Sleep monitoring system and method of human physiological signals

Abstract

The present invention belongs to the technical field of sleep monitoring. A sleep monitoring system and method for human physiological signals are disclosed. First, human physiological signal data including real-time heart rate of the current user is obtained, and characteristic values ​​of heart rate fluctuations, such as the amplitude and speed of heart rate changes, are extracted. At the same time, identity characteristic data such as user gender, age, and weight are obtained. The time to fall asleep is determined by comparing with a first preset value determined by a pre-trained first model, and the first model is trained based on historical data of other users with similar identity characteristics to the current user. The method utilizes heart rate fluctuation characteristics and identity characteristics, and determines personalized preset values ​​for the current user with the help of a big data training model, thereby improving the accuracy of assessing the time it takes for a human body to fall asleep.

Description

Technical Field

[0001] The present invention relates to the technical field of sleep monitoring, and in particular to a sleep monitoring system and method for human physiological signals. Background Art

[0002] Sleep is an indispensable part of maintaining human health. Sleep quality assessment can effectively reflect the body's ability to repair and integrate memory, and is of great significance to the diagnosis and assessment of human health.

[0003] As technology continues to develop and progress, more and more people use smart wearable devices to monitor sleep, such as smart watches and smart bracelets. Smart wearable devices usually evaluate sleep quality based on nighttime sleep parameters such as total sleep duration, duration of each sleep stage, proportion of a sleep stage, segmented duration and total duration, number of sleep interruptions, and daytime sleep parameters such as duration and frequency of daytime naps. The determination of bedtime affects the determination of the above-mentioned nighttime and daytime sleep parameters, and these sleep parameters determine the evaluation of sleep quality.

[0004] The time of falling asleep can be determined by judging whether the heart rate change meets the preset threshold, but the preset threshold is usually a fixed value, which does not take into account the individual differences of different users and the differences in different states, making the judgment of the time of falling asleep not accurate enough.

[0005] In summary, how to accurately assess the time it takes for a human body to fall asleep is a problem that needs to be solved urgently in this field. Summary of the invention

[0006] In view of the above technical problems, the present invention provides a sleep monitoring system, method, electronic device and computer storage medium for human physiological signals to at least partially solve the above technical problems.

[0007] The present invention discloses a sleep monitoring method of human physiological signals, the method comprising the following steps: Step S101, obtaining human physiological signal data of a current user, the physiological signal data comprising a real-time heart rate.

[0008] Step S102: extracting the heart rate fluctuation characteristic value.

[0009] Step S103, obtaining the current user's identity feature data, including gender, age, and weight.

[0010] Step S104, obtaining the current user's exercise characteristics within a preset time before the heart rate fluctuation occurs, including exercise type, exercise time, and average heart rate during exercise.

[0011] Step S105, determining whether the motion feature is acquired, if not, proceeding to step S106, and if so, proceeding to step S107.

[0012] Step S106, determining whether the heart rate fluctuation characteristic value meets a first preset value, if so, the time corresponding to the heart rate fluctuation is the current user's sleeping time.

[0013] Step S107, determining whether the heart rate fluctuation characteristic value meets a second preset value, if so, the time corresponding to the heart rate fluctuation is the current user's sleeping time.

[0014] Among them, the first preset value represents the heart rate characteristic value of a human body from being awake to falling asleep, the first preset value is obtained by a pre-trained first model, and the first model is trained through historical sleep monitoring data of other users. The other users are users with similar identity characteristic data to the current user, and the second preset value is a correction value of the first preset value.

[0015] Optionally, a sleep monitoring method for human physiological signals also includes, wherein the heart rate fluctuation value includes a time domain fluctuation value and / or a frequency domain fluctuation value, wherein the time domain fluctuation value includes the heart rate change amplitude, the heart rate change speed, the heart rate mean, the heart rate standard deviation, and the frequency domain fluctuation value includes the frequency component obtained by Fourier transform.

