A Method and System for Assessing Sleep Fatigue Recovery Quality Based on Heart Rate Variation

Abstract

This invention relates to the field of data processing, specifically to a method and system for assessing the quality of fatigue recovery during sleep based on heart rate changes. The method includes: obtaining a neurological recovery index for each sleep cycle based on the difference between the instantaneous heart rate at each acquisition time within each sleep cycle and the mean of the instantaneous heart rate at all acquisition times within each sleep cycle, and the variance of the composite vector of the body's three-axis acceleration at each acquisition time within each sleep cycle and the mean of the instantaneous heart rate at all acquisition times within each sleep cycle; and performing a quality assessment based on the difference between the mean of the instantaneous heart rate at multiple acquisition times before each sleep cycle and the difference between the instantaneous heart rate at multiple acquisition times before reaching a low point within each sleep cycle, and the interval between the first acquisition time within each sleep cycle and the acquisition time reaching the low point. This invention aims to solve the problem of relatively low accuracy in current fatigue recovery assessments.

Description

Technical Field

[0001] This invention relates to the field of data processing, and more specifically to a method and system for assessing the quality of fatigue recovery during sleep based on heart rate changes. Background Technology

[0002] In the current field of sleep physiological monitoring, mainstream fatigue recovery assessment systems mostly rely on single indicators such as overnight average heart rate variability or static minimum nighttime heart rate. This means that existing technologies generally analyze indicators from different physiological dimensions in isolation, lacking consideration of the synergistic effect between macroscopic cardiovascular deceleration and microscopic neural repair. This single-point and static assessment logic has systemic flaws and is highly susceptible to physiological noise interference, limiting the scientific rigor and accuracy of the assessment results. Consequently, in the current objective scenarios of sleep physiological monitoring, existing sleep assessment algorithms based on heart rate variability are insufficient, leading to a decline in the scientific rigor and accuracy of wearable devices in fatigue recovery assessment. Summary of the Invention

[0003] This invention provides a method and system for assessing the quality of fatigue recovery during sleep based on heart rate variability, in order to address the shortcomings of existing sleep assessment algorithms based on heart rate variability in objective scenarios of sleep physiological monitoring, which result in relatively low accuracy of assessment results obtained by existing technologies.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] At fixed intervals, the system collects the user's instantaneous heart rate and body triaxial acceleration vector, and obtains the start and end times of each sleep cycle.

[0006] Based on the difference between the instantaneous heart rate at each acquisition time within each sleep cycle and the mean instantaneous heart rate at all acquisition times within each sleep cycle, and the variance of the composite vector of the body's three-axis acceleration at each acquisition time within each sleep cycle and the surrounding acquisition times, the stability tendency factor of the user at each acquisition time within each sleep cycle is obtained; all adjacent heartbeat intervals of the user within each sleep cycle are obtained, and based on the stability tendency factor of the user at each acquisition time within each sleep cycle, and the difference between each adjacent heartbeat interval and the next adjacent heartbeat interval within each sleep cycle, the neurofunctional recovery index of the user in each sleep cycle is obtained.

[0007] Based on the distribution of the user's instantaneous heart rate at all sampling times during each sleep period, the sampling time when the user reaches the lowest point during each sleep period is obtained; based on the mean of the user's instantaneous heart rate at multiple sampling times before each sleep period and the difference between the instantaneous heart rate at multiple sampling times before the sampling time when the user reaches the lowest point during each sleep period, and the interval between the first sampling time and the sampling time when the user reaches the lowest point during each sleep period, the heart rate unloading gradient of the user during each sleep period is obtained.

[0008] Based on the user's neurological recovery index and heart rate unloading gradient during each sleep period, a recovery quality index is obtained for each sleep period, and a quality assessment is performed.

[0009] Furthermore, the specific calculation steps for collecting the user's instantaneous heart rate and body triaxial acceleration vector at fixed intervals, and obtaining the start and end times of each sleep cycle are as follows:

[0010] With the collection interval as Seconds, real-time acquisition of two dimensions of data: the user's instantaneous heart rate and the composite vector of body triaxial acceleration at each acquisition time; among which, Indicates the preset data collection interval;

[0011] Obtain the start and end times of each user's sleep session.

[0012] Furthermore, the specific calculation steps for obtaining the stability tendency factor for each acquisition time within each sleep cycle, based on the difference between the instantaneous heart rate at each acquisition time within each sleep cycle and the mean of the instantaneous heart rate at all acquisition times within each sleep cycle, and the variance of the composite vector of the body's three-axis acceleration at each acquisition time within each sleep cycle and the surrounding acquisition times, are as follows:

[0013] The user's first The first sleep during the second sleep The collection time is up to the first The first sleep during the second sleep During each collection time The collection time is denoted as the [number]th collection time. The first sleep during the second sleep The collection time within the time period of each collection time; among which... This indicates the number of data collection times contained within a preset time period;

[0014] According to the user in the The first sleep during the second sleep The instantaneous heart rate at each sampling time within the time period of the sampling time and the user's heart rate at the first sampling time. The difference in the mean instantaneous heart rate across all sampling times during the next sleep period, and the user's heart rate during the first sleep period. The first sleep during the second sleep The variance of the composite vector of the body's three-axis acceleration at all acquisition times within the time period of the acquisition time is obtained to obtain the user's three-axis acceleration vector at the first acquisition time. The first sleep during the second sleep Stability tendency factor for each collection time.

