CN113812965A - Sleep state recognition method, device, electronic device and storage medium - Google Patents
Sleep state recognition method, device, electronic device and storage medium Download PDFInfo
- Publication number
- CN113812965A CN113812965A CN202110953665.4A CN202110953665A CN113812965A CN 113812965 A CN113812965 A CN 113812965A CN 202110953665 A CN202110953665 A CN 202110953665A CN 113812965 A CN113812965 A CN 113812965A
- Authority
- CN
- China
- Prior art keywords
- sleep state
- electroencephalogram
- signal
- neural network
- sleep
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000003062 neural network model Methods 0.000 claims abstract description 72
- 238000012549 training Methods 0.000 claims description 47
- 238000004590 computer program Methods 0.000 claims description 14
- 238000001914 filtration Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 10
- 238000013528 artificial neural network Methods 0.000 abstract description 8
- 210000004556 brain Anatomy 0.000 description 12
- 238000010586 diagram Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 6
- 238000000605 extraction Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 230000008667 sleep stage Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 210000003205 muscle Anatomy 0.000 description 2
- 230000003860 sleep quality Effects 0.000 description 2
- 241001669679 Eleotris Species 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 210000001061 forehead Anatomy 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/389—Electromyography [EMG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/398—Electrooculography [EOG], e.g. detecting nystagmus; Electroretinography [ERG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Surgery (AREA)
- Public Health (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Psychiatry (AREA)
- Psychology (AREA)
- Ophthalmology & Optometry (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Signal Processing (AREA)
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The application relates to a sleep state identification method, wherein the sleep state identification method comprises the following steps: acquiring an electroencephalogram signal of an acquired person; acquiring signal quality information based on the electroencephalogram signals; and inputting the electroencephalogram signal and the signal quality information into a trained first neural network model to obtain the sleep state of the acquired person. By the method and the device, the problem that the sleep state output by the neural network in real time is unstable and has high error rate is solved, the stable output of the sleep state is ensured without additionally adding other sleep state judgment equipment, and the output error rate of the sleep state is reduced.
Description
Technical Field
The present application relates to the field of electroencephalogram identification technologies, and in particular, to a sleep state identification method, an apparatus, an electronic apparatus, and a storage medium.
Background
With the development of science and technology, the EEG (Electroencephalogram) signal recognition technology is continuously explored, experienced medical personnel can effectively judge the sleep stage of a sleeper according to the characteristics of EEG signals in different sleep states and stages, targeted treatment is performed according to different sleep conditions, besides manual judgment by the medical personnel, some related researches try to input the EEG signals into a neural network by using a deep learning method, and deep analysis is performed on the EEG signals by using the neural network, so that sleep stage states similar to the sleep stage states in medicine are obtained.
However, in an actual situation, during the process of acquiring the EEG signal by the EEG signal monitoring device, the EEG signal is easily interfered by eye electricity, muscle electricity, and the like, and the monitored EEG signal sometimes has the situations of unstable signal and poor signal quality, which may cause the sleep state output by the neural network in real time to be unstable, and the error rate to be high, and particularly at the junction of two sleep states, the phenomenon of unstable output state is more serious, which causes a great influence in some scenes that utilize the sleep state to control the sleep environment.
Aiming at the problems of unstable sleep state and high error rate of real-time output of a neural network in the related technology, no effective solution is provided at present.
Disclosure of Invention
The embodiment provides a sleep state identification method, a sleep state identification device, an electronic device and a storage medium, so as to solve the problems that the sleep state output by a neural network in real time is unstable and the error rate is high in the related art.
In a first aspect, a sleep state identification method is provided in this embodiment, including: acquiring an electroencephalogram signal of an acquired person; acquiring signal quality information based on the electroencephalogram signals; and inputting the electroencephalogram signal and the signal quality information into a trained first neural network model to obtain the sleep state of the acquired person.
In one embodiment, before inputting the electroencephalogram signal and the signal quality information into the trained first neural network model, the method further comprises: and carrying out band-pass filtering processing on the electroencephalogram signals to obtain the electroencephalogram signals within a preset frequency range.
In another embodiment, the inputting the brain electrical signal and the signal quality information into the trained first neural network model comprises: extracting time domain characteristics and frequency domain characteristics of the electroencephalogram signals, and inputting the time domain characteristics, the frequency domain characteristics and the signal quality information into a trained first neural network model.
In one embodiment, the extracting the time-domain feature and the frequency-domain feature of the electroencephalogram signal includes: inputting the electroencephalogram signal into a trained second neural network model to obtain the time domain characteristics; and carrying out fast Fourier transform on the electroencephalogram signals to obtain the frequency domain characteristics.
In another embodiment, said obtaining signal quality information based on said brain electrical signal comprises: and acquiring the signal quality information based on the amplitude information and the energy information of the electroencephalogram signal.
In one embodiment, before inputting the electroencephalogram signal and the signal quality information into the trained first neural network model, the method further comprises: acquiring first training data and an initial first neural network model, wherein the first training data comprises a training electroencephalogram signal, training signal quality information corresponding to the training electroencephalogram signal and a sleep state corresponding to the training electroencephalogram signal; training the initial neural network model based on the first training data, resulting in the trained first neural network model.