[0016] Optionally, a sleep monitoring method for human physiological signals also includes determining whether the number of users with identity feature data similar to the current user meets a first quantity threshold; if so, selecting historical data accumulated by similar users within a first time period to train the first model; if not, selecting historical data accumulated by similar users within a second time period to train the first model, wherein the second time period is greater than the first time period.

[0017] Optionally, a sleep monitoring method for human physiological signals further includes: the first preset value is obtained by a pre-trained second model, and the second model is trained by historical sleep monitoring data of the current user.

[0018] The present invention also discloses a sleep monitoring system for human physiological signals, the system includes a sleep monitoring device, characterized in that the sleep monitoring device includes: a first acquisition unit, used to acquire human physiological signal data of the current user, the physiological signal data including real-time heart rate; an extraction unit, used to extract the characteristic value of the heart rate fluctuation; a second acquisition unit, used to acquire identity feature data of the current user, including gender, age, and weight; a third acquisition unit, used to acquire the current user's motion characteristics within a preset time before the heart rate fluctuation occurs, including the type of exercise, exercise time, and average heart rate during exercise; a first judgment unit, used to judge whether the motion characteristics are acquired; and a second judgment unit, used to determine whether the motion characteristics are acquired if the motion characteristics are not acquired. When the motion feature is obtained, it is determined whether the heart rate fluctuation characteristic value meets the first preset value. If so, the time corresponding to the heart rate fluctuation is the time when the current user falls asleep; a third judgment unit is used to determine whether the heart rate fluctuation characteristic value meets the second preset value when the motion feature is obtained. If so, the time corresponding to the heart rate fluctuation is the time when the current user falls asleep; wherein, the first preset value represents the heart rate characteristic value of a human body from being awake to falling asleep, the first preset value is obtained by a pre-trained first model, and the first model is trained through historical sleep monitoring data of other users, the other users are users with similar identity characteristic data to the current user, and the second preset value is a correction value of the first preset value.

[0019] Optionally, a sleep monitoring system for human physiological signals further includes, wherein the heart rate fluctuation characteristic value includes a time domain fluctuation value and / or a frequency domain fluctuation value, wherein the time domain fluctuation value includes the heart rate change amplitude, the heart rate change speed, the heart rate mean, the heart rate standard deviation, and the frequency domain fluctuation value includes the frequency component obtained by Fourier transform.

[0020] Optionally, a sleep monitoring system for human physiological signals also includes a fourth judgment unit, which is used to judge whether the number of users with similar identity feature data to the current user meets a first quantity threshold. If so, historical data accumulated by similar users within a first time period is selected to train the first model. If not, historical data accumulated by similar users within a second time period is selected to train the first model, wherein the second time period is greater than the first time period.

[0021] Optionally, a sleep monitoring system for human physiological signals further includes: the first preset value is obtained by a pre-trained second model, and the second model is trained by historical sleep monitoring data of the current user.

[0022] The present invention also discloses an electronic device, comprising: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement any of the above methods.

[0023] The present invention also discloses a computer storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement any of the above methods.

[0024] The present invention discloses a sleep monitoring method of human physiological signals. First, human physiological signal data including real-time heart rate of the current user is obtained, and characteristic values ​​of heart rate fluctuations, such as heart rate variation amplitude and speed, are extracted. At the same time, identity characteristic data such as user gender, age, and weight are obtained. The time to fall asleep is determined by comparing with a first preset value determined by a pre-trained first model, and the first model is trained based on historical data of other users with similar identity characteristics to the current user. The method uses the heart rate fluctuation characteristics and identity characteristics, and determines the personalized preset value for the current user with the help of a big data training model, thereby improving the accuracy of assessing the time to fall asleep. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 The present invention is a flowchart of a sleep monitoring method for human physiological signals disclosed in an embodiment of the present invention.

[0027] Figure 2 Schematic diagram of a sleep monitoring device disclosed in an embodiment of the present invention.

[0028] Figure 3 It is a schematic diagram of an electronic device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0031] Figure 1An embodiment of the present invention provides a thread asynchronous processing method based on demand adaptive adjustment. The method can be executed by a thread asynchronous processing system based on demand adaptive adjustment. The system can be implemented in the form of hardware and / or software and can be configured in an electronic device, which can be a server. Figure 1 As shown, the method includes: step S101, obtaining physiological signal data of the current user, wherein the physiological signal data includes a real-time heart rate.

