Apnea determination device

Abstract

To provide an apnea determination device which has a novel structure and accurately determines an apnea determination state of a sleeper.SOLUTION: There is provided an apnea determination device 10 for determining an apnea state of a sleeper A, the device comprises: (a) a body motion detector 12 for detecting as a respiratory signal, input by the respiration of the sleeper A on the basis of the body motion of the sleeper A; (b) an envelope curve calculation part 46 for calculating a first envelope curve Ef in a first time window corresponding to a period of the respiration on the basis of the respiratory signal, and calculating a second envelope curve Es in a second time window corresponding to a period of apnea state; (c) and a determination part 50 for determining the apnea state of the sleeper A, on the basis of a duration d of a deviation state of the second envelope curve Es and the first envelope curve Ef, and the deviation between the second envelope curve Es and the first envelope curve Ef.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an apnea determination device for determining an apnea state of a sleeper based on the body movement of the sleeper. [Background technology]

[0002] Conventionally, a method for detecting a sleep state based on a sleeper's brain waves, eyeball reaction, etc. using a sleep polygraph has been known. Also, as disclosed in JP 2001-037742 A (Patent Document 1), an apnea determination device has been proposed that includes a sensor sheet having a plurality of pressure-sensitive elements arranged under bedding, generates a body movement signal based on the load acting on the pressure-sensitive elements of the sensor sheet, and detects an apnea state based on the body movement signal. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2001-037742 A Summary of the Invention [Problem to be solved by the invention]

[0004] Incidentally, the apnea determination device described in Patent Document 1 is designed to determine an apnea state based on a sudden increase in the amplitude of a body movement signal waveform, focusing on the fact that large breathing occurs after an apnea state. However, for example, the amplitude of the body movement signal waveform can suddenly increase when the sleeper turns over in his / her sleep, which may lead to erroneous determination of an apnea state, and therefore further improvement in determination accuracy has been desired.

[0005] Furthermore, the amplitude of the signal waveform of the body movement signal may change due to, for example, changes in the sleeping posture of the sleeper, the position on the sensor sheet, etc. Therefore, the method of Patent Document 1, which determines apnea when the amplitude change of the body movement signal waveform is greater than a threshold value, may not be able to follow the change in signal level due to the change in sleeping posture, etc., and in such a case, it is thought that a decrease in the determination accuracy may become a problem.

[0006] An object of the present invention is to provide an apnea determination device having a novel structure that can more accurately determine the apnea state of a sleeper. [Means for solving the problem]

[0007] The following describes preferred embodiments for understanding the present invention, but the embodiments described below are merely illustrative and may be combined with one another as appropriate, and the multiple components described in each embodiment may be recognized and used independently as far as possible, and may also be combined with any of the components described in another embodiment as appropriate. As a result, the present invention is not limited to the embodiments described below, and various other embodiments may be realized.

[0008] A first aspect is an apnea determination device for determining whether a sleeping person is in an apnea state, and includes a body movement detection device that detects input due to the breathing of the sleeping person as a respiratory signal based on the body movement of the sleeping person, an envelope calculation unit that calculates a first envelope in a first time window corresponding to the respiratory cycle based on the respiratory signal, and calculates a second envelope in a second time window corresponding to the apnea cycle, and a determination unit that determines whether the sleeping person is in an apnea state based on the deviation between the second envelope and the first envelope and the duration of the deviation between the second envelope and the first envelope.

[0009] In a normal breathing state, there is no large discrepancy (discrepancy state) between the first envelope of a first time window corresponding to the breathing cycle and the second envelope of a second time window corresponding to an apnea state cycle longer than the first time window, which may be judged as apnea, but in an apnea state, the discrepancy state continues. Therefore, in an apnea determination device constructed according to the present embodiment, the apnea state of a sleeper is determined based on the discrepancy between the first envelope and the second envelope and the duration of the discrepancy state between the first envelope and the second envelope, thereby preventing erroneous determination due to body movements such as turning over in bed and accurately determining the apnea state.

[0010] For example, when the input due to breathing detected by the body movement detection device changes depending on the sleeping position, etc., the values ​​of both the first envelope and the second envelope change, so that the influence on the determination based on the relative deviation between the first envelope and the second envelope is suppressed. Therefore, even if the detection level of the input due to breathing changes due to the change in the sleeping position, etc., it is possible to make an appropriate determination.

[0011] A second aspect is an apnea detection device described in the first aspect, wherein the detection unit detects an apnea state when the absolute value of the numerical value of the second envelope is greater than the absolute value of the numerical value of the first envelope.

[0012] The apnea determination device constructed according to this embodiment can achieve more accurate apnea determination. That is, in an apnea state, the input due to breathing becomes small or disappears, and the respiratory signal becomes small, so that the absolute value of the numerical value of the second envelope in the time window (long time window) corresponding to the apnea cycle becomes larger than the absolute value of the numerical value of the first envelope in the time window (short time window) corresponding to the respiratory cycle. Therefore, by adding to the requirements for apnea determination that the absolute value of the numerical value of the second envelope is larger than the absolute value of the numerical value of the first envelope, it is possible to more accurately prevent erroneous determination caused by input such as body movement.

