JP5180599B2 - Method and system for determining sleep state - Google Patents

Description

  The present invention relates to a sleep state determination method and system based on a body motion signal generated by breathing motion, and relates to, for example, a sleep state determination method and system for determining a sleep state based on harmonic distribution in the frequency spectrum of the body motion signal.

  The sleep state is divided into a REM sleep period (REM) in which the brain is awake while the body is sleeping, and a non-REM sleep period (NON-REM) in which the brain is also resting. Furthermore, it is known that biological information changes according to the depth of sleep state, and a method for determining the depth of sleep state based on biological information such as skin temperature, body movement, heart rate, etc. is proposed. (For example, patent document 1).

  An apparatus that enables determination of a sleep stage using only a respiratory signal is also proposed. For example, Patent Document 2 discloses a respiratory signal detection unit that detects a variation in a respiratory signal of a human body, and a variation in the respiratory signal. A sleep stage determination means for determining a sleep stage using, preferably, the sleep stage determination means for each unit section set by dividing the fluctuation of the respiratory signal every predetermined time. A sleep stage determination apparatus characterized by determining the above is disclosed.

In addition, the sensor for detecting body movement is composed of a piezoelectric film flexed in a bridge shape proposed by the present inventor and a member that can be expanded and contracted shorter than the piezoelectric film connecting both ends of the piezoelectric film. There is a body motion detection sensor in which the bending of a piezoelectric film changes following the expansion and contraction of a member that can be expanded and contracted (Patent Document 3).
Japanese Patent Laid-Open No. 2004-254827 JP 2006-280686 A JP 2006-340820 A

The device disclosed in Patent Document 1 cannot determine the sleep stage using only the respiratory signal.
Further, the device disclosed in Patent Document 2 enables determination of the sleep stage in the supine state, and does not enable determination of the sleep state in the standing state or the sitting state.

  The present invention relates to a method and system for determining a sleep state using a body motion detection sensor, which makes it possible to determine a sleep step using only a respiratory signal and to determine a sleep step in a non-supposed state. The purpose is to provide.

According to a first aspect of the present invention, a body motion signal from a breathing motion in a unit time is stored in a storage unit by a body motion detection sensor, and a calculation unit obtains a frequency spectrum of the body motion signal stored in the storage unit, and a harmonic in the frequency spectrum A sleep state determination method using a computer for determining a non-sleep state and a sleep state based on a wave distribution, wherein the calculation unit is configured to perform REM sleep based on a distribution of second and third harmonics in a frequency spectrum. A sleep state determination method characterized by determining non-REM sleep.
According to a second invention, in the first invention, the calculation unit further determines the non-sleep state and the sleep state based on the intensity of the body motion signal stored in the storage unit .
A third invention is a sleep state determination method using a computer, in which a body motion signal from a breathing motion in a unit time is stored in a storage unit by a body motion detection sensor, and a calculation unit The peak period is calculated, the period is within a certain range, the stationary region exists in the waveform in the middle of the upper peak and lower peak in the stored body motion signal, and the intensity of the stored body motion signal is A sleep characterized by determining a non-sleep state and a sleep state based on a combination of being equal to or greater than a reference value and the presence of a second or third harmonic in the frequency spectrum of a stored body motion signal This is a state determination method.
The fourth invention is characterized in that, when the arithmetic unit is determined to be in the sleep state, the REM sleep and the non-REM sleep are determined based on the distribution of the second harmonic and the third harmonic in the frequency spectrum. This is a method for determining a sleep state.
According to a fifth aspect of the present invention, there is provided a body motion detection sensor, a signal conversion means for converting a charge induced by the body motion detection sensor into a voltage signal and converting the voltage signal into a digital signal, and a display means for displaying a signal from the signal conversion means. A sleep state determination system comprising: a calculation unit that performs the sleep state determination method according to any one of the first to fourth aspects of the invention based on a signal from the signal conversion means.
6th invention has the piezoelectric film arrange | positioned including the part bent in the cover member, and the band member connected to the piezoelectric film, and it can carry it by mounting | wearing a human trunk with a band member. A body motion detection sensor, a signal conversion means for converting a charge induced by the body motion detection sensor into a voltage signal and converting it into a digital signal, a display means for displaying a signal from the signal conversion means, and a signal conversion means A sleep state determination system including a calculation unit that performs a sleep state determination method based on a signal,
The body motion signal is stored by the body motion detection sensor in the unit time standing or sitting subject's breathing motion, the peak period in the stored body motion signal is calculated, and the period is in a certain range, There is a static region in the waveform of the region between the upper peak and the lower peak in the stored body motion signal, the strength of the stored body motion signal is greater than the reference value, and the frequency spectrum of the stored body motion signal The sleep state determination system is characterized in that a non-sleep state and a sleep state are determined based on a combination of the presence or absence of the second or third harmonic.
7th invention WHEREIN: When it determines with a sleep state in 6th invention, REM sleep and non-REM sleep are determined based on the distribution of the 2nd harmonic and the 3rd harmonic in the said frequency spectrum, And

