JP2009153550A - Sleep determining apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enhance a precision of the determination of a sleep state in a sleep determining apparatus for determining the depth of sleep of a subject. <P>SOLUTION: A processing part 42 of the sleep condition determining apparatus calculates a ratio of a pulse rate to amplitude of a pulse wave at a pulse rate-amplitude ratio calculating part 54. From the calculated result, a noise component which is checked by a pulse wave amplitude trend calculating part 52 and is caused by the variability of the pulse wave is eliminated at an amplitude processing part 57. In addition, a body motion of the subject is detected by a body motion sensor, and the noise caused by the body motion is eliminated at a body motion processing part 56. A sleep condition determining part 58 compares data after eliminating the noise with a threshold value prepared in advance, and determines that the sleep condition is "deep", "slightly deep" or "light". <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a sleep determination device that determines a sleep depth of a subject.

As biological information, the pulse rate, the heart rate, and the like are known, and a method for checking changes in the sleep state by acquiring such information has been developed. For example, in the sleep state determination method disclosed in Patent Document 1, a change in heart rate is examined, a trend curve representing an increase / decrease trend in time series is created, and an increase in heart rate relative to the trend curve is calculated. Further, the degree of variation is obtained using the difference between the measured heart rate and the heart rate measured one minute before. A sleep state is determined by comparing a sleep state appearance probability density distribution table prepared in advance with a combination of an increase in heart rate and a variation degree. In addition, as a use application of the sleep state determined in this way, for example, the indoor environment such as indoor brightness, sound, temperature, and humidity can be controlled corresponding to the sleep state, and more comfortable sleep can be realized. It is possible to do so.
JP 2001-61820 A

During sleep, body movement may be caused by unconsciously hitting and turning. When body movement occurs, noise is superimposed on the output signal of the sensor that measures the heart rate. Such noise affects the increase in the heart rate and the degree of variation. In the sleep determination method disclosed in Patent Literature 1, the determination result of the sleep state has changed discontinuously due to a temporary increase in the heart rate considered to be due to body movement.
This invention is made in view of such a situation, and makes it a main objective to improve the determination accuracy of a sleep state.

The invention according to claim 1 of the present invention that solves the above-described problems includes a pulse wave measurement unit that calculates a pulse rate and a pulse wave amplitude, a pulse wave amplitude trend calculation unit that calculates a temporal change of the pulse wave amplitude, and sleep From a pulse rate / amplitude ratio calculation unit that calculates a ratio of a pulse rate to a pulse wave amplitude as a depth determination value, a sleep depth determination value, and a time change of the pulse wave amplitude determined by the pulse wave amplitude trend calculation unit A sleep determination device including a sleep state determination unit that determines sleep depth.
This sleep determination device pays attention to the fact that the pulse rate decreases or becomes constant when the sleep is deep, and the pulse wave amplitude increases or becomes constant, and by calculating the ratio between the two, the living body that changes depending on the sleep depth Capture the behavior of information more clearly.

The invention according to claim 2 is the sleep determination device according to claim 1, wherein the body movement determination unit that determines the presence or absence of body movement of the subject and the body movement determination unit detect the body movement and then sleep at that time. A body motion processing unit that corrects the sleep depth determination value so as to remove noise caused by body motion from the depth determination value.
This sleep determination device removes the noise component due to body movement and eliminates the sleep depth so that erroneous determination does not occur when the pulse rate and pulse wave amplitude are calculated in a state where noise due to body movement is included. judge.

According to a third aspect of the present invention, in the sleep determination device according to the first or second aspect, when the time change of the pulse wave amplitude calculated by the pulse wave amplitude trend calculating unit tends to decrease, Sleep depth so that noise due to amplitude change is removed from the sleep depth judgment value at the time when the pulse wave amplitude decreases during the time when the pulse wave amplitude tends to increase. It has the amplitude processing part which corrects the thickness judgment value.
For example, the sleep determination device calculates an average value of the pulse wave amplitude for a predetermined time, and examines the tendency of the change of the pulse wave amplitude from the time change of the average value. Furthermore, when the trend of the change in pulse wave amplitude is increasing or decreasing, if the data of one or two average values show the opposite behavior to the overall trend along the way, the data is Let it be noise.