[0032] Among them, a suitable wearable sensor device is connected to the user's body, and the device has a high-precision heart rate monitoring function, which can continuously obtain real-time heart rate information from the current user's human physiological signal data. For example, a smart bracelet or smart watch with a photoelectric sensor can be used, which monitors the heart rate through photoplethysmography (PPG) technology.

[0033] Step S102: extracting the heart rate fluctuation characteristic value.

[0034] After acquiring the real-time heart rate data, the heart rate fluctuation characteristic value is extracted.

[0035] Preferably, the heart rate fluctuation value includes time domain fluctuation value and / or frequency domain fluctuation value, wherein the time domain fluctuation value includes the heart rate change amplitude, the heart rate change speed, the heart rate mean, the heart rate standard deviation, and the frequency domain fluctuation value includes the frequency component obtained by Fourier transform.

[0036] Specifically, for the heart rate mean in the time domain fluctuation value, within a certain monitoring period (such as a 1-minute time window), all the collected heart rate data are summed up and then divided by the number of data points. For example, within this period, 20 heart rate data points are collected, namely 65, 66, 68, 67, etc. These data are added together to get the total, and then divided by 20 to get the heart rate mean.

[0037] Among them, the heart rate standard deviation is used to measure the dispersion of heart rate data around the mean. The larger the standard deviation, the more unstable the heart rate fluctuation, and vice versa. When calculating the heart rate standard deviation, first calculate the square of the difference between each heart rate data point and the mean, sum these square values ​​and divide them by the number of data points minus 1, and finally take the square root. For example, if the mean heart rate in a certain period of time is 70 beats / minute, the sum of the squares of the differences between each data point and the mean is 100, and the number of data points is 20, then the standard deviation is During the process from being awake to falling asleep, the standard deviation of heart rate usually decreases gradually, because the physiological state of the body tends to be stable after falling asleep, and the fluctuation of heart rate also becomes smaller.

[0038] For the heart rate variation, the difference between the highest heart rate and the lowest heart rate in the set time window (such as 1 minute) is calculated, and this difference is the heart rate variation. For example, within a certain 1 minute, the highest heart rate reaches 80 beats / minute and the lowest heart rate is 65 beats / minute, then the heart rate variation is 15 beats / minute.

[0039] Optionally, the heart rate change rate can be extracted by calculating the difference between two consecutive heart rate measurements and dividing it by the measurement time interval. Assuming that the heart rates of two consecutive measurements are 70 beats / minute and 72 beats / minute, respectively, and the measurement time interval is 10 seconds, then the heart rate change rate is (72-70) ÷ (10 ÷ 60) = 12 beats / minute.

[0040] Specifically, for the frequency domain fluctuation value, the heart rate signal in the time domain is converted into a frequency domain signal through Fourier transform. Fourier transform can convert the heart rate signal in the time domain (heart rate sequence that changes over time) into a frequency domain signal. In the frequency domain, the signal is represented as a combination of different frequency components. For the heart rate signal, its frequency domain components are mainly concentrated in the low frequency and very low frequency range.

[0041] Generally speaking, the frequency domain of the heart rate signal is roughly between 0-0.5Hz. Among them, the low-frequency component (LF, usually in the range of 0.04-0.15Hz) is mainly related to the joint regulation of the sympathetic and parasympathetic nerves, while the high-frequency component (HF, usually in the range of 0.15-0.4Hz) mainly reflects the activity of the parasympathetic nerves. The LF / HF ratio can be used to measure the balance of the sympathetic-parasympathetic nerves.

[0042] In the transition phase from wakefulness to sleep (N1 phase), the frequency domain characteristics will change significantly. The parasympathetic nerve activity will further increase, the high-frequency component (HF) will continue to increase, and the low-frequency component (LF) will continue to decrease. The LF / HF ratio will decrease significantly.