[0015] Furthermore, the statement based on the user in the first... The first sleep during the second sleep The instantaneous heart rate at each sampling time within the time period of the sampling time and the user's heart rate at the first sampling time. The difference in the mean instantaneous heart rate across all sampling times during the next sleep period, and the user's heart rate during the first sleep period. The first sleep during the second sleep The variance of the composite vector of the body's three-axis acceleration at all acquisition times within the time period of the acquisition time is obtained to obtain the user's three-axis acceleration vector at the first acquisition time. The first sleep during the second sleep The specific calculation steps for the stability tendency factor at each collection time are as follows:

[0016] ;

[0017] In the formula, Indicates the user's position in the first month. The first sleep during the second sleep Stability tendency factor for each collection time, Indicates the user's position at the time. The first sleep during the second sleep The first time period within the collection time period Instantaneous heart rate at each acquisition time, Indicates the user's position at the time. The mean of instantaneous heart rate at all sampling times during the subsleep period. Represents the absolute value function. It is an exponential function with the natural constant as its base. Indicates the user's position in the first month. The first sleep during the second sleep The number of collection times within the time period of each collection time. Indicates the user's position at the time. The first sleep during the second sleep The variance of the composite vector of body triaxial accelerations at all acquisition times within a given acquisition time period. This represents a logarithmic function with the natural constant as its base. It represents the acceleration due to gravity.

[0018] Furthermore, the specific calculation steps for obtaining the user's neurological function recovery index for each sleep cycle, based on the user's stability tendency factor at each sampling time within each sleep cycle and the difference between each adjacent heartbeat interval and the next adjacent heartbeat interval within each sleep cycle, are as follows:

[0019] The user in the first The first sleep during the second sleep The collection time is up to the first The first sleep during the second sleep During each collection time The collection time is recorded as the user's collection time in the [number]th [year]. The first sleep during the second sleep The collection time within the time period of each collection time; among which... This indicates the number of data collection times contained within a preset time period;

[0020] According to the user in the The first sleep during the second sleep The first and last collection times within the time period of each collection time are used to obtain the user's data at the [number]th collection time. The first sleep during the second sleep The time period in which each data collection time is located;

[0021] Get users in the first The specific formula for calculating the neurological recovery index of the second sleep cycle is as follows:

[0022] ;

[0023] ;

[0024] In the formula, Indicates the user's position at the time. Neurological recovery index of the second sleep cycle Indicates the user's position at the time. The first sleep during the second sleep The time period in which each data collection time is located Root mean square of the difference Indicates the user's position at the time. The first sleep during the second sleep Within the time period of each collection time Quantity, Indicates the user's position at the time. The first sleep during the second sleep The first time period within the collection time period indivual The value, Indicates the user's position at the time. The first sleep during the second sleep The first time period within the collection time period indivual The value, Indicates the user's position at the time. The first sleep during the second sleep Stability tendency factor for each collection time, Indicates the user's position at the time. The second sleep and the first Before the next sleep The mean of the stable tendency factor for all data collection times across all sleep periods within a day. Indicates the user's position at the time. The number of data collection times during each sleep cycle, among which... This indicates the preset period length.

[0025] Furthermore, the specific calculation steps for obtaining the time when the user reaches the lowest point in each sleep cycle based on the distribution of the user's instantaneous heart rate across all sampling times during each sleep cycle are as follows:

[0026] Contain in each window The user's data collection time will be the first consecutive time period. All data acquisition times within a single sleep period are divided into several windows; however, different windows do not contain the same data acquisition time. The number of data collection times included in a preset window;

[0027] The user in the first The first sleep during the second sleep The average instantaneous heart rate at all acquisition times in each window is denoted as the user's heart rate at the [number]th window. The first sleep during the second sleep Overall heart rate in each window;

[0028] The user in the first The window with the lowest overall heart rate among all windows during the next sleep cycle is recorded as the user's heart rate during the first sleep cycle. Low point window during the second sleep;

[0029] The user in the first The time of minimum instantaneous heart rate within all data acquisition timeframes during the low-point window of the next sleep cycle is denoted as the user's heart rate during the first sleep cycle. The time it takes for the data to reach its lowest point during the next sleep cycle.

[0030] Furthermore, the specific calculation steps for obtaining the user's heart rate unloading gradient during each sleep cycle, based on the difference between the average instantaneous heart rate at multiple sampling times before each sleep cycle and the instantaneous heart rate at multiple sampling times before reaching the low point within each sleep cycle, and the interval between the first sampling time and the sampling time reaching the low point within each sleep cycle, are as follows:

[0031] The user in the first Before the next sleep The average instantaneous heart rate over all data collection times within a minute is denoted as the user's heart rate at the 1st minute. Normal heart rate before the next sleep cycle, among which... Indicates the preset time length;

[0032] The user in the first Of all the data collection times during the next sleep period, the time shorter than the user's first sleep period... The time when the data reaches its lowest point during the next sleep cycle is recorded as the user's data collection time during the first sleep cycle. Data acquisition time during the rapid descent process in the second sleep;

[0033] Get users in the first The specific formula for calculating the heart rate unloading gradient during the next sleep phase is as follows:

[0034] ;

[0035] In the formula, Indicates the user's position at the time. Heart rate unloading gradient during sleep, Indicates the user's position at the time. The number of data collection times during the rapid descent process within the second sleep phase. Indicates the user's position at the time. Normal heart rate before sleep. Indicates the user's position at the time. The rapid descent process during the second sleep phase Instantaneous heart rate at each acquisition time, Indicates the user's position at the time. Instantaneous heart rate at the time of data collection when the heart rate reaches its lowest point during the second sleep cycle. Indicates the user's position at the time. The first data collection time during the next sleep cycle. Indicates the user's position at the time. The time it takes for the sample to reach its lowest point during the next sleep cycle. Indicates the user's position at the time. The last data collection time during the next sleep period is related to the user's... The time interval of the first data collection time within the next sleep period It is an exponential function with the natural constant as its base.