In another embodiment, said obtaining the sleep state of the subject further comprises: and matching and acquiring sleep music, sleep light and sleep temperature corresponding to the sleep state in a sleep environment adjustment database based on the sleep state, and applying the sleep music, the sleep light and the sleep temperature to the current environment.
In a second aspect, there is provided a sleep state identification apparatus in the present embodiment, including: an electroencephalogram signal acquisition module: the electroencephalogram acquisition system is used for acquiring an electroencephalogram signal of an acquired person; a signal quality information acquisition module: for obtaining the signal quality information based on the electroencephalogram signal; a sleep state identification module: and the first neural network model is used for inputting the electroencephalogram signal and the signal quality information into the trained first neural network model to obtain the sleep state of the acquired person.
In a third aspect, in this embodiment, there is provided an electronic apparatus, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the processor implements the sleep state identification method according to the first aspect.
In a fourth aspect, in the present embodiment, there is provided a storage medium having stored thereon a computer program which, when executed by a processor, implements the sleep state identification method of the first aspect described above.
Compared with the related art, the sleep state identification method provided by the embodiment acquires the electroencephalogram signal of the acquired person; acquiring signal quality information based on the electroencephalogram signals; the electroencephalogram signals and the signal quality information are input into the trained first neural network model to obtain the sleep state of the acquired person, the problems that the sleep state output by the neural network in real time is unstable and the error rate is high are solved, the stable output of the sleep state is ensured without additionally adding other sleep state judgment equipment, and the output error rate of the sleep state is reduced.
The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below to provide a more thorough understanding of the application.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a block diagram of a hardware structure of a terminal in a sleep state identification method according to an embodiment of the present application.
Fig. 2 is a flowchart illustrating a sleep state identification method according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a reference electroencephalogram acquisition site of electroencephalogram acquisition equipment in the sleep state identification method in the embodiment of the present application.
Fig. 4 is a schematic diagram of a posterior judgment strategy of a sleep state identification method in an embodiment of the present application.
Fig. 5 is a block diagram of a sleep state identification apparatus according to an embodiment of the present application.
Detailed Description
For a clearer understanding of the objects, aspects and advantages of the present application, reference is made to the following description and accompanying drawings.
Unless defined otherwise, technical or scientific terms used herein shall have the same general meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The use of the terms "a" and "an" and "the" and similar referents in the context of this application do not denote a limitation of quantity, either in the singular or the plural. The terms "comprises," "comprising," "has," "having," and any variations thereof, as referred to in this application, are intended to cover non-exclusive inclusions; for example, a process, method, and system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to the listed steps or modules, but may include other steps or modules (elements) not listed or inherent to such process, method, article, or apparatus. Reference throughout this application to "connected," "coupled," and the like is not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. Reference to "a plurality" in this application means two or more. "and/or" describes an association relationship of associated objects, meaning that three relationships may exist, for example, "A and/or B" may mean: a exists alone, A and B exist simultaneously, and B exists alone. In general, the character "/" indicates a relationship in which the objects associated before and after are an "or". The terms "first," "second," "third," and the like in this application are used for distinguishing between similar items and not necessarily for describing a particular sequential or chronological order.
The method embodiments provided in the present embodiment may be executed in a terminal, a computer, or a similar computing device. For example, the method is executed on a terminal, and fig. 1 is a block diagram of a hardware structure of the terminal according to the sleep state identification method in an embodiment of the present application. As shown in fig. 1, the terminal may include one or more processors 102 (only one shown in fig. 1) and a memory 104 for storing data, wherein the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA. The terminal may also include a transmission device 106 for communication functions and an input-output device 108. It will be understood by those of ordinary skill in the art that the structure shown in fig. 1 is merely an illustration and is not intended to limit the structure of the terminal described above. For example, the terminal may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.
The memory 104 may be used to store a computer program, for example, a software program and a module of application software, such as a computer program corresponding to the sleep state identification method in the present embodiment, and the processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, so as to implement the method described above. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the terminal over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device 106 is used to receive or transmit data via a network. The network described above includes a wireless network provided by a communication provider of the terminal. In one example, the transmission device 106 includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.
In this embodiment, a sleep state identification method is provided, and fig. 2 is a schematic flowchart of a sleep state identification method in an embodiment of the present application, and as shown in fig. 2, the flowchart includes the following steps:
step S201, acquiring electroencephalogram signals of the acquired person.
It can be understood that, in the present embodiment, the electroencephalogram signal of the acquired person is acquired by identifying the electroencephalogram signal of the acquired person, so that the electroencephalogram signal of the acquired person needs to be acquired first, in this embodiment, the method for acquiring the electroencephalogram signal of the acquired person may be to acquire the electroencephalogram signal of the acquired person through an electroencephalogram acquisition device, through an electrode in contact with a brain, a site for acquiring electroencephalograms through contacting the brain may be in any position, as shown in fig. 3, fig. 3 is a schematic diagram of a reference electroencephalogram acquisition site of the electroencephalogram acquisition device in the sleep state identification method in the embodiment of the present application. In this embodiment, the Fpz site is selected as a reference to collect a single-channel forehead EEG brain signal at the Fp1 site, it can be understood that in this embodiment, the Fp1 site is selected only for a certain brain electrical collecting device to be suitable for acquiring brain electrical signals of a collected person, in other embodiments, other collecting sites of brain electrical collecting devices of other forms can be selected to acquire brain electrical signals of a collected person, and it is only necessary to ensure that the brain electrical data of the collected person can be relatively better collected through the collecting sites.