[0013] In a third aspect, in the apnea detection device described in the second aspect, the detection unit interrupts the detection of an apnea state when the absolute value of the numerical value of the second envelope is smaller than the absolute value of the numerical value of the first envelope.

[0014] According to an apnea determination device constructed in accordance with the present embodiment, the absolute value of the numerical values ​​of the second envelope curve is greater than the absolute value of the numerical values ​​of the first envelope curve in an apnea state. Therefore, when the absolute value of the numerical values ​​of the second envelope curve is smaller than the absolute value of the numerical values ​​of the first envelope curve, the determination of the apnea state is interrupted, thereby reducing the load of the determination process.

[0015] In addition, the determination of the apnea state may be interrupted by, for example, changing the processing so that the determination of the apnea state is not performed while the absolute value of the numerical value of the second envelope is smaller than the absolute value of the numerical value of the first envelope, or by performing processing to pre-delete the portion of the respiratory signal where the absolute value of the numerical value of the second envelope is smaller than the absolute value of the numerical value of the first envelope.

[0016] A fourth aspect is an apnea determination device according to any one of the first to third aspects, wherein the first time window is set within a range of 3 seconds or more and 6 seconds or less, and the second time window is set within a range of 10 seconds or more and 120 seconds or less.

[0017] With an apnea detection device constructed according to the present embodiment, the first time window is set within the range of 3 to 6 seconds, which is the breathing cycle of an average adult, and the second time window is set within the range of 10 to 120 seconds, which is the apnea cycle of an average adult, thereby making it possible to effectively detect apnea in average people, excluding those with special breathing patterns such as infants.

[0018] In a fifth aspect, in the apnea detection device described in any one of the first to fourth aspects, a determination threshold for the deviation state between the first envelope and the second envelope is set within a range of 20% to 50% of the numerical value of the second envelope.

[0019] According to an apnea determination device constructed in accordance with the present embodiment, an apnea state is determined to exist when the difference between the numerical values ​​of the first envelope curve and the second envelope curve is greater than a determination threshold set within a range of 20% to 50% of the numerical value of the second envelope curve. This makes it possible to accurately distinguish between inputs due to normal breathing, inputs in an apnea state, and large inputs due to body movement or the like, and determine an apnea state.

[0020] A sixth aspect is an apnea determination device according to any one of the first to fifth aspects, wherein the duration for which the determination unit determines that the sleeper is in an apnea state is within a range between a lower duration threshold and an upper duration threshold.

[0021] According to an apnea determination device constructed in accordance with the present embodiment, not only a lower duration threshold but also an upper duration threshold is set as the duration threshold. Therefore, if the level of the respiratory signal changes due to, for example, a change in sleeping posture, even if the deviation between the first envelope and the second envelope continues for a long time in the transition region until the first and second envelopes follow the change, it is determined that no apnea is occurring because the duration of the deviation exceeds the upper duration threshold.

[0022] In a seventh aspect, in the apnea determination device described in any one of the first to sixth aspects, a signal rectification unit is provided that calculates a rectified respiratory signal based on the respiratory signal, and the envelope calculation unit calculates the first envelope and the second envelope based on the rectified respiratory signal.

[0023] According to the apnea determination device constructed in accordance with the present embodiment, the first and second envelopes are calculated based on the rectified respiratory signal, making it possible to more accurately determine an apnea state based on the deviation between the first and second envelopes.

[0024] An eighth aspect is an apnea determination device described in any one of the first to seventh aspects, wherein the envelope calculation unit calculates the first envelope based on a first set value that is smaller than a maximum value of the respiratory signal in the first time window, and calculates the second envelope based on a second set value that is smaller than a maximum value of the respiratory signal in the second time window.

[0025] According to the apnea determination device constructed according to this aspect, the influence of noise in the respiratory signal can be further suppressed, and apnea determination can be performed with high accuracy.

[0026] A ninth aspect is an apnea determination device described in any one of the first to eighth aspects, in which the body movement detection device is a piezoelectric sensor having a flexible sheet-like sensor sheet body that is laid on the bedding and on which the sleeper rests.

[0027] According to the apnea determination device constructed in accordance with this embodiment, the body movement detection device is composed of a piezoelectric sensor having a flexible sheet-like sensor sheet body that is laid on the bedding, so that there is no need to attach electrodes or the like to the sleeper's body, and apnea during sleep can be determined without placing a burden on the sleeper. Effect of the Invention

[0028] According to the present invention, the apnea determination device can more accurately determine the apnea state of a sleeper. [Brief description of the drawings]

[0029] [Figure 1] FIG. 1 is an explanatory diagram for explaining an apnea determination device according to a first embodiment of the present invention; [Diagram 2] FIG. 2 is an explanatory diagram for explaining a piezoelectric sensor constituting the apnea determination device of FIG. 1. [Diagram 3] FIG. 3 is a plan view showing a sensor sheet body constituting the piezoelectric sensor of FIG. 2; [Figure 4] IV-IV cross section of Figure 3 [Diagram 5]Graph showing vibration waveform including respiratory motion acquired by a piezoelectric sensor [Figure 6] A flowchart showing an example of an apnea determination process performed by an apnea determination device. [Figure 7] A graph of the respiratory component extracted from the vibration waveform in Figure 5 and converted to a logarithmic value [Figure 8] A graph showing an enlarged portion of Figure 7 [Figure 9] Graph showing changes in the waveform of the respiratory signal due to changes in sleeping position, etc. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0031] 1 shows an apnea determination device 10 according to a first embodiment of the present invention. The apnea determination device 10 determines an apnea state based on the body movements of a sleeper A while sleeping. The apnea determination device 10 includes a piezoelectric sensor 12 as a body movement detection device that receives minute body movements associated with breathing, heartbeat, etc. of the sleeper A and relatively large body movements such as turning over in bed, and outputs detection signals corresponding to the input body movements (vibrations), and an analysis device 14 that analyzes the detection signals output from the piezoelectric sensor 12.

[0032] The piezoelectric sensor 12 has a structure as shown in Fig. 2, for example, and includes a generally rectangular flexible sensor sheet body 16. As shown in Figs. 3 and 4, the sensor sheet body 16 includes a piezoelectric layer 18, a pair of electrode layers 20a, 20b, and a pair of protective layers 22a, 22b.

[0033] The piezoelectric layer 18 may be made of ceramics, synthetic resin, rubber elastic body (including elastomer), etc., but in this embodiment, it is made of a rubber elastic body. The rubber elastic body used as the material of the piezoelectric layer 18 is preferably one or more selected from crosslinked rubber and thermoplastic elastomer, such as urethane rubber, silicone rubber, nitrile rubber (NBR), hydrogenated nitrile rubber (H-NBR), acrylic rubber, natural rubber, isoprene rubber, ethylene-propylene-diene rubber (EPDM), ethylene-vinyl acetate copolymer, ethylene-vinyl acetate-acrylic acid ester copolymer, butyl rubber, styrene-butadiene rubber, fluororubber, epichlorohydrin rubber, etc. Also, an elastomer modified by introducing a functional group, etc. may be used. As the modified elastomer, hydrogenated nitrile rubber having one or more selected from a carboxyl group, a hydroxyl group, and an amino group is preferable.

[0034] The piezoelectric layer 18 also contains piezoelectric particles. The piezoelectric particles are particles of a compound having piezoelectricity. Ferroelectrics having a perovskite crystal structure are known as compounds having piezoelectricity, and one or a mixture of two or more of barium titanate, strontium titanate, potassium niobate, sodium niobate, lithium niobate, potassium sodium niobate, potassium sodium lithium niobate, lead zirconate titanate (PZT), barium strontium titanate (BST), bismuth lanthanum titanate (BLT), and bismuth strontium tantalate (SBT) can be suitably used.

[0035] The electrode layers 20a, 20b preferably have flexibility that allows them to deform in accordance with the piezoelectric layer 18. Such electrode layers 20a, 20b can be formed, for example, from a conductive material in which a conductive material is mixed with a binder, conductive fibers, or the like. As the binder, the same materials as the crosslinked rubber and thermoplastic elastomer constituting the piezoelectric layer 18 described above can be used. The electrode layers 20a, 20b can also be formed, for example, from a metal foil.

[0036] In addition, the conductive material to be mixed into the electrode layers 20a, 20b is not limited, but may be appropriately selected from metal particles made of silver, gold, copper, nickel, rhodium, palladium, chromium, titanium, platinum, iron, and alloys thereof, metal oxide particles made of zinc oxide, titanium oxide, and the like, metal carbide particles made of titanium carbonate, and the like, metal nanowires made of silver, gold, copper, platinum, nickel, and the like, carbon black, carbon nanotubes, graphite, thin-layer graphite, graphene, and other conductive carbon materials.

[0037] As the material for the protective layers 22a and 22b, the same materials as the crosslinked rubber and thermoplastic elastomer constituting the piezoelectric layer 18 described above can be used.

[0038] In this embodiment, the piezoelectric layer 18, the electrode layers 20a, 20b, and the protective layers 22a, 22b are all formed in a thin rectangular plate shape, and the electrode layers 20a, 20b are fixed to both sides of the piezoelectric layer 18 in the thickness direction, and the protective layers 22a, 22b are fixed to both sides of the piezoelectric layer 18 and the electrode layers 20a, 20b in the thickness direction. As a result, the piezoelectric layer 18 and the electrode layers 20a, 20b are embedded inside the protective layers 22a, 22b without being exposed to the outside. With this structure, the sensor sheet main body 16 is formed in a thin, approximately rectangular sheet shape.

[0039] In the widthwise central portion of the sensor sheet main body 16, the area where the piezoelectric layer 18 and the electrode layers 20a, 20b overlap in the thickness direction is designated as a pressure-sensitive portion 24, and an electric charge is generated when a load is applied to the pressure-sensitive portion 24.

[0040] In addition, the piezoelectric sensor 12 of this embodiment is equipped with a controller 26 and a connector 28, and the electrode layers 20a, 20b and the controller 26 are electrically connected by wiring 30a, 30b, and the controller 26 is electrically connected to the analysis device 14 via the connector 28.

[0041] The controller 26 has, for example, an amplifier and an A / D converter. The amplifier amplifies the output (voltage) of the piezoelectric sensor 12. The A / D converter converts the output amplified by the amplifier from an analog signal to a digital signal. However, the output of the piezoelectric sensor 12 may be converted from an analog signal to a digital signal by the A / D converter and then amplified by the amplifier. The controller 26 may also include, for example, a standardization unit 32 that performs a standardization calculation to correct the inherent characteristics of the piezoelectric sensor 12.

[0042] The analysis device 14 only needs to be capable of calculating the signal controlled by the controller 26 (i.e., the digital signal converted by the A / D conversion unit), and may be realized by, for example, a computer including a monitor unit 34 and a computer program installed in the computer to perform operations. The analysis device 14 may be a mobile device such as a tablet terminal or a smartphone. The piezoelectric sensor 12 and the analysis device 14 may be connected to each other by wire or wirelessly.

[0043] In the apnea determination device 10 including such a piezoelectric sensor 12 and an analysis device 14, for example, the piezoelectric sensor 12 is laid so as to extend in the width direction of a bed 36 as bedding on which a sleeper A lies. In particular, in this embodiment, the piezoelectric sensor 12 includes a belt portion 38 whose length is adjustable, and the belt portion 38 is wrapped around a mattress of the bed 36 to fix the piezoelectric sensor 12 on the mattress. The mattress on which the piezoelectric sensor 12 is fixed is covered with, for example, a sheet, so that the sleeper A lies on the piezoelectric sensor 12 without directly contacting the piezoelectric sensor 12. In this embodiment, the piezoelectric sensor 12 is provided near the chest of the sleeper A, so that the body movement due to breathing of the sleeper A can be easily obtained even if the posture of the sleeper A changes slightly due to, for example, turning over in bed.

[0044] When minute or large body movements (vibrations) are input to the pressure-sensitive portion 24 of the piezoelectric sensor 12 due to breathing, heartbeat, turning over, etc. of the sleeper A, an electric charge is generated in the piezoelectric layer 18, and the generated electric charge is detected by the controller 26 as a change in voltage or current. A specific example of the relationship between the electric charge (signal) detected by the controller 26 and time is shown in the graph of FIG. 5. That is, the detection signal shown in the graph of FIG. 5 is derived from body movements such as breathing, heartbeat, turning over, etc. Note that the detection signal shown in FIG. 5 is processed data that has been subjected to processing and correction such as standardization calculation by the controller 26.

[0045] Hereinafter, a specific example of a procedure for determining an apnea state of a sleeping person A using the apnea determination device 10 will be described with reference to the flowchart of Fig. 6. The apnea determination process shown in the flowchart of Fig. 6 includes a data processing part of S1 to S5 and an apnea determination part of S6 to S8.

[0046] First, in S1 of Fig. 6, the charge (signal) generated in the piezoelectric layer 18 of the piezoelectric sensor 12 is detected as a micromovement signal. This micromovement signal is original data including not only input due to breathing but also input due to bodily movements such as heartbeat and turning over in bed. The sampling rate fs of the bodily movements is not particularly limited, but is set to, for example, 100 Hz. In S1, the micromovement signal is acquired over a predetermined data acquisition time. The data acquisition time may be, for example, a preset time, or may be set based on the sleep time between falling asleep and waking determined based on the body movement waveform.

[0047] Next, in S2 of Fig. 6, a standardization operation is performed on the body movement signal in order to correct the inherent characteristics of the piezoelectric sensor 12. A standardization unit 32 that performs the standardization operation on the body movement signal may be provided in the controller 26, or in the computer of the analysis device 14. Fig. 5 shows the signal waveform after the standardization process in S2 has been performed.

[0048] Next, in S3 of Fig. 6, a bandpass filter (respiratory body movement filter) that passes the frequency components of the body movement due to respiration in the body movement signal is used to extract the components due to respiration (respiratory signal) from the body movement signal (Extraction of respiration). Generally, body vibration due to respiration is vibration of 0.08 Hz or more and 0.5 Hz or less. Therefore, by using a bandpass filter that passes the frequency components of 0.08 Hz to 0.5 Hz, the components of body movement due to heartbeat, turning over in bed, etc. are removed from the body movement signal, and the components of body movement due to respiration (respiratory signal) are extracted. However, the cutoff frequency of the bandpass filter is not limited and may be set appropriately. It should be noted that a lowpass filter may be used to extract the respiration signal instead of the bandpass filter.

[0049] The bandpass filter is provided, for example, in the controller 26 of the piezoelectric sensor 12, and the respiratory signal is acquired in the piezoelectric sensor 12. Therefore, a respiratory signal acquisition unit 42 that extracts a respiratory signal corresponding to respiration from the detection signal of the piezoelectric sensor 12 may be provided in the controller 26. Thereafter, the respiratory signal converted into a digital signal by the A / D conversion unit may be transmitted to the computer of the analysis device 14, and the subsequent calculations may be performed on the computer.

[0050] Next, in S4 of FIG. 6, a rectification process is performed to logarithmize the signal value obtained through a band-pass filter that transmits respiratory components. That is, the signal value obtained through the band-pass filter is squared, converted into a logarithmic value, and then further multiplied by 1 / 2 to perform logarithmization (rectification and log transformation) of the absolute value of the signal value to calculate a rectified respiratory signal. FIG. 7 shows a waveform of the logarithmized respiratory signal, and FIG. 8 shows an enlarged view of a part of the rectified respiratory signal in FIG. 7, including an apnea state. The signal rectification unit 44 that performs the logarithmic (rectification) process of the respiratory signal may be provided in the controller 26, and in that case, the rectified rectified respiratory signal can be transmitted to the computer of the analysis device 14, and subsequent calculations can be performed on the computer. The signal rectification unit 44 may be configured to include the computer of the analysis device 14 and a program installed on the computer, and in that case, the respiratory signal can be transmitted to the computer of the analysis device 14, and subsequent calculations, including the rectification process, can be performed on the computer.

[0051] Next, in S5 of FIG. 6, the envelope calculation unit 46 of the computer of the analysis device 14 calculates a first envelope Ef (fast upper envelope) and a second envelope Es (slow upper envelope) based on the rectified respiratory signal of FIG. 7 rectified by the signal rectification unit 44. The first envelope Ef is an envelope in a first time window that is a time window corresponding to a respiratory cycle, and is shown by a dashed line in FIG. 7 and FIG. 8. The second envelope Es is an envelope in a second time window that is a time window corresponding to an apnea cycle longer than the first time window, and is shown by a two-dot dashed line in FIG. 7 and FIG. 8. Preferably, the first time window is set within a range of 3 seconds or more and 6 seconds or less based on a typical respiratory cycle of an adult, and the second time window is set within a range of 10 seconds or more and 120 seconds or less based on a typical apnea cycle of an adult. In this embodiment, the first time window is set to 6 seconds, and the second time window is set to 60 seconds. The set values ​​of the first and second time windows are stored in the storage unit 48 of the computer of the analysis device 14, for example.

[0052] In this embodiment, the first envelope Ef is the calculation result of the first set value (95th percentile value) of the rectified respiratory signal in the range of 3 seconds before and after an arbitrary time point t (each half of the first time window). Similarly, the second envelope Es is the calculation result of the second set value (95th percentile value) of the rectified respiratory signal in the range of 30 seconds before and after an arbitrary time point t (each half of the second time window). Therefore, as shown in FIG. 7, the first envelope Ef has a higher tracking ability to the rectified respiratory signal than the second envelope Es, and the numerical change is larger than that of the gentle second envelope Es. The first and second envelopes Ef and Es are both envelopes on the upper side of the signal waveform. Note that the first envelope Ef of the 6-second window may be calculated based on the maximum value of the rectified respiratory signal in the range of 3 seconds before and after an arbitrary time point t, but it is preferable to calculate it based on a first set value smaller than the maximum value, and in this embodiment, the first set value is the 95th percentile value of the rectified respiratory signal in the first time window. Similarly, the second envelope Es of the 60-second window may be calculated based on the maximum value of the rectified respiratory signal in a range of 30 seconds before and after any time point t, but it is preferable to calculate it based on a second set value that is smaller than the maximum value, and in this embodiment, the second set value is the 95th percentile value of the rectified respiratory signal in the second time window.

[0053] 6, the judgment unit 50 of the computer of the analysis device 14 judges whether or not the difference between the numerical values ​​of the second envelope Es and the first envelope Ef at time t is greater than a preset deviation judgment threshold. The deviation judgment threshold is desirably set within a range of 20% to 50% of the numerical value of the second envelope Es, and is set to, for example, 30%. The set value of the deviation judgment threshold is stored, for example, in the memory unit 48 of the computer of the analysis device 14.

[0054] In S6, for example, if the real-value of the first envelope Ef at a certain time t is Eft and the real-value of the second envelope Es is Est, In(Est)-In(Eft)>In(1 / (1-0.3)) The deviation state is determined based on whether or not the following is satisfied. In the description of this embodiment, the values ​​of the first and second envelopes Ef and Es are, in principle, antilogarithms of logarithmic values. Also, In(Es) and In(Ef) are referred to as logarithmic values.

[0055] In addition, it is preferable to perform a process of comparing the logarithmic value of the first envelope Ef and the logarithmic value of the second envelope Es at time t prior to the judgment of S6. If the logarithmic value of the second envelope Es is greater than the logarithmic value of the first envelope Ef, the judgment of S6 (including the following S7 to S9) is performed, and if the logarithmic value of the second envelope Es is equal to or less than the logarithmic value of the first envelope Ef, the judgment of S6 is not performed, and the comparison process between the logarithmic value of the first envelope Ef and the logarithmic value of the second envelope Es is performed again with t=t+1 / fs. In this way, if the envelope deviation judgment is stopped when the logarithmic value of the second envelope Es is equal to or less than the logarithmic value of the first envelope Ef, the influence of input due to body movement and noise in the respiratory waveform can be further suppressed, and a highly accurate envelope deviation judgment can be realized. Note that t is set, for example, to 0, which is the start time of acquiring the body movement waveform in S1. The initial value of t is not particularly limited, but is set to 0, for example.

[0056] In the judgment of S6, if it is judged that the difference in logarithm between the second envelope Es and the first envelope Ef is equal to or less than the deviation judgment threshold and the first and second envelopes Ef and Es do not diverge (the degree of deviation is small), the judgment of S6 is executed again with t=t+1 / fs. Note that the magnitude of 1 / fs, which is the time interval of the judgment process, is not particularly limited, but is set to, for example, 1 second.

[0057] In the judgment of S6, when it is judged that the difference between the logarithm of the second envelope Es and the first envelope Ef is greater than the deviation judgment threshold, the duration d of the deviation state (see FIG. 8) is calculated in S7. The duration d (duration of respiratory decline) is the duration of the deviation state in which the difference between the logarithm of the second envelope Es and the first envelope Ef is greater than the deviation judgment threshold. In S7, for example, the same deviation judgment as in S6 is repeated at a predetermined time interval (1 / fs in this embodiment), and the elapsed time from the time when the first and second envelopes Ef and Es are judged to be in a deviation state in the judgment of S6 to the time when they are next judged not to be in a deviation state is set as the duration d of the deviation state. Note that the duration d can also be calculated as the product of the number of times when it is judged to be in a deviation state consecutively and 1 / fs, which is the time interval of the judgment process.

[0058] Next, in S8 of Fig. 6, the determination unit 50 determines whether the duration d is within a duration threshold range (Duration criteria). As the duration thresholds, it is desirable to set a lower duration threshold and an upper duration threshold. The lower duration threshold is preferably set within a range of 5 to 30 seconds, for example, 10 seconds. The upper duration threshold is preferably set within a range of 40 to 90 seconds, for example, 70 seconds.

[0059] In the determination of S8, if the duration d is equal to or less than the duration lower threshold or equal to or more than the duration upper threshold, it is determined that the sleeper A is not in an apneic state, and t is set to t+d, and the process from S6 onwards is executed again. Note that the duration d is reset to the initial value 0 after t=t+d is set.

[0060] In the judgment of S8, when it is judged that the duration d is greater than the duration lower limit threshold and less than the duration upper limit threshold, that is, when the duration d is between the duration lower limit threshold and the duration upper limit threshold, the judgment unit 50 judges in S9 that the sleeper A is in an apnea state (respiration decline) (respiration event). In short, the judgment unit 50 judges the sleeper A in an apnea state when the deviation between the first and second envelopes Ef and Es is greater than the deviation judgment threshold and the duration d of the deviation state is within a predetermined duration threshold range. When the judgment unit 50 judges apnea in S9, for example, an apnea response process may be executed, such as indicating that the waveform diagrams of Figs. 7 and 8 displayed on the monitor are in an apnea judgment region. The apnea state in this embodiment includes both a hypopnea state in which breathing becomes excessively shallow and a complete apnea state in which breathing stops.

[0061] After the apnea determination in S9, t is set to t+d, and the processes from S6 onward are executed again. Note that the duration d is reset to the initial value 0 after t=t+d.

[0062] The apnea determination process can be ended based on, for example, the elapsed time of the determination process, a preset end time, etc. In this embodiment, the apnea determination process ends based on the end time of acquisition of the body movement waveform in S1.

[0063] In the apnea determination process of this embodiment shown in Fig. 6, when the second envelope Es of the second time window set based on the apnea cycle and the first envelope Ef of the first time window set based on the respiratory cycle deviate from each other by a larger deviation than the deviation determination threshold of the first and second envelopes Ef, Es, and the duration of the deviation state is within a predetermined range, the sleeper A is determined to be in an apnea state. In this way, not only the first and second envelopes Ef, Es deviate from each other by a larger deviation than the deviation determination threshold, but also the duration of the deviation state is added to the conditions for apnea determination, making it easier to prevent erroneous determination due to body movement or noise, and enabling highly accurate apnea determination.

[0064] Since the logarithmic difference between the first and second envelopes Ef, Es is compared with the deviation determination threshold set based on the logarithmic value of the second envelope Es, even if there is an amplitude change in the body movement waveform caused by a change in the sleeping posture of the sleeper A as shown in Fig. 9, the effect on the logarithmic difference between the first and second envelopes Ef, Es and the deviation determination threshold is small. Therefore, apnea can be stably and accurately determined regardless of the sleeping posture of the sleeper A or the contact state with the piezoelectric sensor 12.

[0065] In this embodiment, when the second envelope Es is smaller than the first envelope Ef, the deviation determination between the first and second envelopes Ef, Es is stopped, and the deviation determination is performed only when the second envelope Es is larger than the first envelope Ef. Therefore, when the first envelope Ef temporarily becomes larger than the second envelope Es due to a large body movement or the like, it is possible to prevent an erroneous determination of the deviation state or an error due to an inability to determine the deviation state from occurring.

[0066] The duration d of the deviation state between the first and second envelopes Ef and Es is set as a condition for apnea determination to be longer than the lower duration threshold and shorter than the upper duration threshold, thereby improving the accuracy of apnea determination. That is, by setting the condition for apnea determination to be longer than the lower duration threshold, it is possible to prevent the deviation between the first and second envelopes Ef and Es due to a short-term large input due to body movement or the like from being determined as apnea. In addition, by setting the condition for apnea determination to be shorter than the upper duration threshold, it is possible to prevent the deviation from being determined as apnea when the first and second envelopes Ef and Es are in a deviation state for a long period of time due to a change in amplitude of the body movement waveform caused by a change in the sleeping position of the sleeper A as shown in FIG. 9.

[0067] The signal waveform in Fig. 9 shows a case where the patient goes from a normal sleeping state (resting state) with a large signal amplitude to a normal sleeping state with a small signal amplitude through body movement such as turning over in bed. When the patient goes from a sleeping state with a large signal amplitude to a sleeping state with a small signal amplitude in the signal waveform in Fig. 9, it is easy to erroneously determine that apnea has occurred due to the difference in amplitude of the signal waveform. However, according to the apnea determination device 10 of this embodiment, even if there is a change in the amplitude of the signal waveform as shown in Fig. 9, it is difficult to erroneously determine that apnea has occurred as described above.

[0068] Since the first and second envelope curves Ef, Es are calculated based on the 95th percentile values, which are first and second set values ​​that are smaller than the maximum values ​​in the first and second time windows, they are less likely to pick up noise that occurs as sudden extreme values ​​and are less susceptible to the effects of noise, making it possible to accurately determine apnea based on the first and second envelope curves Ef, Es.

[0069] The first time window is set within a range of 3 to 6 seconds based on a typical respiratory cycle of an adult, and the second time window is set within a range of 10 to 120 seconds based on a typical apnea cycle of an adult. This allows the apnea state of an adult to be accurately determined by the apnea determination process of this embodiment. When determining apnea of ​​an infant or a mammal other than a human, whose respiratory cycle or apnea cycle is different from that of an adult, for example, the settings of the first and second time windows may be changed.

[0070] Furthermore, by performing logarithmic processing (S4) on the detection signal (body movement signal) of the piezoelectric sensor 12 by the signal rectification unit 44 before determining whether or not an apnea state is occurring (S6, S8), the effect of body movement such as turning over on the apnea determination can be reduced, and the apnea determination can be performed with greater accuracy.

[0071] The body movement detection device that detects the body movement of the sleeper A is a flexible sheet-like piezoelectric sensor 12, and can detect the body movement of the sleeper A while it is placed on the mattress of the bed 36, without being attached to the body of the sleeper A. Therefore, it is unlikely to cause discomfort to the sleeper A and disturb his / her sleep, and it is possible to perform apnea determination in a normal sleeping state while reducing the burden on the mind and body of the sleeper A.

[0072] Although the embodiment of the present invention has been described above in detail, the present invention is not limited to the specific description. For example, in the above embodiment, the first and second envelopes Ef and Es are calculated based on the rectified respiratory signal obtained by performing logarithmic rectification on the respiratory signal extracted from the body movement signal, but such rectification is not essential. In addition, the rectification of the respiratory signal by the signal rectification unit 44 is not limited to logarithmic rectification, and may be, for example, a process of squaring the respiratory signal and then calculating the square root to make the signal output value positive (absolute value), or a process of ignoring the part where the signal output value is negative.

[0073] The portions of the respiratory waveform where the signal output value is negative may be converted to absolute values ​​by squaring the value and then calculating the square root as in the above embodiment, but for example, the first and second envelopes Ef, Es may be calculated using only the portions where the output value is positive, ignoring the portions where the signal output value is negative.

[0074] In the above embodiment, an example was shown in which the deviation between the first and second envelopes Ef, Es was determined based on the difference between the logarithmic values ​​of the first and second envelopes Ef, Es. However, for example, the first and second envelopes can be set for the output value of a respiratory signal that is not logarithmically converted, and the deviation between the first and second envelopes can be determined based on the difference or ratio between the values ​​of the first and second envelopes.

[0075] In addition, apnea determination based on the deviation between the first and second envelopes Ef, Es and the duration of the deviation state can also be made to determine apnea when, for example, the integral value of the difference between the first and second envelopes Ef, Es in the range determined to be a deviation state (the area of ​​the light-shaded portion in Figure 8) is greater than a preset apnea determination threshold.

[0076] The deviation between the first and second envelopes Ef and Es does not necessarily mean that the difference between the first and second envelopes Ef and Es at a certain time t is large. For example, when the difference between the slope of the first envelope Ef and the slope of the second envelope Es in the graph of FIG. 7 is larger than a deviation determination threshold, the first and second envelopes Ef and Es may be determined to be in a deviation state. Also, the deviation between the first and second envelopes Ef and Es may be determined based on the ratio of the values ​​of the first envelope Ef and the second envelope Es.

[0077] The first and second envelopes used in apnea determination can also be lower envelopes based on a set value on the bottom side of the signal waveform, for example, when rectification processes such as squaring, logarithmic conversion, and truncation of negative output values ​​are not performed.

[0078] In the above embodiment, an example is shown in which apnea determination is performed later based on the body movement signals of the sleeping person A acquired over a predetermined period of time. However, for example, while the sleeping person A is sleeping, the apnea state can be determined in approximately real time by executing the apnea determination part (S6 to S9 in FIG. 6) for determining apnea in parallel with the data processing part (S1 to S5 in FIG. 6) including the acquisition of the body movement signals. In this case, for example, the entire apnea determination process including not only the apnea determination part but also the data processing part such as the acquisition of the body movement signals, the extraction of the respiratory signal, and the rectification of the respiratory signal may be repeated over time. In addition, when the apnea determination process is performed in real time, it is possible to issue a warning by sound or monitor display, or send a notice such as an e-mail to a mobile device of a medical professional, etc., in the apnea response process when apnea is determined. In addition, when the apnea determination process is performed in real time, the process can be terminated based on, for example, the number of times the apnea determination is performed, the awakening detection of the sleeping person A, etc.

[0079] In the above embodiment, an example was shown in which a complete apnea state where breathing stops and a hypopnea state where breathing level is reduced are both determined as apnea without being distinguished, but a complete apnea state and a hypopnea state in an apnea state can also be distinguished. That is, in the above embodiment, only one deviation judgment threshold is set in S6 of the flow chart in FIG. 6, but a first deviation judgment threshold for judging whether or not there is an apnea state and a second deviation judgment threshold for distinguishing between a hypopnea state and a complete apnea state are set. Then, when the deviation between the second envelope and the first envelope is continuously between the first deviation judgment threshold and the second deviation judgment threshold, it is determined to be a hypopnea state, and when the deviation is continuously larger than the second deviation judgment threshold, it is determined to be a complete apnea state, thereby making it possible to distinguish between a hypopnea state and a complete apnea state. The first deviation judgment threshold corresponds to the deviation judgment threshold in the above embodiment, and is desirably set within a range of 20 to 50% of the value of the second envelope. The second deviation determination threshold is set to a value larger than the first deviation determination threshold, and is set, for example, within a range of 60 to 90% of the numerical value of the second envelope.

[0080] For example, an apnea state may be determined to exist when the duration d of the deviation state between the first and second envelopes Ef and Es is greater than the duration lower limit threshold without setting an upper limit threshold, for example. In this way, for example, apnea that continues longer than normal is also determined to exist, so that apnea states are less likely to be overlooked.

[0081] In the above embodiment, the piezoelectric sensor 12 that detects body movement based on the piezoelectricity of the piezoelectric layer 18 has been exemplified as an example of the body movement detection device, but it is also possible to employ, for example, a capacitance sensor that detects body movement based on a change in capacitance caused by the approach / separation of a pair of electrode layers 20a, 20b, or an air pressure sensor that has a bag-shaped sensor sheet body with air sealed inside and detects body movement based on a change in air pressure inside the sensor sheet body due to input. Also, for example, a sheet-shaped conductive rubber made of a rubber material mixed with a conductive filler may be applied to the sensor sheet body of the body movement detection device to detect the body movement of the sleeper A based on a change in electrical resistance caused by deformation of the conductive rubber. [Explanation of symbols]

[0082] 10 Apnea determination device (first embodiment) 12 Piezoelectric sensor (body movement detection device) 14 Analysis device 16 Sensor sheet body 18 Piezoelectric layer 20a,20b electrode layer 22a,22b Protective layer 24 Pressure sensing part 26 Controller 28 Connectors 30a, 30b Wiring 32 Standardization Department 34 Monitor section 36 Beds (Bedding) 38 Belt section 42 Breathing signal acquisition section 44 Signal rectification section 46 Envelope calculation section 48 Memory section 50 Judgment section fs Sampling rate Ef First envelope Es second envelope t time d Duration A. Sleeper

Claims

1. An apnea determination device for determining an apnea state of a sleeper, comprising: a body movement detection device for detecting an input due to breathing of the sleeping person as a breathing signal based on the body movement of the sleeping person; an envelope calculation unit that calculates a first envelope in a first time window corresponding to a period of breathing based on the respiratory signal, and calculates a second envelope in a second time window corresponding to a period of an apnea state; a determination unit that determines whether the sleeper is in an apnea state based on a deviation between the second envelope and the first envelope and a duration of the deviation between the second envelope and the first envelope; An apnea detection device comprising:

2. The apnea determining device according to claim 1 , wherein the determining unit determines an apnea state when an absolute value of the numerical value of the second envelope is greater than an absolute value of the numerical value of the first envelope.

3. The apnea determining device according to claim 2 , wherein the determining unit suspends determination of an apnea state when an absolute value of the numerical value of the second envelope is smaller than an absolute value of the numerical value of the first envelope.

4. The apnea determination device according to claim 1 or 2, wherein the first time window is set within a range of 3 seconds or more and 6 seconds or less, and the second time window is set within a range of 10 seconds or more and 120 seconds or less.

5. 3. The apnea determination device according to claim 1 or 2, wherein a threshold for determining the deviation state between the first envelope and the second envelope is set within a range of 20% to 50% of the numerical value of the second envelope.

6. The apnea determining device according to claim 1 or 2, wherein the duration for which the determining unit determines that the sleeper is in an apnea state is within a range from a lower duration threshold to an upper duration threshold.

7. A signal rectification unit that calculates a rectified respiratory signal based on the respiratory signal, 3. The apnea determining device according to claim 1, wherein the envelope calculation unit calculates the first envelope and the second envelope based on the rectified respiratory signal.

8. The apnea determination device according to claim 1 or 2, wherein the envelope calculation unit calculates the first envelope based on a first set value that is smaller than the maximum value of the respiratory signal in the first time window, and calculates the second envelope based on a second set value that is smaller than the maximum value of the respiratory signal in the second time window.

9. 3. The apnea determining device according to claim 1, wherein the body movement detecting device is a piezoelectric sensor having a sensor sheet body in the form of a flexible sheet that is laid on bedding and on which the sleeper rests.

Images