[0043] For example, the HF component may increase from about 40% in the wakefulness and relaxation stage to 50%-60%, while the LF component may decrease from about 60% to 40%-50%, and the LF / HF ratio may decrease to about 1.0-1.2. This change reflects the neural regulation process of the brain entering the early stage of sleep, the physiological state of the body changes from wakefulness to light sleep, and the heart rate regulation mechanism relies more on the parasympathetic nerves, making the heart rate show the characteristics of enhanced high-frequency components in the frequency domain.

[0044] Step S103, obtaining the current user's identity feature data, including gender, age, and weight.

[0045] Among them, in some embodiments, when a user uses the matching sleep monitoring device for the first time, he needs to enter his gender, age, weight and other identity characteristics information. For example, a user is a male, 35 years old, and weighs 75 kilograms. It is understandable that the system can also bind the sleep monitoring device based on the user identity characteristic data obtained from other channels, which is not limited in this embodiment.

[0046] Step S104, obtaining the current user's exercise characteristics within a preset time before the heart rate fluctuation occurs, including exercise type, exercise time, and average heart rate during exercise.

[0047] In some embodiments, exercise before bedtime may affect heart rate fluctuations, which in turn affects the judgment of sleep time. Therefore, in order to further improve the accuracy of judging sleep time, while monitoring the user's heart rate fluctuations, the system will trace back the user's exercise data within a preset time (for example, 30 minutes) to obtain the current user's exercise characteristics. This is achieved through connection and data interaction with smart bracelets or other wearable exercise monitoring devices.

[0048] Among them, wearable devices have built-in sensors such as accelerometers and gyroscopes, which can identify the user's movement pattern. Common types of movement include walking, running, standing, sitting for a long time, going up and down stairs, etc. For example, when the accelerometer detects regular periodic vibrations, and the vibration amplitude and frequency meet the walking characteristics, the movement type is determined to be walking; if a more violent and high-frequency vibration is detected, it is judged to be running in combination with the gyroscope data. For exercise time recording, the duration of the exercise is accurately recorded from the start of the movement to the end. For example, if the user starts running at 9:30 pm and ends running at 10 pm, the exercise time is 30 minutes. For the calculation of the average heart rate during exercise, the heart rate data is continuously collected during the exercise process and its average value is calculated. For example, during the above 30-minute running process, a total of 60 heart rate data points are collected, and the sum of these data points divided by 60 is the average heart rate during exercise. Assuming the sum is 4200 times / minute, the average heart rate is 70 times / minute.

[0049] Step S105, determining whether the motion feature is acquired, if not, proceeding to step S106, and if so, proceeding to step S107.

[0050] Step S106, determining whether the heart rate fluctuation characteristic value meets a first preset value, if so, the time corresponding to the heart rate fluctuation is the current user's sleeping time.

[0051] Among them, the first preset value represents the heart rate characteristic value of a human body from being awake to falling asleep, and the first preset value is obtained by a pre-trained first model, and the first model is trained through historical sleep monitoring data of other users, and the other users are users with similar identity characteristic data to the current user.

[0052] The first preset value is obtained through a pre-trained first model. The training data of the first model comes from a large amount of sleep monitoring history data of other users with similar identity characteristics to the current user.

[0053] In some embodiments, in order to improve the accuracy of the output results of the first model, it is necessary to increase the diversity of sleep monitoring historical data of other users as much as possible, while taking into account the efficiency and cost of training. Therefore, an appropriate amount of historical data is required for model training. Specifically, if the number of users with identity feature data similar to the current user meets the first quantity threshold (for example, greater than or equal to 200 people), the accumulated historical data of similar users within the first time period (for example, 3 months) can be selected to train the model. Conversely, if the number of users with identity feature data similar to the current user does not meet the first quantity threshold (for example, less than 200 people), the accumulated historical data of similar users within the second time period (for example, 5 months) can be selected to train the model; this can both meet the accuracy of model prediction and reduce the cost of model training. It can be understood that the second time threshold can be flexibly adjusted according to the specific number of similar users, and the present disclosure does not limit it.

[0054] In some embodiments, for the user's first use of the sleep monitoring device or the initial period of time (e.g., less than 3 months), the first preset value can be obtained by training a big data model based on other similar users. Specifically, for gender factors, the users are divided into two groups, male and female, for analysis. For age factors, 5 years is used as an age group, such as 20-25 years old, 25-30 years old, etc. For weight factors, 10 kilograms is used as a weight group, such as 60-70 kilograms, 70-80 kilograms, etc. For example, for a current user who is 35 years old, male, and weighs 75 kilograms, the training data on which the first model is based comes from many other users who are 30-40 years old, male, and weigh 70-80 kilograms.

[0055] The first model may adopt a recurrent neural network model such as a long short-term memory network (LSTM) or a gated recurrent unit (GRU), which is not limited in this embodiment.

[0056] During the training process, data such as the heart rate change amplitude and speed of these similar users from waking up to falling asleep during sleep monitoring are collected. By statistically analyzing these data, the heart rate characteristic value range from waking up to falling asleep for such users with similar identity characteristics is determined as the first preset value.

[0057] Among them, taking the heart rate fluctuation characteristics including the heart rate variation range and the heart rate variation speed as an example, assuming that after a large amount of data statistical analysis, the model outputs that for this type of user group, the first preset value range of the heart rate variation range is 10-20 times / minute, and the first preset value range of the heart rate variation speed is 3-8 times / minute; compare the heart rate variation range and speed extracted by the current user with the above first preset value range. If the heart rate variation range A of the current user satisfies beats / minute, and the heart rate change rate meets beats / minute, the time corresponding to the heart rate fluctuation is determined to be the current user's sleep time. beats / minute and heart rate change rate times / minute, if it meets the first preset value range, the corresponding time, for example, 23:00, can be determined as the sleeping time.

[0058] Step S107, determining whether the heart rate fluctuation characteristic value meets a second preset value, if so, the time corresponding to the heart rate fluctuation is the current user's sleeping time.

[0059] The second preset value is a correction value of the first preset value.

[0060] In some embodiments, exercise before bedtime may affect heart rate fluctuations, thereby affecting the judgment of falling asleep time. This is because if there is no exercise before going to bed, the body is in a normal awake state and the heart rate is relatively stable. When falling asleep, the heart rate will gradually decrease, and its change curve is relatively gentle. For example, a healthy adult's heart rate before going to bed is 70-80 beats / minute, and the heart rate may gradually drop to about 60 beats / minute during falling asleep. After exercising before going to bed, the heart rate will rise sharply due to the excitement of the sympathetic nerves during exercise. After the exercise, the heart rate will not return to normal immediately, but will slowly decrease. When falling asleep, the heart rate may still be higher than the normal level when falling asleep, and in the early stage of sleep, the heart rate will drop by a larger amplitude and faster speed. For example, high-intensity exercise before going to bed, the heart rate may reach 140 beats / minute at the end of the exercise, and may still be 90 beats / minute when falling asleep, and then gradually decrease.

[0061] Optionally, taking the heart rate fluctuation characteristics including the heart rate change amplitude and the heart rate change speed as an example, if the user has previously performed high-intensity exercise (such as long-term running), the average heart rate during exercise is high and the exercise time is long, then when judging the time to fall asleep, the first preset value range of the heart rate change amplitude and speed is corrected and adjusted accordingly. Assuming that the first preset value change amplitude range is 10-20 times / minute under normal circumstances (no exercise), if it is detected that the user has performed 30 minutes of high-intensity running within the preset time (such as 30 minutes before going to bed), it may be adjusted to the second preset value of 15-25 times / minute. Similarly, the preset value of the heart rate change speed will also be corrected and adjusted according to the exercise situation. Assuming that the original first preset value change speed is 3-8 times / minute, it may be adjusted to the second preset value of 5-12 times / minute after high-intensity exercise. It can be understood that the above second preset value range is only an example, and the second preset value can be flexibly adjusted according to the type of exercise, exercise time and average heart rate during exercise, and the present disclosure is not limited.