Step S202, acquiring signal quality information based on the electroencephalogram signals.
In this embodiment, after acquiring the electroencephalogram of the person to be acquired, signal quality information of the electroencephalogram is also required to be acquired, that is, the signal quality of the electroencephalogram is determined, it can be understood that, when acquiring the electroencephalogram, not only electroencephalograms, but also electric waves such as eye waves, muscle waves and the like are acquired, and interference of some noises is also caused, and the presence of these non-electroencephalograms interferes with subsequent sleep state identification according to the identification of the sleep state, so that the identified sleep state changes too frequently, greatly or inaccurately, on the other hand, the signal quality level also reflects information such as some actions and action amplitudes appearing in the process of being acquired by the user, and these information also contribute to the determination and identification of the sleep state, and therefore, in this embodiment, the signal quality determination of the electroencephalogram is required to be performed on the acquired electroencephalogram, the method for judging the signal quality can be according to the occupation ratio of the electroencephalogram signal in all signals or other methods, and only the signal quality of the electroencephalogram signal can be judged according to the acquired electroencephalogram signal.
Step S203, the electroencephalogram signal and the signal quality information are input into the trained first neural network model, and the sleep state of the acquired person is obtained.
It can be understood that, after the electroencephalogram signal and the signal quality information of the electroencephalogram signal are acquired, the sleep state of the acquired person can be identified and judged based on the acquired electroencephalogram signal and the signal quality information of the electroencephalogram signal, in the embodiment, the sleep state of the person to be collected is judged through a trained neural network model, the trained neural network model can input electroencephalogram signals and signal quality information of the electroencephalogram signals, then the sleep state of the acquired person is obtained based on the electroencephalogram signal and the signal quality information of the electroencephalogram signal, which is easy to understand, through the electroencephalogram signal and the signal quality information, the neural network model can be more sensitive to the EEG signal with higher signal quality information when the sleep state is judged, so that the neural network model can judge and identify the sleep state more accurately, and the more accurate sleep state of the acquired person is obtained.
Through the steps, acquiring an electroencephalogram signal of the acquired person; acquiring signal quality information based on the electroencephalogram signals; the electroencephalogram signals and the signal quality information are input into the trained first neural network model to obtain the sleep state of the acquired person, the problems that the sleep state output by the neural network in real time is unstable and the error rate is high are solved, other sleep state judgment devices do not need to be additionally arranged, the stable output of the sleep state is guaranteed, and the output error rate of the sleep state is reduced.
In one embodiment, before inputting the electroencephalogram signal and the signal quality information into the trained first neural network model, the method further comprises: and performing band-pass filtering processing on the electroencephalogram signals to obtain the electroencephalogram signals within a preset frequency range.
It can be understood that some drift and power frequency noise still exist in the EEG signal obtained based on EEG collection equipment, in order to reduce the identification to unnecessary information and improve the identification precision, before the EEG signal is input to the neural network model and sleep state is judged, still need carry out filtering process to it, in this embodiment, use band-pass filtering module to obtain filtering process to the EEG signal, obtain the EEG signal within 2 ~ 45Hz scope, can filter most drift and power frequency noise, only keep the main signal data of EEG signal. By carrying out band-pass filtering processing on the electroencephalogram data, the data quality of the electroencephalogram signals can be improved, unnecessary and unnecessary interference is removed, and the accuracy of subsequent sleep state identification can be improved.
In another embodiment, inputting the brain electrical signal and the signal quality information into the trained first neural network model comprises: and extracting time domain characteristics and frequency domain characteristics of the electroencephalogram signals, and inputting the time domain characteristics, the frequency domain characteristics and signal quality information into the trained first neural network model.
In this embodiment, before the electroencephalogram signal and the signal quality information are input to the neural network model, it is further required to extract the time domain feature, the frequency domain feature, and the feature of the electroencephalogram signal, which can be understood that, in the electroencephalogram signal, the feature that can best reflect the current sleep state of the acquired person is the time domain feature and the frequency domain feature, and when the current sleep state of the acquired person is identified, if all the electroencephalogram signals are identified, the occupied computation amount and time are too long, based on this, in order to improve the efficiency of identifying the sleep state and reduce the computation load, the time domain feature and the frequency domain feature can be respectively extracted from the electroencephalogram signal, and then the time domain feature, the frequency domain feature, and the signal quality information are simultaneously input to the trained first neural network model, and it can be understood that the first neural network model can input the time domain feature, the frequency domain feature, and the time domain feature, The frequency domain characteristics and corresponding signal quality information are output, the corresponding sleep state is output, the time domain characteristics and the frequency domain characteristics are extracted from the signal quality information to judge the sleep state, and the efficiency of identifying the sleep state of the acquired person is improved.
