KR20200046598A - Apparatus and method for inducing sleep using neuro-feedback - Google Patents

Abstract

The present invention relates to a technology for sleep induction using a neurofeedback. A sleep induction device of the present invention pre-builds and stores sleep prediction models for multiple users, selects a sleep prediction model corresponding to a subject by receiving at least one of brainwave or user characteristic information of the subject measured using a brainwave measurement means, generates a sleep induction stimulus to induce the subject′s sleep, measures, by using the brainwave measurement means, the brainwave of the subject responding to the sleep induction stimulus to determine a sleep-wake state according to the sleep prediction model, and generates a feedback signal for reducing a stimulus causing wakefulness to adjust the sleep induction stimulus and allow the subject to recognize a decrease in the stimulus, when the sleep-wake state corresponds to a sleepiness pattern as a result of the determination.

Description

Apparatus and method for inducing sleep using neuro-feedback}

The present invention relates to a sleep inducing technique for improving sleep, in particular, by using the neuro-feedback technology to measure the user's brain waves to predict the sleep state, in accordance with the characteristics of the individual device that helps to enter the sleep, It relates to a method and a recording medium recording the method.

Insomnia complains of difficulty in starting or maintaining sleep, or sleep that is not rejuvenated. 27% of the modern adult population are known to experience temporary insomnia and 9% have chronic insomnia.

The causes of insomnia are various, but psychological and behavioral responses to irregular life due to increased stress are considered to be one of the main causes. The problems brought by insomnia cause various accidents and abnormal psychiatric problems such as depression, anxiety, and bipolar disorder, and cardiovascular and immune system abnormalities. It is very difficult to calculate the direct and indirect social burdens caused by insomnia arithmetically, but it is very difficult to calculate the labor force due to insomnia, increased medical use, decreased cognitive function, and resulting traffic accidents, industrial accidents and other accidents, and other physical diseases and depression. Considering the increase in mental illness, the social burden is predicted to be astronomical. However, the current treatment of insomnia has several problems.

The most commonly used sleeping pill prescription is the ease of sleeping, but it has side effects. In other words, resistance and dependence occur, and the prescription dose gradually increases and the drug is difficult to stop. In addition, the symptoms of sleep apnea that are often accompanied by increased cardiovascular side effects, problems such as decreased memory during the day, increased risk of falls, and abnormal behaviors such as binge eating, violent behavior, and suicide while intoxicated Can occur.

For this reason, the sleep medicine community recommends non-pharmacological cognitive behavior therapy (CBT) as the first recommended treatment for insomnia. In other words, it is a treatment method that helps insomnia patients to sleep by resolving their chronic misconceptions about sleep, excessive expectation, anxiety, wrong sleeping habits, irregular sleep schedules, and hyperactivity and tension. CBT consists of sleep hygiene education, sleep limit therapy, stimulation limit therapy, cognitive therapy, relaxation therapy, etc., which change thoughts and behaviors through repetitive education and training. However, the problem with CBT is that treatment is time consuming and expensive. It takes a lot of time to meet with experts and train them and get behavior correction.

The core of the CBT treatment process is the process of creating a relaxation state that can relieve anxiety, tension, and sleep when insomnia occurs. If a method for effectively accomplishing this is developed, it will provide a breakthrough in the treatment of insomnia.

In this regard, development of various tools has been attempted in the past with similar thinking. Previously, a therapeutic device using neuro-feedback has been proposed, which is a relaxation training through feedback from an EEG, so as to relieve the usual anxiety and tension, and does not lead to actual sleep using this device. In addition, a device has been developed to create the synchronization of the EEG through external stimulation, but this is also not a method for giving an insomniac patient a direct feedback to sleep, and its effectiveness has not been scientifically verified.

On the other hand, in applying neurofeedback to improving sleep disorders, it has been pointed out that the prior studies have not considered changes in EEG with age. The following prior art documents introduce that it is possible to consider the effect of age and gender on sleep induction.

Carrier, Julie, et al. "The effects of age and gender on sleep EEG power spectral density in the middle years of life (ages 20-60 years old)." Psychophysiology 38.2 (2001): 232-242.

The technical problem to be solved by the present invention is to prevent the side effects of sleeping pills prescription for the treatment of insomnia patients in the related art, and in applying the non-pharmacological cognitive behavioral therapy, sleep induction technology using the existing neurofeedback according to the age of each individual brain Because the characteristics of the signal are not taken into account, the determination of the sleep state according to the subject solves the inaccurate problem, and when the subject progresses from the awake state to the sleep state, a sudden stimulus is applied to the subject to prevent side effects that hinder sleep induction. .

In order to solve the above technical problem, a sleep inducing method according to an embodiment of the present invention includes: (a) a sleep inducing device pre-building and storing a sleep prediction model for a plurality of users; (b) the sleep inducing device receiving at least one of the EEG or user characteristic information of the subject measured using the EEG measurement means and selecting a sleep prediction model corresponding to the subject; (c) the sleep induction device generating a sleep induction stimulus for inducing the subject's sleep; (d) determining, by the sleep-inducing device, a sleep-wake state according to the sleep prediction model by measuring the brain waves of the subject responding to the sleep-inducing stimulus using the brain-wave measuring means; And (e) when the sleep-wake state corresponds to a sleepiness pattern as a result of the discrimination, the sleep-inducing device generates a feedback signal to reduce the stimulus that causes arousal, thereby controlling the sleep-inducing stimulus so that the subject And inducing a reduction in stimulation to be recognized.

In the sleep induction method according to an embodiment, the step of constructing and storing the sleep prediction model in advance includes: (a1) measuring brain waves for a plurality of users, but classifying the measured brain waves into age groups of users; (a2) extracting the properties of the EEG from the classified EEG by using the power spectrum of the frequency band; And (a3) deriving a sleep prediction model representing a sleep-wake state according to the attribute value for each user's age group using a machine learning algorithm for the extracted brain wave attributes.

In the sleep induction method according to an embodiment, the step of classifying the (a1) measured brainwaves by the age group of the user may include a polysomnography device for measuring the state of sleep and a device for measuring the brainwaves of the user at the same time Measure brain signals by utilizing, but can synchronize the measured sleep state.

In the sleep induction method according to an embodiment, the step of extracting the properties of the (a2) brainwaves may include the average and standard deviation of the power spectrum values of the frequency bands from the brainwaves by the age group, alpha / beta The properties of the EEG can be extracted by calculating the ratio of each average value according to the combination of (beta) / delta / theta.