[0036] Furthermore, the specific calculation steps for obtaining the user's recovery quality index for each sleep cycle based on the user's neurological recovery index and heart rate unloading gradient are as follows:

[0037] ;

[0038] In the formula, Indicates the user's position at the time. The quality index of sleep recovery. Indicates the user's position at the time. The neurological recovery index of the second sleep cycle Represents the absolute value function. Indicates the user's position at the time. Heart rate unloading gradient during sleep.

[0039] The beneficial effects of the technical solution of the present invention are:

[0040] First, a time-series dynamic stability tendency factor was proposed. By introducing heart rate dynamic deviation and body motion variance as physical gating conditions, complex sleep signals were precisely filtered, effectively eliminating non-repairing physiological artifacts and achieving purified extraction of the effective repair depth of microscopic parasympathetic nerves.

[0041] Second, a geometric integral model of the heart rate unloading trajectory was constructed, which transforms the dynamic process and time-consuming characteristics of heart rate decay from high potential energy to low potential energy into a quantifiable heart rate unloading gradient, making up for the shortcomings of traditional algorithms that ignore the time dimension and the decrease characteristics.

[0042] Third, a collaborative and balanced evaluation framework was established, which uses the natural mathematical deviation of macro-unloading and micro-repair characteristics to construct a consistency penalty mechanism, abandoning the traditional fixed-weight fusion method.

[0043] This system not only constrains the assessment bias caused by abnormal single indicators, but also eliminates innate physical differences by introducing individual historical dynamic baselines, and finally outputs a comprehensive fatigue recovery quality index with high confidence and objective physiological guidance. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a flowchart illustrating the steps of a sleep fatigue recovery quality assessment method based on heart rate changes, according to an embodiment of the present invention. Detailed Implementation

[0046] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the sleep fatigue recovery quality assessment method and system based on heart rate changes proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0048] Example 1:

[0049] This invention provides a method for assessing the quality of fatigue recovery during sleep based on heart rate changes, specifically as follows: Figure 1 As shown, it includes:

[0050] Step S001: Collect the user's instantaneous heart rate and body triaxial acceleration vector at fixed intervals, and obtain the start and end times of each sleep cycle.

[0051] Specifically, wearable devices equipped with optical heart rate monitoring and microelectromechanical systems (MEMS) triaxial accelerometers are used to collect data at intervals of [missing information]. The system acquires data in real-time, second by second, of the user's instantaneous heart rate and the composite vector of body triaxial acceleration at each acquisition time. The preset acquisition interval in this embodiment... This example will be used to illustrate the concept; other values ​​can be set in other implementations. When The larger the value, the smaller the computational cost, but the lower the reliability of the calculation results; when... The smaller the value, the greater the computational load, but the more reliable the calculation results. Obtaining the synthesized vector of the user's three-axis body acceleration at each acquisition time is a well-known existing technique, and will not be described in detail in this embodiment.

[0052] Furthermore, the start and end times of each sleep cycle are obtained through a wearable device equipped with optical heart rate monitoring and a microelectromechanical system (MEMS) triaxial accelerometer. The automatic identification of whether the user is asleep or not at any given time is a well-known technology and will not be elaborated upon in this embodiment.

[0053] At this point, we have obtained the user's sleep time and a data sequence of the user's two dimensions.

[0054] Step S002: Based on the difference between the instantaneous heart rate of the user at each acquisition time within each sleep cycle and the mean of the instantaneous heart rate at all acquisition times within each sleep cycle, and the variance of the composite vector of the body's three-axis acceleration at each acquisition time within each sleep cycle and the surrounding acquisition times, obtain the stability tendency factor of the user at each acquisition time within each sleep cycle; obtain all adjacent heartbeat intervals of the user within each sleep cycle, and based on the stability tendency factor of the user at each acquisition time within each sleep cycle, and the difference between each adjacent heartbeat interval and the next adjacent heartbeat interval within each sleep cycle, obtain the neurofunctional recovery index of the user in each sleep cycle.

[0055] It should be noted that existing sleep assessment algorithms based on heart rate variability tend to calculate the average heart rate variability throughout the night, i.e., the average... Quantifying sleep recovery quality by applying indiscriminate global averaging of data may result in the loss of significant real-world information about vagal nerve regulation, thus failing to accurately reflect the body's recovery status. Furthermore, assessments based on heart rate variability only extract the lowest or average heart rate throughout the night, ignoring the dynamic decrease and duration of heart rate troughs from sleep onset. While these two dimensions of data are systematically related in assessing fatigue recovery quality, existing algorithms are not only one-sided in their assessment of individual points but also fail to consider the synergistic effect between these two dimensions. Therefore, the assessment results obtained by current methods are not only scientifically insufficient but also relatively inaccurate. Thus, improvements are needed to the methods for assessing fatigue recovery quality during sleep.