In one embodiment, extracting the time domain feature and the frequency domain feature of the electroencephalogram signal comprises: inputting the electroencephalogram signal into a trained second neural network model to obtain time domain characteristics; and carrying out fast Fourier transform on the electroencephalogram signals to obtain frequency domain characteristics.
In this embodiment, the method for extracting the time domain feature may be inputting the electroencephalogram signal into a trained second neural network model, and then extracting the time domain signal of the electroencephalogram signal based on the trained second neural network model, it can be understood that the trained second neural network model may be a convolutional neural network model, may also be one of convolutional neural network layers in the neural network model, may also be any one of feature extractors that can extract the time domain feature of the electroencephalogram signal, only needs to ensure that the time domain feature corresponding to the electroencephalogram signal can be extracted based on the electroencephalogram signal, the method for extracting the frequency domain feature may be performing fast fourier transform on the electroencephalogram signal to extract the frequency domain feature of the electroencephalogram signal, in addition, in other embodiments, there may also be other methods for extracting the frequency domain feature, which is not specifically limited in this embodiment, the corresponding frequency domain characteristics can be obtained based on the electroencephalogram signals only by ensuring, time domain characteristic extraction without manual calculation is realized by selecting the connected second neural network model, the efficiency of time domain characteristic extraction can be improved, and the extraction cost can be reduced.
In another embodiment, obtaining signal quality information based on the brain electrical signal comprises: and judging the signal quality of the electroencephalogram signal based on the amplitude information and the energy information, and obtaining signal quality information.
It can be understood that the electroencephalogram signal quality information is obtained based on the amplitude information and the energy information of the electroencephalogram signal, in this embodiment, the amplitude information and the energy information of the electroencephalogram signal are obtained first, and then the signal quality is classified according to the amplitude information and the energy information, wherein the signal quality is classified into three grades of "good", "very poor" and "poor", the signal quality information obtained based on the amplitude information and the energy information can objectively reflect the signal quality of the electroencephalogram signal to be an effective basis for identifying the sleep state, in other embodiments, there may be a method for judging the signal quality based on other information in the electroencephalogram signal, for example, judging the signal quality only according to the energy information of the signal or only according to the amplitude information of the signal, or according to other signal characteristic information, there may be other classifications for the signal quality, for example, the signal level is divided into 1, 2, 3, 4, and 5 levels, and then the signal level is determined according to the characteristics of the signal, but this embodiment only exemplifies one of the levels, and does not specifically limit other modes.
In one embodiment, before inputting the electroencephalogram signal and the signal quality information into the trained first neural network model, the method further comprises: acquiring first training data and an initial first neural network model, wherein the first training data comprises a training electroencephalogram signal, training signal quality information corresponding to the training electroencephalogram signal and a sleep state corresponding to the training electroencephalogram signal; an initial neural network model is trained based on the first training data to obtain a trained first neural network model.
In this embodiment, before the electroencephalogram signal and the signal quality information are input into the trained first neural network model, the initial first neural network model needs to be trained to obtain the trained first neural network model capable of obtaining the sleep state based on the electroencephalogram signal and the signal quality information, therefore, the initial first neural network model needs to be obtained first, then the initial first neural network model needs to be trained based on first training data, the first training data comprises the training electroencephalogram signal, the training signal quality information corresponding to the training electroencephalogram signal, and the sleep state corresponding to the training electroencephalogram signal, and based on this, the initial first neural network model needs to be trained through the first training data until the first neural network model can obtain the accurate sleep state through the electroencephalogram signal and the signal quality information, and in other embodiments, the sleep state can be obtained through the time domain feature, the signal quality information, and the signal quality information of the electroencephalogram signal, The frequency domain characteristics and the signal quality information are used as first training data to obtain a first neural network model capable of judging the sleep state through the time domain characteristics, the frequency domain characteristics and the signal quality information, the initial first training model is trained through the first training data to obtain the trained first neural network model, and the accuracy of judging the sleep state is guaranteed.
In another embodiment, the method after acquiring the sleep state further comprises: and matching and acquiring sleep music, sleep light and sleep temperature corresponding to the sleep state in the sleep environment adjustment database based on the sleep state, and applying the sleep music, the sleep light and the sleep temperature to the current environment.
It is easy to understand that in different sleep states, the sleep quality of the user in different sleep states can be improved or the user can enter the next sleep state more quickly through different sleep music, sleep light and sleep temperature, therefore, in the method, after the sleep state is obtained, the corresponding sleep music, sleep light and sleep temperature suitable for the current sleep state of the user in the sleep environment adjustment database are matched based on the sleep state and applied to the current environment, wherein the sleep environment adjustment database should include various sleep music, sleep light and sleep temperature suitable for the current sleep state, and the corresponding sleep music, sleep light and sleep temperature are applied to the environment where the user is located based on the sleep state, so that the sleep quality of the user can be improved.