[0062] Specifically, if the heart rate fluctuation characteristic value meets the second preset value range, then the time corresponding to the heart rate fluctuation is the current user's sleep time. For example, after correction, the second preset value range has a heart rate variation range of 15-25 beats / minute and a heart rate variation rate of 5-12 beats / minute. If the current user's heart rate variation range is 21 beats / minute and the heart rate variation rate is 10 beats / minute, then the time corresponding to the heart rate fluctuation can be determined as the sleep time.

[0063] In some embodiments, after the user has used the sleep monitoring device for a long enough time, for example, after the user has used it for three months, enough current user data can be accumulated, and the preset value obtained by the second model trained based on the historical data of the current user can be used as the first preset value. The second model can use the same or different neural network type as the first model, which will not be described in detail in this embodiment.

[0064] Through the disclosed embodiment, human physiological signal data including real-time heart rate of the current user is first obtained, and characteristic values ​​of heart rate fluctuations, such as the amplitude and speed of heart rate changes, are extracted. At the same time, identity characteristic data such as gender, age, and weight of the user are obtained. The time to fall asleep is determined by comparing with a first preset value determined by a pre-trained first model, which is trained based on historical data of other users with similar identity characteristics to the current user. This method utilizes the characteristics of heart rate fluctuations and identity characteristics, and determines personalized preset values ​​for the current user with the help of a big data training model, thereby improving the accuracy of assessing the time it takes for a person to fall asleep.

[0065] Another exemplary embodiment of the present disclosure provides a sleep monitoring system for human physiological signals, the system at least comprising a sleep monitoring device for human physiological signals, such as Figure 2As shown, the device includes: a first acquisition unit 201, used to acquire the current user's human physiological signal data, the physiological signal data including real-time heart rate; an extraction unit 202, used to extract the heart rate fluctuation characteristic value; a second acquisition unit 203, used to acquire the current user's identity feature data, including gender, age, and weight; a third acquisition unit 204, used to acquire the current user's motion characteristics within a preset time before the heart rate fluctuation occurs, including the type of exercise, exercise time, and average heart rate during exercise; a first judgment unit 205, used to judge whether the motion characteristics are acquired; a second judgment unit 206, used to judge whether the heart rate fluctuation is acquired when the motion characteristics are not acquired. Whether the motion characteristic value satisfies the first preset value, if so, the corresponding time of the heart rate fluctuation is the time when the current user falls asleep; the third judgment unit 207 is used to judge whether the heart rate fluctuation characteristic value satisfies the second preset value when the motion characteristic is acquired, if so, the corresponding time of the heart rate fluctuation is the time when the current user falls asleep; wherein, the first preset value represents the heart rate characteristic value of the human body from waking up to falling asleep, the first preset value is obtained by a pre-trained first model, and the first model is trained through the sleep monitoring historical data of other users, the other users are users with similar identity characteristic data to the current user, and the second preset value is a correction value of the first preset value.

[0066] Optionally, the heart rate fluctuation value includes a time domain fluctuation value and / or a frequency domain fluctuation value, wherein the time domain fluctuation value includes the heart rate change amplitude, the heart rate change speed, the heart rate mean, the heart rate standard deviation, and the frequency domain fluctuation value includes the frequency component obtained by Fourier transform.

[0067] Optionally, it also includes a fourth judgment unit, which is used to judge whether the number of users similar to the current user's identity feature data meets a first quantity threshold. If so, historical data accumulated by similar users within a first time period is selected to train the first model. If not, historical data accumulated by similar users within a second time period is selected to train the first model, wherein the second time period is greater than the first time period.

[0068] Optionally, the first preset value is obtained by a pre-trained second model, and the second model is trained by historical sleep monitoring data of the current user.

[0069] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0070] Through the disclosed embodiment, human physiological signal data including real-time heart rate of the current user is first obtained, and characteristic values ​​of heart rate fluctuations, such as the amplitude and speed of heart rate changes, are extracted. At the same time, identity characteristic data such as gender, age, and weight of the user are obtained. The time to fall asleep is determined by comparing with a first preset value determined by a pre-trained first model, which is trained based on historical data of other users with similar identity characteristics to the current user. This method utilizes the characteristics of heart rate fluctuations and identity characteristics, and determines personalized preset values ​​for the current user with the help of a big data training model, thereby improving the accuracy of assessing the time it takes for a person to fall asleep.

