CN113520339B

CN113520339B - Sleep data validity analysis method and device and wearable device

Info

Publication number: CN113520339B
Application number: CN202010288567.9A
Authority: CN
Inventors: 许德省; 李靖; 周林峰
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2020-04-14
Filing date: 2020-04-14
Publication date: 2022-10-11
Anticipated expiration: 2040-04-14
Also published as: CN113520339A

Abstract

The sleep data effectiveness analysis method comprises the steps of obtaining sleep data in a sleep cycle of a user; performing sleep staging according to the sleep data to obtain a sleep staging result; and when the ratio of at least one of the rapid eye movement period, the deep sleep period and the light sleep period to the sleep period exceeds a preset threshold, determining that the sleep data in the sleep period of the user are abnormal. The method and the device can quickly and effectively judge whether the sleep data in the sleep state are abnormal.

Description

Sleep data validity analysis method and device and wearable device

Technical Field

The application relates to the technical field of data processing, in particular to a sleep data validity analysis method and device and wearable equipment.

Background

At present, most wearable devices have a sleep monitoring function, and the sleep monitoring is to monitor the actions of people through a body motion recorder to obtain posture data, so as to judge the sleep state through the posture data.

When sleeping deeply, the muscles of the person can be relaxed, and the limbs can not move greatly, even can not move, and when sleeping shallowly, the person can move slightly to a certain extent. The principle of the wearable device is that the wearable device is in a sleep state by monitoring the motion state of a human body and slowly changing from a light motion mode to a motionless mode at night.

However, many users take off the wearable device before sleeping and place the wearable device at a bedside table or a pillow, so that when the wearable device is not worn, the wearable device still mistakenly thinks that the human body has entered sleep, and a large amount of meaningless sleep data are acquired. Therefore, it is necessary to analyze whether the collected sleep data is really effective.

Disclosure of Invention

The embodiment of the invention provides a sleep data validity analysis method, a sleep data validity analysis device and wearable equipment, wherein the sleep data validity analysis method comprises the steps of acquiring sleep data in a sleep state, obtaining a sleep staging result according to the sleep data, and judging whether the ratio exceeds a preset threshold value according to the ratio of a deep sleep period, a light sleep period or a rapid eye movement period in the sleep staging result to a sleep cycle, so that whether the sleep data in the sleep state is abnormal or not is judged quickly and effectively.

In a first aspect, an embodiment of the present invention provides a sleep data validity analysis method, where the method includes: acquiring sleep data in a sleep cycle of a user; performing sleep staging according to the sleep data to obtain a sleep staging result; and when the ratio of at least one of the rapid eye movement period, the deep sleep period and the light sleep period to the sleep period exceeds a preset threshold, determining that the sleep data in the sleep period of the user are abnormal.

Generally, in the whole sleep cycle, the rapid eye movement period accounts for about 20%, the light sleep period accounts for about 55%, and the deep sleep period accounts for about 25%. It can be understood that, when the proportion of any one of the periods (for example, the deep sleep period accounts for 40%) is greater than the preset threshold, the sleep period result is considered to be incorrect, so that the abnormality of acquiring the sleep data in the sleep cycle is confirmed, and the analysis efficiency is improved.

With reference to the first aspect, in one possible implementation manner, the sleep data includes heart rate data and body movement data; the step of performing sleep staging according to the sleep data to obtain a sleep staging result comprises the following steps:

counting the total duration of the sleep cycle of the user according to the sleep data; and when the total duration exceeds a preset value, sleep staging is carried out by utilizing the heart rate data and the body movement data to obtain a sleep staging result.

It can be understood that when the user does not actually enter the sleep state and the wearable device mistakenly determines that the wearable device enters the sleep state, the sleep staging result can be found to be abnormal after the sleep staging. For example, when a user reads or reads a book, the wearable device misjudges that the user enters a sleep state, and at the moment, the heart rate data and the body movement data are jointly used as the basis of sleep staging, so that the accuracy of sleep data staging can be improved according to the difference of the heart rate data and the body movement data along with the change of the sleep state.

and when the total duration does not exceed the preset value, performing sleep staging by using the body movement data to obtain a sleep staging result.

It can be understood that, for example, when the total duration is less than 2 hours, the data amount of the acquired heart rate data is small due to the short sleep duration, and at this time, if the heart rate data is used as the basis for sleep staging, the result of sleep staging is prone to be inaccurate. It can understand ground, wearable equipment utilizes the body movement data judgement user in a certain time quantum of acceleration sensor collection whether to get into sleep, if get into sleep, just can trigger the sleep monitoring procedure, and PPG sensor just can begin to gather heart rate data, so when the sleep is long shorter, heart rate data often is less, can not cover whole sleep cycle. For example, user 12 noon: 30 to 14:00 afternoon nap, wherein the heart rate data acquisition time period is 13-14 points: 00, the missing part accounts for a larger proportion of the whole sleep cycle, so when the sleep duration is less than a preset value, only the body movement data is used for sleep staging, the accuracy and the efficiency of the sleep staging can be improved, and the influence of the heart rate data of the missing part on the whole sleep staging is avoided.

With reference to the first aspect, in a possible implementation manner, before performing sleep staging according to the sleep data to obtain a sleep staging result, the method further includes:

and preprocessing the sleep data, wherein the preprocessing comprises at least one of smoothing and denoising, removing abnormal values and filling missing values.

The smoothing and denoising may be performed by using gaussian filtering, fourier transform filtering, convolution smoothing algorithm, and the like, which is not specifically limited in this embodiment of the present application. The elimination of the abnormal values can delete the obviously abnormal values, for example, whether the acquired data is abnormal can be judged by configuring a data threshold in advance. And (3) supplementing the missing value, namely supplementing the actual value according to a theoretical rule and improving the feasibility of calculation.

With reference to the first aspect, in a possible implementation manner, before confirming that sleep data in the sleep cycle of the user is abnormal when a ratio of at least one of a fast eye movement period, a deep sleep period, and a light sleep period to the sleep cycle exceeds a preset threshold, the method further includes:

receiving preset thresholds set by the user, wherein the preset thresholds comprise a first preset threshold, a second preset threshold and a third preset threshold; the first preset threshold is a ratio range of the deep sleep period to the sleep cycle; the second preset threshold is a ratio range of the light sleep period to the sleep cycle; the third preset threshold is a ratio range of the rapid eye movement period to the sleep cycle.

The user can set various threshold values according to the actual conditions of the user, so that the personalized requirements of the user can be met when the sleep data are judged abnormally, and the data analysis result is accurate and effective.

With reference to the first aspect, in a possible implementation manner, before the determining that sleep data in the sleep cycle of the user is abnormal when a ratio of at least one of a fast eye movement period, a deep sleep period, and a shallow sleep period to the sleep cycle exceeds a preset threshold, the method further includes:

acquiring a plurality of normal sleep data of the user; respectively carrying out sleep staging on the plurality of normal sleep data to obtain a plurality of normal sleep staging results; respectively calculating the ratio of the deep sleep period to the sleep cycle, the ratio of the light sleep period to the sleep cycle and the ratio of the rapid eye movement period to the sleep cycle in each normal sleep stage result; determining a range formed by the ratios of the deep sleep periods to the sleep cycle as a first preset threshold; determining a range formed by the ratios of the light sleep periods to the sleep cycle as a second preset threshold; and determining a range formed by the ratios of the rapid eye movement periods to the sleep cycle as a third preset threshold.

The system calculates the preset threshold of the ratio of each period of the user to the sleep cycle according to the sleep data in the normal sleep state confirmed by the user, so that the individualized requirements of the user can be met when the sleep data to be analyzed is judged, and the data analysis result is accurate and effective.

With reference to the first aspect, in a possible implementation manner, the step of counting a total duration of the sleep cycle of the user according to the sleep data includes:

extracting a plurality of characteristic values in the body motion data; determining whether the user enters sleep or stops sleeping according to the characteristic value of the body movement data; and calculating the sleep duration according to the time of entering the sleep and the time of finishing the sleep.

With reference to the first aspect, in a possible implementation manner, after the determining that sleep data in the sleep cycle of the user is abnormal when a ratio of at least one of a fast eye movement period, a deep sleep period, and a shallow sleep period to the sleep cycle exceeds a preset threshold, the method further includes: and processing the sleep data by adopting a preset processing mode, wherein the processing mode comprises at least one of retention, cutting and deletion.

