JP6518056B2 - Sleep state determination device, sleep state determination method and program - Google Patents

Description

  The present invention relates to a sleep state determination device and the like.

  2. Description of the Related Art Conventionally, there is known a technique for determining a sleep state by a sleep polysomnograph (PSG) from heart rate / respiration information. In recent years, among sleep states, there is a need to determine REM sleep and non-REM sleep by methods other than PSG.

  For example, an invention is disclosed that determines REM sleep and non-REM sleep using a standard deviation of heart rates every minute (see, for example, Patent Document 1).

  There is also known a sleep state estimation device that estimates a sleep state from the variation coefficient of variance of peak intervals of the respiratory waveform and the variation coefficient of variance of peak values (see, for example, Patent Document 2).

  There is also known an apparatus for determining a sleep state in stages by utilizing a respiration rate, a heart rate and the like (see, for example, patent documents 3 and 4).

Patent No. 3733133 gazette Patent No. 4551148 Patent No. 4582642 Patent No. 4356680

  Although the prior art uses respiratory cycle / respiratory interval and respiratory peak value for evaluation, there is a problem that it is difficult to accurately detect the sleep state from the respiratory waveform because it is input together with various noises .

  Further, depending on the mounting position in the wearable sensor, and the relative positional relationship between the sensor in the sleep position and the human body in the non-wearable sensor, the respiration waveform changes separately from the factor that is inherent to the human body. When using a large amount of measurement information, if all measurement items can be measured with high accuracy, the accuracy of the final evaluation result may be improved, but if there are measurement items with low accuracy, the measurement items concerned There is a problem that the accuracy of the final evaluation result is lowered.

  For example, in Patent Document 1, Patent Document 2, and Patent Document 3, there is a problem that individual differences and judgment errors are likely to occur because of being based on the respiration waveform. In Patent Document 4, a section of 20% or more from the large standard deviation of the overnight standard deviation is determined as REM sleep, but the rem appearance rate is not necessarily 20%, and a large error occurs.

  Furthermore, in patent document 1, although shallow sleep and deep sleep were discriminate | determined, it was not necessarily able to determine non-REM sleep and rem sleep.

  In view of the above-described problems, the present invention aims to simplify and accurately perform non-REM sleep easily and accurately by using, for example, the respiration rate within a predetermined time as biological information within a predetermined time of a measurement subject. It is providing the sleep state determination apparatus etc. which can perform determination of REM sleep.

In order to solve the problems described above, the sleep state determination device of the present invention is
Respiration detection means for detecting respiration as biological information of the subject;
A respiration rate output unit that outputs a respiration rate included in the respiration rate calculation time from the detected respiration;
Respiratory rate variation coefficient calculation means for calculating a respiratory rate variation coefficient using an average value of the respiration rates included in a variation coefficient calculation section and a standard deviation of the respiration rates;
An average respiratory rate variation coefficient calculating means for calculating an average respiratory rate variation coefficient which is an average value of the respiratory rate variation coefficients from bedtime to getting up;
Sleep state determining means for determining the sleep state of the subject by comparing the respiratory rate variation coefficient with the average respiratory rate variation coefficient;
And the like.

The sleep state determination method of the present invention is
A respiration detection step of detecting respiration as biological information of the subject;
A respiration rate output step for outputting a respiration rate included in the respiration rate calculation time from the detected respiration;
A respiratory rate variation coefficient calculating step of calculating a respiratory rate variation coefficient using an average value of the respiration rates included in a variation coefficient calculation section and a standard deviation of the respiration rates;
An average respiratory rate variation coefficient calculating step of calculating an average respiratory rate variation coefficient which is an average value of the respiratory rate variation coefficients from bedtime to getting up;
A sleep state determining step of determining a sleep state of the subject by comparing the respiratory rate variation coefficient with the average respiratory rate variation coefficient;
And the like.

  According to the sleep state determination apparatus of the present invention, the respiration of the subject is detected, and the respiration rate included in the respiration rate calculation time is output from the detected respiration. Also, the respiratory rate variation coefficient is calculated from the average value and the standard deviation of the respiratory rates included in the variation coefficient calculation section. The sleep state of the subject can be determined by comparing the respiratory rate variation coefficient with the average respiratory rate variation coefficient, which is the average value of the respiratory rate variation coefficients from bedtime to wake-up.

  Thereby, the determination of the sleep state can be performed using the respiration rate as the biological information of the person to be measured. This makes it possible to appropriately determine the sleep state by using a simple device that calculates the respiration rate.

It is a figure for demonstrating the whole in 1st Embodiment. It is a figure for demonstrating the function structure in 1st Embodiment. It is a figure which shows an example of a data structure of the respiration rate table in 1st Embodiment. It is a figure showing an example of data composition of a variable table in a 1st embodiment. It is a figure which shows the data flow of the area respiration rate output process in 1st Embodiment. It is a figure which shows the data flow of the sleep state determination processing in 1st Embodiment. It is a figure for demonstrating the agreement of the determination in 1st Embodiment. It is the figure which showed the respiration rate in 1st Embodiment, a coefficient of variation, and the plot of the determination result. It is a figure for demonstrating the agreement of the determination in 1st Embodiment. It is a figure for demonstrating the agreement of the determination in 2nd Embodiment. It is the figure which showed the respiration rate in a 2nd embodiment, a coefficient of variation, and a plot of a decision result. It is a figure for demonstrating the agreement of the determination in 2nd Embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that the present embodiment describes an example of the invention, and does not limit the contents of the invention.

[1. First embodiment]
[1.1 System appearance]
First, the first embodiment will be described. FIG. 1 is a diagram for explaining how to use the sleep determination system. As shown in FIG. 1, the sleep determination system 1 includes a processing unit 5 for processing a value output from the detection device 3 placed between the floor of the bed 10 and the mattress 20 and the detection device 3. It is configured to be equipped. The detection device 3 and the processing device 5 constitute a sleep state determination device.

  When the subject P is present on the mattress 20, the detection device 3 detects body movement (vibration) of the subject P via the mattress 20. Then, based on the vibration detected as the biological information, the respiration (the respiration rate within a predetermined time) of the person to be measured P is detected (output). In the present embodiment, the detected respiration (number) is transmitted to the processing device 5 (for example, a computer etc.) via radio, but it is integrally formed by providing the display device etc. in the detection device 3, for example It is also good. In addition, since the processing device 5 may be a general-purpose device, the processing device 5 is not limited to an information processing device such as a computer, and may be configured by a device such as a tablet or a smartphone.

  Here, the detection device 3 is configured in a sheet shape so as to be thinner. As a result, even when placed between the bed 10 and the mattress 20, the subject P can be used without feeling discomfort.

  In addition, the detection apparatus 3 should just be able to detect respiration of the to-be-measured person P as biological information. In the present embodiment, the respiration is detected based on the body movement, but for example, the sound of breathing is detected using an audio sensor, or the body movement of the person to be measured P is detected by an infrared sensor or the like. A gauged actuator may be used.

[1.2 Functional configuration]
Subsequently, the mechanism configuration of the sleep determination system 1 will be described with reference to FIG. The sleep determination system 1 in the present embodiment is configured to include the detection device 3 and the processing device 5, and each functional unit (process) may be realized by any means other than the respiration detection unit 200.

  In the sleep determination system 1, a respiration detection unit 200, a storage unit 300, an input unit 400, and a display unit 500 are connected to the control unit 100 via a bus.

  The control unit 100 is a functional unit for controlling the operation of the sleep determination system 1 and is configured by a control circuit necessary for the sleep determination system 1 such as a CPU. The control unit 100 realizes various processes by reading and executing various programs stored in the storage unit 300. In the present embodiment, the control unit 100 operates as a whole, but is provided in each of the detection device 3 and the processing device 5.

  The respiration detection unit 200 is a functional unit for detecting the respiration of the subject. In this embodiment, it is a sensor part for detecting a vibration. The respiration of the subject is detected from the vibration (body movement) detected by the sensor unit. In addition, by using the sensor unit, it is possible to detect body motions other than the respiration of the person to be measured P, such as turning over or heartbeat.

  The respiration detection unit 200 in the present embodiment detects, for example, the vibration (body movement) of the subject using a pressure sensor and detects respiration from the vibration, but the load sensor detects the center of gravity of the subject (body movement) The respiration may be detected by a change in the above, or may be detected based on the sound picked up by the microphone by providing the microphone. That is, in order to detect the respiration rate of the subject, any conventional sensor may be used.

  The storage unit 300 is a functional unit that stores various data and programs for the sleep determination system 1 to operate. The control unit 100 realizes various functions by reading and executing the program stored in the storage unit 300. Here, the storage unit 300 is configured of, for example, a semiconductor memory, a magnetic disk device, or the like.