In addition, in some other embodiments, the sleep state directly determined by the neural network model has a certain error rate, and particularly, at the boundary between two sleep states, the output sleep state may have an unstable phenomenon, in order to improve the stability of the sleep state output and the accuracy of the analysis conclusion of the subsequent sleep state, in this embodiment, a final sleep state determination policy is provided, which is a posterior determination policy after the neural network model outputs the sleep state, as shown in fig. 4, fig. 4 is a posterior determination policy diagram of the sleep state identification method in the embodiment of the present application. The sleep state output by the neural network model is S0, the length of each small block is a trigger step length T of the algorithm, the broken line in the graph is a final sleep state output after posterior judgment based on the sleep state output by the neural network model, the broken line in the graph is S1, when measurement is started, S1 is an awake state, when the S1 switches the sleep state is judged according to the change of S0 in the process of acquiring and processing signals in real time, a sliding window is adopted, W small blocks (5 in the graph) are intercepted, when the S0 state corresponding to K (3 in the graph) small blocks in the W blocks in the sliding window is inconsistent with the current S1 state, S1 is made to be S0, in other embodiments, the values of K and W can be set by comprehensively considering the instability, the judgment accuracy and the actual allowable delay of the neural network model output without additional requirements, in this embodiment, after passing through the a posteriori decision strategy, the maximum delay of the decision states of the output final sleep states S1 and S0 is D ═ W × T, and in the method, the maximum delay D ═ 5 × 3 ═ 15 seconds, so that the accuracy of outputting the sleep states is greatly improved, the jitter of the sleep state output is reduced, and the real-time experience is optimized by a small amount of delay.
It should be noted that the steps illustrated in the above-described flow diagrams or in the flow diagrams of the figures may be performed in a computer system, such as a set of computer-executable instructions, and that, although a logical order is illustrated in the flow diagrams, in some cases, the steps illustrated or described may be performed in an order different than here.
In this embodiment, a sleep state identification apparatus is further provided, and the apparatus is used to implement the foregoing embodiments and preferred embodiments, and the details that have been already described are not repeated. The terms "module," "unit," "subunit," and the like as used below may implement a combination of software and/or hardware for a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
Fig. 5 is a block diagram of a sleep state identification apparatus according to an embodiment of the present application, and as shown in fig. 5, the apparatus includes: the system comprises an electroencephalogram signal acquisition module 10, a signal quality information acquisition module 20, a sleep state identification module 30, a filtering processing module, a feature extraction module, a first neural network model training module and a sleep environment adjustment module.
An electroencephalogram signal acquisition module 10: the electroencephalogram acquisition system is used for acquiring an electroencephalogram signal of an acquired person;
the signal quality information acquisition module 20: for obtaining signal quality information based on the electroencephalogram signal.
Sleep state identification module 30: the first neural network model is used for inputting the electroencephalogram signals and the signal quality information into the trained first neural network model to obtain the sleep state of the acquired person.
A filtering processing module: the method is used for carrying out band-pass filtering processing on the electroencephalogram signals to obtain the electroencephalogram signals within a preset frequency range.
A feature extraction module: the neural network model is used for extracting time domain characteristics and frequency domain characteristics of the electroencephalogram signals and inputting the time domain characteristics, the frequency domain characteristics and signal quality information into the trained first neural network model.
A feature extraction module: the first training data comprise training electroencephalogram signals, training signal quality information corresponding to the training electroencephalogram signals and sleep states corresponding to the training electroencephalogram signals; an initial neural network model is trained based on the first training data to obtain a trained first neural network model.
The first neural network model training module: the first training data comprise training electroencephalogram signals, training signal quality information corresponding to the training electroencephalogram signals and sleep states corresponding to the training electroencephalogram signals; an initial neural network model is trained based on the training data to obtain a trained first neural network model.
A sleep environment adjustment module: and the system is used for matching and acquiring sleep music, sleep light and sleep temperature corresponding to the sleep state in the sleep environment adjustment database based on the sleep state, and applying the sleep music, the sleep light and the sleep temperature to the current environment.
The above modules may be functional modules or program modules, and may be implemented by software or hardware. For a module implemented by hardware, the modules may be located in the same processor; or the modules can be respectively positioned in different processors in any combination.
There is also provided in this embodiment an electronic device comprising a memory having a computer program stored therein and a processor arranged to run the computer program to perform the steps of any of the above method embodiments.
Optionally, the electronic apparatus may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.
Optionally, in this embodiment, the processor may be configured to execute the following steps by a computer program:
and S1, acquiring the electroencephalogram signal of the acquired person.
And S2, acquiring signal quality information based on the electroencephalogram signals.
And S3, inputting the electroencephalogram signal and the signal quality information into the trained first neural network model to obtain the sleep state of the acquired person.
It should be noted that, for specific examples in this embodiment, reference may be made to the examples described in the foregoing embodiments and optional implementations, and details are not described again in this embodiment.
In addition, in combination with the sleep state identification method provided in the foregoing embodiment, a storage medium may also be provided to implement in this embodiment. The storage medium having stored thereon a computer program; the computer program, when executed by a processor, implements any of the sleep state recognition methods in the above embodiments.
It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to be limiting. All other embodiments, which can be derived by a person skilled in the art from the examples provided herein without any inventive step, shall fall within the scope of protection of the present application.