In the sleep derivation method according to an embodiment, the step of deriving the sleep prediction model (a3) is performed by a user using a machine learning algorithm including at least one of a logistic regression or a random forest algorithm. The sleep prediction model representing the sleep-wake state according to the attribute value for each age group may be derived in the form of a regression equation or a decision tree.

In the sleep induction method according to an embodiment, the step of classifying the (a1) measured brainwaves by the age group of the user, (a4) using a logistic regression algorithm, has the largest odds ratio of the sleep prediction model The method may further include selecting an attribute as an attribute having a relatively high effect on sleep-wake state prediction.

In the sleep induction method according to an embodiment, the step (b) of selecting a sleep prediction model corresponding to the subject may include (b1) inputting user-specific information including the age of the subject, or a pretest process Receiving the EEG of the subject measured through; And (b2) selecting a sleep prediction model corresponding to the subject from a pre-built sleep prediction model for a plurality of users.

In the sleep induction method according to an embodiment, the (d) determining the sleep-wake state according to the sleep prediction model may include (d1) measuring the EEG of the subject in response to the sleep-inducing stimulus to determine the frequency of each frequency band. Calculating an average for the power spectrum values; And (d2) determining a sleep-wake state of a subject from the sleep prediction model according to a ratio of an average power spectrum value of the alpha wave band and an average power spectrum value of the theta wave band.

In the sleep induction method according to an embodiment, the step of controlling the sleep-inducing stimulus by generating a feedback signal to reduce the stimulus causing arousal (e1), (e1) the sleep-wake state is an elevation pattern from the awakening pattern Detecting a point in time when entering; And (e2) generating a feedback signal that reduces the stimulus causing arousal from the sensed time point and controlling the sleep-induced stimulus by performing gradual signal control until it approaches the sleep pattern. In addition, (e3) through the ratio of the average power spectrum of the alpha-band power spectrum and the average power spectrum of the theta-wave band, checking the progress of the sleepiness state may further include.

A sleep induction method according to an embodiment includes: (f) when the sleep-wake state of the subject reaches a sleep pattern, the sleep induction device stops feedback; And (g) the sleep inducing device updating the sleep induction feedback method by measuring a time required for the subject to sleep.

On the other hand, hereinafter, a recording medium readable by a computer recording a program for executing the sleep-inducing method described above on a computer is provided.

In order to solve the above technical problem, an apparatus for inducing sleep according to an embodiment of the present invention includes: a model storage unit for pre-constructing and storing a sleep prediction model for a plurality of users; An input unit that receives the brainwaves of the subject measured using the EEG measurement means, the sleep state of the subject, and user characteristic information; And receiving at least one of the subject's EEG or user characteristic information through the input unit, selecting a sleep prediction model corresponding to the subject from the model storage unit, generating a sleep-inducing stimulus for inducing the subject's sleep, The sleep-awakening state according to the sleep prediction model is determined by measuring the brainwave of the subject responding to the sleep-inducing stimulus using the EEG measuring means, and awakening when the sleep-wakening state corresponds to a sleepy pattern It includes; a processing unit for generating a feedback signal to reduce the stimulation causing the control to induce the subject to recognize the reduction of the stimulation by adjusting the sleep-induced stimulation.

In the sleep inducing apparatus according to an embodiment, the model storage unit measures the EEG for a plurality of users, classifies the measured EEG according to the user's age group, and power spectrum of the frequency band from the classified EEG according to the age band ) To extract the properties of the EEG, and extract the properties of the extracted EEG by using a machine learning algorithm to derive and store the sleep prediction model representing the sleep-wake state according to the property value for each age group of the user. have. In addition, the model storage unit, the average and standard deviation of the power spectrum (power spectrum) value of the frequency band from the brainwaves by age group, alpha (alpha) / beta (delta) / delta (theta) EEG combination Attributes of brain waves are extracted by calculating the ratio of each average value according to, but the attribute having the highest odds ratio of the sleep prediction model using a logistic regression algorithm has a relatively high effect on sleep-wake state prediction It can be selected as an attribute.

In the sleep inducing apparatus according to an embodiment, the processing unit receives user-specific information including the age of the subject through the input unit, or receives the brainwaves of the subject measured through a pretest process, and multiple It is possible to select a sleep prediction model corresponding to the subject from a sleep prediction model previously constructed for a user of.

In the sleep inducing apparatus according to an embodiment, the processing unit measures the brain waves of the subject responding to the sleep-inducing stimulus to calculate an average of power spectrum values of each frequency band, and an average power spectrum value of the alpha wave band The sleep-wake state of the subject can be determined from the sleep prediction model according to the ratio of the average power spectrum of the theta-wave band.

In the sleep inducing apparatus according to an embodiment, the processing unit detects a point in time when the sleep-wake state enters the elevation pattern from the awakening pattern, and generates a feedback signal to reduce the stimulus causing awakening from the detected time point The sleep-inducing stimulus is adjusted by performing gradual signal control until it approaches the sleep pattern, but the progress of the sleepiness state can be examined through a ratio of the average power spectrum of the alpha-wave band and the average power spectrum of the theta-wave band.

In the sleep guidance device according to an embodiment, the processor stops feedback when the sleep-wake state of the subject reaches a sleep pattern, and updates the sleep guidance feedback method by measuring the time it takes for the subject to sleep can do.

In the sleep inducing apparatus according to an embodiment, the sleep-inducing stimulus includes any one of sound, light, or vibration, and the feedback signal is obtained by changing any one of the intensity, frequency, period, stimulus type, or shape of the stimulus It can induce the subject to notice a decrease in stimulation.

According to embodiments of the present invention, a sleep-wake state discrimination model for each age group is provided to a plurality of users suffering from insomnia to select a predictive model that matches the characteristics of an individual's age or brain signal, thereby It is possible to more accurately determine the sleep-wake state, and provides sleep-inducing feedback to sleepy subjects to continuously evaluate sleep progress, but by inducing the subject to recognize the reduction of the stimulus that causes wakefulness, it is fast and effective insomnia. It is possible to improve, and by measuring the time required to reach the input of the subject, the sleep guidance feedback is updated to optimize the personalized sleep guidance for the individual subject.