It can be understood that, when the sleep data is abnormal, the user may select to delete the sleep data of the current sleep cycle, so as to avoid the abnormal data from being stored, thereby affecting the subsequent sleep-related data analysis. The user can also cut out real and effective sleep data from the original sleep data to be stored. Of course, the system may also automatically delete abnormal sleep data. Accuracy of sleep data within a sleep cycle is improved.

In a second aspect, an embodiment of the present invention provides a sleep data validity analysis apparatus, where the apparatus includes: the first acquisition unit is used for acquiring sleep data in a sleep cycle of a user; the sleep staging unit is used for performing sleep staging according to the sleep data to obtain a sleep staging result, wherein the sleep staging result comprises a rapid eye movement period, a deep sleep period and a light sleep period; the judging unit is used for judging whether the ratio of at least one of the rapid eye movement period, the deep sleep period and the light sleep period to the sleep period exceeds a corresponding preset threshold value; and the determining unit is used for determining that the sleep data in the sleep cycle of the user are abnormal if the ratio exceeds the corresponding preset threshold.

With reference to the second aspect, in one possible implementation manner, the sleep data includes heart rate data and body movement data; the sleep staging unit includes: the statistic subunit is used for counting the total duration of the sleep cycle of the user according to the sleep data; the judging subunit is used for judging whether the total duration exceeds a preset value; and the first staging subunit is used for performing sleep staging by utilizing the heart rate data and the body movement data when the total duration exceeds the preset value to obtain a sleep staging result, wherein the sleep staging result comprises a rapid eye movement period, a deep sleep period and a shallow sleep period.

The preset value may be, for example, 2 hours, 3 hours, 4 hours, and the like, for example, when the total duration is longer than 3 hours, the probability of a user sleeping in the afternoon or nap is small, at this time, if only the body movement data is used for sleep staging, and the user does not wear a wearable device, the body movement data is always close to 0, and then the sleep staging is easily inaccurate, so that the body movement data and the heart rate data are jointly used as the basis of the sleep staging by using the difference of the heart rate data changing along with the sleep state, and the accuracy of the sleep staging can be improved.

With reference to the second aspect, in one possible implementation manner, the sleep staging unit further includes: and the second staging subunit is used for performing sleep staging by utilizing the body movement data when the total duration does not exceed the preset value to obtain a sleep staging result, wherein the sleep staging result comprises a rapid eye movement period, a deep sleep period and a light sleep period.

For example, when the total duration is less than 3 hours, the probability of a user sleeping in the afternoon or nap is high, the sleep staging is performed by using the body movement data, a large amount of invalid sleep data generated due to the fact that the wearable device is not worn is not prone to occurring, only the human body movement data is used, the data volume of the sleep staging is reduced, and the staging efficiency is improved.

With reference to the second aspect, in one possible implementation manner, the apparatus further includes: and the preprocessing unit is used for preprocessing the sleep data, and the preprocessing comprises at least one of smoothing and denoising, removing abnormal values and filling up missing values.

The smoothing and denoising may be performed by using a smoothing and denoising algorithm, for example, a gaussian filter, a fourier transform filter, a convolution smoothing algorithm, and the like, which is not specifically limited in this embodiment of the present application. The elimination of the abnormal values can delete the obviously abnormal values, for example, whether the acquired data is abnormal can be judged by configuring a data threshold in advance. And (3) supplementing the missing value, namely supplementing the actual value according to a theoretical rule and improving the feasibility of calculation.

With reference to the second aspect, in one possible implementation manner, the apparatus further includes: the receiving unit is used for receiving preset thresholds set by the user, and the preset thresholds comprise a first preset threshold, a second preset threshold and a third preset threshold; the first preset threshold is a ratio range of the deep sleep period to the sleep cycle; the second preset threshold is a ratio range of the light sleep period to the sleep cycle; the third preset threshold is a ratio range of the rapid eye movement period to the sleep cycle.

With reference to the second aspect, in one possible implementation manner, the apparatus further includes: a second acquiring unit, configured to acquire a plurality of normal sleep data of the user; the sleep staging unit is further configured to perform sleep staging on the plurality of normal sleep data to obtain a plurality of normal sleep staging results; the calculating unit is used for respectively calculating the ratio of the deep sleep period to the sleep cycle, the ratio of the light sleep period to the sleep cycle and the ratio of the rapid eye movement period to the sleep cycle in each normal sleep stage result; the first confirming unit is used for confirming a range formed by the ratios of the deep sleep periods to the sleep cycle as a first preset threshold; the second confirming unit is used for confirming a range formed by the ratios of the plurality of the light sleep periods to the sleep cycle as a second preset threshold; and the third confirming unit is used for confirming a range formed by the ratios of the plurality of rapid eye movement periods to the sleep cycle as a third preset threshold.

With reference to the second aspect, in one possible implementation manner, the apparatus further includes: and the processing unit is used for processing the sleep data by adopting a preset processing mode, wherein the processing mode comprises at least one of reservation, cutting and deletion.

In a third aspect, an embodiment of the present invention provides a wearable device, including a sensor module and a processor, where the sensor module is configured to collect sleep data in a sleep cycle of a user; the processor is used for acquiring sleep data acquired by the sensor module; performing sleep staging according to the sleep data to obtain sleep staging results, wherein the sleep staging results comprise a rapid eye movement period, a deep sleep period and a light sleep period; judging whether the ratio of at least one of the rapid eye movement period, the deep sleep period and the light sleep period to the sleep period exceeds a corresponding preset threshold value or not; and if the ratio exceeds the corresponding preset threshold, determining that the sleep data in the sleep cycle of the user is abnormal.

Generally, in the whole sleep cycle, the rapid eye movement period accounts for about 20%, the light sleep period accounts for about 55%, and the deep sleep period accounts for about 25%. It can be understood that, when the proportion of any one of the periods (for example, the deep sleep period accounts for 40%) is greater than the preset threshold, the sleep period result may be regarded as a wrong result, so that it is determined that the sleep data collected in the sleep cycle is abnormal, and the analysis efficiency is improved.

With reference to the third aspect, in one possible implementation manner, the sensor module includes a heart rate sensor and an acceleration sensor; the heart rate sensor is used for collecting heart rate data of the user, and the acceleration sensor is used for collecting body movement data of the user; the processor is further configured to: counting the total duration of the sleep cycle of the user according to the sleep data; judging whether the total duration exceeds a preset value; and when the total duration exceeds the preset value, performing sleep staging by using the heart rate data and the body movement data to obtain sleep staging results, wherein the sleep staging results comprise a rapid eye movement period, a deep sleep period and a light sleep period.

With reference to the third aspect, in a possible implementation manner, the processor is further configured to: receiving a preset threshold value set by the user, wherein the preset threshold value comprises a first preset threshold value, a second preset threshold value and a third preset threshold value; the first preset threshold is a ratio range of the deep sleep period to the sleep cycle; the second preset threshold is a ratio range of the light sleep period to the sleep cycle; the third preset threshold is a ratio range of the rapid eye movement period to the sleep cycle.

With reference to the third aspect, in a possible implementation manner, the processor is further configured to: acquiring a plurality of normal sleep data of the user; respectively carrying out sleep staging on the plurality of normal sleep data to obtain a plurality of normal sleep staging results; respectively calculating the ratio of the deep sleep period to the sleep cycle, the ratio of the light sleep period to the sleep cycle and the ratio of the rapid eye movement period to the sleep cycle in each normal sleep stage result; determining a range formed by the ratios of the deep sleep periods to the sleep cycle as a first preset threshold; determining a range formed by the ratios of the light sleep periods to the sleep cycle as a second preset threshold; determining a range composed of ratios of the plurality of fast eye movement periods to the sleep cycle as a third preset threshold.

With reference to the third aspect, in one possible implementation manner, the processor is further configured to: and processing the sleep data by adopting a preset processing mode, wherein the processing mode comprises at least one of reservation, cutting and deletion.

In a fourth aspect, an embodiment of the present invention provides an electronic device, where the electronic device includes a memory and a processor, the memory may be a non-volatile storage medium, the memory stores a computer program therein, and the processor is connected to the memory and executes the computer program to implement the method in the first aspect or any possible implementation manner of the first aspect.

In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium storing program code for execution by a device, where the program code includes instructions for performing the method in the first aspect or any possible implementation manner of the first aspect.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive labor.