According to a fourth aspect of the present invention, in the sleep determination device according to any one of the first to third aspects, the sleep state determination unit uses a sleep depth determination value calculated as a rest state before falling asleep as a threshold value. It is determined that the sleep state is deep when the sleep depth determination value calculated after falling asleep is equal to or less than a threshold value and the temporal change of the pulse wave amplitude is increasing or substantially constant.
This sleep determination device improves the determination accuracy of the sleep depth even when there is an individual difference by determining the sleep depth based on the situation before falling asleep.

The invention according to claim 5 is the sleep determination apparatus according to claim 2, wherein an optical type that measures a pulse rate and a pulse wave amplitude in a case in which a finger can be inserted from the tip to the first joint and the tip is closed. Sensor unit having a body motion sensor for detecting body motion and a transmission / reception unit for delivering the output of the optical sensor to the pulse wave measurement unit and delivering the output of the body motion sensor to the body motion determination unit It is characterized by having.
In this sleep determination device, the case whose tip is blocked shields external light, and the pulse wave measurement accuracy by the optical sensor is improved. Since a body motion sensor is also provided in the case of the sensor unit, it is possible to measure pulse waves and body motions simply by attaching the sensor unit to a finger.

  According to the present invention, the ratio of the pulse rate to the pulse wave amplitude is used as a value for determining the sleep depth, so that data more suitable for the change in the sleep depth of the subject can be obtained, and the sleep depth can be obtained. It becomes possible to determine with high accuracy.

The best mode for carrying out the invention will be described in detail with reference to the drawings.
FIG. 1 shows a schematic configuration of the sleep determination device. The sleep determination device 1 includes a sensor unit 2 attached to a human body, a transmission / reception unit 3 connected to the sensor unit 2, and a display terminal 4 capable of communicating with the transmission / reception unit 3.

As shown in the usage pattern in FIG. 2, a fingertip is inserted into the sensor unit 2. A cable 6 extends from the sensor unit 2 and is connected to the transmission / reception unit 3. The transmission / reception unit 3 is attached to the wrist.
The sensor unit 2 has a cylindrical case 12 provided with one insertion slot 11 for inserting a finger, and the distal end side of the case 12 is closed. The length of the case 12 is such a length that the first joint (also referred to as a distal interphalangeal joint) can be fixed from the tip of the finger when the finger is inserted so that the fingertip hits the innermost part of the case 12. . As the material constituting the case 12, a material that absorbs or blocks light is used. When the surface of the case 12 is treated with black coating or the like, light transmission can be further prevented.

  As shown in FIG. 3, a sheet 13 is attached to the insertion slot 11 so as to partially block the insertion slot 11. The sheet 13 is fixed to the peripheral edge of the insertion slot 11 and extends to a substantially central portion of the insertion slot 11 and is elastically deformed according to the shape of the finger when the finger is inserted. Such a sheet 13 is manufactured from a material capable of shielding light from outside the case 12. Further, the central portion of the sheet 13 is cut out in a substantially V shape. This facilitates the insertion of the finger and facilitates the close contact with the finger, improving the light shielding property.

As shown in FIGS. 4 and 5, an anti-slip material 14 is attached in the case 12 in the vicinity of the insertion slot 11 at a position opposite to the attachment position of the sheet 13, that is, a position on the belly side of the finger. The anti-slip material 14 prevents the finger from coming off the case 12 during measurement, and is manufactured from, for example, a resin film. A pulse wave sensor 15 is disposed on the innermost side of the anti-slip material 14. The pulse wave sensor 15 is disposed so as to be located at the center of the terminal node of the finger when the finger is abutted against the innermost part. The pulse wave sensor 15 is an optical sensor having a light receiving element 17 mounted on a substrate 16 and a light emitting element 18 having an irradiation angle of 45 to 90 ° with respect to the light receiving element 17. The light emitting element 18 is an LED (light emitting diode) that emits infrared light or red light. The light receiving element 17 is disposed toward the internal space of the case 12, and for example, an optical frequency conversion element (LFC element) is used. Since the light receiving element 17 outputs a digital signal whose frequency changes according to the magnitude of the pulse pressure, the circuit configuration is simplified and noise resistance is improved as compared with the case where a light receiving element that outputs an analog signal is used. Thus, it becomes unnecessary to consider the saturation of the amplitude of the pulse wave signal and the delay due to the filter time constant. In order to improve the mounting property of the sensor unit 2, the light receiving element 17 is covered with a resin film having high transparency and flexibility such as silicone resin.
Further, a body motion sensor 19 is mounted on the back surface of the substrate 16. As the body motion sensor 19, a sensor that outputs a signal according to the movement of the body of the subject, such as an acceleration sensor, is used.

