline is obtained statistically based on the resting heart rate and / or respiratory rate in the historical sleep data; when the reference data is the preset general data, the cardiopulmonary function baseline is a general baseline parameter. Cardiopulmonary function scores are calculated based on the baseline cardiopulmonary function indicators and the cardiopulmonary function indicators themselves. The cardiopulmonary function scores include resting heart rate score and / or respiratory rate score.

11. The method according to claim 10, characterized in that, The baseline cardiopulmonary function indicators were obtained in the following manner: The set of cardiopulmonary parameters located within the analysis period in the reference data is traversed, and the set of cardiopulmonary parameters includes at least one of resting heart rate parameters and respiratory rate parameters; Calculate the average parameter value of the cardiopulmonary parameter set, wherein the average parameter value includes at least one of average resting heart rate and average respiratory rate; A baseline for cardiopulmonary function indicators is set based on the average parameter values, and the baseline for cardiopulmonary function indicators includes at least one of a resting heart rate baseline and a respiratory rate baseline.

12. The method according to claim 11, characterized in that, The calculation of the cardiopulmonary function score based on the baseline cardiopulmonary function indicators and the cardiopulmonary function indicators includes: Calculate the first deviation of the resting heart rate relative to the resting heart rate baseline; If the resting heart rate is lower than the resting heart rate baseline, and the first deviation value is less than or equal to a preset resting heart rate deviation judgment threshold, the resting heart rate score is calculated according to the first deviation value within a first value range; the minimum value of the first value range is greater than or equal to 0. If the resting heart rate is higher than the resting heart rate baseline, or the first deviation value is greater than the preset resting heart rate deviation judgment threshold, the resting heart rate score is calculated according to the first deviation value within a second value range; the maximum value of the second value range is less than 0; And / or, Calculate the second deviation of the respiratory rate relative to the respiratory rate baseline; If the respiratory rate is lower than the respiratory rate baseline, and the second deviation value is less than or equal to a preset respiratory rate deviation judgment threshold, the respiratory rate score is calculated according to the two deviation values ​​within a first value range; the minimum value of the first value range is greater than or equal to 0. If the respiratory rate is higher than the respiratory rate baseline, or the two deviation values ​​are greater than the preset respiratory rate deviation judgment threshold, the respiratory rate score is calculated according to the two deviation values ​​within a second value range; the maximum value of the second value range is less than 0.

13. The method according to any one of claims 1 to 12, characterized in that, The method further includes: In response to a preset operation to display the recovery score, a target interface corresponding to the recovery score is displayed, the target interface including preset elements corresponding to the recovery score; The display strategy of the preset element is set according to the recovery score; Based on the display strategy, the preset elements are displayed in the target interface.

14. A wearable device, characterized in that, include: Biosensors are configured to collect physiological parameters; A controller is connected to the biosensor; the controller is configured to: Obtain sleep monitoring data and reference data; The sleep monitoring data includes physiological parameters collected during the user's sleep. Based on the sleep monitoring data, recovery evaluation indicators are determined. These indicators include core indicators and cardiopulmonary function indicators. The core indicators include sleep heart rate variability parameters and sleep scores. The cardiopulmonary function indicators include resting heart rate and / or respiratory rate. The core indicator score is calculated based on the reference data and the core indicator, and the cardiopulmonary function indicator score is calculated based on the reference data and the cardiopulmonary function indicator. The recovery score is calculated based on the scores of the core indicators and the cardiopulmonary function indicators.

Images