[0056] It should be further noted that in real sleep scenarios, the true physiological repair period is accompanied by extremely stable heart rate under central nervous system control and a significant disappearance of body movement. Therefore, this invention utilizes the stability of heart rate fluctuations and the extremely low level of body movement as two objective physical characteristics to map the user's sleep status. Thus, based on the distribution of heart rate data and the composite vector of body triaxial acceleration at each acquisition time and multiple subsequent acquisition times, a stability tendency factor is obtained for each acquisition time.

[0057] Specifically, the user's first The first sleep during the second sleep The collection time is up to the first The first sleep during the second sleep During each collection time The collection time is denoted as the [number]th collection time. The first sleep during the second sleep The collection time is the collection time within a time period that is included in the collection time. In this embodiment, the number of collection times is preset within a time period. This example will be used to illustrate the concept; other implementations may use different values. Specifically, if the user's... Of all the data collection times during the user's sleep period, the one located in the user's sleep period is... The first sleep during the second sleep The number of collection times after each collection time is insufficient. The first one will be the user's first... The first sleep during the second sleep The collection time is up to the first All acquisition times within the last acquisition time of a sleep cycle are denoted as the th acquisition time. The first sleep during the second sleep The collection time within the time period of the first collection time; and at this time the first collection time... The first sleep during the second sleep The collection time within the time period of the first collection time includes the user's collection time in the first collection time. The first sleep during the second sleep The first and last data collection times. (The sentence is incomplete and likely refers to a data collection time.) The first sleep during the second sleep The first collection time within the time period of the first collection time is used as the first collection time. The first sleep during the second sleep The start time of the time period in which the first collection time falls will be the first time period in the ... The first sleep during the second sleep The last collection time within the time period of the first collection time is used as the first collection time. The first sleep during the second sleep The deadline of the time period in which the data is collected is obtained to determine the user's data collection time in the first time period. The first sleep during the second sleep The invention specifies the time period within which each collection time occurs. Furthermore, when acquiring each time period, the invention ensures that each time period contains at least two heartbeat intervals. In this embodiment, the chosen time period is... This is because heart rate is usually measured in beats per minute, while the sampling interval in this embodiment is 20 beats per minute, so one minute is used as a time period.

[0058] Furthermore, acquiring user information in the first... The first sleep during the second sleep The specific calculation formula for the stability tendency factor for each collection time is as follows:

[0059] ;

[0060] In the formula, Indicates the user's position at the time. The first sleep during the second sleep Stability tendency factor for each collection time, Indicates the user's position at the time. The first sleep during the second sleep The first time period within the collection time period Instantaneous heart rate at each acquisition time, Indicates the user's position at the time. The mean of instantaneous heart rate at all sampling times during the subsleep period. Represents the absolute value function. As an exponential function with the natural constant as its base, this embodiment uses it to represent an inverse proportional relationship and performs normalization processing. Indicates the user's position at the time. The first sleep during the second sleep The number of collection times within the time period of each collection time. Indicates the user's position at the time. The first sleep during the second sleep The variance of the composite vector of body triaxial accelerations at all acquisition times within a given acquisition time period. This represents a logarithmic function with the natural constant as its base; in this embodiment, it is used for smoothing. Represents gravitational acceleration, which is Furthermore, in this invention, the user in the first... The first sleep during the second sleep The collection time is less than the user's in the first... The first sleep during the second sleep Each collection time.

[0061] It should be noted that, This means that in the first The first sleep during the second sleep The difference between the instantaneous heart rate at each sampling time within the time period of the sampling period and the global average heart rate over a complete sleep cycle, when the sampling time is the first... The more chaotic the heart rate is within the time period in which the data is collected, the better. The closer the value is to If the heart rate remains relatively stable during this time period, The closer the result value is to ;for part, The larger the value, the more likely the user is to be on the first day. The first sleep during the second sleep The greater the range of motion within a given time period, the more likely the sleep posture is to be in a non-static state. The smaller the value, the more stable the user's sleeping position. The larger the value, the better. Adding 1 to the denominator prevents the denominator from being zero. Divide by It is used to eliminate dimensions, so that The input to the function is a dimensionless number.

[0062] Furthermore, acquiring user information in the first... The specific formula for calculating the neurological recovery index of the second sleep cycle is as follows:

[0063] ;

[0064] ;