1 is a view illustrating a state of brain waves according to stimulation in the field of sleep induction technology in which embodiments of the present invention are implemented.
2 is a flowchart illustrating a sleep induction method using neuro feedback according to an embodiment of the present invention.
3 is a flowchart more specifically illustrating a process of pre-establishing a sleep prediction model in the sleep derivation method of FIG. 2 according to an embodiment of the present invention.
FIG. 4 is a diagram illustrating property values of brain waves to be used as learning data adopted by embodiments of the present invention and the sleep-wake state accordingly.
FIG. 5 exemplifies the performance comparison result of a model trained by including the performance of the model (f-measure) and the standard deviation of the average value and the ratio of the model trained with the power spectrum average value of delta / theta / alpha / beta for subjects in their 30s. It is one drawing.
FIG. 6 is a diagram illustrating a result showing how each property of an EEG can be determined with a relatively high probability of sleep-wake state through a logistic regression algorithm.
7 is a diagram illustrating a part of a decision model built by a random forest algorithm.
8 is a diagram illustrating a result of predicting a sleep state by a logistic regression and a random forest algorithm.
9 is a flowchart illustrating a process of updating a sleep guidance feedback method in the sleep guidance method of FIG. 2 according to an embodiment of the present invention.
10 is a diagram illustrating changes in alpha and theta powers in the process from awakening to sleeping.
11 is a diagram illustrating a change in brain waves in the wakeful state.
12 is a view for explaining a change in the alpha wave in FIG. 11.
13 is a view for explaining the pie of EEG according to age.
14 is a block diagram showing a sleep inducing apparatus using neuro feedback according to an embodiment of the present invention.

Before explaining the embodiments of the present invention, after reviewing the outline of the neurofeedback technology and the problems indicated by the insomnia treatment technology utilizing the technology, an overview of the technical means adopted by the embodiments of the present invention to solve these problems To introduce.

EEG, published by German Hans Berger (1873-1941) in 1929, is an important biosignal showing the activity of the human brain. EEG is classified into delta waves (1 ~ 4Hz), theta waves (4 ~ 8Hz), alpha waves (8 ~ 13Hz), beta waves (13 ~ 30Hz), and gamma waves (30 ~ 120Hz) according to the mental activity state. Delta waves are brain waves generated during deep sleep, theta waves are generated during normal sleep, and are the basic brain waves when dreaming. Alpha waves are brain waves that appear when you are resting. Close your eyes and rest while your consciousness is awake. It comes out strongly when there is. Beta waves are brain waves that occur when the brain is doing some mental work, like learning. Gamma waves are brain waves that appear when cognitive functions occur by combining information scattered in various parts of the brain.

EEG is a normal mental action, the vibration is fast or slow. If not, it means that the brain's function is abnormal. That is, since the normal brain and the abnormal brain exhibit distinct features in the EEG, it is possible to determine whether the brain is abnormal by measuring the EEG.

In general, it is known that the body's functions controlled by the autonomic nervous system, such as the rhythm of the brain, cannot be controlled by us. However, in the 1950s, Dr. Miller at Yale University in the United States discovered that muscles that could not be controlled by our will, such as the muscles of the intestine or heart (the involuntary muscle), or the autonomic nervous system, can be controlled by our will according to conditions. Biofeedback is the technology that controls the involuntary muscles or autonomic nervous system with our will. In particular, biofeedback technology that controls brain waves is called neurofeedback by combining it with the neuro- prefix, neuro.

The discovery of neurofeedback dates back to 1934, when Dr. Matthew and Adrian of the University of Cambridge, England, produced an EEG to measure the EEG. One day, while measuring the EEG, I tried to make the sound come out of the speaker only when the alpha wave came out. Then I discovered that the alpha wave is getting stronger. This is the principle of neurofeedback.

By measuring the user's brain waves and notifying the user that a specific brain wave has occurred, the brain automatically strengthens the specific brain wave. This is like the conditional reflection discovered by Dr. Pavlov. If you strike a bell while feeding the dog, it will drool even if you hit the bell later. This means that the brain's unconditional reflex circuit, which looks like it's drooling, is connected to the bell, creating a conditional society. If you keep repeating this, the newly created conditional reflection circuit will be strengthened to make it a habit. Then, even if you hit the bell, you will drool. With this principle, if a specific brain wave is notified every time a circuit is generated, the circuit by the brain wave is developed in the brain, and if it is repeatedly repeated, the circuit is strengthened and the specific brain wave is increased. In other words, neuro feedback refers to a method of strengthening the circuit of the brain wave by strengthening the activity of a specific frequency band of the brain wave through learning and training, and repeat training by giving a stimulus to recognize it when a desired brain wave comes out In the process of performing it and repeating it, it becomes possible to control the state of the EEG by itself.

In 1958, Dr. Joe Kamiya of the University of Chicago, USA, conducted the first experiment to change the state of the mind by adjusting brain waves according to the principle of neurofeedback. This experiment showed that the brainwaves that can be arbitrarily controlled can be controlled at will, which is the first neurofeedback experiment in a true sense.

In 1971, Dr. Barry Sturman of UCLA, USA, succeeded in treating epilepsy with neurofeedback using SMR waves (12–16 Hz brain waves generated in the sensory motor cortex of the brain). This is the world's first application of neurofeedback to disease treatment. In 1976, Dr. Ruva of Tennessee, USA reported a successful example of treating ADD and ADHD with neurofeedback using SMR and beta waves.

In 1979, Dr. Davidson of the United States found that when the alpha-wave balance between the left and right brains broke, he became depressed. He discovered an electroencephalographic abnormality on the so-called emotional disorder. In 1995, Dr. Rosenfeld reported successful experiments in treating depression through neurofeedback training to balance the left and right brain alpha waves. Meanwhile, in 1989, Dr. Peniston and Dr. Kulkoski reported that it was effective in treating neurotoxicity and post-traumatic stress disorder (PTSD) by providing neurofeedback training.

On the other hand, the existing method of treating insomnia using the neurofeedback technique has proposed a method of training to induce relaxation state well through usual training. However, even if training is done in this way, it is a completely different situation to actually sleep, and rather, attempting to ask for a real sleep in a state where there is not enough relaxation training may cause anxiety and adversely affect the brain's wakefulness. . This is because the sleep controlled by the autonomic nerve cannot be controlled by an effort to sleep consciously. Therefore, it is necessary to give feedback from a person who sleeps through neurofeedback about changes in brain waves caused by autonomic nerves.