[0071] It should be noted that: the sleep monitoring device provided in the above embodiment only uses the division of the above functional units as an example when performing sleep monitoring. In actual applications, the above functions can be assigned to different functional units as needed. In addition, the thread asynchronous processing device provided in the above embodiment and the thread asynchronous processing method embodiment based on demand adaptive adjustment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0072] Figure 3 Schematic diagram of an electronic device 300 according to an embodiment of the present application. Figure 3 As shown, the electronic device 300 further includes: a sleep monitoring module 303, a communication module 304, an input unit 305 and a power supply 306. The processor 301 is electrically connected to the sleep monitoring module 303, the communication module 304, the input unit 305 and the power supply 306 respectively. Those skilled in the art will understand that Figure 3 The electronic device structure shown in the figure does not constitute a limitation of the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0073] The sleep monitoring module 303 can be used for sleep monitoring based on human physiological signals.

[0074] The communication module 304 can be used to communicate with other devices.

[0075] The input unit 305 may be used to receive input numbers, character information or user feature information (such as fingerprint, iris, facial information, etc.), and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0076] The power supply 306 is used to supply power to various components of the electronic device 300. Optionally, the power supply 306 can be logically connected to the processor 301 through a power management system, so that the power management system can manage charging, discharging, power consumption, and other functions. The power supply 306 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0077] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0078] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0079] To this end, the embodiment of the present application provides a computer-readable storage medium, which stores multiple computer programs, and the computer program can be loaded by a processor to execute the steps of a sleep monitoring method for human physiological signals provided in the embodiment of the present application. For example, the computer program can execute the following steps: Step S101, obtaining the current user's human physiological signal data, the physiological signal data including the real-time heart rate.

[0080] Step S102: extracting the heart rate fluctuation characteristic value.

[0081] Step S103, obtaining the current user's identity feature data, including gender, age, and weight.

[0082] Step S104, obtaining the current user's exercise characteristics within a preset time before the heart rate fluctuation occurs, including exercise type, exercise time, and average heart rate during exercise.

[0083] Step S105, determining whether the motion feature is acquired, if not, proceeding to step S106, and if so, proceeding to step S107.

[0084] Step S106, determining whether the heart rate fluctuation characteristic value meets a first preset value, if so, the time corresponding to the heart rate fluctuation is the current user's sleeping time.

[0085] Step S107, determining whether the heart rate fluctuation characteristic value meets a second preset value, if so, the time corresponding to the heart rate fluctuation is the current user's sleeping time.

[0086] Among them, the first preset value represents the heart rate characteristic value of a human body from being awake to falling asleep, the first preset value is obtained by a pre-trained first model, and the first model is trained through historical sleep monitoring data of other users. The other users are users with similar identity characteristic data to the current user, and the second preset value is a correction value of the first preset value.

[0087] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0088] Since the computer program stored in the storage medium can execute the steps in any one of the sleep monitoring methods for human physiological signals provided in the embodiments of the present application, the beneficial effects of any one of the sleep monitoring methods for human physiological signals provided in the embodiments of the present application can be achieved. For details, please refer to the previous embodiments and will not be repeated here.

[0089] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present application. These computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0090] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0091] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0092] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions and variations of these embodiments are made without departing from the principles and spirit of the present invention, and still fall within the scope of protection of the present invention.