Fig. 1 is a schematic diagram of an architecture of a sleep monitoring system according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of a wearable device provided in an embodiment of the present invention;

fig. 3 is a schematic flowchart of a sleep data validity analysis method according to an embodiment of the present invention;

fig. 4 is another schematic flowchart of a sleep data validity analysis method according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of a graphical user interface of an electronic device according to an embodiment of the present invention;

fig. 6 is a schematic diagram of a graphical user interface of a wearable device provided by an embodiment of the invention;

FIG. 7 is a schematic diagram of another graphical user interface of a wearable device provided by embodiments of the present invention;

FIG. 8 is a schematic view of another graphical user interface of an electronic device provided by an embodiment of the invention;

FIG. 9 is a schematic diagram of another graphical user interface of an electronic device provided by an embodiment of the invention;

fig. 10 is a schematic block diagram of a sleep data validity analysis apparatus according to an embodiment of the present invention.

Detailed Description

For better understanding of the technical solutions of the present invention, the following detailed descriptions of the embodiments are provided with reference to the accompanying drawings.

It should be understood that the described embodiments are only some embodiments of the invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The terminology used in the embodiments is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the examples of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

It should be understood that the term "and/or" as used herein is merely one type of association that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.

It should be understood that although the terms first, second, third, etc. may be used in the embodiments to describe the terminals, these terminals should not be limited by these terms. These terms are only used to distinguish one terminal from another. For example, a first terminal may also be referred to as a second terminal, and similarly, a second terminal may also be referred to as a first terminal, without departing from the scope of embodiments.

The word "if" as used herein may be interpreted as "at 8230; \8230;" or "when 8230; \8230;" or "in response to a determination" or "in response to a detection", depending on the context. Similarly, the phrase "if determined" or "if detected (a stated condition or event)" may be interpreted as "upon determining" or "in response to determining" or "upon detecting (a stated condition or event)" or "in response to detecting (a stated condition or event)", depending on the context.

In addition, the system architecture and the service scenario described in this embodiment are for more clearly illustrating the technical solution in the embodiment of the present application, and do not limit the technical solution provided in this embodiment, and it can be known by those skilled in the art that along with the evolution of the system architecture and the appearance of a new service scenario, the technical solution provided in this embodiment is also applicable to similar technical problems.

Since the present embodiment relates to the application of terms, for the convenience of understanding, the following description will first describe related terms and related concepts such as neural networks, etc. that may be related to the present embodiment.

The sleep staging refers to dividing the sleep of a person into different periods according to the form of the person appearing during the sleep. The sleep is classified into two classes, three classes, four classes, five classes and six classes according to the sleep standard. In the two categories, the sleep and waking are classified; in the three categories, the eye Movement is classified into a conscious period, a Rapid Eye Movement (REM) period and a Non-Rapid eye Movement (NREM) period; in the four categories, the four categories are classified into a waking period, a rapid eye movement period, a light sleep period and a deep sleep period; in the fifth classification, the sleep period is divided into a waking period, a rapid eye movement period, a sleep period N1, a sleep period N2 and a sleep period N3, wherein the N1 and the N2 are light sleep periods, and the N3 is deep sleep periods; in the six categories, the sleep stage is divided into a waking period, a REM period, a sleep stage S1, a sleep stage S2, a sleep stage S3 and a sleep stage S4, wherein S1 and S2 are light sleep periods, and S3 and S4 are deep sleep periods. At present, four classifications, five classifications and six classifications are all common sleep staging modes, and in the present embodiment, four classifications are used as sleep staging modes. Generally, in the whole sleep cycle, the rapid eye movement period accounts for about 20%, the light sleep period accounts for about 55%, and the deep sleep period accounts for about 25%. In the fifth category, REM period is about 20%, N1, N2 about 55%, and N3 about 25%; in the six classifications, S1 is about 5%, S2 is about 50%, and S3, S4 are about 25% in total. The sleep staging results can be used to assess sleep quality, analyze characteristics of individual sleep stages, thereby assisting in diagnosing sleep-related diseases, providing sleep-aid solutions, and the like.

Photoplethysmography (PPG) is a non-invasive method for detecting changes in blood volume in living tissue by means of an electro-optical technique. When a light beam with a certain wavelength irradiates the surface of the skin, the light beam is transmitted to the photosensitive sensor in a transmission or reflection mode, and the light intensity detected by the photosensitive sensor is weakened due to the absorption attenuation effect of skin muscles and blood at the detection end in the process, wherein the absorption of the skin muscles, tissues and the like to the light is kept constant in the whole blood circulation, the volume of the blood in the skin is changed in a pulsating mode under the action of the heart, the peripheral blood volume is the largest when the heart contracts, and the maximum detected light intensity of the light absorption quantity is the smallest; when the heart is in diastole, on the contrary, the detected light intensity is the maximum, so that the light intensity received by the photosensitive sensor is in pulsatile change, the light intensity change signal is converted into an electric signal (AC signal), the change of volume pulse blood flow can be obtained, and the electric signal (AC signal) is further converted into a human heart rate signal.

The heart rate is the number of heartbeats per minute of a normal person in a resting state, and is also called resting heart rate or resting heart rate.

Heart Rate Variability (HRV), refers to the variation in the difference between successive heart cycles.

A photoplethysmography (PPG) sensor for acquiring a human heart rate.

An acceleration sensor (ACC) is used to detect the magnitude of acceleration in various directions (typically three axes, x, y and z axes).

For convenience of understanding, an application scenario to which the embodiments of the present disclosure may be applied is introduced first, and please refer to fig. 1, which is a schematic structural diagram of a sleep monitoring system provided in the embodiments of the present disclosure. The sleep monitoring system includes a wearable device 100, a server 200, and an electronic device 300.

The wearable device may be, for example, a smart watch, and in use, the wearable device surrounds a wrist of a user, the display screen is disposed on a surface of the wearable device, the user can directly perform a touch operation on the display screen or perform a selection input by using a physical button, and after inputting an instruction, the wearable device 100 may collect sleep data of the user, and in a specific implementation, the sleep data includes, but is not limited to, data of the following items: heart rate data, body motion data, etc. The wearable device 100 can perform validity determination on the acquired sleep data, upload the sleep data confirmed to be valid to the server 200, or synchronize the sleep data confirmed to be valid to the electronic device 300.

Specifically, the sleep data may be analyzed by a computer program built in the wearable device 100 to obtain a sleep staging result; or analyzing the sleep data by using a computer program in the server 200 to obtain a sleep staging result; or analyzing the sleep data by using a computer program built in the electronic device 300 to obtain a sleep staging result; and judging whether the sleep data in the sleep cycle of the user is abnormal or not according to the sleep staging result. And when the acquired sleep data are abnormal, sending a prompt to the user.

Fig. 2 is a schematic structural diagram of a wearable device according to an embodiment of the present invention. As shown in fig. 2, the wearable device 100 may specifically include a watch body and a wrist band connected to each other, where the watch body includes a front case (not shown in fig. 2), a touch panel 101 (also called a touch screen), a display screen 102, a bottom case (not shown in fig. 2), and a processor 103, a memory 105, a Microphone (MIC) 106, a Bluetooth (BT) 108, a PPG sensor 109, an acceleration sensor 110, a light sensor 111, a power supply 112, a power supply management system 113, and the like. Although not shown, the wearable device 100 may further include: an antenna, a WiFi module, a Near Field Communication (NFC) module, a Global Positioning System (GPS) module, a speaker, and the like.

The following describes the functional components of the wearable device 100:

the touch screen 101, also referred to as a touch panel, may collect touch operations of a watch user thereon (e.g., operations of the user on or near the touch panel using any suitable object or accessory such as a finger, a stylus, etc.) and drive a responsive connection device according to a preset program. Alternatively, the touch panel 101 may include two parts, i.e., a touch detection device and a touch controller. The touch detection device detects the touch direction of a user, detects a signal brought by touch operation and transmits the signal to the touch controller; the touch controller receives touch information from the touch sensing device, converts the touch information into touch point coordinates, sends the touch point coordinates to the processor 103, and can receive and execute commands sent by the processor 103. In addition, the touch panel may be implemented in various types such as a resistive type, a capacitive type, an infrared ray, and a surface acoustic wave. In addition to the touch screen 101, the wearable device 100 may also include other input devices, which may include, but are not limited to, function keys (such as volume control keys, switch keys, etc.).

