A sleep quality analysis system based on human HRV measurement

Abstract

The present invention discloses a sleep quality analysis system based on human HRV measurement, which relates to the technical field of sleep quality. The preliminary analysis module can quickly judge the fluctuation of the user's sleep quality through the calculation of the heart rate variability factor HRV. When the HRV value exceeds the preset threshold, the system automatically sends a sleep quality fluctuation command to the cloud platform, so as to respond to the change of the sleep state in a timely manner. In addition, the neural activity module analyzes the balance state of the parasympathetic nerve and the sympathetic nerve to construct the neural activity ratio Jhb, which can clarify the activity bias of the nervous system of the user in the sleep state and provide a reliable basis for the in-depth analysis of the sleep state. The sleep state comprehensive analysis module further combines the heart rate recovery ratio Xhb and the neural activity ratio Jhb, and calculates the sleep quality evaluation index Sgzs through the trained sleep quality prediction model, further realizing the quantitative evaluation of the user's sleep quality.

Description

Technical Field

[0001] The present invention relates to the technical field of sleep quality, and particularly to a sleep quality analysis system based on human HRV measurement. Background Art

[0002] With the increasing demand for personalized health management, sleep monitoring has gradually become one of the important research directions in modern health technology. As a physiological parameter, human HRV (heart rate variability) can reflect the activity of the cardiac autonomic nervous system, especially the balance state between the sympathetic nerve and the parasympathetic nerve. By analyzing HRV, the health status and nerve activity of the human body during sleep can be deeply understood, and thus a reliable basis can be provided for evaluating sleep quality.

[0003] There are still some deficiencies in the current analysis of the impact of physiological parameters such as human HRV on sleep quality. Most of the existing sleep monitoring systems rely on simple sleep duration and the distribution of sleep stages, lacking an assessment of the deep relationship between the activity of the autonomic nervous system and sleep quality, lacking real-time monitoring and feedback on HRV fluctuations and nerve activity states, and being prone to ignoring subtle changes in sleep quality, resulting in insufficient accuracy and personalization of the assessment results. In addition, many systems only focus on HRV changes while ignoring the importance of parameters such as heart rate recovery and sleep state, and it is difficult to provide sufficient feedback and compensation measures for the overall sleep health status of users. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides a sleep quality analysis system based on human HRV measurement, which solves the problems in the above background art.

[0005] To achieve the above object, the present invention is realized through the following technical solutions: A sleep quality analysis system based on human HRV measurement, comprising a data monitoring module, a preliminary analysis module, a nerve activity module, a comprehensive sleep state analysis module, and a feedback module;

[0006] The data monitoring module is used to monitor the health status of a user according to a plurality of groups of monitoring instruments, so as to obtain relevant physiological data, relevant electroactivity data, and relevant personalized data of the user during sleep;

[0007] The preliminary analysis module is used to generate a heart rate variability factor HRV according to the relevant physiological data. If the heart rate variability factor HRV exceeds a preset threshold, a sleep quality fluctuation command is sent to the cloud platform;

[0008] The nerve activity module is used to analyze the balance state between the parasympathetic nerve and the sympathetic nerve according to relevant electroactivity data after receiving the sleep quality fluctuation command, so as to construct the nerve activity ratio Jhb, and based on the nerve activity ratio Jhb, judge the nerve activity bias of the current user in the sleep state;

[0009] The sleep state comprehensive analysis module is used to construct the heart rate recovery ratio Xhb according to relevant personalized data, and combine the nerve activity ratio Jhb with the trained sleep quality prediction model to fit and obtain the sleep quality evaluation index Sgzs;

[0010] The feedback module is used to preset the evaluation threshold Q, and compare and analyze the sleep quality evaluation index Sgzs with the evaluation threshold Q to comprehensively evaluate and judge the sleep quality state of the current user, and take corresponding compensation means according to the quality state.

[0011] Preferably, the data monitoring module includes a deployment unit, a first acquisition unit, a second acquisition unit and a third acquisition unit;

[0012] The deployment unit is used to deploy several groups of monitoring instruments around the user before the user goes to sleep. Among them, several groups of monitoring instruments include a health bracelet, an electrocardiogram sensor, a respiration monitoring sensor, an accelerometer, a camera and a gyroscope;

[0013] The first acquisition unit is used to monitor and record in real time the relevant physiological data of the user in the sleep state. Among them, the relevant physiological data includes the RR interval Rjj and the respiration difference Hxc in each monitoring period;

[0014] The second acquisition unit is used to monitor and record in real time the relevant electroactivity data of the user in the sleep state. Among them, the relevant electroactivity data includes the power of the heart rate variability signal in the high-frequency range and the power of the heart rate variability signal in the low-frequency range;

[0015] The third acquisition unit is used to monitor and record in real time the relevant personalized data of the user in the sleep state. Among them, the relevant personalized data includes the maximum heart rate value Xz max at the maximum heart rate value and then the heart rate value Xz(t) when entering the recovery state t, the resting heart rate value Xz in the sleep state rest the number of times Hzsc of changing body posture during sleep, the total sleep duration Zssc and the number of times Jxcs of waking up during sleep.