[0065] In the formula, Indicates the user's position at the time. Neurological recovery index of the second sleep cycle Indicates the user's position at the time. The first sleep during the second sleep The time period in which each data collection time is located Root mean square of the difference Indicates the user's position at the time. The first sleep during the second sleep Within the time period of each collection time Quantity, Indicates the user's position at the time. The first sleep during the second sleep The first time period within the collection time period indivual The value, Indicates the user's position at the time. The first sleep during the second sleep The first time period within the collection time period The value of the interval between adjacent heartbeats. Indicates the user's position at the time. The first sleep during the second sleep Stability tendency factor for each collection time, Indicates the user's position at the time. Number of data collection times during each sleep cycle Indicates the user's position at the time. The second sleep and the first Before the next sleep The mean of the stability tendency factor for all data collection times across all sleep periods within a day. In other embodiments, a preset period length may also be used. When the user's sleep is more stable, the value can be set smaller, because in this case, a baseline of the user's sleep over a period of time can be obtained within a shorter timeframe. Conversely, when the user's sleep is more unstable, the value can be set larger. This embodiment sets... The value is 7 because 7 days is a relatively complete life cycle under normal circumstances. The adjacent heartbeat interval is a core term on an electrocardiogram (ECG), referring to the time interval between the R-wave peaks of two adjacent QRS complexes. Obtaining the adjacent heartbeat interval is a well-known existing technique, and will not be elaborated upon in this embodiment. Furthermore, if the first acquisition time out of all acquisition times coincides with the user's acquisition time on the [missing information]... If the time interval between the first data collection time before the next sleep period is less than 7 days, then the user will be included in the data collection time at the next sleep period. The second sleep and the first The mean of the stability tendency factor across all data collection times within all sleeps prior to the next sleep period, as... .

[0066] It should be noted that, for part, As an indicator, it is widely recognized as a highly specific indicator reflecting the rapid regulatory activity of the cardiac parasympathetic nervous system, especially the vagus nerve. From a physiological perspective, when the body is in a state of deep fatigue or chronic stress, the sympathetic nervous system remains overactive while parasympathetic nerve activity is suppressed. This autonomic imbalance leads to decreased heart rate variability, manifested as… When the heart rate is low, the heart rate rhythm exhibits rigid characteristics. However, during periods of rest and recovery, such as quality sleep, the parasympathetic nervous system is activated and dominates the rest-digestion-repair process. This rapid and precise regulation of the sinoatrial node causes greater fluctuations in the intervals between adjacent heartbeats, directly manifesting as… The significant increase in the value, and the plausible explanation for this indicator's contribution to recovery, lies in the fact that this parasympathetic-driven high-frequency heart rate fluctuation causes rhythmic changes in cardiac output and arterial blood flow. Theoretically, this can generate more dynamic shear stress in the microcirculation, and pulsatile shear stress is a key physical stimulus for maintaining and improving vascular endothelial function, helping to enhance tissue fluid exchange and metabolic waste removal efficiency. The elevation of this level is not only an important marker of parasympathetic activation and entering a repair state, but the corresponding physiological process itself tends to be one of the intrinsic mechanisms that promote fatigue recovery.

[0067] It should be further explained that in the actual physical environment of sleep, the following factors lead to... There are two other typical reasons for elevated heart rate: First, when a person is dreaming, especially during a nightmare, the sympathetic nervous system experiences paroxysmal strong impulses, causing extremely irregular breathing. These irregular changes in intrathoracic pressure can trigger respiratory sinus arrhythmia. It can also increase; secondly, during sleep, the body may unconsciously turn over or experience muscle twitches. To support this momentary physical movement, the cardiovascular system will immediately make reflexive adjustments, leading to violent oscillations between heartbeats. It will also increase. However, in both states, the body does not undergo deep tissue repair and fatigue elimination. The result reflects the body's tendency to maintain a stable state. The larger the value, the more pronounced the tendency of the body to be in a state of absolute physical rest and extremely low heart rate. Therefore, only when... When larger The higher the confidence level, the greater the confidence level, and the overall result of this part of the formula will tend to be larger, and the corresponding neural function recovery index should also be larger. This reflects the user's first Average situation during a sleep cycle.

[0068] It should be further explained that, for Part of, with As a baseline level for users over a period of time, Multiplying by this part is equivalent to comparing it with the individual's recent neurological function recovery effect by ratio. When the neurological function recovery effect of this sleep tends to be better, the ratio is larger, reflecting a more significant improvement above the baseline level, and therefore the corresponding neurological function recovery index is larger.

[0069] This gives the user's neurological recovery index for each sleep cycle.

[0070] Step S003: Based on the distribution of the user's instantaneous heart rate at all acquisition times during each sleep period, obtain the acquisition time when the user reaches the lowest point during each sleep period; based on the difference between the average instantaneous heart rate at multiple acquisition times before each sleep period and the instantaneous heart rate at multiple acquisition times before reaching the lowest point during each sleep period, and the interval between the first acquisition time and the acquisition time reaching the lowest point during each sleep period, obtain the user's heart rate unloading gradient during each sleep period.

[0071] It's important to note that after the body enters sleep, the cardiovascular system and overall body fluid metabolism undergo a transition from high to low load. Therefore, for a healthy and efficient sleep, the heart rate should rapidly and steeply drop to its lowest point during the night, exhibiting a steep slope in the first half of the time series. This slope reflects a rapid decrease in core body temperature, a sharp drop in metabolic rate, and a rapid retreat of the sympathetic nervous system. However, if the heart rate drops slowly, gradually, or even delays reaching its lowest point, daytime fatigue is more likely, indicating ineffective metabolism or persistent sympathetic nervous system excitation. Therefore, based on the changes in heart rate data before and during sleep, a heart rate unloading gradient is derived for each sleep cycle.

[0072] It should be further noted that, since data collection may be affected by noise, this invention first divides the collection time into multiple windows when obtaining the collection time of the user reaching the lowest point during each sleep cycle. Then, within the window containing the collection time of the lowest point, the collection time of the lowest point is selected. This invention reduces the impact of noise on obtaining the collection time of the lowest point by using multiple collection times to determine the window containing the collection time of the lowest point.