Display screen 102 may be used to display information entered by or provided to the user as well as various menus for the watch. Alternatively, the Display screen 102 may be configured in the form of a Liquid Crystal Display (LCD), an organic light-Emitting Diode (OLED), or the like.

Further, the touch panel 101 may cover the display screen 102, and when the touch panel 101 detects a touch operation on or near the touch panel 101, the touch operation is transmitted to the processor 103 to determine the type of the touch event, and then the processor 103 provides a corresponding visual output on the display screen 102 according to the type of the touch event. Although in fig. 2, touch panel 101 and display screen 102 are implemented as two separate components to implement the input and output functions of the watch, in some embodiments, touch panel 201 may be integrated with display screen 202 to implement the input and output functions of the wearable device.

The processor 103 is used for performing system scheduling, controlling the display screen and the touch screen, and processing data of the microphone 106 and the bluetooth 108. The processor 103 is used to perform operations on the sensor data, such as: the processor 103 may be further configured to determine a sleep staging result according to the acquired sleep data in the sleep cycle, finally determine whether the sleep data is abnormal according to the sleep staging result, and instruct the prompter to send a prompt message when the sleep data is abnormal, so that the user can check the related information conveniently.

A microphone 106, also referred to as a microphone. The microphone 106 may convert the collected sound signals into electrical signals, which are received by the audio circuit and converted into audio data; the audio circuit can also convert the audio data into an electric signal, transmit the electric signal to a loudspeaker, and convert the electric signal into a sound signal to be output by the loudspeaker.

Bluetooth 108: the wearable device 100 can interact information with other electronic devices (such as a mobile phone, a tablet computer, etc.) through bluetooth, and is connected to a network through the electronic devices, connected to a server, and used for processing functions such as voice recognition.

The sensor may be a PPG sensor 109, an acceleration sensor 110, a light sensor 111, a temperature sensor, a motion sensor, or other sensor. In particular, the light sensor may include an ambient light sensor and a proximity sensor. As for other sensors such as a gyroscope, a barometer, a hygrometer, a thermometer, and an infrared sensor, which may be further configured to the wearable device 100, detailed descriptions thereof are omitted.

The memory 105 is used for storing software programs and data (for example, motion information), and the processor 103 executes various functional applications and data processing of the watch by running the software programs and data stored in the memory. The memory 105 mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system and application programs (such as a sound playing function and an image playing function) required by at least one function; the stored data area may store data created from use of the watch (e.g., audio data, a phone book, etc.). Further, the memory may include high speed random access memory, and may also include non-volatile memory, such as a magnetic disk storage device, flash memory device, or other volatile solid state storage device.

A timer (not shown in fig. 2), the time length of which can be dynamically adjusted, for example: when the timer is started, the sensor starts to acquire the sleep data, and the timer can also be used for controlling the time length for acquiring the sleep data by the sensor.

A prompt unit (not shown in fig. 2), which may include: indicator lights, display screens, etc. The prompting mode of the prompting unit can comprise one or more of the following modes: vibration prompt, indicator light prompt, text prompt and voice prompt.

The watch also comprises a power supply 112 (for example a battery) for supplying power to the various components, the power supply 112 being preferably logically connected to the processor 103 via a power management system 113, so as to manage the functions of charging, discharging and power consumption management, etc., via the power management system 113.

The following describes a method for analyzing sleep data validity provided by the present application.

At present, the threshold of the wearing detection algorithm of most wearable devices is set to be greater than or equal to an 8mm threshold, for example, the distance between the bottom PPG sensor of the smart watch and the wrist surface is controlled to be about 8-10mm, and when the distance between the PPG sensor and the wrist surface is greater than the 8mm threshold, the smart watch can determine that the wearable device is not worn. In the unworn state, the wearable device does not trigger the sleep monitoring procedure.

However, sometimes, the user does not wear the wearable device, and the wearable device still detects the heart rate value, for example, when a light beam with a certain wavelength irradiates the fiber surface, the light beam is easily similar to the heart rate signal by transmitting or reflecting the signal transmitted back to the PPG sensor due to interference of the concave-convex waveform on the fiber surface, so that the wearing detection algorithm has a misjudgment phenomenon, and the user wears the wearable device. At this time, if the wearable device determines that the user enters a sleep state according to the collected user movement data, a sleep monitoring program is triggered, so that the collected sleep data is abnormal, and the sleep data accumulated by the cloud server is wrong or meaningless. Therefore, effectiveness judgment is performed on the acquired sleep data to reduce interference of invalid data on effective data, so that the sleep state of the user can be analyzed by using the effective sleep data accumulated by the cloud server, the sleep quality of the user is evaluated, diseases related to sleep are assisted to diagnose, a sleep assisting scheme is provided, and the like.

Fig. 3 is a flowchart of a sleep data validity analysis method provided in this embodiment, and fig. 4 is another flowchart of the sleep data validity analysis method provided in this embodiment, as shown in fig. 3 and 4, the method includes:

step S102, acquiring sleep data in a sleep cycle of a user.

Specifically, the sleep data comprises heart rate data and body movement data, the body movement data is data generated by recording the movement of limbs of the human body, and the heart rate data is data for recording the heartbeat information of the human body.

The sleep data may also include electromyographic data (data that records bioelectrical information of human muscles), respiratory data (e.g., respiratory rate, snoring), blood flow rate, electrocardiographic data, and the like.

In this embodiment, various types of sleep data are collected through a sensor built in the wearable device 100, for example, various types of sleep data are collected through a smart watch worn on a wrist of a user, and this embodiment of the present application does not limit a specific form of the wearable device 100, and for example, the wearable device may also be an earband-type wearable device, a smart ring, or the like.

In this embodiment, the body movement data of the user is collected by an acceleration sensor on the wearable device. Heart rate data of the user is acquired by the PPG sensor. In other embodiments, the body movement data of the user may be collected through a gyroscope, and the heart rate data of the user may be collected through a heart rate sensor, which is not limited herein.

Further, the wearable device 100 can calculate various feature values of the body movement data according to the collected body movement data, and determine whether the user goes to sleep according to the feature values. For example, when the average value of the acceleration of the user is continuously lower than a set threshold value for a prescribed time or more, the user is determined to go to sleep. After determining that the user goes to sleep, the processor of the wearable device 100 can control the PPG sensor and the accelerometer to acquire heart rate data of the user according to a preset acquisition frequency and duration.

In one embodiment, the characteristic values of the body motion data include acceleration maximum values, acceleration minimum values, acceleration average values, acceleration accumulated values, displacements, body motion amplitudes, and the like. For example, the acceleration data in the X, Y and Z axis directions of the physical ground collected by the acceleration sensor are calculatedAverage value of acceleration in each direction

Wherein A represents an average value of acceleration, A _x Represents the acceleration in the X-axis direction, A _y Representing acceleration in the direction of the Y-axis, A _z Indicating acceleration in the Z-axis direction. The average value of the acceleration is used to determine the amount of activity of the user, such as a large amount of activity, a medium amount of activity, and a small amount of activity. The acceleration sensor may be a piezoresistive type, a piezoelectric type or a capacitive type, and is not limited herein.

Further, the characteristic values of the heart rate data include a maximum heart rate value, a minimum heart rate value, an average heart rate value, an HRV value, a variance heart rate, and the like.

In other embodiments, the flow rate of the blood in the human body can be acquired by a biological impedance sensor; an electrocardiogram is acquired by an electrocardiograph sensor, and the like, which is not limited herein.

During data acquisition, the sleep data of the user can be acquired according to the preset acquisition frequency. For example, sleep data is collected once every 30 seconds, and assuming that sleep data of a user between 22.

And step S104, performing sleep staging according to the sleep data to obtain sleep staging results, wherein the sleep staging results comprise a rapid eye movement period, a deep sleep period and a shallow sleep period.

Specifically, the sleep data may be analyzed by a computer program built in the wearable device 100 to obtain a sleep staging result; or synchronizing the sleep data collected by the wearable device 100 to the cloud server 200, and analyzing the sleep data by using a computer program in the cloud server to obtain a sleep staging result; or synchronizing the sleep data collected by the wearable device 100 to the electronic device 300, and analyzing the sleep data by using a computer program built in the electronic device to obtain a sleep staging result; and judging whether the sleep data in the sleep cycle of the user is abnormal or not according to the sleep staging result. And when the acquired sleep data is abnormal, sending a prompt to the user.