[0016] Preferably, the preliminary analysis module includes an HRV analysis unit and a preliminary judgment unit;

[0017] The HRV analysis unit extracts the RR intervals Rjj within each monitoring period from the relevant physiological data, and uses a statistical algorithm to obtain the heart rate variability factor HRV. The heart rate variability factor HRV is specifically obtained in the following manner:

[0018]

[0019] In the formula, N represents the number of RR intervals, i = 1, 2, 3,..., N, and Rjj i represents the i-th RR interval, and Rjj avg represents the average RR interval.

[0020] Preferably, the preliminary judgment unit is used to preset a threshold, and compare the heart rate variability factor HRV with the set threshold to preliminarily judge the heart rate fluctuation difference of the current user in the sleep state. The specific judgment content is as follows:

[0021] If the heart rate variability factor HRV exceeds the set threshold, it will be preliminarily judged that the heart rate fluctuation of the current user in the sleep state is in a normal state at this time;

[0022] If the heart rate variability factor HRV does not exceed the set threshold, it will be preliminarily judged that the heart rate fluctuation of the current user in the sleep state is not in a normal state at this time, and a sleep quality fluctuation command will be sent to the cloud platform.

[0023] Preferably, the neural activity module includes a frequency domain analysis unit and a neural bias selection unit;

[0024] The frequency domain analysis unit is used to analyze the balance state between the parasympathetic nerve and the sympathetic nerve according to the relevant electrical activity data, so as to obtain the power of the heart rate variability signal in the high-frequency range and the power of the heart rate variability signal in the low-frequency range respectively, and record them as HF power P 1 and LF power P 2 , and the specific acquisition method is as follows:

[0025]

[0026] In the formula, Ppz(f) represents the power spectral density; the integration range of HF power P 1 is the high-frequency range, that is, 0.15Hz - 0.4Hz; the integration range of LF power P 2 is the low-frequency range, that is, 0.04Hz - 0.15Hz; f represents the frequency.

[0027] Preferably, the neural bias selection unit is used to according to HF power P 1 and LF power P 2, the neural activity ratio Jhb is calculated, and the neural activity ratio Jhb is obtained through the following formula:

[0028]

[0029] Based on the numerical value of the neural activity ratio Jhb, the neural activity bias of the current user in the sleep state is judged. The specific judgment content is as follows:

[0030] If the neural activity ratio Jhb > 1, it is judged that the activity of the parasympathetic nerve of the current user dominates in the sleep state at this time;

[0031] If the neural activity ratio Jhb < 1, it is judged that the activity of the sympathetic nerve of the current user dominates in the sleep state at this time;

[0032] If the neural activity ratio Jhb = 1, it is judged that the activities of the sympathetic nerve and the parasympathetic nerve of the current user are in a relatively stable state in the sleep state at this time.

[0033] Preferably, the sleep state comprehensive analysis module includes a recovery ability unit, a change unit and a comprehensive analysis unit;

[0034] The recovery ability unit is used to construct a heart rate recovery ratio Xhb according to relevant personalized data, and is specifically obtained through the following formula:

[0035]

[0036] In the formula, Xz max represents the maximum heart rate value in the sleep state, Xz rest represents the resting heart rate value in the sleep state, and Xz(t) represents the heart rate value at time t when entering the recovery state after the maximum heart rate value.

[0037] Preferably, the change unit is used to respectively obtain a body position change rate Tbz and a sleep fragmentation rate Spz according to relevant personalized data, and is specifically obtained through the following formula:

[0038]

[0039] In the formula, Hzsc represents the number of times of changing body posture during sleep, Jxcs represents the number of times of waking up during sleep, and Zssc represents the total sleep duration.

[0040] Preferably, the comprehensive analysis unit is used to construct a sleep quality prediction model using a convolutional neural network, and input the neural activity ratio Jhb, the body position change rate Tbz, the sleep fragmentation rate Spz and the heart rate recovery ratio Xhb into the sleep quality prediction model. After linear normalization processing, the sleep quality evaluation index Sgzs is obtained by fitting, and is specifically obtained through the following method:

[0041]

[0042] In the formula, Hxc represents the respiratory difference value, and a 1 、a 2 、a 3 、a 4 and a 5 respectively represent the weight values of the heart rate recovery ratio Xhb, the neural activity ratio Jhb, the sleep fragmentation rate Spz, the body position change rate Tbz, and the respiratory difference value Hxc, and V represents a correction constant.

[0043] Preferably, the feedback module includes a comparison unit and a compensation unit;

[0044] The comparison unit is used to preset an evaluation threshold Q, and by comparing and analyzing the evaluation threshold Q with the sleep quality evaluation index Sgzs, to comprehensively evaluate and judge the quality state of the current user's sleep. The specific content is as follows:

[0045] If the sleep quality evaluation index Sgzs exceeds the evaluation threshold Q, it will be judged that the quality of the current user's sleep is in a normal state at this time;

[0046] If the sleep quality evaluation index Sgzs does not exceed the evaluation threshold Q, it will be judged that the quality of the current user's sleep is not in a normal state at this time;

[0047] The compensation unit is used to take corresponding compensation measures according to the corresponding sleep quality obtained in the comparison unit. The specific content is as follows:

[0048] If it is in a normal state, it will be recommended that the user maintain the current sleep environment, and at the same time prompt the user to perform mild relaxation activities before going to bed. The relaxation activities include meditation, deep breathing exercises, and stretching exercises;

[0049] If it is not in a normal state, it will help the user regulate the sympathetic nerve activity and the parasympathetic nerve activity, and guide the user to perform meditation, progressive muscle relaxation, and deep breathing exercises, and at the same time remind the user to adjust the work and rest time to maintain a regular sleep schedule.