According to the present invention, it is possible to determine a sleep state based on a change in respiration.
Moreover, it becomes possible to determine a sleep stage by analyzing the waveform of the respiratory signal at the time of sleep.

In the best form of the sleep state determination method according to the present invention, the bent piezoelectric film collects body motion data due to respiration by measuring a change in voltage caused by the body motion of the subject, and the collected body motion The non-sleep state (wake state) and the sleep state are determined based on the data.
The determination of the non-sleep state and the sleep state can be broadly divided into two methods: (1) a method for determining based on body movement collection data based on breathing motion and (2) a method for determining based on the frequency spectrum of the collected body motion data. is there.
The present invention is also characterized in that (3) the REM sleep state and the non-REM sleep state are determined based on the distribution of the second harmonic and the third harmonic in the frequency spectrum.

In order to be able to determine the sleeping state in the standing or sitting state in the outside, the body motion detection sensor includes a main memory unit and a power source, and can be worn and carried on the body.
As shown in FIG. 1, a variety of preferable body motion detection sensors are composed of a piezoelectric film bent in a bridge shape, and an elastic member that is shorter than the piezoelectric film connecting both ends of the piezoelectric film and is in contact with the subject. A body motion detection sensor in which the flexure of the piezoelectric film changes following the expansion and contraction of the possible member is exemplified.
One end of the piezoelectric film is connected to a band member inserted into the cover member, and is expanded and contracted as the subject moves. For example, when attached to the chest or abdomen, the air in the lungs is exhaled during exhalation, and the chest circumference / reincarnation becomes smaller. When the chest circumference / reincarnation becomes larger by being sucked, the piezoelectric film is pulled and stretched.

  Since the piezoelectric film is a capacitor, ideally, the voltage remains output if there is no discharge. For example, when PVDF is used, the voltage is large and becomes several volts or more. At the time of discharging, the differential characteristics can be obtained by converting the discharge current into a voltage via a resistor. Here, according to the type and shape (size and thickness) of the piezoelectric film to be used, the resistor can output a voltage as high as possible and obtain a signal having a necessary frequency. Choose the best one.

  The material of the piezoelectric film is not particularly limited as long as it exhibits a piezo effect, but a material that is light and flexible is preferable in order to enhance the wearing feeling. Piezoelectric ceramics such as PZT are well known as piezo materials, but PVDF, which is lightweight, flexible and has good workability, is a preferred material. PVDF has a feature that it has a very wide response band and is difficult to have a specific resonance frequency. Since PVDF is not suitable for use in a high temperature environment, a thin film or a thick film using a piezoelectric ceramic is used when it is used in such an environment.

  The piezoelectric film is preferably protected against noise by shielding with a conductive cloth tape. Conductive cloth tape can be a general one used for shielding electromagnetic waves and static electricity of electronic equipment, shielding of signal cables and connectors, and the adhesive surface is also conductive and conductive even when bonded. A shielding effect can be obtained.