As shown in fig. 5, in the present embodiment, the wearable device 100 synchronizes the acquired sleep data to the electronic device 200, for example, the user's cell phone. It is understood that the wearable device 100 can share data with a mobile phone, a tablet computer, etc. through bluetooth. The user can view the sleep staging results through the sleep monitoring application.

Step S104, specifically including:

and step S1041, counting the total duration of the sleep cycle of the user according to the sleep data.

Specifically, whether the user enters the sleep or finishes the sleep is determined according to the characteristic value of the body movement data, and the sleep duration is calculated according to the time when the user enters the sleep and the time when the user finishes the sleep. For example, if the characteristic value (acceleration average value, acceleration accumulated value, displacement, and body movement amplitude) of the body movement data is continuously lower than the set threshold value for a predetermined time or longer, it is determined that the user's activity amount is small, and it is determined that the user has fallen asleep. The sleep duration of the user is determined by adopting a plurality of characteristic values of the body movement data, the sleep duration can be compared with each preset characteristic value range, the state of the user can be judged quickly, whether the user enters the sleep or not can be judged, and the accuracy and the efficiency of judgment can be improved.

In other embodiments, the method further comprises: and receiving the sleep signal and the sleep ending signal, confirming the time when the sleep signal is received as the sleep starting time, and confirming the time when the sleep ending signal is received as the sleep ending time.

The sleep signal may be, for example, a voice signal, a gesture signal, a touch signal, or the like. As shown in fig. 6 to 7, the user opens the sleep monitoring application program by clicking the program on the display interface of the wearable device 100, and inputs a signal to fall asleep through the touch screen or the touch button. It can be understood that, a user may actively input a sleep signal or a sleep signal ending through the sleep monitoring application APP of the wearable device 100, a graphical user interface of the sleep monitoring application APP includes a first control (start of sleep) and a second control (end of sleep), the user clicks the first control to trigger generation of the sleep signal, and the user clicks the second control to trigger generation of the sleep signal ending.

An application on wearable device 100 having an alarm function may also be monitored to determine an end-of-sleep moment. Specifically, data sharing permission between the alarm clock APP and the sleep monitoring application APP needs to be set, for example, when the alarm clock is determined to be 7 am, the sleep monitoring application APP can read alarm clock data of the alarm clock APP to determine that the sleep termination time is 7 am.

The above is merely an example of how to determine the sleep time of the user, and is not intended to limit the present solution.

And step S1042, when the total duration exceeds a preset value, performing sleep stage by using the heart rate data and the body movement data to obtain a sleep stage result.

The preset value may be, for example, 2 hours, 3 hours, 4 hours, or the like. Understandably, when the user does not actually enter the sleep state and the wearable device is mistakenly judged to enter the sleep state, the sleep stage result can be found to be abnormal after the sleep stage. For example, when a user reads or reads a book, the wearable device misjudges that the user enters a sleep state, and at the moment, the heart rate data and the body movement data are jointly used as the basis of sleep staging, so that the accuracy of sleep data staging can be improved according to the difference of the heart rate data and the body movement data along with the change of the sleep state. It can be understood that if the acquired sleep data includes a large amount of data in a non-sleep state, it is easy to cause that a certain sleep stage (a fast eye movement period, a deep sleep period, and a light sleep period) in the sleep stage result is higher in proportion, for example, the light sleep period accounts for 70%, and thus it is determined that the sleep data is abnormal.

When the user does not wear the wearable equipment, when the wearable equipment misjudges to get into the sleep state, the pseudo heart rate signal similar to the heart rate signal is transmitted back to through the transmission or reflection mode to the light beam of PPG sensor because of the interference of the concave-convex waveform on the fiber surface, and the pseudo heart rate signal can not show the difference along with the change of the sleep state, and can determine that the sleep data is abnormal according to the characteristic value of the heart rate signal.

Further, step S104 further includes:

and step S1043, when the total duration does not exceed a preset value, performing sleep staging by using the body movement data to obtain a sleep staging result.

For example, when the total duration is less than 2 hours, the data amount of the acquired heart rate data is small because the sleep duration is short, and at this time, if the heart rate data is used as the basis of the sleep stage, the result of the sleep stage is easily inaccurate. The wearable device can judge whether the user enters the sleep by utilizing the body movement data in a certain time period acquired by the acceleration sensor, if the user enters the sleep, the sleep monitoring program is triggered, and the PPG sensor starts to acquire the heart rate data, so that the heart rate data are often less when the sleep time is shorter, and the whole sleep cycle cannot be covered. For example, user 12 noon: 30 to 14:00 afternoon nap, wherein the heart rate data acquisition time period is 13 to 14:00, the missing part accounts for a larger proportion of the whole sleep cycle, so when the sleep duration is less than a preset value, only the body movement data is used for sleep staging, the accuracy and the efficiency of the sleep staging can be improved, and the influence of the heart rate data of the missing part on the whole sleep staging is avoided. And after the user carries out sleep stage by using the body movement data, directly outputting a sleep stage result to the user.

Step S104 may further include: and step S1044, automatically deleting the acquired sleep data when the total duration is less than the minimum time limit. For example, the total duration is less than 1 hour, the sleep data collected in the fragmented sleep state has little meaning to the overall sleep health analysis of the user, and the system can automatically delete the data.

It should be noted that the human beings have different ages and sleep for different periods, for example, children have a sleep period of 9 to 10 hours, adults have a sleep period of 7 to 9 hours, elderly people have a sleep period of 6 to 8 hours, and elderly people older than 80 years need 9 to 10 hours. The normal heart rate of an adult is 60-100 times/minute, the heart rate of athletes is about 50 times/minute generally, the heart rate of adult females is faster than that of adult males of the same age generally, the heart rate of old people is slower than that of younger people, and the heart rate of children is faster than that of adults. The sleeping incubation time of the old is often longer than that of the young. Therefore, personalized sleep staging needs should be considered when performing sleep staging according to physiological characteristic data.

Based on sleep data collected by adults in sleep state, corresponding characteristic value ranges are respectively set in the waking period, the light sleep period, the deep sleep period and the rapid eye movement period, for example, the average value range of each axis (X, Y and Z axes) of acceleration in body movement data in the waking period is>250m/s ² (ii) a During the light sleep period, the average value range of each axis of acceleration in the body movement data is 50-150 m/s ² (ii) a During deep sleep, the average value range of each axis of acceleration in the body movement data is 20-100 m/s ² . In the waking period, the range of the heart rate value in the heart rate data is 60-100 times/minute; in the light sleep period, the range of the heart rate value in the heart rate data is 60-90 times/min; the range of heart rate values in the heart rate data during the deep sleep period is 40-80 times/min. For example, the range of the heart rate value of the light sleep period is 60 to 90 times/minute, and when the heart rate value in a certain period of time (23.

In one embodiment, feature values such as a heart rate maximum value, a heart rate minimum value, a heart rate average value and an HRV value in the body movement data and the heart rate data of different types of users can be calculated respectively. And comparing each calculated characteristic value with a preset characteristic value range corresponding to each sleep stage, and similarly, presetting and configuring the fluctuation range of each characteristic value in the corresponding sleep stage (a waking period, a rapid eye movement period, a deep sleep period and a shallow sleep period). For example, the characteristic values of the sleep data of the child are compared with the characteristic value ranges corresponding to the sleep stages of the corresponding child.

The calculation modes of all the characteristic values are different, and each characteristic value has a respective calculation method.

Wherein A represents anAverage value of speed, A _x Represents the acceleration in the X-axis direction, A _y Representing acceleration in the direction of the Y-axis, A _z Indicating acceleration in the Z-axis direction. And is not particularly limited herein.

The sleep data can also be staged by using a preset sleep staging model. The sleep staging model may be a recurrent neural network model, a convolutional neural network model, a hidden markov model, a principal component analysis model, or the like.

In one embodiment, the sleep staging model includes a convolutional neural network model that first needs to be trained. Specifically, the sleep data of a plurality of users in normal sleep conditions all night are respectively collected as training samples, and each training sample is marked with a user type label (for example: adult male, adult female, elderly, teenager). Each training sample is analyzed and distinguished by an expert, corresponding sleep stages are obtained and marked by a sleep label, wherein the sleep label specifically comprises four types of labels of a waking period, a rapid eye movement period, a deep sleep period and a light sleep period. Inputting the training sample (numerical sequence) into the convolutional neural network model, training until the model converges, and storing the network parameters. The convolutional neural network model comprises a network input layer, a hidden layer and a network output layer, wherein the loss function is a cross entropy loss function, and the loss function is trained by using an optimizer.

It should be noted that the training samples need to be processed in advance, for example, the acquired heart rate data and the acquired body movement data are made into a numerical sequence according to a time sequence, and the numerical sequence is labeled with a user type label, so as to obtain the training samples.

The trained sleep staging model can perform sleep staging on the acquired sleep data and output a sleep staging result. Before collecting the user's sleep data, the user may perform personalized parameter settings, such as age, sleep duration, sleep regularity, etc., on wearable device 100 or electronic device 300.

The sleep staging model may also be a principal component analysis model. In the process of model construction, firstly, body movement data of a plurality of users (including adult males, adult females, old people and teenagers) all night under a normal sleep condition are collected, a plurality of characteristic values such as an acceleration maximum value, an acceleration minimum value, an acceleration average value, an acceleration accumulated value, a displacement and a body movement amplitude in the body movement data are extracted, and the plurality of characteristic values form a group of multi-dimensional vectors; analyzing by a principal component analysis method to obtain a principal component analysis matrix; and then, clustering analysis is carried out on the principal component analysis matrix by utilizing a machine learning algorithm to obtain a sleep stage model. Therefore, the acquired body movement data can be subjected to sleep staging by using the sleep staging model, and a sleep staging result is output.

Further, after step S102 and before step S104, the method further comprises:

step S103, preprocessing the sleep data, wherein the preprocessing comprises at least one of smoothing denoising, removing abnormal values and filling missing values.

The smoothing and denoising may be performed by using gaussian filtering, fourier transform filtering, convolution smoothing algorithm, and the like, which is not specifically limited in this embodiment of the present application.

The abnormal values are removed, so that the obviously abnormal values can be deleted, and whether the acquired data are abnormal or not can be judged by configuring a data threshold in advance. For example, the maximum value of the heart rate has a threshold of 120 beats/minute, and the value of the heart rate exceeding the threshold is an abnormal value.

And filling up the missing values refers to filling up the missing values according to a theoretical rule, so that the calculation feasibility is improved.

Step S105, when the ratio of at least one of the fast eye movement period, the deep sleep period and the light sleep period to the sleep period exceeds a preset threshold, determining that the sleep data in the sleep period of the user is abnormal.

Generally, in the whole sleep cycle, the rapid eye movement period accounts for about 20%, the light sleep period accounts for about 55%, and the deep sleep period accounts for about 25%. If the collected sleep data passes through the sleep stage, the result is that the rapid eye movement period accounts for about 10%, the light sleep period accounts for about 70%, and the deep sleep period accounts for about 20%, and thus the sleep stage result is abnormal.

Further, before step S105, the method further comprises:

determining a preset threshold value, wherein the preset threshold value comprises a first preset threshold value, a second preset threshold value and a third preset threshold value; the first preset threshold is a ratio range of the deep sleep period to the sleep cycle; the second preset threshold is a ratio range of the light sleep period to the sleep cycle; the third preset threshold is a ratio range of the rapid eye movement period to the sleep cycle.

The preset threshold may be generated according to a setting.

The method comprises the steps of obtaining a plurality of normal sleep data of a user, and determining preset thresholds according to the normal sleep data, wherein the preset thresholds comprise a first preset threshold, a second preset threshold and a third preset threshold.

The preset threshold of the ratio of each epoch to the sleep cycle can be obtained in the above manner, but is not limited to the above manner.

The method comprises the following specific steps: acquiring a plurality of normal sleep data of a user, and performing sleep staging on the normal sleep data respectively to obtain a plurality of normal sleep staging results; respectively calculating the ratio of the deep sleep period to the sleep cycle, the ratio of the light sleep period to the sleep cycle and the ratio of the rapid eye movement period to the sleep cycle in each normal sleep stage result; determining a range formed by the ratios of the deep sleep periods to the sleep cycle as a first preset threshold; determining a range formed by the ratios of the plurality of light sleep periods to the sleep cycle as a second preset threshold; and determining a range formed by the ratios of the plurality of rapid eye movement periods to the sleep cycle as a third preset threshold.

For example, in the normal sleep data collected during 10 times of sleep of the user, the ratio of the light sleep period to the sleep cycle is 60.5%, 58.2%, 55.6%, 59.7%, 58.8%, 54.5%, 56.2%, 60.1%, 54.3%, and 57.8%, respectively. Then, the second preset threshold of the user's light sleep period in the sleep cycle is 54.3-60.5%. Or the elastic space is increased by 2 points on the basis of the determined second preset threshold value, for example, 52.3-62.5%.

For example: the second preset threshold value is (52.3-62.5%), and when the ratio of the light sleep period to the sleep period is smaller than the first limit value (52.3%), the abnormal sleep data in the sleep period of the user is confirmed; or

And when the ratio of the light sleep period to the sleep period is greater than a second limit value (62.5%), confirming that the sleep data in the sleep period of the user are abnormal.

It can be understood that when the user does not wear the wearable device, for example, the smart watch is placed on a bedside table, and after sleep data of the user collected by the smart watch passes through a sleep period, a ratio of a light sleep period of the user to a sleep period exceeds a normal range (52.3-62.5%), the sleep data in a non-sleep state is doped in the sleep data, that is, the sleep data is abnormal.

Optionally, after step S104, the method further includes:

step S107, when the ratio of all the stages in the sleep cycle does not exceed the corresponding preset threshold, the sleep data in the sleep cycle of the user is confirmed to be normal.

And step S108, outputting a sleep staging result.

In one embodiment, the sleep staging results can be output to the user in a pie chart, bar chart, line chart, perspective view, etc. so that the user can quickly learn his or her sleep state on the display interface of the wearable device 100 or the electronic device 300.

Optionally, after step S104, the method further comprises:

and step S106, processing the sleep data by adopting a preset processing mode, wherein the processing mode comprises at least one of reservation, cutting and deletion.

Specifically, when the processing mode is retention, the sleep data will be preserved, and even if the data is found to be abnormal by the effectiveness analysis, the sleep data will be preserved. And can assess the quality of sleep of the user, diagnose sleep-related ailments, and the like based thereon.

After step S105 and before step S106, a prompt message is output to prompt the user to select a processing mode. In this embodiment, as shown in fig. 8 and 9, when the sleep data is abnormal, the prompt information is output, the display interface may present the prompt information (retention, cutting, and deletion), and the user can select a corresponding processing mode according to the prompt and by combining the situation of the user, so as to avoid storing a large amount of abnormal sleep data. The user inputs the selected processing mode through the display interface on the electronic device 300 or the wearable device 100, and the system performs corresponding operation according to the instruction input by the user.

When the processing mode is reservation, even if the sleep data is abnormal, the sleep data is also saved.

When the processing mode is clipping, the sleep data is output in a sequence form, and the user can clip the sleep data through a clipping tool, such as a video clip. For example, when the sensor collects the sleep data of the user as 21:00, because the user does not sleep in the period of 21 to 00 to 22.

When the processing mode is deleting, the abnormal sleep data is automatically deleted. When the user selects the delete processing method, "remember the current operation" occurs, and after the setting, the system autonomously deletes all the sleep data determined to be abnormal.

Fig. 10 is a schematic block diagram of a sleep data validity analysis apparatus according to an embodiment of the present invention. As shown in fig. 10, an embodiment of the present invention provides a sleep data validity analysis apparatus, which is configured to execute the sleep data validity analysis method in the foregoing embodiment, that is, the following specific working processes of each product may refer to the corresponding processes in the foregoing method embodiments.

The sleep data validity analysis device provided by the embodiment of the application may be a wearable device 100 such as a smart band, a smart watch, a smart ring, and smart glasses, or may also be an electronic device 300 used in conjunction with the wearable device, where the electronic device may be a mobile phone, a tablet computer, a Personal Computer (PC), a Personal Digital Assistant (PDA), a netbook, and the like, and may also be a server 200, and the server 200 may be implemented by an independent server or a server cluster formed by multiple servers. The embodiment of the present application does not specifically limit the specific form of the electronic device.