[0050] The present invention provides a sleep quality analysis system based on human HRV measurement, which has the following beneficial effects:

[0051] (1) The data monitoring module can obtain the user's physiological data, electroactivity data, and personalized data through a variety of monitoring instruments to comprehensively capture the physiological changes during sleep; the preliminary analysis module can quickly judge the fluctuation of the user's sleep quality by calculating the heart rate variability factor HRV. When the HRV value exceeds the preset threshold, the system automatically sends a sleep quality fluctuation command to the cloud platform to timely respond to the change of the sleep state. In addition, the neural activity module can analyze the balance state of the parasympathetic nerve and the sympathetic nerve to construct the neural activity ratio Jhb, which can clarify the activity bias of the nervous system of the user in the sleep state and provide a reliable basis for the in-depth analysis of the sleep state. The sleep state comprehensive analysis module further combines the heart rate recovery ratio Xhb and the neural activity ratio Jhb, and through the trained sleep quality prediction model, calculates the sleep quality evaluation index Sgzs to further realize the quantitative evaluation of the user's sleep quality. Finally, the feedback module can quickly judge whether the user's sleep state is normal by comparing Sgzs with the preset evaluation threshold Q, and provide personalized compensation measures, such as adjusting the sleep environment and providing relaxation training, etc., when the sleep quality is poor to help the user improve the sleep quality. Through this system, the accuracy and personalization of sleep monitoring can be effectively improved, and the sleep quality problems can be timely discovered and feedback, so as to help users better manage and improve their own sleep health status, which has significant practical application value.

[0052] (2) Through the setting of the preliminary analysis module, the rapid and accurate judgment of the heart rate fluctuation of the user in the sleep state is further realized. The HRV analysis unit can extract the RR interval of each monitoring period from the relevant physiological data and calculate the heart rate variability factor HRV by using statistical algorithms. This HRV factor not only comprehensively reflects the user's heart rate fluctuation through an accurate calculation formula based on the number of RR intervals and their mean value, but also can provide a direct evaluation of the activity of the autonomic nervous system, especially the balance of the sympathetic nerve and the parasympathetic nerve.

[0053] (3) The nerve bias selection unit calculates the nerve activity ratio Jhb based on the HF power and LF power, so as to further analyze the nerve activity bias of the user during sleep. Specifically, if the nerve activity ratio Jhb is greater than 1, it indicates that the parasympathetic nerve activity dominates, which may mean that the user has a better relaxation state and higher sleep quality; if the nerve activity ratio Jhb is less than 1, the sympathetic nerve activity dominates, which may indicate that the user is in a tense or stressed state, which may affect the sleep quality; if the nerve activity ratio Jhb is equal to 1, it indicates that the sympathetic and parasympathetic nerve activities are relatively balanced, and the user's nerve state is in a relatively stable state. This analysis ability enables the system to understand the nervous system state of the user during sleep in detail, helps to timely discover and adjust the factors that may affect sleep quality, and thus provides personalized sleep optimization suggestions for the user. Through accurate nerve activity monitoring and analysis, the present invention can more effectively support sleep health management and improve the user's sleep quality and overall quality of life.

[0054] (4) The calculation result of the heart rate recovery ratio can reveal the user's heart rate recovery after sleep, provides profound insights into sleep quality, and helps to evaluate the impact of sleep on heart health; the change unit calculates the body position change rate and sleep fragmentation rate through further analysis of personalized data; the calculation of the body position change rate involves the number of changes in the body posture during sleep, and the sleep fragmentation rate reflects the number of awakenings during sleep and the total sleep duration. These indicators can reveal the frequency of body position adjustment and the continuity of sleep of the user during sleep, and further reflect the stability and depth of sleep. A higher body position change rate and sleep fragmentation rate may indicate poor sleep quality, lack of sleep coherence or being disturbed. The comprehensive analysis unit integrates these data to provide a multi-faceted sleep quality assessment, which can reveal the user's sleep recovery ability and changes during sleep. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a block diagram of a sleep quality analysis system based on human HRV measurement according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0057] Embodiment 1

[0058] Please refer to Figure 1, the present invention provides a sleep quality analysis system based on human HRV measurement, including a data monitoring module, a preliminary analysis module, a neural activity module, a comprehensive sleep state analysis module and a feedback module;

[0059] The data monitoring module is used to monitor the health status of the user according to a number of monitoring instruments to obtain relevant physiological data, relevant electroactivity data and relevant personalized data of the user during sleep;

[0060] The preliminary analysis module is used to generate a heart rate variability factor HRV according to the relevant physiological data. If the heart rate variability factor HRV exceeds a preset threshold, a sleep quality fluctuation command is sent to the cloud platform;

[0061] The neural activity module is used to analyze the balance state between the parasympathetic nerve and the sympathetic nerve according to the relevant electroactivity data after receiving the sleep quality fluctuation command, so as to construct a neural activity ratio Jhb. Based on the neural activity ratio Jhb, the neural activity bias of the current user in the sleep state is judged;

[0062] The comprehensive sleep state analysis module is used to construct a heart rate recovery ratio Xhb according to the relevant personalized data, and combine it with the neural activity ratio Jhb, and through a trained sleep quality prediction model, the sleep quality evaluation index Sgzs is obtained by fitting;

[0063] The feedback module is used to preset an evaluation threshold Q, and compare and analyze the sleep quality evaluation index Sgzs with the evaluation threshold Q to comprehensively judge the quality state of the current user's sleep, and take corresponding compensation measures according to the quality state.