  The piezoelectric film is preferably housed in a case member as shown in FIG. This is because when the piezoelectric film is housed in the cover member, even if the sensor portion is attached to the chest, the film is not directly in contact with the body, so that noise caused by the heartbeat can also be suppressed.

The body motion detection sensor is connected to a measurement circuit such as a charge amplifier or FET (field drop transistor), and converts the amount of charge induced in the piezoelectric film into a voltage signal. By passing through the measurement circuit, it can be output only when the piezoelectric film charge is induced, so it can be made less susceptible to noise caused by changes in body position and noise from the heartbeat. Judgment can be made instantly.
The analog voltage signal of the measurement circuit is converted into a digital signal (A / D conversion), output on the screen, and transmitted to a dedicated program of the computer, where the presence or absence of body movement is determined. When the voltage signal is below the reference value, it is counted that there is no body movement, and when there is no body movement for a certain period of time, an alarm can be issued to notify the abnormality. In order to realize a portable monitor system, it is preferable to use a small charge amplifier as the measurement circuit.

  Hereinafter, details of the present invention will be described with reference to examples, but the present invention is not limited to these examples.

  As shown in FIG. 1, the body motion detection sensor of Example 1 can be expanded and contracted shorter than the piezoelectric film 2 formed of a 28 μm-thick PVDF film bent in a bridge shape and the piezoelectric film 2 connecting both ends of the piezoelectric film 2. A body motion detection sensor having a sensor unit 1 that includes an elastic member 3 and is fixed to a subject by band members 31 and 32 so that the bending of the piezoelectric film 2 changes following the expansion and contraction of the elastic member 3. It is. Although the cover member is not shown in FIG. 1, the sensor unit 1 is actually housed in the cover member as shown in FIG.

  The circuit unit 4 provided adjacent to the sensor unit 1 has a main storage unit, and the stored data can be transferred to a personal computer (PC) by a USB cable. The electric charge generated according to the respiratory signal in the piezoelectric film 2 is converted into a voltage by the charge amplifier of the circuit unit 4. The voltage signal is A / D converted, and 15.26 digital signals are output to the PC per second. Data was imported into dedicated software (HULINKS FlexPro7) installed on the PC, and the correlation between the condition of the subject and the waveform signal was verified based on the output waveform.

The body movement due to the breathing movement of the subject in the sitting position wearing the body movement detection sensor on the chest was measured.
FIG. 3A is a measurement graph of body movement due to respiration during non-sleeping, and FIG. 3B is an enlarged view of the dotted line in FIG.
The non-sleep state (during activity) is characterized in that the amplitude suddenly increases or decreases. It has been confirmed that the amplitude becomes irregular not only during work but also during conversation.

FIG. 4A is a measurement graph of body movement due to breathing during sleep (non-REM), and FIG. 4B is an enlarged view of the dotted line portion of FIG.
The feature of breathing during sleep is that the amplitude period is constant, and that a static region can be observed as shown by the dotted line in FIG.

From FIG. 3 and FIG. 4, it has confirmed that a sleep state and a non-sleep state can be determined by combining the following conditions independently or in combination.
(A) The peak period in the stored body motion signal is calculated, and the non-sleep state and the sleep state can be determined based on whether the period is in a certain range. In FIG. 3, the period of both the upper peak and the lower peak is not constant, and in FIG. 4, the period of both the upper peak and the lower peak is constant.
(A) The non-sleep state and the sleep state can be determined based on the waveform of the region between the upper peak and the lower peak in the stored body motion signal. In FIG. 4B, a stationary region surrounded by a dotted line in the middle of the upper peak and the lower peak can be observed.
(C) The output signal in the sleep state is larger than the output signal in the non-sleep state.
The analysis based on the body movement waveform can be applied to the REM sleep state (sleeping state). In the REM sleep state and the non-REM sleep state, as will be described later, a difference can be recognized when the FET analysis is performed.