The apparatus includes a first obtaining unit 410, a sleep staging unit 420, and a first determining unit 430.

A first obtaining unit 410, configured to obtain sleep data in a sleep cycle of a user; a sleep staging unit 420, configured to perform sleep staging according to the sleep data to obtain a sleep staging result; a first determining unit 430, configured to determine that sleep data in the sleep cycle of the user is abnormal when a ratio of at least one of a fast eye movement period, a deep sleep period, and a shallow sleep period to the sleep cycle exceeds a preset threshold.

Further, the wearable device 100 can calculate various feature values of the body movement data according to the collected body movement data, and determine whether the user goes to sleep according to the feature values. For example, when the average value of the acceleration of the user is continuously lower than a set threshold value for a prescribed time or more, the user is determined to go to sleep. After determining that the user is asleep, the processor of the wearable device 100 can control the PPG sensor and the accelerometer sensor to acquire heart rate data of the user according to a preset acquisition frequency and duration.

In one embodiment, the characteristic values of the body motion data include acceleration maximum values, acceleration minimum values, acceleration average values, acceleration accumulated values, displacements, body motion amplitudes, and the like. For example, acceleration data in X, Y and Z axis directions of the physical ground collected by an acceleration sensor is used for calculating the average value of the acceleration in each direction

Wherein A represents an average value of acceleration, A _x Denotes the acceleration in the X-axis direction, A _y Representing acceleration in the direction of the Y-axis, A _z Indicating acceleration in the Z-axis direction. The amount of activity of the user, such as a large amount of activity, a medium amount of activity, and a small amount of activity, is determined from the average value of the acceleration. The acceleration sensor may be a piezoresistive type, a piezoelectric type or a capacitive type, and is not limited herein.

Further, the characteristic values of the heart rate data include a heart rate maximum value, a heart rate minimum value, a heart rate average value, an HRV value, a heart rate variance, and the like.

In other embodiments, the blood flow rate of the human body can be acquired through a biological impedance sensor; an electrocardiogram is acquired by the electrocardiograph sensor, and the like, which is not limited herein.

During data acquisition, the sleep data of the user can be acquired according to the preset acquisition frequency. For example, sleep data is collected every 30 seconds, and if sleep data of a user between 22.

In one possible implementation, the sleep staging unit 420 includes a statistics subunit, a first staging subunit.

The statistic subunit is used for counting the total duration of the sleep cycle of the user according to the sleep data; and the first staging subunit is used for performing sleep staging by using the heart rate data and the body movement data when the total duration exceeds a preset value, so as to obtain a sleep staging result.

The statistical subunit is used for determining whether the user enters the sleep or finishes the sleep according to the characteristic value of the body movement data, and calculating the sleep duration according to the time of entering the sleep and the time of finishing the sleep. For example, if the characteristic value (acceleration average value, acceleration accumulated value, displacement, and body movement amplitude) of the body movement data is continuously lower than the set threshold value for a predetermined time or longer, it is determined that the user's activity amount is small, and it is determined that the user has fallen asleep. The sleep duration of the user is determined by adopting a plurality of characteristic values of the body movement data, the sleep duration can be compared with each preset characteristic value range, the state of the user can be judged quickly, whether the user enters the sleep or not can be judged, and the accuracy and the efficiency of judgment can be improved.

In other embodiments, the statistical subunit is further configured to receive the sleep-in signal and the sleep-ending signal, determine a time when the sleep-in signal is received as a sleep start time, and determine a time when the sleep-ending signal is received as a sleep end time; and calculating the sleep duration according to the sleep starting time and the sleep ending time.

The sleep signal may be, for example, a voice signal, a gesture signal, a touch signal, or the like. As shown in fig. 6 to 7, the user opens the sleep monitoring application program by clicking the program on the display interface of the wearable device 100, and inputs a signal to fall asleep through the touch screen or the touch button. As can be appreciated, the user may actively input a sleep signal or a sleep signal ending signal through the sleep monitoring application APP of the wearable device 100, a graphical user interface of the sleep monitoring application APP includes a first control (start sleep) and a second control (end sleep), the user clicks the first control to trigger generation of the sleep signal, and the user clicks the second control to trigger generation of the sleep signal ending.

Further, the preset value may be, for example, 2 hours, 3 hours, 4 hours, or the like. It can be understood that when the user does not actually enter the sleep state and the wearable device mistakenly determines that the wearable device enters the sleep state, the sleep staging result can be found to be abnormal after the sleep staging. For example, when a user reads or reads a book, the wearable device misjudges that the user enters a sleep state, and at the moment, the heart rate data and the body movement data are jointly used as the basis of sleep staging, so that the accuracy of sleep data staging can be improved according to the difference of the heart rate data and the body movement data along with the change of the sleep state. It can be understood that if the acquired sleep data includes a large amount of data in a non-sleep state, it is easy to cause that a certain sleep stage (a fast eye movement period, a deep sleep period, and a light sleep period) in the sleep stage result is higher in proportion, for example, the light sleep period accounts for 70%, and thus it is determined that the sleep data is abnormal. When the user does not wear the wearable equipment, when the wearable equipment misjudges to get into the sleep state, the pseudo heart rate signal similar to the heart rate signal is transmitted back to through the transmission or reflection mode to the light beam of PPG sensor because of the interference of the concave-convex waveform on the fiber surface, and the pseudo heart rate signal can not show the difference along with the change of the sleep state, and can determine that the sleep data is abnormal according to the characteristic value of the heart rate signal.

The process performed by the first stage subunit may be performed in a convolutional neural network model.

In one possible implementation, the sleep staging unit 420 further includes: and the second staging subunit is used for staging the sleep by utilizing the body movement data when the total duration does not exceed the preset value, so as to obtain a sleep staging result.

For example, when the total duration is less than 2 hours, the data amount of the acquired heart rate data is small because the sleep duration is short, and at this time, if the heart rate data is used as the basis of the sleep stage, the result of the sleep stage is easily inaccurate. It can understand ground, wearable equipment utilizes the body movement data judgement user in a certain time quantum of acceleration sensor collection whether to get into sleep, if get into sleep, just can trigger the sleep monitoring procedure, and PPG sensor just can begin to gather heart rate data, so when the sleep is long shorter, heart rate data often is less, can not cover whole sleep cycle. For example, user 12 noon: 30 to 14:00 afternoon nap, wherein the heart rate data acquisition time period is 13-14 points: 00, the missing part accounts for a larger proportion of the whole sleep cycle, so when the sleep duration is less than a preset value, the sleep staging is carried out only by utilizing the body movement data, the accuracy and the efficiency of the sleep staging can be improved, and the influence of the heart rate data of the missing part on the whole sleep staging is avoided.

The process performed by the second staging subunit may be performed in a principal component analysis model.

In one possible implementation, the apparatus 400 further includes: the preprocessing unit 440 is configured to preprocess the sleep data, where the preprocessing includes at least one of smoothing and denoising, removing outliers, and filling missing values.

The smoothing and denoising may be performed by using a smoothing and denoising algorithm, for example, gaussian filtering, fourier transform filtering, convolution smoothing algorithm, and the like, which is not specifically limited in this embodiment of the present application. The elimination of the abnormal values can delete the obviously abnormal values, for example, whether the acquired data is abnormal can be judged by configuring a data threshold in advance. And filling up the missing value refers to filling up the actual value according to a theoretical rule, so that the calculation feasibility is improved.

In one possible implementation, the apparatus further includes: a receiving unit 450, configured to receive a preset threshold set by a user, where the preset threshold includes a first preset threshold, a second preset threshold, and a third preset threshold; the first preset threshold is a ratio range of the deep sleep period to the sleep cycle; the second preset threshold is a ratio range of the light sleep period to the sleep cycle; the third preset threshold is a ratio range of the rapid eye movement period to the sleep cycle.

The user can set various threshold values according to the actual conditions of the user, so that the personalized requirements of the user can be met when the sleep data are judged to be abnormal, and the data analysis result is accurate and effective.