In addition, existing methods of treating insomnia by synchronizing brain waves by stimulating at a frequency of a certain time interval to induce sleep have been attempted, but inducing elevation due to the phenomenon of brain waves lacks scientific evidence, and irritation is rather patient It may cause arousal. Therefore, it is intended to present a technical means for inducing a patient to a comfortable sleeping state by utilizing neuro-feedback technology without such side effects.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. However, in the following description and accompanying drawings, detailed descriptions of well-known functions or configurations that may obscure the subject matter of the present invention are omitted. In addition, it should be noted that the same components throughout the drawings are denoted by the same reference numerals as much as possible.

1 is a view illustrating the state of the brain wave according to the stimulation in the field of sleep induction technology in which embodiments of the present invention is implemented, the red box represents the brain wave of the awakening state, the other indicates the sleep state.

Until now, neurofeedback has been used as a method of training a targeted EEG state through audio-visual stimulation, usually in awake state. Researchers of the present invention, when applying neurofeedback during sleep, do not give a stimulus in the awake state before sleep, but then select a method to repeatedly learn the state by giving appropriate sound stimulation when the body is relaxed and close to sleep Was introduced. However, in the process of testing, it has been found that this method may activate cognitive function by sound stimulation, making it difficult to fall into sleep. In the present invention devised to compensate for this, a method in which sound is reduced as it approaches the desired brain wave state in the course of sleep is differentiated from the conventional control method of neurofeedback, maximizing stability in inducing sleep Experimentally proved that it can.

Referring to FIG. 1, it can be seen from FIG. 1 (a) that awakening a2 occurs again by sound stimulation after reaching the surface a1. On the contrary, in FIG. 1 (b) according to the feedback method of reducing the intensity of the stimulus causing the arousal proposed by the embodiments of the present invention, the sound stimulus is almost extinguished after reaching the sleep b1 and does not cause arousal. You can confirm that there is.

The embodiments of the present invention proposed below using these technical principles are close to sleep by providing feedback to induce the subject to recognize the reduction of the stimulus causing arousal in the sleep induction mechanism of the conventional neurofeedback. After that, the side effects that were awakened again were blocked.

In addition, the embodiments of the present invention note that the state of the brain waves appearing in the process of sleep-wakening differs depending on the individual, thereby providing a basic sleep-inducing stimulus to the subject to test the subject's condition in response thereto (pretest ) By introducing the process, an optimal sleep-awakening model was selected according to the characteristics of the individual, and through this, it was configured to significantly reduce trial and error in clinical applications.

2 is a flowchart illustrating a sleep induction method using neuro feedback according to an embodiment of the present invention.

In step S110, the sleep inducing device pre-builds and stores a sleep prediction model for a plurality of users. In order to provide a sleep-inducing stimulus to improve sleep in a patient suffering from insomnia, it is necessary to measure the brain signal of the subject to determine whether or not he or she has reached a drowsiness state just before sleep. In this process, the brain signals of a large number of users are measured in advance to generate a sleep prediction model according to individual characteristics (age). For example, depending on the age, the maximum value of the brain signal amplitude at awakening may be 12 Hz, while some people are slow at 10 Hz. Therefore, in order to build a more accurate sleep discrimination model, the measured brain signals are classified according to these user characteristics, and after converting the classified brain signals into input variables of a machine learning algorithm through various signal processing algorithms, the user's sleep-wakening A model for determining whether or not is extracted. A more specific process of constructing the sleep prediction model will be described later with reference to FIG. 3.

In step S120, the sleep inducing device receives at least one of the EEG or user characteristic information of the subject measured using the EEG measurement means and selects a sleep prediction model corresponding to the subject. In this process, user-specific information including the age of the subject is input, or the brainwave of the subject measured through a pretest process is input. That is, the user's age is inputted into a wearable brain signal measuring device used by the user (or software including the software implementing the same), or the user's EEG characteristics are measured through a pretest process for measuring the EEG. Discriminate. Since a user may have a brain signal pattern different from a general brain signal pattern of the corresponding age group, before selecting a sleep prediction model, it is necessary to accurately determine the brain signal pattern of the current subject through brain signal measurement in a pretest process. There is.

Then, according to the determined characteristics of the EEG, a sleep prediction model corresponding to the subject may be selected from a sleep prediction model previously built for a plurality of users through step S110. For example, if the user's age is input as 30s, a sleep prediction model in 30s may be selected from the models classified in step S110. Alternatively, when a user measures brain signals through a pretest process, a sleep prediction model for an age group corresponding to the corresponding amplitude value may be selected according to the maximum value of the brain signal amplitude.

In step S130, the sleep inducing device generates a sleep-inducing stimulus for inducing the sleep of the subject. In this process, basic sleep-inducing stimuli (eg, sound, vibration, visual stimuli, etc.) are generated to induce sleep of a user who is experiencing insomnia.

In step S140, the sleep-inducing device measures the brain waves of the subject responding to the sleep-inducing stimulus using the EEG measuring means to determine a sleep-wake state according to the sleep prediction model. First, the brainwaves of the subject responding to the sleep-inducing stimulus are measured to average the power spectrum values of each frequency band. Then, the sleep-wake state of the subject is determined from the sleep prediction model according to the ratio of the average power spectrum of the alpha-wave band and the average power spectrum of the theta-wave band.

More specifically, after separating the EEG through a fast Fourier transform (FFT) for the measured brain signal, among the power spectrum values of each EEG, the average of the power spectrum values of the alpha wave band and the power spectrum of the theta wave band Find the ratio to the mean of the values. From the measured ratio of the alpha wave and theta wave of the user, that is, when the corresponding ratio value increases from 'alpha / theta wave', it can be determined that the awakening state is approached, and when the ratio value decreases, it can be determined that the sleep rate is approached.

As a result of the determination in step S150, when the sleep-wake state corresponds to a sleepiness pattern, the sleep-inducing device generates a feedback signal to reduce the stimulus causing arousal through step S160, thereby controlling the sleep-inducing stimulus The subject is induced to be aware of a decrease in irritation. To this end, it detects a time point when the sleep-wake state enters the elevation pattern from the wake-up pattern, and generates a feedback signal that reduces the stimulus causing wakefulness from the sensed time point to gradually control the signal until it approaches the sleep pattern. It is desirable to control the sleep-induced stimulation by performing. At this time, the progress of the sleepy state may be checked through a ratio of the average power spectrum value of the alpha wave band and the average power spectrum value of the theta wave band.