Claims

1. A sleep monitoring method of human physiological signals, characterized in that: The method comprises the following steps: Step S101, obtaining physiological signal data of the current user, wherein the physiological signal data includes a real-time heart rate; Step S102, extracting heart rate fluctuation characteristic values; Step S103, obtaining the current user's identity feature data, including gender, age, and weight; Step S104, obtaining the current user's exercise characteristics within a preset time before the heart rate fluctuation occurs, including exercise type, exercise time, and average heart rate during exercise; Step S105, determining whether the motion feature is obtained, if not, proceeding to step S106, if yes, proceeding to step S107; Step S106, determining whether the heart rate fluctuation characteristic value meets a first preset value, if so, the time corresponding to the heart rate fluctuation is the current user's sleeping time; Step S107, determining whether the heart rate fluctuation characteristic value meets a second preset value, and if so, the time corresponding to the heart rate fluctuation is the current user's sleeping time; Among them, the first preset value represents the heart rate characteristic value of a human body from being awake to falling asleep, the first preset value is obtained by a pre-trained first model, and the first model is trained through historical sleep monitoring data of other users, or the first preset value is obtained by a pre-trained second model, and the second model is trained through historical sleep monitoring data of the current user, the other users are users with identity characteristic data similar to that of the current user, the second preset value is a correction value of the first preset value, and the second preset value is set based on the motion characteristics.

2. The sleep monitoring method of human physiological signals according to claim 1, characterized in that: in, The heart rate fluctuation characteristic value includes time domain fluctuation value and / or frequency domain fluctuation value, wherein the time domain fluctuation value includes the heart rate change amplitude, the heart rate change speed, the heart rate mean, and the heart rate standard deviation, and the frequency domain fluctuation value includes the frequency component obtained by Fourier transform.

3. The sleep monitoring method of human physiological signals according to claim 1, characterized in that: It also includes determining whether the number of users similar to the current user's identity feature data meets a first quantity threshold. If so, selecting historical data accumulated by similar users within a first time period to train the first model. If not, selecting historical data accumulated by similar users within a second time period to train the first model, wherein the second time period is greater than the first time period.

4. A sleep monitoring system for human physiological signals, the system comprising a sleep monitoring device, characterized in that: The sleep monitoring device comprises: A first acquisition unit, used to acquire physiological signal data of the current user, wherein the physiological signal data includes a real-time heart rate; An extraction unit, used for extracting characteristic values ​​of heart rate fluctuations; The second acquisition unit is used to acquire the identity characteristic data of the current user, including gender, age, and weight; The third acquisition unit is used to acquire the current user's exercise characteristics within a preset time before the heart rate fluctuation occurs, including exercise type, exercise time, and average heart rate during exercise; A first judging unit, used to judge whether the motion feature is acquired; A second judgment unit is used to judge whether the heart rate fluctuation characteristic value meets a first preset value when the motion characteristic is not obtained, and if so, the time corresponding to the heart rate fluctuation is the current user's sleeping time; A third judgment unit is used to judge whether the heart rate fluctuation characteristic value meets a second preset value when the motion characteristic is obtained, and if so, the time corresponding to the heart rate fluctuation is the current user's sleeping time; Among them, the first preset value represents the heart rate characteristic value of a human body from being awake to falling asleep, the first preset value is obtained by a pre-trained first model, and the first model is trained through historical sleep monitoring data of other users, or the first preset value is obtained by a pre-trained second model, and the second model is trained through historical sleep monitoring data of the current user, the other users are users with identity characteristic data similar to that of the current user, the second preset value is a correction value of the first preset value, and the second preset value is set based on the motion characteristics.

5. The sleep monitoring system of human physiological signals according to claim 4, characterized in that: in, The heart rate fluctuation characteristic value includes time domain fluctuation value and / or frequency domain fluctuation value, wherein the time domain fluctuation value includes the heart rate change amplitude, the heart rate change speed, the heart rate mean, and the heart rate standard deviation, and the frequency domain fluctuation value includes the frequency component obtained by Fourier transform.

6. The sleep monitoring system of human physiological signals according to claim 4, characterized in that: It also includes a fourth judgment unit, which is used to judge whether the number of users similar to the current user's identity feature data meets a first quantity threshold. If so, historical data accumulated by similar users within a first time period are selected to train the first model. If not, historical data accumulated by similar users within a second time period are selected to train the first model, wherein the second time period is greater than the first time period.

7. An electronic device comprising: At least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the method according to any one of claims 1 to 3.

8. A computer storage medium storing a computer program, characterized in that: The computer program is executed by a processor to implement the method according to any one of claims 1 to 3.

Images