  Further, in the storage unit 300, a respiration rate table 302 and a variable table 304 are stored. Further, a respiratory rate output program 310 and a sleep state determination program 312 are stored as programs.

  The respiration rate table 302 is a table that calculates and stores the respiration rate per unit time from the respiration detected based on the respiration detection unit 200. Here, as the unit time, the time of the section for calculating the respiration rate is set such as “20” seconds, “30” seconds, “1” minutes, and the like. In the present embodiment, as shown in FIG. 3, the respiration rate every 60 seconds is detected and stored.

The variable table 304 is an area in which various coefficients and variables are stored. For example, the coefficient α (for example, “1.2”) stored in FIG. 4 and the average respiratory rate variation coefficient C (for example, “0.06”) are stored. In addition to this, temporarily used variables are also stored.

  The average respiratory rate variation coefficient C stores the average respiratory rate variation coefficient in one night (one sleep). The coefficient C is calculated from the respiratory rate in one night, but may be calculated and updated constantly, for example, based on the acquired value.

  The input unit 400 is a functional unit for the measurer to instruct / operate on the sleep determination system 1. For example, it is configured by an operation button, a touch panel, a voice input device, and the like.

  The display unit 500 is a functional unit for displaying the sleep state and the evaluation, and displaying the operation of the sleep determination system 1. For example, it is comprised by display apparatuses, such as a liquid crystal display.

[1.3 Flow of processing]
Subsequently, a flow of processing of the sleep determination system 1 in the present embodiment will be described.

[1.3.1 Breathing number output processing]
First, a respiration rate output process for outputting the respiration rate of the subject will be described with reference to FIG. The respiration rate output process is a process that is implemented by reading out the respiration rate output program 310 in FIG.

  The respiration detection unit 200 detects respiration (step S102). Subsequently, it is determined whether the respiratory rate calculation time has elapsed (step S104). Here, when the respiration rate calculation time has elapsed (step S104; Yes), the number of respirations detected in the respiration rate calculation time is output as the respiration rate to the respiration rate table 302 (step S106). This process is repeatedly executed until the respiratory rate output process is completed (Step S108; No → Step S102).

[1.3.2 Sleep state determination processing]
Continuing on, sleep state determination processing will be described using FIG. The sleep state determination process is a process that is realized by the sleep state determination program 312 in FIG. 2 being read out and executed by the control unit 100.

  First, the respiration rate An is calculated by executing the above-described respiration rate calculation process (step S202). Here, the calculated respiration rate is the respiration rate per unit time.

  Subsequently, a respiratory rate variation coefficient Bn is calculated from the calculated respiratory rate (step S204). In the method of calculating the respiratory rate variation coefficient Bn, first, the respiratory rate average value and the standard deviation for each section of the respiratory rate variation coefficient calculation section are calculated. That is, the respiratory rate average value and the respiratory rate standard deviation at time n are calculated based on a plurality of sections before and after a certain time n, and a number obtained by dividing the standard deviation by the average value is calculated as the respiratory rate variation coefficient Bn.

For example, a method of calculating including seven sections before and after (7 minutes before and after) will be described. First, the respiration rate average value Xn of the respiration rate variation coefficient calculation section is calculated from the respiration rate An of each section.
Next, a standard deviation Yn is calculated from the respiration rate An.
In addition, although a denominator calculates a standard deviation using n-1 here, you may calculate using n.
Thereby, the respiratory rate variation coefficient Bn is calculated. The time for calculating the coefficient of variation is usually 3 minutes to 25 minutes.

  At this time, if one section is for 20 seconds, Bn is calculated from the respiration rate in the strong two minutes before and after. Further, if one section is one minute, Bn is calculated from the respiration rate in the seven minutes before and after. The number of sections for calculating Bn may be any number of sections. That is, it is only necessary to calculate the respiration rate variation coefficient Bn based on the respiration rate before and after the time for determining the sleep state.

  Next, an average respiratory rate variation coefficient C is calculated (step S206). Here, the average respiratory rate variation coefficient C is calculated as the average value of the respiratory rate variation coefficients Bn overnight. That is, for the respiratory rate variation coefficient Bn calculated in step S204, an average value for one night is calculated and updated.

Next, the calculated respiratory rate variation coefficient Bn is compared with the average respiratory rate variation coefficient C (step S208). Here, for adjustment, the average respiratory rate variation coefficient C multiplied by the coefficient α is compared with the respiratory rate variation coefficient Bn.

  At this time, if Bn is C × α or more, it is determined as REM sleep as a sleep state (step S208; Yes → step S210). Otherwise, it is determined as non-REM sleep (step S212).

Here, values of “1” to “1.5” are usually used as the coefficients . In the case of “1”, a large amount of REM sleep is detected, and in the case of “1.5”, a small amount of REM sleep is detected. It is preferable to set an appropriate coefficient according to the condition of the subject.

For example, when the subject has a sleep disorder such as an apnea disorder, disorder of breathing occurs when an apnea occurs (as an example, an increase in respiratory rate due to a return to breathing, etc.). In this case, the determination of REM sleep increases. By setting a high coefficient for such a subject, it is possible to more appropriately determine the sleep state.

  Note that the determination of the sleep state may not be performed in the awake state. That is, when the subject is determined to be in the awake state, it may be determined to be the awake state as the sleep state. Here, whether or not the subject is in the awake state can be determined based on the respiration rate, or the awake state can be separately determined by the body movement detection sensor.

[1.4 Example]
Then, the Example using the sleep determination system 1 in this embodiment is demonstrated. In the present embodiment, as shown in FIG. 1, an apparatus using a sheet type vibrometer (SBV) for detecting respiration is placed under the subject, and breathing is performed from the body motion of the subject. Detect and execute.

  FIG. 7 is a diagram for explaining the situation in which the expert determines the sleep state using a sleep polysomnograph (PSG) and a sleep polysomnograph (PSG) determined using the present embodiment (SBV) .

  This data is obtained by acquiring two night data of six subjects. The basic characteristics of the subject are 2 males and 4 females, age 30-40 years, height 155-170 cm, weight 49-62 kg.

  FIG. 7 shows how each of the awakening (Wake), the REM sleep (REM), and the non-REM sleep (NREM) were determined in the data of the 12 nights (6 persons × 2 nights). Here, those in which the determination result using the method of the present embodiment and the determination result of the sleep state in the case of using PSG are identical match in the Wake138 + REM898 + NREM2873 = 3909 section among all 5227 sections, and the matching rate It shows a high agreement rate of 74.8%.

This is an average value, and for example, it is possible to further improve the coincidence rate by adjusting the coefficient α for each individual.

  FIG. 8 (a) is a diagram in which the respiratory rate per unit time in one subject is plotted overnight, where the unit time is one minute. FIG. 8 (b) is a view showing the relationship between the respiratory rate variation coefficient and the average respiratory rate variation coefficient per unit time, and FIG. 8 (c) shows that the sleep state is determined from the respiratory rate variation coefficient. FIG. Here, in FIG. 8C, a thin line indicates the determination of the sleep state by PSG, and a circle (o) indicates three stages of the determination of the sleep state according to the present embodiment "Wake, Rem sleep ( "REM), non-REM sleep (NREM)" is shown. The thin lines indicate awakening (Wake) when using PSG, REM sleep (REM), and non-REM sleep (Stage 1 to Stage 4). Further, FIG. 9 is a table showing the coincidence of each sleep state determined by PSG and SBV in the example shown in FIG.

  Of the total 484 sections, the judgment by PSG and SBV in the Wake13 + REM146 + NREM237 = 396 section is in agreement, showing an extremely high coincidence rate of 81.8%. As described above, according to the present embodiment, it is possible to appropriately determine the sleep state such as REM sleep and non-REM sleep of the measurement subject despite being a simple device.

[2. Second embodiment]
Subsequently, the second embodiment will be described. The second embodiment has the same functional configuration and processing as the first embodiment, but describes a case where not the coefficient of variation but the standard deviation is used as the determination value.

  That is, the determination is made based on the comparison between the standard deviation Yn calculated in the first embodiment and the average standard deviation which is the average for one night. Specifically, the standard deviation Yn is calculated in step S204 of FIG. 6, and the average standard deviation Cy is calculated overnight in step S206.

Then, in step S208, it is determined whether or not the standard deviation Yn is the average standard deviation Cy or more, and it is determined whether the sleep state is REM sleep or non-REM sleep. Here, as in FIG. 6, the average standard deviation Cy may be multiplied by a predetermined coefficient .

  An example of the case of determination in the second embodiment will be described with reference to the drawings. FIG. 10 is a diagram for explaining the situation in which the expert determines the sleep state using a sleep polysomnograph (PSG) and a sleep polysomnograph (PSG) that is determined using the present embodiment (SBV) .

  This data is obtained by acquiring two night data of six subjects. The basic characteristics of the subject are 2 males and 4 females, age 30-40 years, height 155-170 cm, weight 49-62 kg.

  FIG. 10 shows how each of the awakening (Wake), the REM sleep (REM), and the non-REM sleep (NREM) were determined in the data of the 12 nights (6 persons × 2 nights). Here, those in which the determination result using the method of the present embodiment and the determination result in the sleep state when using PSG are identical are coincident in the Wake138 + REM904 + NREM2893 = 3935 section among all 5227 sections, and the coincidence rate It shows a high agreement rate of 75.3%.

  FIG. 11 (a) is a diagram in which the respiration rate per unit time in a certain subject is plotted overnight, where the unit time is one minute. FIG. 11 (b) is a diagram showing the relationship between the standard deviation per unit time and the average standard deviation, and FIG. 11 (c) is a diagram showing that the sleep state is determined from the standard deviation. Here, in FIG. 11C, a thin line indicates the determination of the sleep state by PSG, and a circle (o) indicates three stages of the determination of the sleep state according to the present embodiment “awake (Wake), rem sleep ( "REM), non-REM sleep (NREM)" is shown. The thin lines indicate awakening (Wake) when using PSG, REM sleep (REM), and non-REM sleep (Stage 1 to Stage 4). Further, FIG. 12 is a table showing the coincidence of each sleep state determined by PSG and SBV in the example shown in FIG.

  The determination by PSG and SBV is in agreement in Wake13 + REM153 + NREM240 = 406 sections among all 484 sections, and the agreement rate 83.9% and a very high agreement rate are shown. As described above, according to the present embodiment, by using the standard deviation, it is possible to appropriately determine the sleep state such as REM sleep and non-REM sleep of the measurement subject despite being a simple device. .

[3. Modified example]
The embodiment of the present invention has been described in detail with reference to the drawings, but the specific configuration is not limited to this embodiment, and the design and the like within the scope of the present invention are also claimed. include.

  In addition, in embodiment mentioned above, although it demonstrated using a respiration rate as an example of biometric information, you may evaluate similarly using a heart rate as biometric information of a to-be-measured person. That is, by using the detected respiration as the heartbeat and the respiration rate as the heart rate, it is possible to perform the same sleep determination.

  By using heartbeats (numbers) instead of respirations (numbers) as biological information, it is possible to appropriately evaluate sleep using a conventional heartbeat detection device such as an electrocardiograph, for example. There is.

Reference Signs List 1 sleep determination system 3 detection device 5 processing device 10 bed 20 mattress 100 control unit 200 respiration detection unit 300 storage unit 302 respiration rate table 304 variable table 310 respiration rate output program 312 sleep state determination program 400 input unit 500 display unit

Claims (4)

Respiration detection means for detecting respiration as biological information of the subject;
A respiration rate output unit that outputs a respiration rate included in the respiration rate calculation time from the detected respiration;
Respiratory rate variation coefficient calculation means for calculating a respiratory rate variation coefficient using an average value of the respiration rates included in a variation coefficient calculation section and a standard deviation of the respiration rates;
An average respiratory rate variation coefficient calculating means for calculating an average respiratory rate variation coefficient which is an average value of the respiratory rate variation coefficients from bedtime to getting up;
Sleep state determination means for determining whether the subject is in the sleep state as REM sleep or non-REM sleep by comparing the respiratory rate change coefficient with the average respiratory rate change coefficient;
A sleep state determination apparatus comprising: The sleep state determining means, the respiratory rate variation coefficient, and the average respiratory rate variation coefficient 1 to 1 . The sleep state determination device according to claim 1, characterized in that when it is equal to or greater than a value obtained by multiplying a coefficient within a range of 5, it is determined that REM sleep, and other than that is non-REM sleep. A respiration detection step of detecting respiration as biological information of the subject;
A respiration rate output step for outputting a respiration rate included in the respiration rate calculation time from the detected respiration;
A respiratory rate variation coefficient calculating step of calculating a respiratory rate variation coefficient using an average value of the respiration rates included in a variation coefficient calculation section and a standard deviation of the respiration rates;
An average respiratory rate variation coefficient calculating step of calculating an average respiratory rate variation coefficient which is an average value of the respiratory rate variation coefficients from bedtime to getting up;
A sleep state determination step of determining whether the subject's sleep state is REM sleep or non-REM sleep by comparing the respiratory rate change coefficient with the average respiratory rate change coefficient;
A sleep state determination method comprising: A program that causes a computer to realize the sleep state determination method according to claim 3.