More specifically, when the ratio value of 'alpha / theta wave' in step S150 becomes lower than the existing ratio value, that is, when it is determined that the theta wave increases and the user becomes closer to the drowsiness state, sleep-induced stimulation is gradually reduced. By doing so, it provides the feedback that informs the user that their attempt to rise is successful, so that they fall into sleep naturally. For example, it is possible to give feedback to the user that the attempt of elevation is proceeding well by giving a sound stimulus of 100% in volume through step S130 and then decreasing the volume by 10% each time through step S160. Methods to reduce the stimulus include reducing the volume of the sound, increasing the interval of the sound, how much the volume of the sound is reduced, and how many seconds the interval of the sound is experimentally derived from the clinical field or the operator Can be adjusted. These values can also be automatically adjusted to personal customization by measuring the time to enter the surface in the future.

FIG. 3 is a flowchart illustrating in more detail the process of constructing a sleep prediction model (S110) in advance in the sleep derivation method of FIG. 2 according to an embodiment of the present invention.

In step S111, the sleep inducing device measures brain waves for a plurality of users, but classifies the brain waves measured through step S112 according to a user's age group. This process is a preliminary task to create a sleep prediction model, a polysomnography device that plays a role in determining a user's sleep awakening state, and a wearable brain signal that a user who is experiencing physical insomnia will use to improve sleep. It is preferable to measure and record the sleep brain signal value of the user by simultaneously using the measuring device, but to synchronize the measured sleep state. The polysomnography device more accurately measures sleep-related body signals to determine the state of sleep (awakening, drowsiness, sleep, etc.) by viewpoint, and the wearable brain signal measurement device records brain signals obtained from sensor electrodes attached to the forehead. do. The reason for using these two devices at the same time in making the sleep prediction model is to increase the discrimination accuracy of the wearable device used to derive the actual elevation from the information of the sleep polygraph device having high reliability in determining sleep state.

The brain signal data measured in this way is separately classified into age groups, i.e., 10s, 20s, 30s, 40s, 50s, 60s, 70s, 80s, etc., to create a sleep prediction model according to the user's age group. do.

In step S113, the sleep inducing device extracts the properties of the EEG using the power spectrum of the frequency band from the classified EEGs. To this end, the average and standard deviation of the power spectrum values of the frequency band from the brainwaves by the age group, each average value according to the EEG combination of alpha / alpha / beta / delta / theta By calculating the ratio of, the properties of the EEG can be extracted. From an implementation point of view, the average of the power spectrum of each frequency band (delta wave, theta wave, alpha wave, beta wave), the standard deviation, and the ratio of the average value (for example, alpha / theta, alpha / delta, through fast Fourier transform), Theta / delta, beta / delta, etc.). These attributes are used as input values of the machine learning classification algorithm to be performed in the next step S114.

In operation S114, the sleep inducing device derives a sleep prediction model indicating a sleep-wake state according to attribute values for each age group of a user by using a machine learning algorithm to extract the properties of the extracted EEG. For the machine learning algorithm, for example, a logistic regression or a random forest algorithm may be used, and a sleep prediction model indicating a sleep-wake state according to the attribute value for each user's age group. It can be derived in the form of a regression equation or a decision tree.

FIG. 4 is a diagram illustrating property values of brain waves to be used as learning data adopted by embodiments of the present invention and the sleep-wake state accordingly. The sleep-wake state is a value measured by brain signal measurement equipment and is divided into two types: wake / sleep in the last column.

Referring to FIG. 4, the delta / theta / alpha / beta average is a frequency range of each brain signal in a brain signal through a fast Fourier transform (delta wave 0.2 to 3.99 Hz, theta wave 4 to 7.99 Hz, alpha wave 8 to 12.99 Hz, It shows the average of the power spectrum values measured at beta wave 13 ~ 30Hz). In addition to using the average value of the delta / theta / alpha / beta power spectrum as training data for machine learning algorithms, the standard deviation and ratio of the average value of each frequency range (alpha / theta, alpha / delta, theta / delta, beta / delta) Also, classification performance is improved by using it as a kind of learning data.

FIG. 5 exemplifies the performance comparison result of a model trained by including the performance of the model (f-measure) and the standard deviation of the average value and the ratio of the model trained with the power spectrum average value of delta / theta / alpha / beta for subjects in their 30s. It is one drawing. The machine learning algorithm used for learning is a logistic regression algorithm. In particular, it shows higher performance by adding the standard deviation of the average value of the power spectrum as training data. In order to evaluate the performance of the user's sleep prediction model by age group, it is possible to utilize performance evaluation measures such as precision and recall in addition to the f-feasure illustrated in FIG. 5.

FIG. 6 is a diagram illustrating a result showing how each property of an EEG can be determined with a relatively high probability of sleep-wake state through a logistic regression algorithm.

Referring to FIG. 6, it can be seen that the 'beta / delta' property has a relatively high proportion in determining sleep-wake state compared to other properties. That is, by using the logistic regression algorithm, the property having the highest odds ratio of the sleep prediction model is selected as the property having a relatively high influence on the sleep-wake state prediction, so that the judgment of the sleep-wake state can be greatly assisted. You can. By utilizing the machine learning algorithm, it is possible to grasp an attribute having a higher influence than the average value of the power spectrum by adding the standard deviation and the ratio attribute to the training data, and when predicting the sleep state through the selected sleep prediction model in response to the subject, it is relatively Only minimal data with higher impact can be extracted and used to predict sleep.

Among the brain signal data classified with these attribute data, the brain signal attribute values of subjects in their 30s were learned to build a model with each machine learning algorithm, and the results of predicting sleep state by logistic regression and random forest algorithm were illustrated. As shown in FIG. 8, the model created through the logistic regression algorithm was able to predict the sleep state with an average accuracy of 92.7%, and the random forest classification algorithm was able to predict the sleep state of the training data with an accuracy of 98.9%. The model built from each machine learning algorithm can be used to determine the sleep state through the process of measuring the brain signal of the actual subject and extracting the properties of the brain signal.

In the case of a model constructed by a logistic regression algorithm, the result value is extracted as a linear regression equation, and may be composed of a linear equation of the following equation (1).

Equation 1 is a multiple regression model extracted by performing a logistic regression algorithm, where a, b, ..., etc. represent the regression coefficients of this regression model, and x ₁ , x ₂ , ... are dependent variables y, that is, an independent variable for obtaining a threshold value for determining the awakening-sleep state. Here, x ₁ , x ₂ , ..., etc. mean each attribute value (brain signal spectrum average value, standard deviation, ratio), and each attribute multiplies the coefficients extracted through learning by the logistic regression algorithm. When the subject's brain signal value undergoes attribute transformation and the attribute value of this regression equation is input to the variable x, the value of the linear regression equation may be calculated. When this value is y, the following equation (2) is calculated to obtain a threshold value that can determine the awakening-sleep state.

Equation 2 converts the threshold value obtained from the multiple regression model of Equation 1 into a single regression model and extracts a value between '0' and '1'. When the threshold value of Equation 2 is greater than or equal to '0.5', it can be determined that it is awakened, and when it is less than '0.5', it is a sleep state. Basically, the awakening-sleep state can be grasped based on the threshold value of '0.5'. However, if necessary, the user's sleep state can be determined more strictly or insensitively by adjusting this threshold according to the user's characteristics. have. For example, if the default threshold value is lowered to '0.3', it is determined to be in a sleep state only when the probability value is less than 30%, so the sleep-induced stimulation can be changed more strictly, and the default threshold value is set to '0.7'. If it is high, it is possible to change the sleep-induced stimulation more insensitively because all probability values of 0% to 70% are determined as the sleep state.

In the case of a model constructed by a random forest algorithm, it is possible to determine whether it is finally awake or sleeping through a nested decision tree. FIG. 7 is a diagram illustrating a part of a decision model built by a random forest algorithm using a machine learning tool called 'Weka'. Depending on the condition of the value of each property, the Sleep or Wake state is determined. 'x_mean' represents the average power spectrum value of the brain signal frequency band x, and 'x_stdev' represents the standard deviation of the average value. Finally, 'x / y' represents the ratio of the average power spectral values of x and y.

Through the above process, it was explained that sleep prediction models for various types of users can be built.

9 is a flowchart for additionally explaining a process of updating a sleep guidance feedback method in the sleep guidance method of FIG. 2 according to an embodiment of the present invention. In order to avoid duplication of description, a subsequent signal is continuously described in step S160 of generating a feedback signal that reduces stimulation and controlling sleep-induced stimulation.

However, in the case of step S160, more sophisticated stimulation control is required to induce sleep in terms of implementation. If the user shows a sleepy brain wave pattern, the sleep-inducing stimulus is reduced in step S160 (for example, the volume is adjusted to -10%), depending on the ratio of alpha wave and theta wave in applying this control method in clinical trials. Continuing to reduce stimulation can result in a situation where feedback is interrupted (for example, the volume is 0%) even though the user has not reached a complete sleep pattern. Therefore, it is necessary to more precisely control the process of controlling the sleep-induced stimulation in step S160.

To this end, if the stimulation decreases according to the sleepy brain waves and then reaches the stage immediately before the feedback is completely stopped, the minimum feedback is continued if the theta brain waves in the last 30 seconds are less than 1/3 of the total brain waves (about 33%). You can introduce a way to maintain. In other words, if the feedback is maintained at a minimum level and the theta brainwave exceeds 1/3 of the total brainwave, the process proceeds to step S180 where the feedback is completely stopped, but the sleep pattern condition is satisfied in step S170 even if it does not exceed 1/3. If possible, you can stop the feedback unconditionally. As described above, determining the minimum feedback maintenance through the ratio of theta brain waves to the total brain waves is a criterion for determining whether to sleep deeply. At this time, the ratio '1/3' of theta brainwaves is an example derived from experimental results, and the ratio may be set differently according to implementation needs.

In step S170, when the sleep-wake state of the subject reaches a sleep pattern, the sleep inducing device stops feedback through step S180. According to the age group, it is determined whether the user is in a sleep state or an awake state by a selected sleep-awakeness model. In order to determine the sleep or awakening state of a real user, a polysomnography device that serves to determine the sleep awakening state must be used, but there is already a sleep-wake discrimination model that has been learned and built with the user's wearable brain signal measuring device. Therefore, by observing the output value of this model, it is possible to determine the sleep-wake state even with a wearable brain signal measuring device without a corresponding device. Stop sleeping induction feedback because no more feedback is meaningless when you are asleep.

In step S190, the sleep inducing device updates a sleep induction feedback method by measuring a time taken from the initial sleep induction stimulation application time (step S130) to the sleep. If the time taken so far is shorter than the existing time, the type of sleep-inducing feedback in step S160 or a method or cycle for reducing stimulation is considered to be the most suitable method for the current subject, and the control values are stored in step S160. This can be used not only to personalize with a feedback method suitable for the current subject, but also as a basic sleep-inducing stimulus for a user having the same age or brain signal characteristics. If the time taken so far is longer than the existing time, it is determined that the type of sleep guidance feedback or the stimulus control method of step S160 is not suitable, and the value or control method of the feedback stored in step S160 is different for the next attempt. It is necessary to change it to.

In the embodiments of the present invention presented above, the process of reaching sleep in awakening is evaluated, and sleep-inducing stimulation according to neurofeedback is reduced. In the arousal state, changes in the subject's individual EEG are considered and reflected. For example, the frequency of the alpha wave is slowed and the power is reduced, thereby reducing sleep-induced stimulation (sound). Afterwards, when the alpha wave is lost, focusing on the characteristics of the sleep brain waves previously learned through machine learning, it is confirmed that sleep is achieved when the complete level 1 sleep is reached, and the sleep-inducing stimulus is extinguished. In this case, machine learning data of a similar group is applied by classifying based on an individual's age and EEG characteristics.

10 is a diagram illustrating changes in alpha and theta powers in the process from awakening to sleeping. Referring to FIG. 10, it can be seen that the feedback gradually decreases as the frequency of the alpha wave changes to the awakening-elevation, and when the first stage sleep is reached, the sound is completely extinguished.

In the embodiments of the present invention, a change in feedback in the awakening state is also important. For example, the frequency of the alpha wave becomes slower and the power becomes weaker as the water gets closer to the surface even in the same awakening state. By separating this, it is possible to instill a more stable feeling to the user by adopting a method of appropriately reducing the sound even in the wakeful state, and effective induction of sleep is possible.

FIG. 11 is a diagram illustrating a change in brain waves in the wakeful state, and FIG. 12 is a diagram for explaining a change in alpha waves in FIG. 11.

In FIG. 11, the blue box (c1) represents the arousal state of 10 Hz, and the green box (c2) slowly represents the awakening state (drowsy) of slowing to 9 Hz. 12, the blue box (c1) and the green box (c2) in FIG. 11 show the results of the analysis of the alpha wave, and the closer to the water surface, the slower the peak frequency (peak frequency) and the power (power). Also becomes smaller.