[0064] During the operation of this system, through the collaborative work of multiple modules, it can achieve accurate monitoring, evaluation, and feedback on the user's sleep quality. First, the system's data monitoring module can obtain the user's physiological data, electroactivity data, and personalized data during sleep through multiple monitoring instruments, ensuring the comprehensiveness and multi-dimensionality of data collection and providing a reliable basis for subsequent analysis. The preliminary analysis module effectively captures the quality fluctuations during sleep and has the ability to respond in a timely manner by generating HRV factors in real time and issuing a sleep quality fluctuation command when the HRV exceeds the preset threshold. Secondly, the neural activity module analyzes the balance state between the parasympathetic nerve and the sympathetic nerve to construct the neural activity ratio Jhb, accurately judging the user's neural activity bias. This analysis can delicately reflect the activity of the user's autonomic nervous system during sleep, thus laying a foundation for personalized evaluation. The comprehensive sleep state analysis module further constructs the heart rate recovery ratio Xhb by combining personalized data and fits and generates the sleep quality evaluation index Sgzs through a trained sleep quality prediction model, further realizing a comprehensive quantitative evaluation of the user's sleep quality. Finally, the feedback module can accurately identify the current user's sleep quality state by comparing the sleep quality evaluation index Sgzs with the pre-set evaluation threshold Q and take corresponding compensation measures. This can not only make personalized adjustments and improvements according to the actual sleep situation but also provide timely feedback to the user to help improve their sleep quality. This multi-level systematic method based on physiological data, neural activity, and personalized analysis maximally improves the accuracy and effectiveness of sleep quality evaluation and provides an innovative solution for sleep health management.

[0065] Embodiment 2

[0066] Please refer to Figure 1 , specifically: The data monitoring module includes a deployment unit, a first acquisition unit, a second acquisition unit, and a third acquisition unit;

[0067] The deployment unit is used to deploy several groups of monitoring instruments around the user before the user goes to sleep. Among them, the several groups of monitoring instruments include a health bracelet, an electrocardiogram sensor, a respiratory monitoring sensor, an accelerometer, a camera, and a gyroscope;

[0068] The first acquisition unit is used to monitor and record the relevant physiological data of the user in the sleep state in real time. Among them, the relevant physiological data includes the RR interval Rjj and the respiratory difference Hxc in each monitoring period;

[0069] The second acquisition unit is used to monitor and record the relevant electroactivity data of the user in the sleep state in real time. Among them, the relevant electroactivity data includes the power of the heart rate variability signal in the high-frequency range and the power of the heart rate variability signal in the low-frequency range;

[0070] The third acquisition unit is used to monitor and record in real time relevant personalized data of the user during sleep. Among them, the relevant personalized data includes the maximum heart rate value Xz during sleep. max The heart rate value Xz(t) when entering the recovery state t after the maximum heart rate value, the resting heart rate value Xz during sleep. rest The number of times Hzsc of changing body posture during sleep, the total sleep duration Zssc, and the number of times Jxcs of waking up during sleep.

[0071] The above RR interval Rjj is mainly obtained through an electrocardiogram (ECG) sensor. The ECG sensor can accurately measure the time interval between each heartbeat of the heart, that is, the RR interval. This sensor can be an electrode attached to the skin or a high-precision sensor embedded in a wearable device.

[0072] The respiratory difference Hxc is mainly measured through a respiratory monitoring sensor or a chest strap (such as a strap with a strain gauge on the chest). The respiratory sensor records the respiratory frequency and depth, and then calculates the respiratory difference. It can also be obtained through the internal sensor of an advanced sleep monitoring device.

[0073] The heart rate value is usually obtained through a photoplethysmogram (PPG) sensor. The PPG sensor measures the heart rate by detecting minute optical changes in blood vessels under the skin. This kind of sensor is commonly found in health bracelets and smart watches.

[0074] The number of times of changing body posture during sleep is usually detected by an accelerometer and a gyroscope. The accelerometer monitors the acceleration change of the body, and the gyroscope detects the rotation of the posture. The combination of the two can judge the change of body posture. These sensors are commonly found in smart watches, health bracelets, and specialized sleep monitoring devices.

[0075] The number of times of waking up during sleep is monitored and obtained through a camera.

[0076] The power spectral density is mainly obtained through an electrocardiogram (ECG) monitoring instrument or the built-in analysis function of a sleep monitoring device. The power spectral density is part of the frequency domain analysis.