[0073] Specifically, the user will be in the first Before the next sleep The average instantaneous heart rate over all data collection times within a minute is denoted as the user's heart rate at the [number]th minute. Normal heart rate before the next sleep cycle. The preset time length in this embodiment refers to... The value is 30 because, under normal circumstances, the average instantaneous heart rate over all data collection times within 30 minutes before sleep can be used to quantify the user's overall heart rate before sleep.

[0074] Furthermore, each window contains The user's data collection time will be the first consecutive time period. All data collection time within a single sleep period is divided into several windows. Different windows do not contain the same data collection time. The user's data collection time is recorded in the first... The number of acquisition times contained in the last window during the next sleep cycle may be less than [a certain number]. The user in the The first sleep during the second sleep The acquisition time contained in each window is less than the user's acquisition time in the first window. The first sleep during the second sleep The number of acquisition times contained in a single window. In this embodiment, the preset number of acquisition times contained in a single window is... This example is used for illustration; other values ​​can be set in other implementations. This embodiment selects... This means using the preceding methods while maintaining consistency. I chose 20, but and The values ​​can be different. And the greater the likelihood of noise during data collection, the better. The larger the value, the less likely the data will be to be affected by noise during data collection. The smaller the value, the better.

[0075] Furthermore, the user in the first The first sleep during the second sleep The average instantaneous heart rate at all acquisition times in each window is denoted as the user's heart rate at the [number]th window. The first sleep during the second sleep The overall heart rate of the window.

[0076] Furthermore, the user in the first The window with the lowest overall heart rate among all windows during the next sleep cycle is recorded as the user's heart rate during the first sleep cycle. Low point window during the next sleep period.

[0077] Furthermore, the user in the first The time of minimum instantaneous heart rate within all data acquisition timeframes during the low-point window of the next sleep cycle is denoted as the user's heart rate during the first sleep cycle. The time to reach the lowest point during the next sleep cycle. In extreme cases, the user's data collection time during the first sleep cycle... Within the low-point window of the next sleep cycle, there are multiple sampling times corresponding to the user's minimum instantaneous heart rate. Therefore, among these multiple sampling times corresponding to the user's minimum instantaneous heart rate, the sampling time with the smallest value is selected as the user's minimum heart rate during the first sleep cycle. The time it takes to reach the lowest point during the next sleep cycle.

[0078] Furthermore, the user in the first Of all the data collection times during the next sleep period, the time shorter than the user's first sleep period... The time when the data reaches its lowest point during the next sleep cycle is recorded as the user's data collection time during the first sleep cycle. Data collection time during the rapid descent process within the second sleep phase.

[0079] Furthermore, acquiring user information in the first... The specific formula for calculating the heart rate unloading gradient during the next sleep phase is as follows:

[0080] ;

[0081] In the formula, Indicates the user's position in the first month. Heart rate unloading gradient during sleep, Indicates the user's position in the first month. The number of data collection times during the rapid descent process within the second sleep phase. Indicates the user's position in the first month. Normal heart rate before the next sleep cycle, in units of , Indicates the user's position in the first month. The rapid descent process during the second sleep phase Instantaneous heart rate at each sampling time, unit: , Indicates the user's position in the first month. The instantaneous heart rate at the time of reaching the lowest point during the second sleep cycle, in units of... , Indicates the user's position in the first month. The first data collection time within the sleep cycle, in seconds; Indicates the user's position in the first month. The time taken to reach the lowest point during the next sleep cycle, in seconds; Indicates the user's position in the first month. The last data collection time during the next sleep period is the same as the user's time in the first sleep period. The time interval of the first data collection time within the next sleep period This is an exponential function with a base of the natural constant, used in this embodiment to represent an inverse proportional relationship. Wherein, It is a positive value.

[0082] It should be noted that, The larger the value, the more likely the user is in the [number]th [period]. During the first sleep, the heart rate drops very rapidly. The more it matches the quality of sleep recovery, the steeper the heart rate curve tends to be in the first half of the night, reflecting the rapid decrease in the user's core body temperature and the decrease in the excitability of the sympathetic nervous system. The smaller the value, the more likely the user is in the first... During the second sleep phase, the heart rate drops to a low value relatively quickly. The larger the value, the more it indicates that the user is in the first... The better the recovery during the next sleep period; Indicates before falling asleep The average heart rate over a minute, in a physical sense, reflects the residual load remaining after an individual has experienced stress and fatigue throughout the day. The corresponding heart rate value can also be seen as the initial value of the sleep process, which gradually declines until it reaches the lowest heart rate value. This process is the gradual unloading of stress by the individual, as reflected by the heart rate. This represents the proportion of total sleep time required to reach the trough of heart rate. The higher this proportion, the greater the resistance to reaching deep sleep, and the less likely residual fatigue from the daytime can be effectively recovered. Therefore, the larger this value, the smaller the corresponding heart rate unloading gradient.

[0083] At this point, the user's heart rate unloading gradient during each sleep cycle is obtained.

[0084] Step S004: Based on the user's neurological function recovery index and heart rate unloading gradient during each sleep period, obtain the user's recovery quality index for each sleep period and conduct a quality assessment.