It is obvious that the drawings are only examples or embodiments of the present application, and it is obvious to those skilled in the art that the present application can be applied to other similar cases according to the drawings without creative efforts. Moreover, it should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another.
The term "embodiment" is used herein to mean that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is to be expressly or implicitly understood by one of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the patent protection. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.
Claims (10)
1. A sleep state recognition method, comprising:
acquiring an electroencephalogram signal of an acquired person;
acquiring signal quality information based on the electroencephalogram signals;
and inputting the electroencephalogram signal and the signal quality information into a trained first neural network model to obtain the sleep state of the acquired person.
2. The sleep state recognition method of claim 1, wherein before inputting the electroencephalogram signal and the signal quality information into the trained first neural network model, further comprising:
and carrying out band-pass filtering processing on the electroencephalogram signals to obtain the electroencephalogram signals within a preset frequency range.
3. The sleep state recognition method of claim 1 or 2, wherein the inputting the electroencephalogram signal and the signal quality information into the trained first neural network model comprises:
extracting time domain characteristics and frequency domain characteristics of the electroencephalogram signals, and inputting the time domain characteristics, the frequency domain characteristics and the signal quality information into a trained first neural network model.
4. The sleep state recognition method of claim 3, wherein the extracting the time domain feature and the frequency domain feature of the electroencephalogram signal comprises:
inputting the electroencephalogram signal into a trained second neural network model to obtain the time domain characteristics;
and carrying out fast Fourier transform on the electroencephalogram signals to obtain the frequency domain characteristics.
5. The sleep state recognition method of claim 1, wherein the obtaining signal quality information based on the electroencephalogram signal includes:
and acquiring the signal quality information based on the amplitude information and the energy information of the electroencephalogram signal.
6. The sleep state recognition method of claim 1, wherein before inputting the electroencephalogram signal and the signal quality information into the trained first neural network model, further comprising:
acquiring first training data and an initial first neural network model, wherein the first training data comprises a training electroencephalogram signal, training signal quality information corresponding to the training electroencephalogram signal and a sleep state corresponding to the training electroencephalogram signal;
training the initial neural network model based on the first training data, resulting in the trained first neural network model.
7. The sleep state identification method according to claim 1, wherein the obtaining the sleep state of the subject further comprises:
and matching and acquiring sleep music, sleep light and sleep temperature corresponding to the sleep state in a sleep environment adjustment database based on the sleep state, and applying the sleep music, the sleep light and the sleep temperature to the current environment.
8. A sleep state recognition apparatus, comprising:
an electroencephalogram signal acquisition module: the electroencephalogram acquisition system is used for acquiring an electroencephalogram signal of an acquired person;
a signal quality information acquisition module: for obtaining the signal quality information based on the electroencephalogram signal;
a sleep state identification module: and the first neural network model is used for inputting the electroencephalogram signal and the signal quality information into the trained first neural network model to obtain the sleep state of the acquired person.
9. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, and the processor is configured to execute the computer program to perform the sleep state identification method of any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the sleep state identification method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110953665.4A CN113812965B (en) | 2021-08-19 | 2021-08-19 | Sleep state identification method, sleep state identification device, electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110953665.4A CN113812965B (en) | 2021-08-19 | 2021-08-19 | Sleep state identification method, sleep state identification device, electronic device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113812965A true CN113812965A (en) | 2021-12-21 |
| CN113812965B CN113812965B (en) | 2024-04-09 |
Family
ID=78913291
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110953665.4A Active CN113812965B (en) | 2021-08-19 | 2021-08-19 | Sleep state identification method, sleep state identification device, electronic device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113812965B (en) |