  The power supply to the pulse wave sensor 15 and the body motion sensor 19 and the extraction of signals from these sensors 15 and 19 are performed through the cable 6 through the hole 20 passed through the case 12. The hole 20 opens to the bottom side of the case 12 separately from the insertion port 11.

An elastic body 21 is disposed on the opposite side of the pulse wave sensor 15, that is, on the inner surface of the upper wall of the case 12. The elastic body 21 has a function of contacting the back of the finger and bringing the finger into close contact with the pulse wave sensor 15. The elastic body 21 is gradually increased in thickness from the insertion port 11 side toward the innermost side, that is, the amount of protrusion to the inner hole of the case 12 is increased. Such an elastic body 21 is designed such that the mounting load gradually decreases from the tip of the finger toward the base, improving the adhesion of the finger to the pulse wave sensor 15 and the feeling of pressure when the sensor is mounted. Reduces both.
A lateral slip prevention material 22 is attached to the left and right inner walls of the case 12. The lateral slip prevention material 22 contacts the finger so as not to laterally shift in the left-right direction, and is made of, for example, an elastic body. The lateral slip prevention member 22 is provided as a pair spaced apart from each other on the insertion portion side and the innermost side, but may be a pair of continuous members.
A light shielding film (not shown) such as aluminum is attached to the inner wall of the case 12 to block light other than the light emitting element 18. The light shielding film may be another metal foil, a layer formed by applying a light shielding paint, or a vapor deposition film.

As shown in FIG. 2, the transmission / reception unit 3 is attached with a band 31 near the wrist, and is connected to the sensor unit 2 with a cable 6. The band 31 is provided with a locking portion such as a hook-and-loop fastener (not shown) so that the transmission / reception unit 3 can be easily attached and detached and the tightening strength of the band 31 can be adjusted.
As shown in FIG. 1, the transmission / reception unit 3 includes a CPU (central processing unit) 32 to which a signal from the pulse wave sensor 25 is input, an amplifier 33 that amplifies the signal from the body motion sensor 19, and noise from the amplified signal. A filter 34 to be removed, a battery 35 such as a battery that supplies power to the sensor unit 2, and a wireless communication module 36 are included.

The CPU 32 performs light emission control of the light emitting element 18 of the pulse wave sensor 15. Further, a signal output from the light receiving element 17 is received, a pulse wave signal is generated by applying an edge process or a digital filter, and a process of passing to the wireless communication module 36 is performed. In addition, after the signal output from the body motion sensor 19 is amplified by the amplifier 33, the body motion signal from which noise has been removed by the filter 34 is input, and therefore, the signal is delivered to the wireless communication module 36 in synchronization with the pulse wave signal.
The wireless communication module 36 has a function of transmitting a pulse wave signal and a body motion signal to the display terminal 4 using a known short-range wireless communication protocol.
In addition, when a memory is provided in the transmission / reception unit 3 and the memory is used as a buffer, communication with the display terminal 4 can be performed smoothly.

  The display terminal 4 includes a wireless communication module 41 that performs communication with the transmission / reception module 36, a processing unit 42 that receives data from the wireless communication module 41 and performs data processing, and a display unit 43 that displays data. Have. The display terminal 4 can use a dedicated device or a general-purpose computer, and is realized by causing a computer or the like to execute a sleep determination program. In addition to the display unit 43 or instead of the display unit 43, a device that outputs data to a recording medium such as paper or a device that outputs data to another device may be provided.