For feedback in the wakeful state, it is necessary to analyze in advance the differences in brain waves between individuals. In the state of full awakening, some people have a slightly slower peak frequency of alpha wave (9 Hz), some people have 11 Hz, and there are individual differences in power. It is necessary to perform a pretest before sleep in order to analyze the characteristics of the brain waves of each individual. The pre-test can be conducted for 5 minutes before sleep, and can be performed with eyes closed and stable according to the guidance voice of the neuro feedback software. Personalized neurofeedback may be performed based on the measured awakening-lung stability.

On the other hand, the fact that the EEG reflects mental and physical activity by frequency band is common to users, but there are individual differences. In particular, it is influenced by age, personality, and medical condition, which means that the peak frequency and power may be different even if the alpha wave is predominantly awake-lung stable. EEG during sleep is also affected by various factors, but the fact that alpha waves are lost and theta waves dominate when entering sleep in a large frame is the same for everyone.

13 is a view for explaining the pie of EEG according to age. Referring to FIG. 13, as a result of analyzing the brainwaves of 8 awakening-lung eyes during the teens to 60s, the peaks (yellow boxes) of the alpha wave band are the same, but power tends to decrease with age. The peak frequency varies depending on the individual, but the smaller the power, the faster the tendency.

As described above, the age difference of the EEG is proved through many studies, and the difference between the individual EEGs does not change in the morning, but gradually changes with the aging of several years. Therefore, in the embodiments of the present invention, the pre-test process is to investigate the current EEG characteristics in advance, and once performed, it is not necessary to re-execute for a long time, but may be additionally performed as necessary.

14 is a block diagram illustrating a sleep inducing apparatus 1000 using neuro feedback according to an embodiment of the present invention, and is a reconstruction of a series of processes described with reference to FIG. 2 in terms of device configuration. Therefore, here, only functions / operations of each component are briefly described in order to avoid duplication of description.

The model storage unit 10 is configured to store and predict a sleep prediction model in advance for a plurality of users. The model storage unit 10 measures the EEG for a plurality of users, classifies the measured EEG by age of the user, and uses the power spectrum of the frequency band from the classified EEG according to the age group. The attribute is extracted, and the extracted attribute of the EEG can be derived and stored by using a machine learning algorithm to derive a sleep prediction model representing the sleep-wake state according to the attribute value for each user's age group.

In addition, the model storage unit 10, the average and standard deviation of the power spectrum (power spectrum) values of the frequency band from the brainwaves by age, alpha (alpha) / beta (delta) / theta (theta) ) Extracts the properties of brain waves by calculating the ratio of each average value according to the combination of brain waves, but uses the logistic regression algorithm to set the property with the highest odds ratio of the sleep prediction model relative to sleep-wake state prediction It can be selected as a highly influential attribute.

The input unit 20 is configured to receive the EEG of the subject measured using the EEG measurement means, the state of the subject's sleep, and user characteristic information.

The processor 30 receives at least one of the EEG or user characteristic information of the subject through the input unit 20, selects a sleep prediction model corresponding to the subject from the model storage unit 10, and sleeps the subject Generating a sleep-inducing stimulus for inducing, and measuring the brain waves of the subject responding to the sleep-inducing stimulus using the EEG measuring means to determine the sleep-wake state according to the sleep prediction model, the determination result of the sleep -When the arousal state corresponds to the sleepiness pattern, it generates a feedback signal that reduces the stimulus that causes arousal, and controls the sleep-induced stimulation to induce the subject to recognize the decrease in stimulation. Here, the sleep-inducing stimulus includes any one of sound, light, or vibration, and the feedback signal changes the intensity, frequency, period, type or type of stimulation of the stimulus, thereby allowing the subject to recognize the decrease in stimulation. You can induce them to do it.

The processing unit 30 receives user-specific information including the age of the subject through the input unit 20 or receives the brainwaves of the subject measured through a pretest process, and for multiple users A sleep prediction model corresponding to the subject can be selected from a previously constructed sleep prediction model.

In addition, the processing unit 30 measures the brain waves of the subject responding to the sleep-induced stimulation to calculate an average of the power spectrum values of each frequency band, the power spectrum average value of the alpha wave band and the power spectrum of theta wave band The sleep-wake state of the subject can be determined from the sleep prediction model according to the ratio of the average value.

In addition, the processing unit 30 detects a time point when the sleep-wake state enters the elevation pattern from the wake-up pattern, and generates a feedback signal to reduce the stimulus causing wakefulness from the sensed time point to approach the sleep pattern. The sleep-induced stimulation may be adjusted by performing gradual signal control until the sleepiness state can be examined through a ratio of the average power spectrum of the alpha-wave band and the average power spectrum of the theta-wave band.

Furthermore, the processing unit 30 may stop the feedback when the sleep-wake state of the subject reaches the sleep pattern, and update the sleep-induced feedback method by measuring the time required for the subject to sleep.

According to the above-described embodiments of the present invention, a sleep-wake state discrimination model for each age group is provided to a plurality of users suffering from insomnia by selecting a predictive model that matches the characteristics of an individual's age or brain signal. It is possible to more accurately determine the sleep-wake state according to brain signals, and provide sleep-inducing feedback to sleepy subjects to continuously evaluate the sleep progression, but induce the subject to recognize the decrease in the stimulus that causes wakefulness. By doing so, it is possible to improve sleep quickly and effectively, and by measuring the time required to reach the input of the subject, the sleep guidance feedback is updated to enable personalized sleep guidance optimized for the individual subject.

On the other hand, embodiments of the present invention can be implemented in computer-readable code on a computer-readable recording medium. The computer-readable recording medium includes any kind of recording device in which data readable by a computer system is stored.

Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like. In addition, the computer-readable recording medium can be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. And functional programs, codes, and code segments for implementing the present invention can be easily inferred by programmers in the technical field to which the present invention pertains.

In the above, the present invention has been mainly focused on various embodiments. Those skilled in the art to which the present invention pertains will appreciate that the present invention may be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered in terms of explanation, not limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the equivalent range should be interpreted as being included in the present invention.