Citations (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004097495A (en) * | 2002-09-09 | 2004-04-02 | Yamatake Corp | Sleeping state distinguishing device, and sleeping monitoring system |
| CN101923392A (en) * | 2010-09-02 | 2010-12-22 | 上海交通大学 | Asynchronous brain-computer interactive control method for EEG signal |
| US20110124979A1 (en) * | 2007-08-21 | 2011-05-26 | Conor Heneghan | Method and system for monitoring sleep |
| WO2014123512A1 (en) * | 2013-02-05 | 2014-08-14 | Vivaquant Llc | Detecting fiducial points in physiological signals |
| CN104463101A (en) * | 2014-11-06 | 2015-03-25 | 科大讯飞股份有限公司 | Answer recognition method and system for textual test question |
| CN105126187A (en) * | 2015-08-31 | 2015-12-09 | 杭州回车电子科技有限公司 | Method and system for assisting sleeping through brain waves |
| CN105476631A (en) * | 2015-11-25 | 2016-04-13 | 华南理工大学 | EEG (electroencephalogram) based sleep detection and sleep aid method and device |
| CN105615834A (en) * | 2015-12-22 | 2016-06-01 | 深圳创达云睿智能科技有限公司 | Sleep stage classification method and device based on sleep EEG (electroencephalogram) signals |
| CN106388818A (en) * | 2016-09-21 | 2017-02-15 | 广州视源电子科技股份有限公司 | A feature information extracting method and system for a sleep state monitoring model |
| CN106388778A (en) * | 2016-09-21 | 2017-02-15 | 广州视源电子科技股份有限公司 | A method and a system for electroencephalogram signal preprocessing in sleep state analysis |
| CN106419893A (en) * | 2016-09-18 | 2017-02-22 | 广州视源电子科技股份有限公司 | Method and device for detecting sleep state |
| CN106510641A (en) * | 2016-12-19 | 2017-03-22 | 姚健欣 | Method and system for detecting different sleep stages of human body |
| WO2017092431A1 (en) * | 2015-12-01 | 2017-06-08 | 乐视控股（北京）有限公司 | Human hand detection method and device based on skin colour |
| CN107998499A (en) * | 2017-11-28 | 2018-05-08 | 广州视源电子科技股份有限公司 | Processing method and system, the sleep secondary server system of sleep auxiliary content |
| CN109086700A (en) * | 2018-07-20 | 2018-12-25 | 杭州电子科技大学 | Radar range profile's target identification method based on depth convolutional neural networks |
| CN109602417A (en) * | 2018-11-23 | 2019-04-12 | 杭州妞诺科技有限公司 | Sleep stage method and system based on random forest |
| CN109674468A (en) * | 2019-01-30 | 2019-04-26 | 大连理工大学 | It is a kind of singly to lead brain electricity sleep mode automatically method by stages |
| WO2019158704A1 (en) * | 2018-02-15 | 2019-08-22 | Biosency | Monitoring device for monitoring a physiological parameter and methods thereof |
| CN110210961A (en) * | 2019-05-21 | 2019-09-06 | 平安普惠企业管理有限公司 | Data capture method, server and computer storage medium based on alliance's chain |
| CN110327040A (en) * | 2019-04-24 | 2019-10-15 | 武汉理工大学 | Sleep stage method and system based on cloud platform |
| CN110456846A (en) * | 2019-08-19 | 2019-11-15 | 浙江大学 | A kind of adaptive more sense organ sleeping-assisting systems based on artificial intelligence |
| US10559299B1 (en) * | 2018-12-10 | 2020-02-11 | Apprente Llc | Reconciliation between simulator and speech recognition output using sequence-to-sequence mapping |
| CN110897639A (en) * | 2020-01-02 | 2020-03-24 | 清华大学深圳国际研究生院 | Electroencephalogram sleep staging method based on deep convolutional neural network |
| US20200093425A1 (en) * | 2018-09-21 | 2020-03-26 | Tata Consultancy Services Limited | System and method for non-apnea sleep arousal detection |
| CN111012341A (en) * | 2020-01-08 | 2020-04-17 | 东南大学 | Artifact removal and electroencephalogram signal quality evaluation method based on wearable electroencephalogram equipment |
| CN111202517A (en) * | 2020-02-03 | 2020-05-29 | 山东师范大学 | Sleep automatic staging method, system, medium and electronic equipment |
| CN113180691A (en) * | 2020-12-28 | 2021-07-30 | 天津大学 | Three-channel sleep apnea and hypopnea syndrome recognition device |
-
2021
- 2021-08-19 CN CN202110953665.4A patent/CN113812965B/en active Active
Patent Citations (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004097495A (en) * | 2002-09-09 | 2004-04-02 | Yamatake Corp | Sleeping state distinguishing device, and sleeping monitoring system |
| US20110124979A1 (en) * | 2007-08-21 | 2011-05-26 | Conor Heneghan | Method and system for monitoring sleep |
| CN101923392A (en) * | 2010-09-02 | 2010-12-22 | 上海交通大学 | Asynchronous brain-computer interactive control method for EEG signal |
| WO2014123512A1 (en) * | 2013-02-05 | 2014-08-14 | Vivaquant Llc | Detecting fiducial points in physiological signals |
| CN104463101A (en) * | 2014-11-06 | 2015-03-25 | 科大讯飞股份有限公司 | Answer recognition method and system for textual test question |
| CN105126187A (en) * | 2015-08-31 | 2015-12-09 | 杭州回车电子科技有限公司 | Method and system for assisting sleeping through brain waves |
| CN105476631A (en) * | 2015-11-25 | 2016-04-13 | 华南理工大学 | EEG (electroencephalogram) based sleep detection and sleep aid method and device |
| WO2017092431A1 (en) * | 2015-12-01 | 2017-06-08 | 乐视控股（北京）有限公司 | Human hand detection method and device based on skin colour |
| CN105615834A (en) * | 2015-12-22 | 2016-06-01 | 深圳创达云睿智能科技有限公司 | Sleep stage classification method and device based on sleep EEG (electroencephalogram) signals |
| CN106419893A (en) * | 2016-09-18 | 2017-02-22 | 广州视源电子科技股份有限公司 | Method and device for detecting sleep state |
| CN106388778A (en) * | 2016-09-21 | 2017-02-15 | 广州视源电子科技股份有限公司 | A method and a system for electroencephalogram signal preprocessing in sleep state analysis |