In one possible implementation, the apparatus further includes: a second acquisition unit for acquiring a plurality of normal sleep data of the user; the sleep staging unit is also used for performing sleep staging on the plurality of normal sleep data respectively to obtain a plurality of normal sleep staging results; the calculating unit is used for respectively calculating the ratio of the deep sleep period to the sleep cycle, the ratio of the light sleep period to the sleep cycle and the ratio of the rapid eye movement period to the sleep cycle in each normal sleep stage result; the first confirming unit is used for confirming a range formed by ratios of a plurality of deep sleep periods to a sleep cycle as a first preset threshold; the second confirming unit is used for confirming a range formed by the ratios of the plurality of light sleep periods to the sleep cycle as a second preset threshold; and the third confirming unit is used for confirming a range formed by the ratios of the plurality of rapid eye movement periods to the sleep cycle as a third preset threshold value.

According to the sleep data in the normal sleep state confirmed by the user, the preset threshold of the ratio of each period of the user to the sleep cycle is calculated according to the sleep data, so that when the sleep data to be analyzed is judged, the personalized requirements of the user can be met, and the data analysis result is accurate and effective.

In one possible implementation, the apparatus further includes: the processing unit 460 is configured to process the sleep data by using a preset processing manner, where the processing manner includes at least one of retention, clipping, and deletion.

The application also provides wearable equipment, which comprises a sensor module and a processor, wherein the sensor module is used for acquiring sleep data in a sleep cycle of a user; the processor is used for acquiring sleep data in the sleep cycle of the user; performing sleep staging according to the sleep data to obtain a sleep staging result; and when the ratio of at least one of the rapid eye movement period, the deep sleep period and the light sleep period to the sleep period exceeds a preset threshold value, confirming that the sleep data in the sleep period of the user are abnormal.

The sensor module comprises a heart rate sensor and an acceleration sensor; the heart rate sensor is used for collecting heart rate data of the user, and the acceleration sensor is used for collecting body movement data of the user; the processor is further configured to: counting the total duration of the sleep cycle of the user according to the sleep data; and when the total duration exceeds a preset value, performing sleep staging by using the heart rate data and the body movement data to obtain a sleep staging result.

Further, the processor is further configured to: and when the total duration does not exceed the preset value, performing sleep staging by using the body movement data to obtain a sleep staging result.

Further, the processor is further configured to: receiving a preset threshold value set by the user, wherein the preset threshold value comprises a first preset threshold value, a second preset threshold value and a third preset threshold value; the first preset threshold is a ratio range of the deep sleep period to the sleep cycle; the second preset threshold is a ratio range of the light sleep period to the sleep cycle; the third preset threshold is a ratio range of the rapid eye movement period to the sleep cycle.

Further, the processor is further configured to: acquiring a plurality of normal sleep data of the user; respectively carrying out sleep staging on the plurality of normal sleep data to obtain a plurality of normal sleep staging results; respectively calculating the ratio of the deep sleep period to the sleep cycle, the ratio of the light sleep period to the sleep cycle and the ratio of the rapid eye movement period to the sleep cycle in each normal sleep stage result; determining a range formed by the ratios of the deep sleep periods to the sleep cycle as a first preset threshold; determining a range formed by the ratios of the light sleep periods to the sleep cycle as a second preset threshold; determining a range composed of ratios of the plurality of fast eye movement periods to the sleep cycle as a third preset threshold.

Further, the processor is further configured to:

and processing the sleep data by adopting a preset processing mode, wherein the processing mode comprises at least one of reservation, cutting and deletion.

The application further provides an electronic device, which includes a processor and a memory, where the memory may be a non-volatile storage medium, and a computer program is stored in the memory, and the processor is connected to the memory and executes the computer program to implement the sleep data validity analysis method shown in fig. 3 to 4.

The present application also provides a computer-readable storage medium storing program code for execution by a device, which when run on a computer, causes the computer to perform the steps of the sleep data validity analysis method as described above in fig. 3 to 4.

The present application also provides a computer program product containing instructions which, when run on a computer or any at least one processor, causes the computer to perform the steps of the sleep data validity analysis method as shown in fig. 2-4.

In the above embodiments, the Processor may be a Central Processing Unit (CPU), or may be other general-purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory referred to may be read-only memory (ROM), other types of static memory devices that can store static information and instructions, random Access Memory (RAM), or other types of dynamic memory devices that can store information and instructions, EEPROM, compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, blu-ray disc, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, etc.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

The above embodiments are only specific embodiments of the present application, and any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the present disclosure, and all the changes or substitutions should be covered by the protection scope of the present application. The protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A sleep data validity analysis method, the method comprising:

acquiring sleep data in a sleep cycle of a user;

performing sleep staging according to the sleep data to obtain a sleep staging result;

and when the ratio of at least one of the rapid eye movement period, the deep sleep period and the light sleep period to the sleep period exceeds a preset threshold value, determining that the sleep data in the sleep period of the user is invalid.

2. The sleep data effectiveness analysis method according to claim 1, wherein the sleep data includes heart rate data and body movement data; the step of performing sleep staging according to the sleep data to obtain a sleep staging result comprises the following steps:

counting the total duration of the sleep cycle of the user according to the sleep data;

and when the total duration exceeds a preset value, performing sleep staging by using the heart rate data and the body movement data to obtain a sleep staging result.

3. The sleep data validity analysis method according to claim 2, wherein the step of performing sleep staging according to the sleep data to obtain sleep staging results further comprises:

4. The sleep data validity analysis method according to any one of claims 1 to 3, wherein before the sleep staging according to the sleep data to obtain a sleep staging result, the method further comprises:

and preprocessing the sleep data, wherein the preprocessing comprises at least one of smoothing and denoising, eliminating an invalid value and filling up a missing value.

5. The method for analyzing sleep data validity according to claim 1, wherein before confirming that the sleep data in the sleep cycle of the user is invalid when the ratio of at least one of the fast eye movement period, the deep sleep period and the light sleep period to the sleep cycle exceeds a preset threshold, the method further comprises:

6. The method for analyzing sleep data validity according to claim 1, wherein before confirming that the sleep data in the sleep cycle of the user is invalid when the ratio of at least one of the fast eye movement period, the deep sleep period and the light sleep period to the sleep cycle exceeds a preset threshold, the method further comprises:

acquiring a plurality of normal sleep data of the user;

respectively carrying out sleep staging on the plurality of normal sleep data to obtain a plurality of normal sleep staging results;

respectively calculating the ratio of the deep sleep period to the sleep cycle, the ratio of the light sleep period to the sleep cycle and the ratio of the rapid eye movement period to the sleep cycle in each normal sleep stage result;

determining a range formed by the ratios of the deep sleep periods to the sleep cycle as a first preset threshold;

determining a range formed by the ratios of the light sleep periods to the sleep cycle as a second preset threshold;

determining a range composed of ratios of the plurality of fast eye movement periods to the sleep cycle as a third preset threshold.

7. The sleep data validity analysis method as claimed in claim 2, wherein the step of counting the total duration of the sleep cycle of the user according to the sleep data comprises:

extracting a plurality of characteristic values in the body motion data;

determining whether the user enters sleep or stops sleeping according to the characteristic value of the body movement data;

and calculating the sleep duration according to the time of entering the sleep and the time of finishing the sleep.

8. The method for analyzing sleep data validity according to claim 1, wherein after confirming that the sleep data in the sleep cycle of the user is invalid when the ratio of at least one of the fast eye movement period, the deep sleep period and the light sleep period to the sleep cycle exceeds a preset threshold, the method further comprises:

9. A wearable device comprising a sensor module and a processor, wherein,

the sensor module is used for collecting sleep data in a sleep cycle of a user;

the processor is used for acquiring sleep data in the sleep cycle of the user; performing sleep staging according to the sleep data to obtain a sleep staging result; and when the ratio of at least one of the rapid eye movement period, the deep sleep period and the light sleep period to the sleep period exceeds a preset threshold value, confirming that the sleep data in the sleep period of the user is invalid.

10. The wearable device of claim 9, wherein the sensor module comprises a heart rate sensor and an acceleration sensor; the heart rate sensor is used for collecting heart rate data of the user, and the acceleration sensor is used for collecting body movement data of the user; the processor is further configured to:

and when the total duration exceeds a preset value, sleep staging is carried out by utilizing the heart rate data and the body movement data to obtain a sleep staging result.

11. The wearable device of claim 10, wherein the processor is further configured to:

12. The wearable device of claim 9, wherein the processor is further configured to:

13. The wearable device of claim 9, wherein the processor is further configured to:

acquiring a plurality of normal sleep data of the user;

14. The wearable device of claim 9, wherein the processor is further configured to: