CN108201435A - Sleep stage determines method, relevant device and computer-readable medium - Google Patents

Abstract

The embodiment of the invention discloses a kind of sleep stages to determine method, sleep stage equipment and computer-readable medium, and wherein method includes obtaining the electrocardiosignal of target user；Extract the corresponding fisrt feature data of the electrocardiosignal；The fisrt feature data and second feature data are subjected to data fusion, obtain target signature data, the second feature data are corresponding with the relevant characteristic of sleep stage for the target user；Classified using disaggregated model to the target signature data, obtain the result of sleep stage.In the embodiment of the present invention, the electrocardiosignal of target user is obtained, the fisrt feature data extracted from the electrocardiosignal and second feature data are subjected to data fusion, obtain target signature data；Classified using disaggregated model to the target signature data, obtain the result of sleep stage；Sleep stage can be accurately determined, realizes that simply accuracy rate is higher.

Description

Sleep stage determination method, related equipment and computer readable medium

technical field

The present invention relates to the field of electronic technology, in particular to a method for determining sleep stages, related equipment and a computer-readable medium.

Background technique

Sleep is an essential physiological activity of the human body. It is an important and complex physiological phenomenon, which occupies about one-third of the time in life. Sleep is the process of the body's self-repair and improvement, and plays an important regulatory role in maintaining physical and mental health. Sleep staging is to divide the sleep process into different stages according to the different changes of the human body's physiological signals during sleep. In human sleep, there are about 4 to 6 sleep cycles in one night, which are interconnected and repeated, and each sleep stage has its own specific physiological and behavioral characteristics. According to the different characteristics of the EEG, sleep is mainly divided into non-rapid eye movement (Non-rapid eye movement, NREM) and rapid eye movement (Rapid eye movement, REM), and the NREM period is divided into two periods, light sleep and deep sleep. Light sleep is characterized by shallow breathing, body muscles remain relaxed, and no significant eye movements. Deep sleep is characterized by deep, even and regular breathing without noticeable eye movements. The REM period is characterized by slightly faster and irregular breathing, rapid eye movement, and elevated blood pressure, body temperature, and heart rate.

Sleep staging is a very important part of the whole sleep, because sleep staging is usually related to sleep structure, sleep quality and sleep disorders, so sleep staging can be said to be the basis of the entire field of night medicine. An accurate sleep staging result can help us understand the user's health and stress, and the feedback of the result can also help professionals formulate intervention measures to improve sleep.

The currently used technology is usually based on EEG data to complete the determination of sleep staging, and EEG data is difficult to obtain. Therefore, how to use data and technologies other than EEG data to accurately determine sleep staging has become an urgent matter to be solved.

Contents of the invention

An embodiment of the present invention provides a method for determining sleep stages, which can accurately determine sleep stages, is simple to implement, and has high accuracy.

In a first aspect, an embodiment of the present invention provides a method for determining sleep stages, the method comprising:

Obtain the ECG signal of the target user;

extracting the first feature data corresponding to the ECG signal;

performing data fusion on the first feature data and second feature data to obtain target feature data, where the second feature data is feature data related to sleep stages corresponding to the target user;

The target characteristic data is classified by using a classification model to obtain the result of sleep staging.

In a second aspect, an embodiment of the present invention provides a sleep staging device, and the sleep staging device includes a module for performing the method in the first aspect above.

In the third aspect, the embodiment of the present invention provides another sleep staging device, including a processor, an input device, an output device and a memory, the processor, the input device, the output device and the memory are connected to each other, wherein the memory uses A computer program is stored to support the sleep staging device to execute the above method, the computer program includes program instructions, and the processor is configured to invoke the program instructions to execute the above method in the first aspect.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, the computer storage medium stores a computer program, the computer program includes program instructions, and when the program instructions are executed by a processor, the processing The device executes the method of the first aspect above.

In a fifth aspect, an embodiment of the present invention provides a machine learning-based sleep staging system, which includes the sleep staging device of the second aspect above.

In the embodiment of the present invention, the ECG signal of the target user is obtained, and the first feature data and the second feature data extracted from the ECG signal are fused together to obtain the target feature data; the target feature data is processed using a classification model Classify to get the result of sleep staging; the sleep staging can be accurately determined, which is simple to implement and has a high accuracy rate.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. Ordinary technicians can also obtain other drawings based on these drawings on the premise of not paying creative work.

FIG. 1 is a schematic flowchart of a method for determining sleep stages provided by an embodiment of the present invention;

Fig. 2 is a schematic flowchart of a method for extracting feature data corresponding to ECG signals provided by an embodiment of the present invention;

Fig. 3 is a schematic flowchart of a data fusion method provided by an embodiment of the present invention;

Fig. 4 is a schematic flowchart of processing target characteristic data provided by an embodiment of the present invention;

Fig. 5 is a schematic flowchart of determining sleep stages by using target characteristic data provided by an embodiment of the present invention;

Fig. 6 is a schematic block diagram of a sleep staging device provided by an embodiment of the present invention;

Fig. 7 is a schematic block diagram of a sleep staging device provided by another embodiment of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

It should be understood that when used in this specification and the appended claims, the terms "comprising" and "comprises" indicate the presence of described features, integers, steps, operations, elements and/or components, but do not exclude one or Presence or addition of multiple other features, integers, steps, operations, elements, components and/or collections thereof.

It should also be understood that the terminology used in the description of the present invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the present invention. As used in this specification and the appended claims, the singular forms "a", "an" and "the" are intended to include plural referents unless the context clearly dictates otherwise.

It should also be further understood that the term "and/or" used in the description of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations .

As used in this specification and the appended claims, the term "if" may be construed as "when" or "once" or "in response to determining" or "in response to detecting" depending on the context . Similarly, the phrase "if determined" or "if [the described condition or event] is detected" may be construed, depending on the context, to mean "once determined" or "in response to the determination" or "once detected [the described condition or event] ]” or “in response to detection of [described condition or event]”.

Referring to FIG. 1 , it is a schematic flowchart of a method for determining sleep stages provided by an embodiment of the present invention. As shown in FIG. 1 , the method may include:

101. Obtain the ECG signal of the target user;

The sleep staging device acquires the ECG signal of the target user. The sleep staging device may detect the electrocardiographic signal of the target user to obtain the electrocardiographic signal; it may also obtain the electrocardiographic signal of the target user from the electrocardiographic signal detection device. For example, the sleep staging device can be a computer, a mobile phone, a tablet computer, a server, etc., and can obtain the user's ECG signal from the ECG signal detection device. The ECG signal detection device is a device that can detect the user's ECG signal, such as a mattress, a sleep belt, etc. that can detect the user's ECG signal. The above-mentioned electrocardiographic signal detecting device may send the collected electrocardiographic signal to the above-mentioned sleep staging device. For another example, the sleep staging device can be a smart bracelet, a smart watch, etc., which can detect the user's ECG signal. The acquisition of the target user's ECG signal may be obtained by detecting the target user's ECG signal to obtain the target user's ballistocardiogram (Ballisto Cardio Gram, BCG) and/or body surface electrocardiogram (ElectroCardio Gram, ECG); The electrical signal detection device acquires the EGG and/or BGG of the above-mentioned user.

In the process of collecting, amplifying and transmitting ECG signals, due to the influence of the human body, detection instruments, electromagnetic environment, etc., it is inevitable that there will be interference coupled to the ECG signals. In practice, there are mainly the impacts of power frequency interference and baseline drift on ECG signals. In the embodiment of the present invention, the above-mentioned sleep staging device or the ECG signal detection device can use an adaptive digital notch filter, an averaging filter, etc. to remove the power frequency interference in the above-mentioned ECG signal, and then use median filtering, wavelet transform, etc. method to remove baseline drift. In the embodiment of the present invention, the sleep staging device or the electrocardiographic signal detection device may also process the electrocardiographic signal in other ways, so as to effectively suppress various interferences, so as to obtain accurate electrocardiographic signals. The embodiment of the present invention provides a specific example of preprocessing the acquired ECG signal:

1) Obtain the ECG signal;

2) remove the power frequency interference in the above-mentioned ECG signal;

3) Use the median filter method to remove baseline drift;

4) Elimination of abnormalities;

The ECG signal consists of P waves, QRS complexes and T waves. The crest of the QRS wave is the R crest. In the embodiment of the present invention, the sleep staging device can first shield all sampling points less than a certain voltage, so as to remove the false peaks in the P wave and T wave in advance, and then search for the peak value, so as to ensure that the obtained peak value is in the QRS In the wave, this peak is the peak of the R wave. The aforementioned elimination of abnormal beats refers to checking and removing abnormal points in the R wave peak.

The signal that interferes with the ECG signal can be eliminated or suppressed through the preprocessing operation, so as to ensure the accuracy of the ECG signal.

102. Extracting the first feature data corresponding to the electrocardiographic signal;

In an optional implementation manner, the above-mentioned extraction of the first characteristic data corresponding to the above-mentioned ECG signal includes:

Determining at least one RR interval in the above-mentioned ECG signal;

Time-domain analysis and frequency-domain analysis are performed on the at least one RR interval to obtain the first feature data.

The aforementioned determination of at least one RR interval in the aforementioned electrocardiographic signal may be determining the distance between two adjacent QRS peaks in the aforementioned electrocardiographic signal. In the embodiment of the present invention, a value of the RR interval can be obtained by subtracting the sampling time of the previous QRS peak from the sampling time of the latter QRS peak. Assuming that the ECG signal contains 10 peaks, 9 RR intervals can be determined.

Heart Rate Variability (HRV) refers to the small fluctuations between successive sinus heartbeats. There are five commonly used HRV time-domain indicators: NNVGR, SDNN, RMSSD, SDSD and pNN50. The definitions of the above five indicators are as follows:

NNVGR: the average value of all normal sinus rhythm intervals (NN intervals), in ms;

SDNN: standard deviation, that is, the standard deviation of all normal RR intervals in 24h, in ms;

RMSSD: the root mean square value of the difference between adjacent NN intervals in the whole process, in ms;

SDNNID: the standard deviation of the RR interval every 5 minutes within 24 hours, in ms;

PNN50: In the records of all NN intervals, the ratio of the number of adjacent NN intervals whose difference is greater than 50ms to the total number of NN intervals, expressed as a percentage.

Among them, NNVGR is used to evaluate the overall change level of heart rate; SDNN is used to evaluate the size of the overall change of heart rate, that is, the size of sympathetic and vagal tone; SDNNID is used to describe the size of heart rate variation within 5 minutes. RMSSD and PNN50 reflect the magnitude of rapid changes in heart rate, which are sensitive indicators of parasympathetic tension. The HRV time-domain detection index can also include the parameters obtained by dividing the standard deviation of the continuous normal RR interval per minute by the average normal RR interval of this period, and the number of adjacent NN intervals with a difference greater than 50 ms in all normal sinus intervals , average heart rate, and heart rate standard deviation. The average heart rate describes the average heart rate in 5 minutes, and the standard deviation of heart rate describes the fluctuation of heart rate in 5 minutes. The HRV time-domain detection index may also include other parameters, which are not limited in this embodiment of the present invention.

Studies have found that the heart rate spectrum curve of normal people is between 0-0.4Hz in the basic state, 0.003-0.04Hz is very low frequency (VLF), 0.04-0.15Hz is low frequency (LF), and 0.15-0.4Hz is high frequency (HF). ), 0-0.40Hz is the total power spectrum (TP). HRV frequency domain indicators can include VLF, LF, HF, TP, LF/HF, etc.

The time-domain analysis and frequency-domain analysis of the above-mentioned at least one RR interval to obtain the above-mentioned first characteristic data may be to perform time-domain analysis and frequency-domain analysis on the above-mentioned at least one RR interval to obtain the above-mentioned frequency domain index and the above-mentioned time domain Some or all of the indicators. It can be understood that the above-mentioned first feature data is the HRV time-domain index and HRV frequency-domain index obtained according to the above-mentioned RR interval or the above-mentioned heartbeat signal.

The embodiment of the present invention provides a specific example of extracting the characteristic data corresponding to the ECG signal, as shown in Figure 2, the method may include:

201. The sleep staging device acquires ECG signals;

202. Determine at least one RR interval of the electrocardiographic signal;

203. Calculate the HRV time domain index and the HRV frequency domain index according to the above at least one RR interval and the above electrocardiographic signal;

The above-mentioned sleep staging device can perform time-domain analysis on the above-mentioned at least one RR interval to obtain the above-mentioned HRV time-domain index; it can perform frequency-domain analysis on the above-mentioned ECG signal to obtain the above-mentioned HRV frequency-domain index.

204. Combine the above-mentioned HRV time-domain index and the above-mentioned HRV frequency-domain index to obtain the first characteristic data corresponding to the above-mentioned ECG signal.

In the embodiment of the present invention, the first feature data can be quickly and accurately extracted from the ECG signal.

103. Perform data fusion of the above-mentioned first feature data and second feature data to obtain target feature data, where the above-mentioned second feature data is feature data related to sleep stages corresponding to the above-mentioned target user;

The above-mentioned second feature data may be feature data extracted from respiratory signals and/or electromyographic signals; it may also be feature data related to sleep stages such as gender, age, and body movement data. The above-mentioned sleep staging device can obtain the above-mentioned respiratory signal and/or myoelectric signal, and extract the above-mentioned second feature data.