[0077] In this embodiment, before the user enters the sleep state, the deployment unit can pre-arrange several sets of monitoring instruments (such as health bracelets) to further ensure the full coverage and stability of the monitoring environment. The first acquisition unit can collect and record the user's physiological data in real time, including the RR interval Rjj and the respiratory difference, ensuring continuous monitoring of the activities of the heart and respiratory system and providing detailed time series data for subsequent sleep quality analysis. The second acquisition unit can accurately analyze the activity states of the user's sympathetic and parasympathetic nerves by collecting the power of the heart rate variability signal in the high-frequency and low-frequency ranges, which is of great significance for evaluating the balance of the user's autonomic nervous system and sleep depth. The third acquisition unit focuses on the monitoring of the user's personalized data, including key parameters such as the maximum heart rate value, heart rate recovery value, and resting heart rate value, as well as behavioral data such as the number of changes in body posture during sleep, total sleep duration, and number of awakenings, providing a comprehensive observation and record of the user's overall sleep process. Through these multi-dimensional data acquisitions, the present invention can more accurately evaluate the user's sleep quality and provide customized feedback and improvement suggestions by analyzing the correlations between these data. Such a system can not only monitor physiological and electrical activity data but also combine the user's personalized characteristics, further improving the accuracy and reliability of sleep quality assessment and helping the user optimize their sleep health.

[0078] Embodiment 3

[0079] Please refer to Figure 1 , specifically: The preliminary analysis module includes an HRV analysis unit and a preliminary judgment unit;

[0080] The HRV analysis unit extracts the RR interval Rjj within each monitoring period from the relevant physiological data and uses a statistical algorithm to obtain the heart rate variability factor HRV. The heart rate variability factor HRV is specifically obtained in the following manner:

[0081]

[0082] In the formula, N represents the number of RR intervals, i = 1, 2, 3,..., N, and Rjj i represents the i-th RR interval, and Rjj avg represents the average RR interval; further reflecting the overall heart rate fluctuation;

[0083] The preliminary judgment unit is used to preset a threshold and compare the heart rate variability factor HRV with the set threshold to preliminarily judge the heart rate fluctuation difference of the current user in the sleep state. The specific judgment content is as follows:

[0084] If the heart rate variability factor HRV exceeds the set threshold, it will be preliminarily judged that the heart rate fluctuation of the current user in the sleep state is in a normal state;

[0085] When the heart rate variability factor HRV does not exceed the set threshold, it is preliminarily determined that the heart rate fluctuation of the current user during sleep is not in a normal state, and a sleep quality fluctuation command will be sent to the cloud platform.

[0086] In this embodiment, the beneficial effect of the present invention is that through the design of the preliminary analysis module, it is possible to realize real-time analysis and preliminary judgment of the heart rate fluctuation of the user during sleep. Specifically, the HRV analysis unit can extract the RR interval Rjj of each period from the monitored physiological data, and calculate the heart rate variability factor HRV using statistical algorithms. The HRV factor can reflect the overall heart rate fluctuation, providing an important basis for analyzing the activity of the autonomic nervous system of the user during sleep. Further, the preliminary judgment unit compares the calculated HRV factor with the preset threshold by setting the threshold, and issues a sleep quality fluctuation command according to the comparison result, providing a timely feedback mechanism for subsequent analysis and intervention. Through this mechanism, the system can not only quickly respond to the heart rate fluctuation during the user's sleep, but also judge potential problems at an early stage and initiate further sleep quality analysis and improvement programs. This preliminary judgment function based on HRV analysis provides real-time and accurate sleep monitoring and early warning for users, helping users better manage their sleep health.

[0087] Embodiment 4

[0088] Please refer to Figure 1 , specifically: the neural activity module includes a frequency domain analysis unit and a neural bias selection unit;

[0089] The frequency domain analysis unit is used to analyze the balance state between the parasympathetic nerve and the sympathetic nerve according to the relevant electrical activity data, so as to obtain the power of the heart rate variability signal in the high-frequency range and the power of the heart rate variability signal in the low-frequency range respectively, and record them as HF power P 1 and LF power P 2 , and the specific acquisition method is as follows:

[0090]

[0091] In the formula, Ppz(f) represents the power spectral density; the integration range of the HF power P 1 is the high-frequency range, that is, 0.15Hz - 0.4Hz; the integration range of the LF power P 2 is the low-frequency range, that is, 0.04Hz - 0.15Hz; f represents the frequency.

[0092] The neural bias selection unit is used to based on the HF power P 1 and LF power P 2, the neural activity ratio Jhb is calculated, and the neural activity ratio Jhb is obtained through the following formula:

[0093]

[0094] Based on the numerical value of the neural activity ratio Jhb, the neural activity bias of the current user in the sleep state is judged. The specific judgment content is as follows:

[0095] If the neural activity ratio Jhb > 1, it is judged that the activity of the parasympathetic nerve of the current user dominates in the sleep state at this time;

[0096] If the neural activity ratio Jhb < 1, it is judged that the activity of the sympathetic nerve of the current user dominates in the sleep state at this time;

[0097] If the neural activity ratio Jhb = 1, it is judged that the activity of the sympathetic nerve and the activity of the parasympathetic nerve of the current user are in a relatively stable state in the sleep state at this time.