[0085] Specifically, acquiring users in the first The specific formula for calculating the recovery quality index of the next sleep session is as follows:

[0086] ;

[0087] In the formula, Indicates the user's position in the first month. The quality index of sleep recovery. Indicates the user's position in the first month. The neurological recovery index of the second sleep cycle Represents the absolute value function. Indicates the user's position in the first month. Heart rate unloading gradient during sleep.

[0088] It should be noted that the first part of the formula represents the total baseline of bodily recovery; that is, the product will only be significantly amplified when both dimensions are at a high level. For the second part, when both indicators tend to be larger, it reflects good physiological synergy, indicating rapid heart rate unloading and deep neural repair. In this case, the result value in this part tends to be... The final index obtained It is also larger. When this coordination is disrupted, the difference between the two indices is amplified, causing the result value in this part to tend towards Furthermore, the corresponding first half of the product value will also decrease, ultimately leading to The values ​​also tend to be smaller.

[0089] At this point, the user has obtained the [number] [unit / item / etc.]. The recovery quality index of a sleep session. The higher the recovery quality index of a user's sleep session, the better the fatigue recovery quality during that sleep session.

[0090] Another embodiment of the present invention provides a sleep fatigue recovery quality assessment system based on heart rate changes. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the above method steps S001 to S004.

Claims

1. A method for assessing the quality of fatigue recovery during sleep based on heart rate variability, characterized in that, The method includes the following steps: At fixed intervals, the system collects the user's instantaneous heart rate and body triaxial acceleration vector, and obtains the start and end times of each sleep cycle. Based on the difference between the instantaneous heart rate at each acquisition time within each sleep cycle and the mean instantaneous heart rate at all acquisition times within each sleep cycle, and the variance of the composite vector of the body's three-axis acceleration at each acquisition time within each sleep cycle and the surrounding acquisition times, the stability tendency factor of the user at each acquisition time within each sleep cycle is obtained; all adjacent heartbeat intervals of the user within each sleep cycle are obtained, and based on the stability tendency factor of the user at each acquisition time within each sleep cycle, and the difference between each adjacent heartbeat interval and the next adjacent heartbeat interval within each sleep cycle, the neurofunctional recovery index of the user in each sleep cycle is obtained. Based on the distribution of the user's instantaneous heart rate at all sampling times during each sleep period, the sampling time when the user reaches the lowest point during each sleep period is obtained; based on the mean of the user's instantaneous heart rate at multiple sampling times before each sleep period and the difference between the instantaneous heart rate at multiple sampling times before the sampling time when the user reaches the lowest point during each sleep period, and the interval between the first sampling time and the sampling time when the user reaches the lowest point during each sleep period, the heart rate unloading gradient of the user during each sleep period is obtained. Based on the user's neurological recovery index and heart rate unloading gradient during each sleep period, a recovery quality index is obtained for each sleep period, and a quality assessment is performed. The specific calculation steps for obtaining the stability tendency factor of the user at each acquisition time within each sleep cycle, based on the difference between the instantaneous heart rate of the user at each acquisition time within each sleep cycle and the mean of the instantaneous heart rate at all acquisition times within each sleep cycle, and the variance of the composite vector of the body's three-axis acceleration at each acquisition time within each sleep cycle and the surrounding acquisition times, are as follows: The user's first The first sleep during the second sleep The collection time is up to the first The first sleep during the second sleep During each collection time The collection time is denoted as the [number]th collection time. The first sleep during the second sleep The collection time within the time period of each collection time; among which... This indicates the number of data collection times contained within a preset time period; According to the user in The first sleep during the second sleep The instantaneous heart rate at each sampling time within the time period of the sampling time and the user's heart rate at the first sampling time. The difference in the mean instantaneous heart rate across all sampling times during the next sleep period, and the user's heart rate during the first sleep period. The first sleep during the second sleep The variance of the composite vector of the body's three-axis acceleration at all acquisition times within the time period of the acquisition time is obtained to obtain the user's three-axis acceleration vector at the first acquisition time. The first sleep during the second sleep Stability tendency factor for each collection time; According to the user in the The first sleep during the second sleep The instantaneous heart rate at each sampling time within the time period of the sampling time and the user's heart rate at the first sampling time. The difference in the mean instantaneous heart rate across all sampling times during the next sleep period, and the user's heart rate during the first sleep period. The first sleep during the second sleep The variance of the composite vector of the body's three-axis acceleration at all acquisition times within the time period of the acquisition time is obtained to obtain the user's three-axis acceleration vector at the first acquisition time. The first sleep during the second sleep The specific calculation steps for the stability tendency factor at each collection time are as follows: In the formula, Indicates the user's position in the first month. The first sleep during the second sleep Stability tendency factor for each collection time, Indicates the user's position in the first month. The first sleep during the second sleep The first time period within the collection time period Instantaneous heart rate at each acquisition time, Indicates the user's position in the first month. The mean instantaneous heart rate at all sampling times during the subsleep period. Represents the absolute value function. It is an exponential function with the natural constant as its base. Indicates the user's position in the first month. The first sleep during the second sleep The number of collection times within the time period of each collection time. Indicates the user's position in the first month. The first sleep during the second sleep The variance of the composite vector of body triaxial accelerations at all acquisition times within a given acquisition time period. This represents a logarithmic function with the natural constant as its base. Represents gravitational acceleration; The specific calculation steps for obtaining the user's neurofunctional recovery index for each sleep cycle, based on the user's stability tendency factor at each sampling time within each sleep cycle and the difference between each adjacent heartbeat interval and the next adjacent heartbeat interval within each sleep cycle, are as follows: The user in the first The first sleep during the second sleep The collection time is up to the first The first sleep during the second sleep During each collection time The collection time is recorded as the user's collection time in the [number]th [year]. The first sleep during the second sleep The collection time within the time period of each collection time; among which... This indicates the number of data collection times contained within a preset time period; According to the user in The first sleep during the second sleep The first and last collection times within the time period of each collection time are used to obtain the user's data at the [number]th [time period]. The first sleep during the second sleep The time period in which each data collection time is located; Get users in the first The specific formula for calculating the neurological recovery index of the second sleep cycle is as follows: In the formula, Indicates the user's position in the first month. The neurological recovery index of the second sleep cycle, Indicates the user's position in the first month. The first sleep during the second sleep The time period in which each data collection time is located Root mean square of the difference Indicates the user's position in the first month. The first sleep during the second sleep Within the time period of each collection time Quantity, Indicates the user's position in the first month. The first sleep during the second sleep The first time period within the collection time period indivual The value, Indicates the user's position in the first month. The first sleep during the second sleep The first time period within the collection time period indivual The value, Indicates the user's position in the first month. The first sleep during the second sleep Stability tendency factor for each collection time, Indicates the user's position in the first month. The second sleep and the first Before the next sleep The mean of the stable tendency factor for all data collection times across all sleep periods within a day. Indicates the user's position in the first month. The number of data collection times during each sleep cycle, among which... Indicates the preset period length; The specific steps for calculating the user's heart rate unloading gradient during each sleep cycle are as follows: The difference between the average instantaneous heart rate at multiple sampling times before each sleep cycle and the instantaneous heart rate at multiple sampling times before reaching the lowest point within each sleep cycle, and the interval between the first sampling time and the time when the lowest point is reached within each sleep cycle. The user in the first Before the next sleep The average instantaneous heart rate over all data collection times within a minute is denoted as the user's heart rate at the [number]th minute. Normal heart rate before the next sleep cycle, among which... Indicates the preset time length; The user in the first Of all the data collection times during the next sleep period, the time shorter than the user's first sleep period... The time when the data reaches its lowest point during the next sleep cycle is recorded as the user's data collection time during the first sleep cycle. Data acquisition time during the rapid descent process in the second sleep; Get users in the first The specific formula for calculating the heart rate unloading gradient during the next sleep phase is as follows: In the formula, Indicates the user's position in the first month. Heart rate unloading gradient during sleep, Indicates the user's position in the first month. The number of data collection times during the rapid descent process within the second sleep phase. Indicates the user's position in the first month. Normal heart rate before sleep. Indicates the user's position in the first month. The rapid descent process during the second sleep phase Instantaneous heart rate at each acquisition time, Indicates the user's position in the first month. Instantaneous heart rate at the time of data collection when the heart rate reaches its lowest point during the second sleep cycle. Indicates the user's position in the first month. The first data collection time during the next sleep cycle. Indicates the user's position in the first month. The time it takes for the sample to reach its lowest point during the next sleep cycle. Indicates the user's position in the first month. The last data collection time during the next sleep period is related to the user's... The time interval of the first data collection time within the next sleep period It is an exponential function with the natural constant as its base; The specific calculation steps for obtaining the user's recovery quality index for each sleep cycle based on the user's neurological recovery index and heart rate unloading gradient are as follows: In the formula, Indicates the user's position in the first month. The quality index of sleep recovery. Indicates the user's position in the first month. The neurological recovery index of the second sleep cycle Represents the absolute value function. Indicates the user's position in the first month. Heart rate unloading gradient during sleep.

2. The method for assessing the quality of fatigue recovery during sleep based on heart rate changes according to claim 1, characterized in that, The specific calculation steps for collecting the user's instantaneous heart rate and body triaxial acceleration vector at fixed intervals, and obtaining the start and end times of each sleep cycle are as follows: With the collection interval as The system acquires data in real-time, capturing the user's instantaneous heart rate and three-axis acceleration vector at each acquisition point. Indicates the preset data collection interval; Obtain the start and end times of each user's sleep session.

3. The method for assessing the quality of fatigue recovery during sleep based on heart rate changes according to claim 1, characterized in that, The specific calculation steps for obtaining the time when the user reaches the lowest point in each sleep cycle based on the distribution of the user's instantaneous heart rate across all data collection times during each sleep cycle are as follows: Contain in each window The user's data collection time will be the first consecutive time period. All data acquisition times within a single sleep period are divided into several windows; however, different windows do not contain the same data acquisition time. The number of data collection times included in a preset window; The user in the first The first sleep during the second sleep The average instantaneous heart rate at all acquisition times in each window is denoted as the user's heart rate at the [number]th window. The first sleep during the second sleep Overall heart rate in each window; The user in the first The window with the lowest overall heart rate among all windows during the next sleep cycle is recorded as the user's heart rate during the first sleep cycle. Low point window during the second sleep; The user in the first The time of minimum instantaneous heart rate within all data acquisition timeframes during the low-point window of the next sleep cycle is denoted as the user's heart rate during the first sleep cycle. The time it takes for the data to reach its lowest point during the next sleep cycle.

Images