The embodiment of the present invention provides a specific example of data fusion of the first characteristic data and the second characteristic data. As shown in FIG. 3, the method may include:

301. Acquire first characteristic data and second characteristic data;

The above-mentioned first feature data is the HRV time-domain index and the HRV frequency-domain index determined according to the ECG signal. The above-mentioned second feature data is data related to sleep stages, which may be feature data extracted from respiratory signals and/or myoelectric signals; it may also be feature data related to sleep stages such as gender, age, and body movement data.

The N feature data included in the first feature data and the N feature data included in the second feature data correspond to the first time period to the Nth time period, and the above N is an integer greater than or equal to 1. The N pieces of feature data included in the first feature data may be feature data extracted from N segments of ECG signals corresponding to the first time period to the Nth time period. The N pieces of feature data included in the second feature data may include feature data extracted from N segments of electromyographic signals and/or respiratory signals corresponding to the first time period to the Nth time period. For example, a sleep staging device divides an ECG signal into N parts of equal duration in chronological order, that is, the first time period to the Nth time period, and each time period corresponds to an ECG signal; the sleep staging device extracts The ECG signal corresponding to each time period can obtain N characteristic data.

302. Fuse the first characteristic data and the characteristic data corresponding to the same time period in the second characteristic data to obtain target characteristic data.

In an optional implementation manner, the first characteristic data and the second characteristic data respectively include N characteristic data, and the N characteristic data included in the first characteristic data correspond to the first time period to the Nth For a time period, the N characteristic data included in the second characteristic data correspond to the first time period to the Nth time period; each characteristic data corresponds to a time period, and the N is an integer greater than or equal to 1;

The performing data fusion of the first feature data and the second feature data to obtain the target feature data includes:

Fusing the feature data corresponding to the same time period in the first feature data and the second feature data to obtain the target feature data, and N feature data in the target feature data correspond to the first time period to the Nth time period.

For example, the N pieces of feature data included in the first feature data are feature data corresponding to ECG signals from the first time period to the Nth time period, and each feature data includes a first frequency domain index corresponding to a time period, The second frequency domain index, the first time domain index, and the second time domain index; the N feature data contained in the second feature data are feature data from the first time period to the Nth time period, and each feature data contains a time segment corresponding to the third frequency domain index, the fourth time domain index, gender, age, and body movement; sequentially merge the first feature data and the second feature data in the same feature data corresponding to the time period to obtain N fields, each The first field is the first frequency domain index, the second frequency domain index, the first time domain index, the second time domain index, and the third frequency domain corresponding to the electrocardiographic signal in one time period from the first time period to the Nth time period Indicators, indicators in the fourth time domain, gender, age, and physical activity.

In the embodiment of the present invention, feature data corresponding to the same time period in the first feature data and the second feature data can be quickly combined into one field.

The embodiment of the present invention provides a specific example of processing target characteristic data. As shown in FIG. 4, the method may include:

401. The sleep staging device acquires target feature data;

402. Remove abnormal values and duplicate values of each feature data in the N feature data included in the target feature data;

403. Fill in missing values of each feature data in the above N feature data;

The method of filling missing values can use the method of linear regression.

404. Perform normalization processing on each feature data in the above N feature data;

The method of data normalization is mainly to make all the characteristic data fall in a certain interval after processing. The advantage of this is that the learning effect of the model will not be affected by excessive data fluctuations.

405. Perform feature encoding on each feature data in the aforementioned N feature data.

For example, the feature encoding of age and weight refers to distinguishing age and weight into different intervals, and simplifying the data by encoding several intervals, making it easier for the model to learn the rules.

The processed target feature data will be integrated and passed into the classification model for calculation and analysis.

The embodiment of the present invention can ensure the accuracy of the target feature data and improve the efficiency of sleep staging by processing the target feature data.

104. Use a classification model to classify the above-mentioned target feature data to obtain a result of sleep staging.

The above classification model contains M different classifiers. The above-mentioned sleep staging device can use the extreme gradient enhancement technology in the gradient enhancement decision tree in machine learning to classify the above-mentioned target feature data. The advantages of this technology are: fast operation speed, both interpretability and result optimization, and can automatically learn the characteristics of the data, so that sleep stages can be effectively predicted. In the embodiment of the present invention, other machine learning algorithms or deep learning algorithms may also be used to classify the above target feature data, and then determine the sleep staging results corresponding to each time period.

In an optional implementation, the above-mentioned classification model includes M different classifiers, and the above-mentioned M is an integer greater than 1; the above-mentioned classification model is used to classify the above-mentioned target feature data, and the sleep staging results obtained include:

Input the F-th feature data in the above-mentioned target feature data to the above-mentioned M different classifiers, and the above-mentioned M different classifiers all classify the above-mentioned F-th feature data to obtain M classification results; the above-mentioned F is less than or An integer equal to the above-mentioned N, the above-mentioned M classification results are the results of the sleep staging of the Fth time period determined by the above-mentioned M different classifiers, and each classifier obtains a classification result;

It is determined that the classification result with the largest number among the above M classification results is the result of the sleep stage in the Fth time period.

For example, the classification model in the sleep staging device includes M classifiers, and each classifier inputs the characteristic data corresponding to the Fth time period, and each classifier is based on its currently existing characteristic data and the Fth time period The corresponding feature data determines the result of the sleep stage corresponding to the Fth time period, and obtains M classification results. Assume that the classification model in the sleep staging device contains 50 classifiers, classify the feature data corresponding to the Fth time period, and get 50 classification results, of which, 2 classification results are the first sleep stage, and 3 classification results are the first sleep stage. Two sleep stages, 45 classification results are the third sleep stage; then the result of the sleep stage in the F time period is the third sleep stage.

In the embodiment of the present invention, the ECG signal of the target user is obtained, and the first feature data and the second feature data extracted from the ECG signal are fused together to obtain the target feature data; the target feature data is processed using a classification model Classify to get the result of sleep staging; the sleep staging can be accurately determined, which is simple to implement and has a high accuracy rate.

The embodiment of the present invention provides a specific example of using target feature data to determine sleep stages, as shown in Figure 5, the method may include:

501. Input target feature data into M different classifiers;

The above target characteristic data may include N characteristic data, and each characteristic data corresponds to a time period. A result of sleep staging in a time period corresponding to the characteristic data can be determined through a characteristic data. It can be understood that a classifier can determine the sleep staging results of N time periods through the above target feature data.

502. Each of the above M different classifiers analyzes the N feature data included in the above target feature data, and determines the sleep staging results of the N time periods corresponding to the above N feature data;

503. Determine the result of the sleep stage corresponding to the target feature data according to the results of the N sleep stages determined by each of the above M different classifiers.

Assume that the classification model in the sleep staging device contains 50 classifiers, classify the feature data corresponding to the Fth time period, and get 50 classification results, of which, 2 classification results are the first sleep stage, and 3 classification results are the first sleep stage. Two sleep stages, 45 classification results are the third sleep stage; then the result of the sleep stage in the F time period is the third sleep stage.

The embodiment of the present invention classifies the target feature data through a machine learning algorithm, and then obtains the result of sleep staging, the classification is more accurate, the classification speed is fast, and online learning and large-scale deployment are supported.

In an optional implementation manner, before acquiring the ECG signal of the target user, the above method further includes:

Receive time setting instructions;

The aforementioned acquisition of the ECG signal of the target user includes:

Obtain the electrocardiographic signal of the above-mentioned target user in the time period specified by the above-mentioned time setting instruction.

The receiving the time setting instruction may be receiving the time setting instruction through the input interface of the sleep staging device. The above time setting instruction may include the information of the start time and the end time. Assume that the start time corresponding to the time setting instruction received by the sleep staging device is 22:00, and the end time is 07:00; then the sleep staging device acquires the heartbeat signal of the target user from 22:00 to 07:00 the next day. Assume that the start time corresponding to the time setting instruction received by the sleep staging device is 22:00 and the end time is 22:05; then the sleep staging device acquires the target user's heartbeat signal from 22:00 to 22:05.

In the embodiment of the present invention, the sleep staging device obtains the heartbeat signal of the time period specified by the time setting instruction according to the received time setting instruction, which can meet the needs of different users and apply to different scenarios.

In an optional implementation manner, before the data fusion of the first feature data and the second feature data, the method further includes:

Acquiring the second feature data from user information preset by the sleep staging device;

Or, acquire the second feature data from a server.

The user information preset by the sleep staging device may include various information such as the user's name, age, medical history, health status, and weight. The user information preset by the sleep staging device includes the second feature data. When the sleep staging device determines the sleep stage according to the acquired ECG signal, it can extract the feature data related to the sleep stage corresponding to the target user from the user information preset by the sleep staging device, that is, the first Two feature data.

In practical applications, when a user uses a sleep staging device that does not store the second characteristic data to determine the sleep stage, the user needs to input the second characteristic data, and the operation is complicated and takes a long time. That is to say, each time the user uses a new sleep staging device to determine the sleep stage, he needs to input the second feature data once. In the embodiment of the present invention, the user can store the second characteristic data on the server, and each time a new sleep staging device is used to determine the sleep stage, the user can log in to the target account to obtain the second characteristic data of the target user. Furthermore, the server can also store the historical sleep staging results of the target user, which can be viewed by the user at any time.

In the embodiment of the present invention, the second characteristic data corresponding to the target user may be stored in the server, and the sleep staging device may quickly obtain the second characteristic data.

In an optional implementation manner, the above-mentioned target feature data is classified using the classification model, and after the sleep staging result is obtained, the above-mentioned method further includes:

Display and store the results of the sleep staging described above.

The sleep staging device may display the sleep staging results in real time; it may also store the sleep staging results, and display the sleep staging results after receiving a query instruction for the sleep staging results. Specifically, the results of the above sleep stages may be displayed graphically.

In the embodiment of the present invention, the result of sleep staging can be displayed in real time, and can also be viewed when needed, and the operation is simple.

An embodiment of the present invention also provides a sleep staging device, the sleep staging device is used to execute the modules of any one of the foregoing methods. Specifically, refer to FIG. 6 , which is a schematic block diagram of a sleep staging device provided by an embodiment of the present invention. The sleep staging device of the present embodiment includes:

A signal processing module 601, configured to acquire the ECG signal of the target user;

A feature extraction module 602, configured to extract the first feature data corresponding to the ECG signal;

The data fusion module 603 is configured to perform data fusion of the above-mentioned first feature data and second feature data to obtain target feature data, and the above-mentioned second feature data is feature data related to sleep stages corresponding to the above-mentioned target user;

The prediction module 604 is configured to use a classification model to classify the above-mentioned target feature data to obtain a result of sleep staging.

The specific implementation method is the same as the method in FIG. 1 , and will not be described in detail here.

In an optional implementation manner, the above-mentioned feature extraction module 602 is specifically configured to determine at least one RR interval in the above-mentioned ECG signal; perform time-domain analysis and frequency-domain analysis on the above-mentioned at least one RR interval to obtain the above-mentioned first A characteristic data.

In the embodiment of the present invention, the first feature data can be quickly and accurately extracted from the ECG signal.

In an optional implementation, the N feature data included in the first feature data and the N feature data included in the second feature data correspond to the first time period to the Nth time period, and the above N is greater than or an integer equal to 1;

The above-mentioned signal processing module 601 is further configured to obtain the above-mentioned second feature data;

The above-mentioned data fusion module 603 is specifically used to fuse the above-mentioned first feature data and the above-mentioned second feature data corresponding to the feature data of the same time period to obtain the above-mentioned target feature data, and the N feature data in the above-mentioned target feature data correspond to the above-mentioned The first time period to the above-mentioned Nth time period.

In the embodiment of the present invention, feature data corresponding to the same time period in the first feature data and the second feature data can be quickly combined into one field.

In an optional implementation, the classification model includes M different classifiers, where M is an integer greater than 1;

The prediction module 604 is specifically used to input the F-th feature data in the above-mentioned target feature data to the above-mentioned M different classifiers, and the above-mentioned M different classifiers classify the above-mentioned F-th feature data to obtain M Classification results; the above-mentioned F is an integer less than or equal to the above-mentioned N, and the above-mentioned M classification results are the results of the sleep staging of the Fth time period determined by the above-mentioned M different classifiers, and each classifier obtains a classification result; determine the above-mentioned The classification result with the largest number among the M classification results is the result of sleep staging in the Fth time period.

In the embodiment of the present invention, the sleep stage can be accurately determined, which is simple to implement and has a high accuracy rate.

In an optional implementation manner, the above-mentioned sleep staging device also includes:

A receiving module 605, configured to receive a time setting instruction;

The above-mentioned signal processing module 601 is specifically configured to obtain the electrocardiographic signal of the above-mentioned target user in the time period specified by the above-mentioned time setting instruction.

In the embodiment of the present invention, the sleep staging device obtains the heartbeat signal of the time period specified by the time setting instruction according to the received time setting instruction, which can meet the needs of different users and apply to different scenarios.

An embodiment of the present invention provides a sleep staging system based on machine learning, and the system includes the sleep staging device in the above embodiments.

For example, the sleep staging system based on machine learning includes a server and an ECG signal detection device; the server is a sleep staging device, which obtains the user's ECG signal from the ECG signal detection device and performs sleep staging; the ECG signal The detection device is a device that detects the user's ECG signal, such as a mattress, a sleep belt, etc. that can detect the user's ECG signal. The ECG signal detection device can send the collected ECG signal to the sleep staging device. For another example, the sleep staging system based on machine learning includes a mobile phone and an electrocardiographic signal detection device; the mobile phone is a sleep staging device, and the user's electrocardiogram is obtained from the electrocardiographic signal detection device for sleep staging; The signal detection device can be a smart bracelet, a smart watch, etc., which are used to detect the user's ECG signal and send it to the mobile phone.

Referring to FIG. 7 , it is a schematic block diagram of a sleep staging device provided by another embodiment of the present invention. The sleep staging device in this embodiment as shown in the figure may include: one or more processors 701 ; one or more input devices 702 , one or more output devices 703 and a memory 704 . The aforementioned processor 701 , input device 702 , output device 703 and memory 704 are connected through a bus 705 . The memory 702 is used to store computer programs, and the above computer programs include program instructions, and the processor 701 is used to execute the program instructions stored in the memory 702 . Wherein, the processor 701 is configured to invoke the above-mentioned program instructions to perform: acquiring the ECG signal of the target user; extracting the first feature data corresponding to the above-mentioned ECG signal; performing data fusion of the above-mentioned first feature data and the second feature data, Obtain target feature data, the second feature data is feature data related to sleep stages corresponding to the target user; use a classification model to classify the target feature data, and obtain a sleep stage result.

It should be understood that in the embodiment of the present invention, the so-called processor 701 may be a central processing module (Central Processing Unit, CPU), and the processor may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), dedicated Integrated Circuit (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.

The input device 702 may include a touch panel, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint direction information), a microphone, etc., and the output device 703 may include a display (LCD, etc.), a speaker, and the like.

The memory 704 may include read-only memory and random-access memory, and provides instructions and data to the processor 701 . A portion of memory 704 may also include non-volatile random access memory. For example, memory 704 may also store device type information.

In specific implementation, the processor 701, input device 702, and output device 703 described in the embodiment of the present invention can execute the implementation described in the foregoing embodiments of the present invention, and can also execute the sleep staging device described in the embodiment of the present invention The implementation method will not be repeated here.

In another embodiment of the present invention, a computer-readable storage medium is provided. The above-mentioned computer-readable storage medium stores a computer program, and the above-mentioned computer program includes program instructions. ECG signal; extract the first feature data corresponding to the above-mentioned ECG signal; perform data fusion of the above-mentioned first feature data and second feature data to obtain target feature data, and the above-mentioned second feature data is the sleep stage corresponding to the above-mentioned target user Relevant feature data; using a classification model to classify the above target feature data to obtain sleep staging results.

The above-mentioned computer-readable storage medium may be an internal storage module of the sleep staging device in any of the foregoing embodiments, such as a hard disk or a memory of the sleep staging device. The above-mentioned computer-readable storage medium may also be an external storage device of the above-mentioned sleep staging device, such as a plug-in hard disk equipped on the above-mentioned sleep staging device, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card , Flash Card (Flash Card) and so on. Further, the above-mentioned computer-readable storage medium may also include both an internal storage module of the above-mentioned sleep staging device and an external storage device. The above-mentioned computer-readable storage medium is used to store the above-mentioned computer program and other programs and data required by the above-mentioned sleep staging device. The computer-readable storage medium described above can also be used to temporarily store data that has been output or will be output.

Those of ordinary skill in the art can realize that the modules and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the relationship between hardware and software Interchangeability. In the above description, the composition and steps of each example have been generally described according to their functions. Whether these functions are executed by hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be regarded as exceeding the scope of the present invention.

Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process of the above-described system, sleep staging device and module can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, sleep staging device and method can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices or modules, and may also be electrical, mechanical or other forms of connection.

The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or may be distributed to multiple network modules. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solutions of the embodiments of the present invention.

In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, each module may exist separately physically, or two or more modules may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules.

If the integrated modules are realized in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of software products, and the computer software products are stored in a storage medium In, several instructions are included to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, and other media that can store program codes.

The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person familiar with the technical field can easily think of various equivalents within the technical scope disclosed in the present invention. Modifications or replacements shall all fall within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims (13)

1. A method for determining sleep stages, comprising: Obtain the ECG signal of the target user; extracting the first feature data corresponding to the ECG signal; performing data fusion on the first feature data and second feature data to obtain target feature data, where the second feature data is feature data related to sleep stages corresponding to the target user; The target characteristic data is classified by using a classification model to obtain the result of sleep staging. 2. The method according to claim 1, wherein said extracting the first characteristic data corresponding to said ECG signal comprises: determining at least one RR interval in the electrocardiographic signal; Perform time-domain analysis and frequency-domain analysis on the at least one RR interval to obtain the first characteristic data. 3. The method according to claim 2, wherein the first feature data and the second feature data respectively include N feature data, and the N feature data included in the first feature data correspond to the first time period to the Nth time period, the N feature data contained in the second feature data correspond to the first time period to the Nth time period; each feature data corresponds to a time period, and the N is greater than or an integer equal to 1; The performing data fusion of the first feature data and the second feature data to obtain the target feature data includes: Fusing the feature data corresponding to the same time period in the first feature data and the second feature data to obtain the target feature data, and N feature data in the target feature data correspond to the first time period to the Nth time period. 4. The method according to claim 3, wherein the classification model includes M different classifiers, and the M is an integer greater than 1; the use classification model classifies the target feature data, Results obtained for sleep staging include: Inputting the F-th feature data in the target feature data to the M different classifiers, and the M different classifiers all classify the F-th feature data to obtain M classification results; The F is an integer less than or equal to the N, and the M classification results are the results of the sleep staging of the Fth time period determined by the M different classifiers, and each classifier obtains a classification result; Determining that the classification result with the largest number among the M classification results is the result of the sleep staging in the Fth time period. 5. The method according to any one of claims 1 to 4, characterized in that, before acquiring the ECG signal of the target user, the method further comprises: Receive time setting instructions; The acquisition of the ECG signal of the target user includes: Acquiring the ECG signal of the target user in the time period specified by the time setting instruction. 6. The method according to claim 5, wherein, before performing data fusion on the first feature data and the second feature data, the method further comprises: Acquiring the second feature data from user information preset by the sleep staging device; Or, acquire the second feature data from a server. 7. The method according to claim 5, wherein said use classification model is used to classify said target characteristic data, and after obtaining the result of sleep staging, said method further comprises: The results of the sleep staging are displayed and stored. 8. A sleep staging device, characterized in that, comprising: The signal processing module is used to obtain the ECG signal of the target user; A feature extraction module, configured to extract first feature data corresponding to the ECG signal; A data fusion module, configured to perform data fusion of the first feature data and the second feature data to obtain target feature data, where the second feature data is feature data related to sleep stages corresponding to the target user; The prediction module is configured to use a classification model to classify the target feature data to obtain a sleep staging result. 9. The sleep staging device according to claim 8, wherein the first characteristic data and the second characteristic data respectively comprise N characteristic data, and the N characteristic data contained in the first characteristic data correspond to From the first time period to the Nth time period, the N characteristic data contained in the second characteristic data correspond to the first time period to the Nth time period; each characteristic data corresponds to a time period, and the N is an integer greater than or equal to 1; The data fusion module is specifically used to fuse the feature data corresponding to the same time period in the first feature data and the second feature data to obtain the target feature data, and N pieces of the target feature data The feature data corresponds to the first time period to the Nth time period. 10. The sleep staging device according to claim 9, wherein the classification model comprises M different classifiers, and the M is an integer greater than 1; The prediction module is specifically configured to input the F-th feature data in the target feature data to the M different classifiers, and the M different classifiers classify the F-th feature data , to obtain M classification results; the F is an integer less than or equal to the N, the M classification results are the results of the sleep stages of the Fth time period determined by the M different classifiers, each classification The device obtains a classification result; it is determined that the classification result with the largest number among the M classification results is the result of the sleep stage of the Fth time period. 11. A sleep staging device, characterized by comprising a module for executing the method according to any one of claims 1-7. 12. A sleep staging device, characterized in that it comprises a processor, an input device, an output device and a memory, and the processor, the input device, an output device and a memory are connected to each other, wherein the memory is used to store a computer program, The computer program includes program instructions, and the processor is configured to call the program instructions to execute the method according to any one of claims 1-7. 13. A computer-readable storage medium, wherein the computer storage medium stores a computer program, the computer program includes program instructions, and when executed by a processor, the program instructions cause the processor to execute The method described in any one of claims 1-7.