[0098] In this embodiment, through the design of the neural activity module, the activities of the autonomic nervous system of the user in the sleep state can be effectively analyzed, and the balance state between the sympathetic nerve and the parasympathetic nerve can be determined. Through the frequency domain analysis unit, the system can perform a detailed frequency domain analysis on the heart rate variability signal in the high frequency range and the low frequency range, and calculate the HF power and the LF power respectively. This method of separating and calculating the high and low frequency powers enables the activities of the sympathetic and parasympathetic nerves in the autonomic nervous system to be accurately distinguished, providing detailed neural activity data. In addition, the neural bias selection unit quantifies the ratio into a bias index of the nervous system activity through the neural activity ratio Jhb formula. According to the numerical value of the neural activity ratio Jhb, the system can automatically judge the dominant direction of the neural activity of the user in the sleep state. When the neural activity ratio Jhb > 1, it indicates that the activity of the parasympathetic nerve dominates, meaning that the user is in a deep relaxation state, which is beneficial to deep sleep; while when Jhb < 1, the sympathetic nerve dominates, which means that the user may be in light sleep or even a potential stress state. In addition, when Jhb = 1, it means that the sympathetic and parasympathetic nerves are in a balanced state, which is a relatively stable neural activity state. Through this mechanism, the present invention can not only monitor the neural activity state of the user in real time, but also provide a basis for the subsequent sleep quality assessment, and provide reliable guidance for further sleep intervention and optimization.

[0099] Embodiment 5

[0100] Please refer to Figure 1 , specifically: the sleep state comprehensive analysis module includes a recovery ability unit, a change unit and a comprehensive analysis unit;

[0101] The recovery ability unit is used to construct the heart rate recovery ratio Xhb based on relevant personalized data, and it is specifically obtained through the following formula:

[0102]

[0103] In the formula, Xz max represents the maximum heart rate value in the sleep state, Xz rest represents the resting heart rate value in the sleep state, and Xz(t) represents the heart rate value at time t after entering the recovery state from the maximum heart rate value. This formula is used to quantify the degree of heart rate recovery after exercise.

[0104] The change unit is used to obtain the body position change rate Tbz and the sleep fragmentation rate Spz respectively based on relevant personalized data, and it is specifically obtained through the following formula:

[0105]

[0106] In the formula, Hzsc represents the number of times of changing body postures during sleep, Jxcs represents the number of times of waking up during sleep, and Zssc represents the total sleep duration.

[0107] In this embodiment, the recovery ability unit evaluates the user's recovery ability in the sleep state through the calculation of the heart rate recovery ratio Xhb. The calculation formula of this ratio comprehensively considers the maximum heart rate value, the resting heart rate value, and the heart rate value when entering the recovery state after the maximum heart rate value, so as to provide a comprehensive understanding of the user's heart rate recovery ability. By evaluating the recovery ability, the system can identify whether the user effectively recovers during sleep, so as to judge their sleep quality and health status. Secondly, the change unit further analyzes the sleep quality by obtaining the body position change rate Tbz and the sleep fragmentation rate Spz. The calculation of the body position change rate Tbz takes into account the number of changes in body postures during sleep and the total sleep duration, reflecting the activity level of the user during sleep. The sleep fragmentation rate Spz evaluates the continuity and stability of the user's sleep by calculating the ratio of the number of times of waking up during sleep to the total sleep duration. The comprehensive analysis of these indicators helps to understand the user's sleep fragmentation situation and the impact of body position changes on sleep quality. By comprehensively analyzing these data, the system can provide a more accurate sleep quality assessment, and can identify and adjust the factors that may affect sleep quality, so as to provide personalized improvement suggestions for the user and optimize their sleep health management. This multi-dimensional data analysis ability enables the present invention to more effectively support the scientific assessment and improvement of sleep quality, and improve the user's overall health level and quality of life.

[0108] Embodiment 6

[0109] Please refer to Figure 1, specifically: the comprehensive analysis unit is used to build a sleep quality prediction model using a convolutional neural network, and input the neural activity ratio Jhb, body position change rate Tbz, sleep fragmentation rate Spz, and heart rate recovery ratio Xhb into the sleep quality prediction model. After linear normalization processing, the sleep quality evaluation index Sgzs is obtained by fitting, and it is specifically obtained through the following method:

[0110]

[0111] In the formula, Hxc represents the respiratory difference, a 1 , a 2 , a 3 , a 4 , and a 5 respectively represent the weight values of the heart rate recovery ratio Xhb, neural activity ratio Jhb, sleep fragmentation rate Spz, body position change rate Tbz, and respiratory difference Hxc. Among them, 0 < a 1 ≤1, 0 < a 2 ≤1, 0 < a 3 ≤1, 0 < a 4 ≤1, 0 < a 5 ≤1, and a 1 +a 2 +a 3 +a 4 +a 5 =1, V represents a correction constant.

[0112] The respiratory difference Hxc refers to the difference between the maximum respiratory depth and the minimum reciprocal respiratory depth of the user during sleep.

[0113] The feedback module includes a comparison unit and a compensation unit;

[0114] The comparison unit is used to preset an evaluation threshold Q, and by comparing and analyzing the evaluation threshold Q with the sleep quality evaluation index Sgzs, the quality state of the current user's sleep is comprehensively judged. The specific content is as follows:

[0115] If the sleep quality evaluation index Sgzs exceeds the evaluation threshold Q, it will be judged that the quality of the current user's sleep is in a normal state at this time;

[0116] If the sleep quality evaluation index Sgzs does not exceed the evaluation threshold Q, it will be judged that the quality of the current user's sleep is not in a normal state at this time;

[0117] The compensation unit is used to take corresponding compensation measures according to the corresponding sleep quality obtained in the comparison unit. The specific content is as follows:

[0118] If in the normal state, some mild maintenance compensation measures will be adopted at this time, aiming to help users maintain good sleep quality and prevent the decline of sleep quality. At this time, users will be advised to maintain the current sleep environment, such as environmental conditions like light, noise, temperature and humidity, to ensure continuous good sleep. At the same time, by prompting users to carry out mild relaxation activities before going to bed, including meditation, deep breathing exercises and stretching exercises, to promote a better state of physical relaxation and prevent potential sleep quality decline caused by daily fatigue.

[0119] If not in the normal state, some more proactive intervention measures will be taken at this time, aiming to help users improve the current sleep quality and solve the problems affecting sleep. At this time, it will help users regulate and reduce sympathetic nerve activity and increase parasympathetic nerve activity, and relieve stress by guiding users to carry out meditation, progressive muscle relaxation and deep breathing exercises, to promote a better state of physical relaxation and prevent potential sleep quality decline caused by daily fatigue. At the same time, remind users to adjust their work and rest time, maintain a regular sleep schedule, avoid staying up late or having irregular sleep times, and promote the stability of the biological clock.

[0120] In this embodiment, the comprehensive analysis unit constructs an efficient sleep quality prediction model using a convolutional neural network (CNN). Taking the proportion of neural activity, body position change rate, sleep fragmentation rate, and heart rate recovery ratio as inputs, and through linear normalization processing, a sleep quality assessment index is obtained. This method can comprehensively consider multiple physiological indicators and provide a more accurate and comprehensive sleep quality assessment. This quantitative assessment method can better reflect the user's true sleep state and provide a scientific basis for subsequent feedback measures. The feedback module of the system includes a comparison unit and a compensation unit. The comparison unit compares the sleep quality assessment index with an evaluation threshold Q by setting the evaluation threshold, and can accurately judge the user's sleep quality status. For users with normal sleep quality, the system will recommend maintaining the current good sleep environment and provide mild maintenance compensation measures, such as meditation, deep breathing exercises, etc. These recommendations help maintain the user's good sleep state and prevent further decline in sleep quality. For users whose sleep quality does not meet the standard, the system will take more proactive intervention measures. These measures include reducing sympathetic nerve activity, increasing parasympathetic nerve activity, guiding the user to perform meditation, progressive muscle relaxation, and deep breathing exercises. Through these measures, the user's stress can be effectively reduced and the body can be promoted to relax, thereby improving sleep quality. In addition, the system will also recommend that the user adjust their work and rest time, maintain regular sleep habits, and avoid staying up late or having irregular sleep times, which is of great significance for improving the stability of the biological clock and the overall sleep quality. Through accurate sleep quality assessment and personalized feedback measures, this system can not only improve the user's sleep quality, but also help the user develop healthy living habits, thereby overall improving the user's health level. The intervention measures and recommendations provided by the system can effectively relieve the user's stress and anxiety, improve the sleep environment, and ultimately achieve long-term maintenance of healthy sleep.

[0121] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A sleep quality analysis system based on human HRV measurement, characterized in that: It includes data monitoring module, preliminary analysis module, neural activity module, sleep state comprehensive analysis module and feedback module; The data monitoring module is used to monitor the health status of the user according to a number of monitoring instruments to obtain relevant physiological data, relevant electrical activity data and relevant personalized data of the user during sleep; The preliminary analysis module includes an HRV analysis unit and a preliminary judgment unit; the HRV analysis unit will generate a heart rate variability factor HRV based on relevant physiological data, specifically: Where N is the number of RR intervals, i = 1, 2, 3, ..., N, Rjj i Represented as the i-th RR interval, Rjj avg It is expressed as the average RR interval; the preliminary judgment unit will pre-set a threshold value, and compare the heart rate variability factor HRV with the set threshold value to preliminarily judge the difference in the heart rate fluctuation of the current user in the sleep state. If the heart rate variability factor HRV does not exceed the set threshold value, it will be preliminarily judged that the heart rate fluctuation of the current user in the sleep state is in a normal state. If the heart rate variability factor HRV exceeds the set threshold value, it will be preliminarily judged that the heart rate fluctuation of the current user in the sleep state is not in a normal state, and a sleep quality fluctuation command will be issued to the cloud platform; The neural activity module is used to analyze the balance state between the parasympathetic nerves and the sympathetic nerves according to the relevant electrical activity data after receiving the sleep quality fluctuation command, and to construct the neural activity ratio Jhb according to the power of the heart rate variability signal in the high frequency range and the power of the heart rate variability signal in the low frequency range. Based on the neural activity ratio Jhb, the neural activity bias of the current user in the sleep state is determined, and the neural activity bias indicates the balance state between the parasympathetic nerves and the sympathetic nerves; The sleep state comprehensive analysis module is used to construct the heart rate recovery ratio Xhb based on relevant personalized data, and combine it with the neural activity ratio Jhb and the trained sleep quality prediction model to obtain the sleep quality assessment index Sgzs; The sleep state comprehensive analysis module includes a recovery ability unit, a change unit and a comprehensive analysis unit; The recovery capacity unit is used to obtain the heart rate recovery ratio Xhb as follows: In the formula, Xz max Expressed as the maximum heart rate value in the sleeping state, Xz rest It is represented as the resting heart rate value in the sleeping state, and Xz(t) is represented as the heart rate value when entering the recovery state t after the maximum heart rate value; The change unit is used to obtain the body position change rate Tbz and the sleep fragmentation rate Spz according to the relevant personalized data, which are specifically obtained by the following formula: In the formula, Hzsc represents the number of times the body posture changes during sleep, Jxcs represents the number of times the body wakes up during sleep, and Zssc represents the total sleep duration; the comprehensive analysis unit is used to obtain the sleep quality evaluation index Sgzs, which is: Wherein, Hxc is the breathing difference, which indicates the difference between the maximum breathing depth and the minimum breathing depth of the user during sleep; a1, a2, a3, a4 and a5 respectively represent the weight values ​​of the heart rate recovery ratio Xhb, the neural activity ratio Jhb, the sleep fragmentation rate Spz, the body position change rate Tbz and the breathing difference Hxc, and V represents the correction constant; The feedback module is used to pre-set the evaluation threshold Q, and compare and analyze the sleep quality evaluation index Sgzs with the evaluation threshold Q to comprehensively estimate the quality status of the current user's sleep and take corresponding compensation measures based on the quality status.