1000: sleep induction device
10: model storage
20: input
30: processing unit

Claims (20)

(a) the sleep inducing apparatus pre-built and stored a sleep prediction model for a plurality of users;
(b) the sleep inducing device receiving at least one of the EEG or user characteristic information of the subject measured using the EEG measurement means and selecting a sleep prediction model corresponding to the subject;
(c) the sleep induction device generating a sleep induction stimulus for inducing the subject's sleep;
(d) determining, by the sleep-inducing device, a sleep-wake state according to the sleep prediction model by measuring the brain waves of the subject responding to the sleep-inducing stimulus using the brain-wave measuring means; And
(e) When the sleep-wake state corresponds to a sleepiness pattern as a result of the discrimination, the subject induces a stimulus by adjusting the sleep-inducing stimulus by generating a feedback signal that reduces the stimulus causing the awakening. Inducing so as to recognize the reduction of; sleep induction method comprising a. According to claim 1,
Step (a) is,
(a1) measuring brainwaves for a plurality of users, but classifying the measured brainwaves into age groups of users;
(a2) extracting the properties of the EEG from the classified EEG by using the power spectrum of the frequency band; And
(a3) deriving a sleep prediction model indicating a sleep-wake state according to attribute values for each user's age group using a machine learning algorithm for the extracted brain wave attributes. According to claim 2,
Step (a1) is,
A sleep induction method that simultaneously measures a brain signal using a polysomnography device that measures the state of sleep and a device that measures the user's brain waves, but synchronizes the measured sleep state. According to claim 2,
Step (a2) is,
The ratio of the average and standard deviation of the power spectrum values of the frequency band from the brain waves by age group, and the ratio of each average value according to the EEG combination of alpha / beta / delta / theta A sleep-inducing method that extracts the properties of brain waves by calculating. According to claim 2,
Step (a3) is,
Using a machine learning algorithm including at least one of a logistic regression or a random forest algorithm, a sleep prediction model representing a sleep-wake state according to the attribute value according to the user's age is regression or decision making. Method of deriving sleep in the form of a tree (decision tree). According to claim 2,
Step (a) is,
(a4) using the logistic regression algorithm to select the property having the highest odds ratio (odds ratio) of the sleep prediction model as a property having a relatively high effect on sleep-wake state prediction; . According to claim 1,
Step (b) is,
(b1) receiving user-specific information including the age of the subject, or receiving the brainwaves of the subject measured through a pretest process; And
(b2) selecting a sleep prediction model corresponding to the subject from a sleep prediction model pre-built for multiple users. According to claim 1,
Step (d) is,
(d1) measuring the brainwaves of the subject in response to the sleep-inducing stimulus to calculate an average of power spectrum values of each frequency band; And
(d2) determining a sleep-wake state of a subject from the sleep prediction model according to a ratio of an average power spectrum value of the alpha wave band and an average power spectrum value of the theta wave band. According to claim 1,
Step (e) is,
(e1) detecting a point in time when the sleep-wake state enters the elevation pattern from the wake-up pattern; And
(e2) generating a feedback signal that reduces the stimulus causing arousal from the sensed time point and controlling the sleep-induced stimulus by performing gradual signal control until it approaches the sleep pattern. The method of claim 9,
(e3) examining the progress of the sleepiness state through a ratio of the average power spectrum of the alpha-wave band and the average power spectrum of the theta-wave band; According to claim 1,
(f) when the sleep-wake state of the subject reaches a sleep pattern, the sleep inducing device stops feedback; And
(g) the sleep induction device measuring the time required for the subject to sleep to update the sleep induction feedback method; further comprising a sleep induction method. A computer-readable recording medium recording a program for executing the method of any one of claims 1 to 11 on a computer. A model storage unit for pre-built and stored sleep prediction models for a plurality of users;
An input unit that receives the brainwaves of the subject measured using the EEG measurement means, the sleep state of the subject, and user characteristic information; And
A sleep prediction model corresponding to the subject is selected from the model storage unit by receiving at least one of the subject's EEG or user characteristic information through the input unit, and a sleep inducing stimulus for inducing the subject's sleep is generated. The sleep-awakening state according to the sleep prediction model is determined by measuring the brainwaves of the subject responding to the sleep-inducing stimulus using the EEG measuring means, and as a result of the determination, when the sleep-wakening state corresponds to a sleepiness pattern And a processor configured to generate a feedback signal to reduce the induced stimulus and induce the subject to recognize the decrease in the stimulus by adjusting the sleep-induced stimulus. The method of claim 13,
The model storage unit,
EEG is measured for a plurality of users, but the measured EEG is classified according to the user's age group, and the properties of the EEG are extracted from the classified EEG by using the power spectrum of the frequency band. A sleep inducing apparatus characterized by deriving and storing a sleep prediction model representing a sleep-wake state according to attribute values for each age group of a user using a machine learning algorithm. The method of claim 14,
The model storage unit,
The ratio of the average and standard deviation of the power spectrum values of the frequency band from the brain waves by age group, and the ratio of each average value according to the EEG combination of alpha / beta / delta / theta By extracting the properties of the EEG by calculating, it is characterized by selecting a property having the highest odds ratio (odds ratio) of the sleep prediction model as a property having a relatively high effect on sleep-wake state prediction using a logistic regression algorithm Sleep induction device. The method of claim 13,
The processing unit,
The user-specific information including the age of the subject is input through the input unit, or the brainwaves of the subject measured through a pretest process are input, and from the sleep prediction model pre-built for a plurality of users to the subject And a corresponding sleep prediction model. The method of claim 13,
The processing unit,
The brain wave of the subject responding to the sleep-induced stimulus is measured to calculate an average of power spectrum values of each frequency band, and the sleep prediction model according to the ratio of the power spectrum average value of the alpha-wave band and the power spectrum average value of the theta-wave band Sleep inducing device, characterized in that to determine the sleep-wake state of the subject. The method of claim 13,
The processing unit,
By detecting the time point when the sleep-wake state enters the elevation pattern from the awakening pattern, and generating a feedback signal that reduces the stimulus causing awakening from the sensed time point, performing gradual signal control until it approaches the sleep pattern. A sleep-inducing device characterized in that the sleep-inducing stimulus is controlled, and the progress of the sleepiness state is checked through a ratio of the average power spectrum of the alpha-wave band and the average power spectrum of the theta-wave band. The method of claim 13,
The processing unit,
When the sleep-wake state of the subject reaches the sleep pattern, the feedback is stopped, and the sleep guidance device is characterized by updating the sleep guidance feedback method by measuring the time required for the subject to sleep. The method of claim 13,
The sleep induction stimulus includes any one of sound, light or vibration,
The feedback signal is a sleep inducing device, characterized in that to induce the subject to be aware of the reduction in stimulation by changing any one of the intensity, frequency, period, type or type of stimulation.

Images