The above verification was performed based on the report from the subject and analysis by the measurer of the output waveform, but it goes without saying that automatic determination can be performed by a computer by developing dedicated software. Here, the computer is a processing device including a CPU (central processing unit), a main storage device (RAM, ROM, etc.), an auxiliary storage device (HDD, etc.), and a pointing device for inputting various data and operation commands. A device includes a device (such as a mouse), a character information input device (such as a keyboard) for inputting various data and operation commands, and a display, and a so-called personal computer (PC) can be used.
In software development, it is preferable that an arbitrary reference value for distinguishing between a sleeping state and a non-sleeping state can be set, and a threshold value may be provided to prevent erroneous determination due to noise or the like.

In Example 2, (a) a non-sleep state, (b) a REM sleep state (when asleep), (c) a body movement due to breathing during a non-REM sleep is measured, and an attempt is made to determine which state it is in It was.
The graph on the left in FIG. 5 is a simulation of an ideal respiratory waveform during non-REM sleep. When this was analyzed by FET, it was found that frequency components of the fundamental wave, the second harmonic, and the third harmonic could be obtained as shown in the graph on the right side of FIG. Therefore, when the body motion due to respiration in the same subject (a) non-sleep state, (b) REM sleep state, (c) non-REM sleep state was actually measured and FET analysis was performed, FIG. The results of (c) were obtained. The other measurement conditions are the same as in Example 1.

FIG. 6A is a frequency spectrum of a respiration measurement value in a non-sleep state. Since a strong signal can be confirmed in the vicinity of the 0.3 Hz band, it can be seen that the respiratory cycle is 3.3 seconds. In FIG. 6A, the second harmonic and the third harmonic cannot be detected.
FIG. 6B is a frequency spectrum of the respiration measurement value in the REM sleep state. In FIG. 6B, the second harmonic can be detected with a signal intensity of about 80% of the fundamental wave, and the third harmonic can be detected with a signal intensity of about 50% of the fundamental wave.
FIG.6 (c) is a frequency spectrum of the respiration measurement value of a non-REM sleep state. In FIG. 6C, the second harmonic and the third harmonic can be detected with a strong signal.

From the above, for example, by setting the reference value to 60 to 70% of the signal intensity of the fundamental wave, when neither the second harmonic nor the third harmonic can be detected, it is determined as a non-sleep state, It is disclosed that when only the second harmonic can be detected, it is determined as a REM sleep state, and when the second harmonic and the third harmonic can be detected, it is determined as a non-REM sleep state.
Note that the determination accuracy may be increased by combining the determination method of the present embodiment and the determination method of the first embodiment.

In Example 3, an attempt was made to determine the sleep state by measuring the body movement due to breathing of the subject who is in the passenger seat of the automobile. The instrument used for the measurement is the same as in Examples 1 and 2.
FIG. 7A is a measurement graph of body movement due to respiration during non-sleeping. In FIG. 7A, it can be confirmed that the body movement is irregular and the output is larger than that in FIG.
FIG. 7B shows the FET analysis result (frequency spectrum) of FIG. In FIG. 7B, the second harmonic and the third harmonic cannot be detected.
FIG. 8A is a measurement graph of body movement due to respiration during sleep (non-REM). In FIG. 8A, it can be confirmed that the body movement is regular and the output is smaller than that in FIG.
FIG. 8B shows the FET analysis result (frequency spectrum) of FIG. In FIG. 8B, the second harmonic cannot be detected, but the third harmonic can be detected.

The graph of FIG. 8 indicates that the subject is in non-REM sleep according to the self-report of the subject, but since the second harmonic cannot be detected, the REM sleep state and the non-REM sleep state are shown. It is difficult to judge. However, it was confirmed that it was possible to determine at least whether it was a non-sleep state or a sleep state.
Assuming the case of a driver such as a car, there is no real advantage in determining the REM sleep state and the non-REM sleep state, and it is sufficient if the sleep state and the non-sleep state can be determined. Therefore, when either the second harmonic or the third harmonic can be detected, it is determined that the sleep state is in a broad sense, and preferably, in combination with the signal intensity of the body movement measurement data, the non-sleep state And determining a sleep state.

  INDUSTRIAL APPLICABILITY The present invention can be used not only for use in hospitals and the like, but also for detecting the falling asleep of drivers such as cars and trains.

It is a schematic block diagram of the body movement detection sensor used for the method of Example 1. FIG. 2 is a photograph of a body motion detection sensor used in the method of Example 1. It is drawing which shows the body movement measurement result by the respiration at the time of non-sleep of Example 1. FIG. It is drawing which shows the body movement measurement result by the breathing at the time of sleep of Example 1. FIG. (A) It is drawing which shows the simulation result of the body movement measurement by respiration at the time of non-REM sleep, and the FET analysis result (frequency spectrum) of (b) and (a). It is drawing which shows the FET analysis result (frequency spectrum) of the respiration measurement value of Example 2 at the time of (a) non-sleep, (b) REM sleep, and (c) non-REM sleep. It is drawing which shows the (a) respiration measurement result at the time of non-sleep of Example 3, and the FET analysis result (frequency spectrum) of (b) and (a). It is drawing which shows the respiration measurement result at the time of sleep of Example 3, and the FET analysis result (frequency spectrum) of (b) and (a).

Explanation of symbols

1 Sensor part (cover member)
2 Piezoelectric film 3 Elastic member 4 Circuit part 31, 32 Band member

Claims (7)

The body motion signal from the breathing motion in the unit time is stored in the storage unit by the body motion detection sensor, and the calculation unit obtains the frequency spectrum of the body motion signal stored in the storage unit , and non-based on the harmonic distribution in the frequency spectrum. A sleep state determination method using a computer for determining a sleep state and a sleep state,
A sleep state determination method, wherein the calculation unit determines REM sleep and non-REM sleep based on a distribution of second harmonics and third harmonics in a frequency spectrum. Furthermore, a calculating part determines a non-sleep state and a sleep state based on the intensity | strength of the body motion signal memorize | stored in the memory | storage part, The sleep state determination method of Claim 1 characterized by the above-mentioned. A method for determining a sleep state using a computer,
The body motion signal from the breathing motion in the unit time is stored in the storage unit by the body motion detection sensor, and the calculation unit calculates the peak period in the stored body motion signal and stores that the period is in a certain range. There is a stationary region in the waveform in the middle of the upper and lower peaks in the body motion signal, the strength of the stored body motion signal is greater than the reference value, and the frequency spectrum of the stored body motion signal is A method for determining a sleep state, wherein a non-sleep state and a sleep state are determined based on a combination of the presence of second or third harmonics. The sleep state according to claim 3, wherein when the calculation unit is determined to be a sleep state, the sleep state of claim 3 is determined based on a distribution of second harmonics and third harmonics in the frequency spectrum. Judgment method.   A body motion detection sensor, a signal conversion means for converting a charge induced by the body motion detection sensor into a voltage signal and converting it into a digital signal, a display means for displaying a signal from the signal conversion means, and a signal conversion means A sleep state determination system comprising: a calculation unit that performs the sleep state determination method according to any one of claims 1 to 4 based on a signal. A body motion detection sensor having a piezoelectric film disposed including a bent portion in the cover member, and a band member connected to the piezoelectric film, and can be carried by being worn on a human torso by the band member; The signal induced by the body motion detection sensor is converted into a voltage signal and converted into a digital signal, a display means for displaying the signal from the signal converting means, and a sleep state based on the signal from the signal converting means. A sleep state determination system including a calculation unit that performs a determination method,
The body motion signal is stored by the body motion detection sensor in the unit time standing or sitting subject's breathing motion, the peak period in the stored body motion signal is calculated, and the period is in a certain range, There is a static region in the waveform of the region between the upper peak and the lower peak in the stored body motion signal, the strength of the stored body motion signal is greater than the reference value, and the frequency spectrum of the stored body motion signal A sleep state determination system, wherein a non-sleep state and a sleep state are determined based on a combination of the presence of a second or third harmonic.   The sleep state determination according to claim 6, wherein when it is determined as a sleep state, REM sleep and non-REM sleep are determined based on a distribution of second harmonics and third harmonics in the frequency spectrum. system.