2. A sleep quality analysis system based on human HRV measurement according to claim 1, characterized in that: The data monitoring module includes a deployment unit, a first collection unit, a second collection unit and a third collection unit; The deployment unit is used to deploy several groups of monitoring instruments around the user before the user falls asleep, wherein the several groups of monitoring instruments include a health bracelet, an electrocardiogram sensor, a respiratory monitoring sensor, an accelerometer, a camera and a gyroscope; The first acquisition unit is used to monitor and record relevant physiological data of the user in a sleeping state in real time, wherein the relevant physiological data includes the RR interval Rjj and the breathing difference Hxc in each monitoring period; The second acquisition unit is used to monitor and record relevant electrical activity data of the user in a sleeping state in real time, wherein the relevant electrical activity data includes the power of the heart rate variability signal in a high frequency range and the power of the heart rate variability signal in a low frequency range; The third acquisition unit is used to monitor and record relevant personalized data of the user in the sleep state in real time, wherein the relevant personalized data includes the maximum heart rate value Xz in the sleep state max , the heart rate value Xz(t) when entering the recovery state after the maximum heart rate value, and the resting heart rate value Xz in the sleep state rest , the number of times the body posture changes during sleep Hzsc, the total sleep duration Zssc and the number of times you wake up during sleep Jxcs.

3. A sleep quality analysis system based on human HRV measurement according to claim 2, characterized in that: The neural activity module includes a frequency domain analysis unit and a neural bias selection unit; The frequency domain analysis unit is used to analyze the balance state between the parasympathetic nerves and the sympathetic nerves according to the relevant electrical activity data, so as to obtain the power of the heart rate variability signal in the high frequency range and the power of the heart rate variability signal in the low frequency range, and record them as HF power P1 and LF power P2 respectively. The specific acquisition method is as follows: Wherein, Ppz(f) represents the power spectrum density; the integration range of HF power P1 is the high frequency range, i.e., 0.15 Hz-0.4 Hz; the integration range of LF power P2 is the low frequency range, i.e., 0.04 Hz-0.15 Hz; and f represents the frequency.

4. A sleep quality analysis system based on human HRV measurement according to claim 3, characterized in that: The neural bias selection unit is used to calculate the neural activity ratio Jhb according to the HF power P1 and the LF power P2. The neural activity ratio Jhb is obtained by the following formula: Based on the value of the neural activity ratio Jhb, the neural activity bias of the current user in the sleep state is judged. The specific judgment content is as follows: If the nerve activity ratio Jhb>1, it is determined that the parasympathetic nerve activity is dominant in the current user's sleep state; If the nerve activity ratio Jhb is less than 1, it is determined that the sympathetic nerve activity of the current user is dominant in the sleep state; If the nerve activity ratio Jhb=1, it is determined that the sympathetic nerve activity and the parasympathetic nerve activity of the current user in the sleep state are in a relatively stable state.

5. The sleep quality analysis system based on human HRV measurement according to claim 1, characterized in that: The feedback module includes a comparison unit and a compensation unit; The comparison unit is used to pre-set an evaluation threshold Q, and compare and analyze the evaluation threshold Q with the sleep quality evaluation index Sgzs to comprehensively estimate the quality of the current user's sleep. The specific contents are as follows: If the sleep quality evaluation index Sgzs exceeds the evaluation threshold Q, it is estimated that the current user's sleep quality is in a normal state; If the sleep quality evaluation index Sgzs does not exceed the evaluation threshold Q, it is estimated that the current user's sleep quality is not in a normal state; The compensation unit is used to take corresponding compensation measures according to the corresponding sleep quality obtained in the comparison unit, and the specific contents are as follows: If the sleep state is normal, the user will be advised to maintain the current sleep environment and be prompted to do some light relaxation activities before going to bed, including meditation, deep breathing exercises and stretching exercises. If it is not in a normal state, it will help users regulate sympathetic and parasympathetic nervous activities, and guide users to meditate, progressive muscle relaxation and deep breathing exercises, while reminding users to adjust their work and rest time and maintain a regular sleep schedule.

Images