  FIG. 6 shows a block diagram in which the processing unit is divided into functions. The pulse wave measurement unit 51 acquires a pulse wave signal along a time series, and calculates the pulse rate and the amplitude of the pulse wave. The pulse wave amplitude trend calculation unit 52 calculates a section average value of pulse wave amplitude data, and creates pulse wave amplitude trend data as biological information trend data. The pulse rate trend calculation unit 53 calculates a section average value of the pulse rate data, and creates pulse rate trend data as biometric information trend data. The pulse rate / amplitude ratio calculation unit 54 calculates the sleep depth determination value by dividing the pulse rate trend data by the pulse wave amplitude trend data. The body movement determination unit 55 to which body movement signals are input in time series determines the presence or absence of body movement and outputs the determination result. This determination result is input to the body motion processing unit 56 and used for correcting the sleep depth determination value calculated by the pulse rate / amplitude ratio calculation unit 54. The amplitude processing unit 57 further corrects the sleep depth determination value using the pulse wave amplitude trend data. The sleep state determination unit 58 determines the sleep state using the sleep depth determination value after noise removal, and outputs the sleep state to the display unit 43.

Next, the sleep determination process in this embodiment will be described in detail.
As shown in FIG. 2, the transmission / reception unit 3 is attached to the hand, a finger is inserted into the sensor unit 2, and the fingertip is abutted against the innermost part. The anti-slip material 14, the elastic body 21, and the lateral displacement prevention material 22 cause the finger pad to be in close contact with the light receiving element 17, and external light is shielded by the sheet 13 and the like.
When pulse measurement is started, the light emitting element 18 emits light based on a command from the CPU 32, and the light reflected by the blood vessel of the fingertip is received by the light receiving element 17. The light receiving element 17 outputs a signal having a higher frequency as the amount of received light increases. Further, the body motion signal output from the body motion sensor 19 is amplified by the amplifier 33, noise is removed through the filter 34, and then input to the CPU 32. The body motion signal is, for example, as shown in FIG. The horizontal axis is the elapsed time, and the vertical axis is the signal intensity. The peak that appears in the middle is caused by body movement. These signals are transmitted to the display terminal 4 so as to be synchronized in the acquired order.

The display terminal 4 passes the received signal to the processing unit 42. In the processing unit 42, the body motion signal is processed by the body motion determination unit 55. The body movement determination unit 55 determines the presence or absence of body movement from the change in the signal. For example, in the example of FIG. 7, it is determined that body movement has occurred in the peak portion.
On the other hand, the pulse wave signal is input to the pulse wave measuring unit 51, and the pulse rate and the pulse wave amplitude are calculated. For example, the pulse wave signal is captured for a certain period of time to create time-series data, and noise superimposed on the pulse wave by frequency filtering is reduced and low-frequency swell is reduced. Then, the pulse rate is calculated by examining the period of the pulse wave with the data from which noise has been removed. Further, the pulse wave amplitude is calculated by examining the minimum value and the maximum value of the signal intensity of the data.

The pulse wave amplitude trend calculation unit 52 creates pulse wave amplitude trend data by taking an average of intervals for a predetermined time, for example, every 5 minutes, for pulse wave amplitude data. FIG. 8 shows an example of pulse wave amplitude trend data. In FIG. 8, a graph indicated by a thick line indicates pulse wave amplitude trend data, and a graph indicated by a thin line indicates data before taking the section average. The pulse wave amplitude repeatedly increases and decreases over time.
The pulse rate trend calculation unit 53 creates pulse rate trend data by taking an average of intervals for a predetermined time, for example, every 5 minutes, for the pulse rate data. FIG. 9 shows an example of pulse rate trend data. In FIG. 9, a graph indicated by a thick line indicates pulse rate trend data, and a graph indicated by a thin line indicates data before taking the section average. The pulse rate has decreased since the start of measurement, and a small increase or decrease has occurred around 50 minutes. As described above, since the frequency filter process is performed and the averaging process is performed, a steep peak such as noise does not occur. In general, the deeper the sleep, the smaller the pulse rate, and it is considered that the deepest sleep comes immediately after falling asleep. In this case, it is considered that the person has fallen asleep in the vicinity of 50 minutes.

The pulse rate / amplitude ratio calculation unit 54 obtains the values of the respective trend data in synchronization, divides the value of the pulse rate trend data by the value of the pulse wave amplitude trend data, and calculates the pulse rate / amplitude ratio (= sleep depth). (Determination value) is calculated.
Since the sleep depth determination value data includes a noise component due to body movement, the body movement processing unit 56 removes noise using the determination result of the body movement determination unit 55. That is, when the body motion determination unit 55 detects body motion, the sleep depth determination value data at that time is regarded as noise, and the data is removed.
After removing noise due to body movement, the amplitude processing unit 57 removes noise that can be read from the pulse wave amplitude. In this case, for example, the tendency of change of 10 pieces of the trend data of the pulse wave amplitude obtained by taking a 5-minute interval average is examined, and one or two pieces of data in which the pulse wave amplitude decreases during the increase tendency. If there is one or two data in which the pulse wave amplitude increases in the course of the decrease, these data are regarded as noise, and the data of the sleep depth judgment value at that time is removed.

  In this way, the sleep state determination unit 58 determines the sleep depth using the sleep depth determination value from which noise has been removed. Here, the sleep depth is classified into three types: “deep”, “slightly deep”, and “shallow”. Note that “slightly deep” refers to a state between “deep” and “shallow”. The sleep state determination unit 58 has two threshold values K1 and K2 as a reference when making these three determinations. The first threshold value K1 is an average value of the pulse rate / amplitude ratio when the person is in a resting state immediately after entering the bed. The average value of the pulse rate / amplitude ratio when in a resting state immediately after entering the floor is the time period during which the body motion signal is stable (for example, T1 in FIG. 7). ) Pulse rate and pulse wave amplitude ratio, for example, an average value for 5 minutes is adopted. The second threshold value K2 is a value obtained by halving the first threshold value K1. Since the threshold values K1 and K2 are calculated from the data in the resting state before falling asleep, accurate determination can be made even when there are individual differences.

Here, the criteria for determining each sleep depth will be described.
The determination criterion for the sleep state “deep” is that the pulse rate / amplitude ratio is equal to or lower than the second threshold K2 and the pulse wave amplitude trend is increased or constant.
The criterion for determining the “slightly deep” sleep state is when the pulse rate / amplitude ratio is equal to or less than the first threshold K1 and the pulse wave amplitude trend is increased or constant.
The criterion for determining whether the sleep state is “shallow” is when the pulse rate / amplitude ratio is greater than or equal to the first threshold value K1.

  These criteria are set by paying attention to the fact that a deep sleep state is a relaxed state and a shallow sleep state is a relatively tense state. The pulse rate indicates a change in the movement of the heart and decreases when relaxed and increases when tensioned. For this reason, it decreases or becomes constant when sleep is deep, and increases when sleep is shallow. The magnitude of the pulse wave amplitude increases when relaxed and decreases when nervous. For this reason, it increases or becomes constant when the sleep is deep, and decreases when the sleep is shallow. In this way, the pulse rate and the pulse wave amplitude show opposite behavior depending on the sleep state, but if the pulse rate / amplitude ratio is adopted as an index, the value of the ratio decreases when sleep is deep (that is, when relaxing) If the sleep is shallow (that is, when the person is nervous), the value of the ratio increases, so that the sleep state can be grasped.

FIG. 10 shows a graph for explaining the effect of this embodiment. This graph is output data of the pulse rate / amplitude ratio calculation unit 54 and corresponds to data in a state where correction based on body motion and pulse wave amplitude is not performed.
The small peak at the position marked with a triangle in FIG. 10 could be removed by correction of the body motion processing unit 56. In addition, the small peak at the position marked with the rhombus mark could be removed by the correction of the amplitude processing unit 57. As a result of removing the peaks at the positions marked with triangles and rhombuses as noise, the peaks exceeding the second threshold K2 are in the time zone enclosed by the square in FIG. These time zones are determined as “slightly deep” and “shallow” by the sleep state determination unit 58. When this determination result was compared with a time zone in which the sleep estimated from the electroencephalogram measurement was shallow, both agreed well.

  In this embodiment, by using the pulse rate / amplitude ratio as an index for determining the sleep state, it has become possible to determine the depth of sleep that has been difficult to determine by the variation or increase / decrease of the pulse rate. In addition, by removing the noise due to body movement, the consistency with the sleep state determined from the electroencephalogram could be improved. Furthermore, since the sleep depth judgment value is calculated by dividing the pulse rate by the pulse wave amplitude, the judgment result is likely to be affected by the error of the pulse wave amplitude. The consistency with the sleep state determined from the electroencephalogram could be further improved.

Note that the present invention can be widely applied without being limited to the above-described embodiment.
For example, the second threshold value K2 is not limited to ½ times the first threshold value K1. If a learning function for adjusting the coefficient using data immediately before waking up or immediately before awakening is provided, the accuracy of sleep determination can be further improved.
Data exchange between the sensor unit 2 and the transmission / reception unit 3 may be performed wirelessly. Data exchange between the transmission / reception unit 3 and the display terminal 4 may be performed by wire.

It is a block diagram which shows schematic structure of the sleep state determination apparatus which concerns on embodiment of this invention. It is a figure which shows the usage pattern of a sleep state determination apparatus. It is an external view of a sensor unit. It is sectional drawing along the AA line of FIG. It is sectional drawing along the BB line of FIG. It is the block diagram which divided the processing part of the display apparatus into functions. It is a graph which shows an example of a body movement signal. It is a graph which shows an example of pulse wave amplitude trend data. It is a graph which shows an example of pulse rate trend data. It is a graph for demonstrating the effect of this Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Sleep determination apparatus 2 Sensor unit 3 Transmission / reception unit 4 Display apparatus 12 Case 15 Pulse wave sensor (optical sensor)
DESCRIPTION OF SYMBOLS 19 Body motion sensor 51 Pulse wave measurement part 52 Pulse wave amplitude trend calculation part 54 Pulse rate / amplitude ratio calculation part 55 Body motion determination part 56 Body motion processing part 57 Amplitude processing part 58 Sleep state determination part K1 1st threshold value

Claims (5)

A pulse wave measurement unit for calculating a pulse rate and a pulse wave amplitude;
A pulse wave amplitude trend calculating unit for calculating a time change of the pulse wave amplitude;
A pulse rate / amplitude ratio calculator that calculates a ratio of the pulse rate to the pulse wave amplitude as a sleep depth determination value;
A sleep state determination unit that determines a sleep depth from a sleep depth determination value and a time change of the pulse wave amplitude determined by the pulse wave amplitude trend calculation unit;
A sleep determination device comprising: A body motion determination unit that determines the presence or absence of body motion of the subject;
When the body movement determination unit detects body movement, a body movement processing unit that corrects the sleep depth determination value so as to remove noise caused by body movement from the sleep depth determination value at that time;
The sleep determination apparatus according to claim 1, comprising:   When the pulse wave amplitude calculated by the pulse wave amplitude trend calculator tends to decrease, the pulse wave amplitude increases midway, or when the pulse wave amplitude time change tends to increase 3. An amplitude processing unit that corrects a sleep depth determination value so as to remove noise due to an amplitude change from the sleep depth determination value at that time when the amplitude decreases. The sleep determination apparatus according to.   The sleep state determination unit holds the sleep depth determination value calculated in the resting state before falling asleep as a threshold, the sleep depth determination value calculated after falling asleep is equal to or less than the threshold, and the time change of the pulse wave amplitude increases. The sleep determination device according to any one of claims 1 to 3, wherein the sleep state is determined to be deep when the tendency or the state is substantially constant. A sensor unit in which a finger can be inserted from the tip to the first joint and an optical sensor for measuring the pulse rate and pulse wave amplitude and a body motion sensor for detecting body motion are disposed in a case where the tip is closed;
A transmission / reception unit that delivers the output of the optical sensor to the pulse wave measurement unit, and delivers the output of the body motion sensor to the body motion determination unit;
The sleep determination device according to claim 2, comprising:

Images