| CN106388818A (en) * | 2016-09-21 | 2017-02-15 | 广州视源电子科技股份有限公司 | A feature information extracting method and system for a sleep state monitoring model |
| CN106510641A (en) * | 2016-12-19 | 2017-03-22 | 姚健欣 | Method and system for detecting different sleep stages of human body |
| CN107998499A (en) * | 2017-11-28 | 2018-05-08 | 广州视源电子科技股份有限公司 | Processing method and system, the sleep secondary server system of sleep auxiliary content |
| WO2019158704A1 (en) * | 2018-02-15 | 2019-08-22 | Biosency | Monitoring device for monitoring a physiological parameter and methods thereof |
| CN109086700A (en) * | 2018-07-20 | 2018-12-25 | 杭州电子科技大学 | Radar range profile's target identification method based on depth convolutional neural networks |
| US20200093425A1 (en) * | 2018-09-21 | 2020-03-26 | Tata Consultancy Services Limited | System and method for non-apnea sleep arousal detection |
| CN109602417A (en) * | 2018-11-23 | 2019-04-12 | 杭州妞诺科技有限公司 | Sleep stage method and system based on random forest |
| US10559299B1 (en) * | 2018-12-10 | 2020-02-11 | Apprente Llc | Reconciliation between simulator and speech recognition output using sequence-to-sequence mapping |
| CN109674468A (en) * | 2019-01-30 | 2019-04-26 | 大连理工大学 | It is a kind of singly to lead brain electricity sleep mode automatically method by stages |
| CN110327040A (en) * | 2019-04-24 | 2019-10-15 | 武汉理工大学 | Sleep stage method and system based on cloud platform |
| CN110210961A (en) * | 2019-05-21 | 2019-09-06 | 平安普惠企业管理有限公司 | Data capture method, server and computer storage medium based on alliance's chain |
| CN110456846A (en) * | 2019-08-19 | 2019-11-15 | 浙江大学 | A kind of adaptive more sense organ sleeping-assisting systems based on artificial intelligence |
| CN110897639A (en) * | 2020-01-02 | 2020-03-24 | 清华大学深圳国际研究生院 | Electroencephalogram sleep staging method based on deep convolutional neural network |
| CN111012341A (en) * | 2020-01-08 | 2020-04-17 | 东南大学 | Artifact removal and electroencephalogram signal quality evaluation method based on wearable electroencephalogram equipment |
| CN111202517A (en) * | 2020-02-03 | 2020-05-29 | 山东师范大学 | Sleep automatic staging method, system, medium and electronic equipment |
| CN113180691A (en) * | 2020-12-28 | 2021-07-30 | 天津大学 | Three-channel sleep apnea and hypopnea syndrome recognition device |
Non-Patent Citations (1)
| Title |
|---|
| 农天振: "智慧医疗***中基于心冲击信号的睡姿监测及睡眠质量分布", 《中国优秀硕士学位论文全文数据库》, 15 May 2021 (2021-05-15), pages 054 - 63 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113812965B (en) | 2024-04-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3040018A1 (en) | Method and apparatus for authenticating user based on biosignal | |
| CN105640545B (en) | Fetal electrocardiosignal extraction method and device | |
| WO2021219039A1 (en) | Gesture recognition method and apparatus, man-machine interaction method and apparatus, and device and storage medium | |
| CN113349778B (en) | Emotion analysis method and device based on transcranial direct current stimulation and electronic device | |
| CN105852850A (en) | Method and related device for extracting respiratory signals from ECG (electrocardiogram) signals | |
| CN110688959A (en) | Unmanned aerial vehicle signal identification method and device, electronic equipment and storage medium | |
| Saini et al. | Effective automated method for detection and suppression of muscle artefacts from single‐channel EEG signal | |
| WO2016104498A1 (en) | Drowsiness estimation device and drowsiness estimation program | |
| CN113812965A (en) | Sleep state recognition method, device, electronic device and storage medium | |
| CN113208632A (en) | Attention detection method and system based on convolutional neural network | |
| CN115774856A (en) | Dreaming emotion recognition method and device, electronic equipment and storage medium | |
| CN113869107A (en) | Signal denoising method, signal denoising device, electronic device and storage medium | |
| Kaleem et al. | Telephone-quality pathological speech classification using empirical mode decomposition | |
| CN114947852A (en) | Multi-mode emotion recognition method, device, equipment and storage medium | |
| CN113180663A (en) | Emotion recognition method and system based on convolutional neural network | |
| CN113208623A (en) | Sleep staging method and system based on convolutional neural network | |
| Almusallam et al. | Feature-based ECG sensing scheme for energy efficiency in WBSN | |
| KR20220158462A (en) | EMG signal-based recognition information extraction system and EMG signal-based recognition information extraction method using the same | |
| CN113208633A (en) | Emotion recognition method and system based on EEG brain waves | |
| Matiko et al. | Real time emotion detection within a wireless sensor network and its impact on power consumption | |
| CN105310655A (en) | Health detection system | |
| Mazzoni et al. | An Optimized Heart Rate Detection System Based on Low-Power Microcontroller Platforms for Biosignal Processing | |
| CN104918539A (en) | Processing of skin conductance signals to mitigate noise and detect signal features | |
| CN113397565A (en) | Depression identification method, device, terminal and medium based on electroencephalogram signals | |
| CN117918862B (en) | Attention assessment method, terminal and medium based on original electroencephalogram signals |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |