JP6909608B2 - Respiratory measuring device - Google Patents

Description

The present invention relates to a respiration measuring device that measures the respiration status of a sleeping person.

Conventionally, in the fields of medical care and long-term care, it has been practiced to measure the respiratory state at bedtime as an index for grasping the sleep state and the health state of a patient or the like. For example, Japanese Patent No. 5679971 (Patent Document 1) measures the tension in which a band wrapped around a sleeping person's chest or abdomen is stretched by a breathing motion, or a pressure-sensitive sensor on a mat laid under the sleeping person. It has been proposed to measure the pressure fluctuation accompanying the breathing motion of the sleeping person by providing the above to measure the breathing waveform of the sleeping person. Further, in Japanese Patent No. 4554476 (Patent Document 2), an optical fiber type sensor laid on a bedding plane is used, and a respiratory waveform is measured from a detected value of polarization fluctuation caused by a change in the shape of an optical fiber accompanying a respiratory motion. A device has been proposed.

By the way, in the conventional respiration measuring devices described in Patent Documents 1 and 2, the time series data of the output values of the sensors and the like that fluctuate with the respiration movement are Fourier transformed, and the period related to respiration is obtained from the obtained power spectrum. It is common practice to detect vibrations.

However, since the periodic vibration obtained by this method requires the average amplitude (frequency spectrum) over the time interval assuming that the data at a certain time interval is constant, the instantaneous vibration of the respiration is instantaneous. It does not represent a value. Therefore, it is difficult to detect a sudden change in the respiratory state, and there is a possibility that the apnea state cannot be detected effectively, and there is still room for improvement in order to grasp the accurate respiratory state.

Japanese Patent No. 5679971 Japanese Patent No. 4554476

The present invention has been made in the background of the above circumstances, and the problem of solving the present invention is to provide a respiratory measuring device having a novel structure capable of more accurately measuring changes in the respiratory state of a sleeping person. There is.

Hereinafter, aspects of the present invention made to solve such problems will be described. The components adopted in each of the following aspects can be adopted in any combination as much as possible.

A first aspect of the present invention is a pressure sensor disposed on a floor supporting a sleeping person to detect a pressure fluctuation accompanying a sleeping person's breathing motion, and a pressure fluctuation time series data detected by the pressure sensor. performs CDM processing on the frequency components associated with respiratory comprises data generating means for generating time-series data of the instantaneous frequency, and output means for outputting the data obtained by the data generating means, said The data generation means includes a normal range setting unit for setting a normal range of the instantaneous frequency and an abnormal value determination unit for determining an abnormal value when the value of the instantaneous frequency deviates from the preset normal range. and said that you are.

According to the respiration measuring device having a structure according to this aspect, the frequency component related to respiration in the pressure fluctuation time series data accompanying the respiration movement of the sleeping person detected by the pressure sensor is subjected to CDM (Complex Demodulation) in the data generation means. ) By performing the processing, the time-series data of the instantaneous frequency can be obtained, and the time-series data of the obtained instantaneous frequency can be output via the output means by an arbitrary method. As a result, it is possible to measure the instantaneous frequency related to respiration, which was not obtained by the conventional Fourier transform process, and it is possible to obtain the respiratory rate and respiratory vibration of the sleeping person as instantaneous values instead of the average value in a certain section. .. As a result, the breathing state of the sleeping person can be measured more accurately, and the sleep state is judged (REM / non-REM sleep) and the apnea state is detected instantly from the data of the breathing interval fluctuation obtained in real time. It becomes possible.

Further , according to this aspect, if the output frequency in the time-series data of the instantaneous frequency is within the range of 0.1 Hz to 0.5 Hz, the pressure fluctuation waveform due to respiration at rest is reflected, and if the output frequency deviates from that range. Based on the findings found by the present inventors that it is possible to detect the occurrence of body movements such as apnea and turning over, for example, the range of the instantaneous frequency based on the pressure fluctuation due to breathing at rest is set as the normal range, and the range is set. By determining the deviating value as an abnormal value, it is possible to obtain an accurate instantaneous respiratory rate at rest and detect the occurrence of an abnormality.

A second aspect of the present invention is the one described in the first aspect, wherein when the abnormal value determination unit exceeds the upper limit value of the normal range, the breathing of the sleeping person. It is designed to be judged as an abnormal value due to body movement other than movement.

According to this aspect, when the instantaneous frequency exceeds the normal range, it can be distinguished as a fluctuation due to body movement, and the instantaneous respiratory rate at rest can be obtained by removing such noise.

A third aspect of the present invention is the one described in the first or second aspect, wherein the data generation means further generates time-series data of instantaneous amplitude by the CDM processing, and the normal range. The setting unit further sets the average value of the instantaneous amplitude at rest, and the abnormal value determination unit determines that the value of the instantaneous frequency is equal to or less than the lower limit of the normal range and the value of the instantaneous amplitude. When is 1/3 or less of the average value at rest, it is determined to be an abnormal value due to apnea.

According to this aspect, in the data generation means, by further generating time-series data of instantaneous amplitude by CDM processing, the value of instantaneous amplitude is used, and such a value is preset in the normal range setting unit at rest. When it becomes 1/3 or less of the average value of, it can be determined that the apnea is apnea, and the occurrence of the apnea can be promptly confirmed.

In the fourth aspect of the present invention, in any one of the first to third aspects, the frequency component related to respiration in the pressure fluctuation time series data is 0.1 to 0.5 Hz. belongs to.

According to this aspect, by setting the frequency component to 0.1 to 0.5 Hz, it is possible to widely sample the frequency band related to respiration while removing the frequency component related to the heartbeat. This makes it possible to easily confirm the apnea state in addition to the instantaneous respiratory rate at rest.

A fifth aspect of the present invention is the one described in any one of the first to fourth aspects, wherein the pressure sensor has a sheet shape in which a large number of pressure sensitive points are dispersedly arranged, and the bedtime. It is configured by a multi-point detection pressure-sensitive sensor which is arranged on a support surface for supporting a person and on which an upper body extending from at least the chest to the abdomen of the sleeping person is placed.

According to this aspect, by using a sheet-shaped multi-point detection pressure sensor in which a large number of pressure-sensitive points are dispersedly arranged, the sleeping person's breathing motion such as at bedtime can be performed without restraining the sleeping person's body. Based on the pressure fluctuation, it can be obtained stably. Moreover, since the seat-shaped multi-point detection pressure-sensitive sensor is designed to mount at least the upper body of the sleeping person, which fluctuates due to the respiration movement, from the chest to the abdomen, the pressure fluctuation based on the respiration movement is the most. you can select the pressure sensing points which can advantageously be obtained the pressure fluctuation during-series data including a frequency component related to respiration can advantageously be obtained a.

In the sixth aspect of the present invention, in the fifth aspect, the data generation means includes a pressure fluctuation calculation unit, and the pressure fluctuation calculation unit includes a respiratory frequency among the large number of pressure sensitive points. The pressure fluctuation time series data is calculated from the average value of the output values obtained from the plurality of pressure sensitive points selected in descending order of the output value of the band.

According to this aspect, since the pressure fluctuation time series data is calculated based on the average value of the output values obtained from the plurality of pressure sensitive points selected in descending order of the output value of the respiratory frequency band, it is based on the respiratory motion. It is possible to advantageously acquire pressure fluctuation time series data that most favorably represents pressure fluctuation.

In the seventh aspect of the present invention, in the fifth aspect, the data generation means includes a pressure fluctuation calculation unit, and the pressure fluctuation calculation unit includes an acquisition pressure among the large number of pressure sensitive points. The pressure fluctuation time series data is calculated from the average value of the output values obtained from the plurality of pressure sensitive points selected in descending order of the ratio of the respiratory frequency band to the total area of the power spectrum of.

According to this aspect, the pressure fluctuation time series data is calculated from the average value of the output values obtained from a plurality of pressure sensitive points selected in descending order of the ratio of the respiratory frequency band to the total area of the power spectrum of the acquired pressure. Accordingly, it is possible to obtain a pressure variation time-series data including a frequency component related to respiration more reliably. The number of pressure sensitive points to be selected is preferably 2 to 5, more preferably about 4 because the parts where the body movement corresponding to breathing appears may be dispersed. It is desirable to do.

According to the present invention, the frequency component related to respiration in the pressure fluctuation time series data accompanying the respiration movement of the sleeping person detected by the pressure sensor is subjected to CDM processing in the data generation means to perform instantaneous frequency time series data. Can be obtained. As a result, the respiratory rate and respiratory vibration of the sleeping person can be obtained as an instantaneous value instead of the average value of a certain section as in the conventional case. As a result, it is possible to instantly determine the sleep state and detect the apnea state from the respiratory interval fluctuation data obtained in real time.

The perspective view which shows the bed provided with the respiration measuring apparatus as one Embodiment of this invention. Top view of FIG. The plan view which shows the respiration measuring apparatus of this embodiment. FIG. 3 is an enlarged cross-sectional view of IV-IV in FIG. The flowchart which shows the control method of this invention. A flowchart showing a pressure sensitive point selection process. A flowchart showing an abnormal value detection process. The figure which shows the pressure fluctuation time series data, the instantaneous frequency data and the instantaneous amplitude data which processed it by CDM (at rest). The figure which shows the instantaneous amplitude data and the instantaneous frequency data which CDM processed the pressure fluctuation time series data (at the time of apnea).

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

1 and 2 show a respiration measuring device 10 having a structure according to an embodiment of the present invention. The respiration measuring device 10 includes, for example, a sheet-shaped multipoint detection pressure sensor 20 as a pressure sensor arranged on a support surface 18 of a floor portion 16 of a bed 14 that supports a sleeping person 12, and a multipoint detection pressure sensitive sensor 10. It is configured to include a control device 22 electrically connected to the sensor 20. More specifically, as shown in FIGS. 1 and 2, the multipoint detection pressure sensor 20 is arranged in a part of a region on the support surface 18 of the floor portion 16 provided with, for example, a mattress. The upper body 28 extending from the chest 24 to the abdomen 26 of the sleeping person 12 is placed on the multipoint detection pressure sensor 20. Then, using the multi-point detection pressure sensor 20, pressure fluctuations associated with the breathing motion of the sleeping person 12 are detected. In the following description, the upper part means the upper part in FIG. 1, the lower part means the lower part in FIG. 1, the front part means the upper part in FIG. 2, and the rear part means the lower part in FIG. And. Further, in FIGS. 1 and 2, the sleeping person 12 is shown by a virtual line for easy understanding.

As shown in FIG. 3, the multipoint detection pressure sensor 20 constitutes a surface pressure distribution sensor by arranging, for example, strain gauges at four corners using a strain gauge, a load cell using a magnetic strain body, or the like. However, in the present embodiment, the multi-point detection pressure sensor 20 as a pressure sensor is configured by the sheet-shaped capacitance type sensor. As the capacitance type sensor, conventionally known ones can be appropriately adopted.

3 and 4 schematically show a capacitance type sensor constituting the multipoint detection pressure sensitive sensor 20. In FIG. 3, the dielectric layer 32 and the front base material 34, which will be described later, are shown through, and the pressure sensitive points 42, which will be described later, are shown by diagonally shaded squares.

The capacitance type sensor constituting the multipoint detection pressure sensitive sensor 20 includes a dielectric layer 32, front side electrodes 01X to 32X, back side electrodes 01Y to 25Y, front side wiring 01x to 32x, and back side wiring 01y to 25y. It includes a front side base material 34, a back side base material 36, a front side wiring connector 38, and a back side wiring connector 40.

The dielectric layer 32 is made of, for example, a urethane foam as an elastomer, has a substantially rectangular flat plate-shaped sheet shape in a plan view, and is elastically deformable. Further, the front base material 34 is made of rubber, for example, and has a substantially rectangular flat plate shape in a plan view, and is laminated above the dielectric layer 32 (front side). Further, the back side base material 36 is made of rubber, for example, and has a substantially rectangular flat plate shape in a plan view, and the back side base material 36 is laminated below (back side) of the dielectric layer 32.

As shown in FIG. 4, the outer edge of the front base material 34 and the outer edge of the back side base material 36 are joined, and the front side base material 34 and the back side base material 36 are bonded together in a bag shape. As a result, the dielectric layer 32 is housed in the bag. In addition, the four upper corners of the dielectric layer 32 are spot-bonded to the lower four corners of the front base material 34, while the lower four corners of the dielectric layer 32 are spot-bonded to the upper four corners of the back base material 36. Has been done. Therefore, the dielectric layer 32 is positioned with respect to the front base material 34 and the back side base material 36 so as not to wrinkle during use. The dielectric layer 32 is elastically deformable in the horizontal direction (front-back and left-right directions) with respect to the front-side base material 34 and the back-side base material 36 in a state where the four corners are adhered.

As shown in FIG. 3, a total of 32 front side electrodes 01X to 32X are arranged on the upper surface of the dielectric layer 32, and each of them is formed including, for example, acrylic rubber and conductive carbon black. Has been done. Further, the front side electrodes 01X to 32X each have a substantially band shape, and are formed to be flexibly expandable and contractible. Further, the front side electrodes 01X to 32X extend in the horizontal direction (horizontal direction in FIG. 3) and are separated from each other in the vertical direction (vertical direction in FIG. 3) at predetermined intervals. They are arranged so as to be substantially parallel.

A total of 32 front-side wirings 01x to 32x are arranged on the upper surface of the dielectric layer 32, and each of them is formed including, for example, acrylic rubber and silver powder. Further, the front side wirings 01x to 32x each have a substantially linear shape. Further, the front side wiring connector 38 is arranged at the corners of the front side base material 34 and the back side base material 36, while the front side wirings 01x to 32x are the ends of the front side electrodes 01X to 32X and the front side wiring connectors, respectively. 38 and are connected.

On the other hand, a total of 25 back side electrodes 01Y to 25Y are arranged on the lower surface of the dielectric layer 32. The back side electrodes 01Y to 25Y are formed including, for example, acrylic rubber and conductive carbon black, respectively. Further, the back side electrodes 01Y to 25Y each have a band shape, and are formed to be flexibly expandable and contractible. Further, the back side electrodes 01Y to 25Y extend in the vertical direction (vertical direction in FIG. 3) and are separated from each other in the horizontal direction (horizontal direction in FIG. 3) at predetermined intervals. They are arranged so as to be substantially parallel. As described above, the front side electrodes 01X to 32X and the back side electrodes 01Y to 25Y are arranged in a substantially grid pattern orthogonal to each other when viewed from above or below.

A total of 25 back-side wirings 01y to 25y are arranged on the lower surface of the dielectric layer 32. The back side wirings 01y to 25y are formed including, for example, acrylic rubber and silver powder, respectively. Further, the back side wirings 01y to 25y each have a linear shape. Further, the back side wiring connector 40 is arranged at the corners of the front side base material 34 and the back side base material 36, while the back side wirings 01y to 25y are the ends of the back side electrodes 01Y to 25Y and the back side wiring connectors, respectively. 40 and are connected.

The pressure sensitive points 42 are arranged at a portion (overlapping portion) where the front side electrodes 01X to 32X and the back side electrodes 01Y to 25Y intersect in the vertical direction, as shown by the diagonally shaded squares in FIG. Therefore, they are arranged substantially evenly in the vertical and horizontal directions over substantially the entire surface of the dielectric layer 32. That is, the multi-point detection pressure sensor 20 as a pressure sensor has a structure in which a large number (32 × 25 = 800 in this embodiment) of pressure-sensitive points 42 are dispersedly arranged. Further, the pressure sensitive points 42 include a part of the front side electrodes 01X to 32X, a part of the back side electrodes 01Y to 25Y, and a part of the dielectric layer 32, respectively.

As shown in FIG. 3, the control device 22 is electrically connected to the front side wiring connector 38 and the back side wiring connector 40. The control device 22 includes a CPU (Central Processing Unit) 44, a ROM (Read Only Memory) 46, a RAM (Random Access Memory) 48, a power supply circuit 50, a monitor 52, a timer, a counter, and an input / output interface (not shown). It is equipped with I / O) and the like. The ROM 46 stores a control program or the like based on a control method described later, while the RAM 48 temporarily stores the calculated value or the like of the control program. The control device 22 includes a data generation means A, a pressure fluctuation calculation unit a ₁ , a normal range setting unit a ₂ , and an abnormal value determination unit a _3, which will be described later, and these are stored in the ROM 46 and executed by the CPU 44. It is realized by the software that is used.

The power supply circuit 50 scans and sequentially applies a periodic rectangular wave voltage to the pressure sensitive point 42. Further, the ROM 46 stores in advance a map showing the correspondence between the capacitance and the body pressure (pressure) of the capacitor configured at the pressure sensitive point 42. On the other hand, the RAM 48 temporarily stores the capacitance of the pressure sensitive point 42 input from the front side wiring connector 38 and the back side wiring connector 40. Then, the CPU 44 detects the body pressure acting on the pressure sensitive point 42 from the capacitance of the pressure sensitive point 42 stored in the RAM 48 based on the map stored in the ROM 46. The work contents and measurement results of the control device 22 are output to the monitor 52.

Next, when the sleeping person 12 lies on the floor portion 16 of the bed 14 and goes to bed using the breathing measuring device 10 of the present embodiment, the breathing of the sleeping person 12 is related to the pressure fluctuation accompanying the breathing motion of the sleeping person 12. A method of obtaining instantaneous frequency data will be described.

FIG. 5 shows the processing contents of the data generation means A of the control device 22. In this control process, for example, the user operates an operation button (the operation button may be a GUI or the like displayed on the monitor 52 provided on the control device 22) provided on the control device 22 (the operation button may be a GUI or the like displayed on the monitor 52 provided on the control device 22). It is supposed to be started by. First, the data generation means A acquires pressure data in S1. The acquisition of the pressure data (S1) is to acquire the time series data of the pressure fluctuation in an arbitrary section N (seconds) at all the pressure sensitive points 42.

After the acquisition of the pressure data (S1 in FIG. 5) is completed, the data generation means A executes the pressure sensitive point selection process in S2.

The processing content of the pressure sensitive point selection process (S2) will be described with reference to FIG. First, in S20, the data generation means A performs FFT analysis (Fast Fourier Transform) on the pressure data at all the pressure sensitive points 42 acquired in S1 for each frequency at all the pressure sensitive points 42. Power spectrum of. Subsequently, in S21, the integrated value in the entire area, that is, the entire frequency band and the integrated value in the respiratory frequency band (for example, 0.1 to 0.5 Hz) are calculated using the power spectra at all the pressure sensitive points 42. For example, four pressure sensitive points 42 are selected in descending order of the ratio of the integrated value in the respiratory frequency band to the integrated value in the entire frequency band, and the pressure sensitive point selection process (S2) is completed. The number of pressure sensitive points 42 to be selected may be one, but in consideration of the case where the body movement corresponding to breathing is dispersed over a plurality of pressure sensitive points 42, a plurality of pressure sensitive points 42 such as 2 to 5 are selected. It is preferable, and more preferably about four.

(In FIG. 5, S2) pressure sensing points selecting process after is finished, the pressure fluctuation calculating unit a ₁ data generating unit A, at S3, selected for example, four pressure sensitive at pressure sensing points selection process (S2) The average value of the output values obtained from the point 42 is calculated and acquired as pressure fluctuation time series data.

After the calculation / acquisition of the pressure fluctuation time series data (S3 in FIG. 5) is completed, the data generation means A detects the pressure fluctuation time series by the multipoint detection pressure sensor 20 as the pressure sensor in S4. Using the data, CDM processing is performed on the frequency components related to respiration, that is, the frequency components in the range of 0.1 to 0.5 Hz, for example.

Next, the details of the CDM processing in S4 will be described below. CDM processing is a signal processing method that analyzes temporal changes in the instantaneous frequency and instantaneous amplitude of frequency components within a predetermined range in a time series. This time, for the pressure fluctuation time series data: X (t), which is the given time series, the component of interest (component related to respiration): x (t) instantaneous frequency: F (t) and instantaneous amplitude : The procedure for obtaining A (t) will be described. Given that the frequency components associated with breathing are known to be in the range 0.1-0.5 Hz or (0.3 Hz (center frequency: f ₀ ) ± 0.2 Hz ( _{w w)), then instantaneous.} If the deviation between the frequency: F (t) and the center frequency: f ₀ is regarded as a change in the phase: P (t), there is a relationship shown in Equation 1.

As a result, if the component of interest: x (t) has an instantaneous amplitude: A (t), x (t) is expressed as in Equation 2.

On the other hand, the original pressure fluctuation time series data: X (t) contains a component other than x (t): z (t), and is therefore expressed as in Equation 3.

The complex number representation of Equation 3 is expressed as Equation 4.

Here, it is assumed that z (t) does not include a component in the range of 0.1 to 0.5 Hz, that is, (0.3 Hz (center frequency: f _{0) ± 0.2 Hz).} There is. Next, if Y (t) is a signal obtained by multiplying both sides of Equation 4 by exp (−i2πf _{0 t) and shifting by −f 0} on the frequency axis, it is expressed as Equation 5.

As a result, the instantaneous frequency of the desired component exists within the range _{of ± f w in Y (t).} Therefore, when Y (t) is applied to _{a low-pass filter having a frequency of f w} , the output time series data is expressed as in Equation 6.

Using the time series data of the number 6 thus obtained: y (t), it is possible to obtain the instantaneous amplitude: A (t) of the component of interest (component related to respiration): x (t). can. That is, from Equation 6, since the amplitude of y (t) = | y (t) | = A (t) / 2, the instantaneous amplitude: A (t) can be obtained from Equation 7.

In addition, using the time series data of Equation 6: y (t), the instantaneous frequency of the component of interest (component related to respiration): x (t): F (t) can also be obtained. First, P (t) is obtained from Equation 6, that is, using the time series data of Equation 6, y (t), exp (from y (t) / | y (t) | as shown in Equation 8). After finding iP (t)), find P (t) using Equation 9.

Then, by using the equation 9, the instantaneous frequency: F (t) can be obtained from the equation 1. As described above, the instantaneous frequency: F (t) and the instantaneous amplitude: A (t) can be obtained.

By performing the CDM processing in S4, the data generation means A generates time-series data of the instantaneous frequency and the instantaneous amplitude of the frequency component related to respiration in S5, and the instantaneous frequency / amplitude acquisition is executed. The data obtained by the data generation means is visualized by, for example, a graph display, and is output to the monitor 52.

After the instantaneous frequency / amplitude acquisition (S5 in FIG. 5) is completed, the data generation means A executes the abnormal value detection process in S6.

The processing content of the abnormal value detection process (S6) will be described with reference to FIG. 7. First, in S60, the normal range of the instantaneous frequency and the average value of the instantaneous amplitude at rest are preset in the normal range setting unit a _{2 of the data generation means A.} In the present embodiment, the normal range of the instantaneous frequency is set to 0.1 to 0.5 Hz, which is a value assumed in advance as a range of pressure fluctuation based on a general resting breathing motion. However, it may be set based on the range of the instantaneous frequency at rest of the sleeping person 12 acquired in advance. On the other hand, the average value of the instantaneous amplitude at rest is set by using the average value of the instantaneous amplitude of the sleeping person 12 acquired in advance at rest. In this embodiment, the average value: 0.02 kPa (kilopascal) is set.

Abnormal value determination unit a ₃ data generating means A, in S60～S64, the value of the instantaneous frequency is adapted to determine an abnormal value when departing from the normal range. More specifically, the abnormal value determination unit a ₃ of the data generation means A determines in S60 whether or not the instantaneous frequency acquired in S5 is 0.1 to 0.5 Hz or more, which is the normal range, and determines whether or not the instantaneous frequency is 0.1 to 0.5 Hz or more. When is 0.5 Hz or higher (S60 = Yes), that is, when it is an abnormal value, it is assumed that body movements other than the breathing movements of the sleeping person 12 are occurring, and in S61, the occurrence of body movements is monitored, for example, 52. The abnormal value detection process (S6) is terminated after the data is output by changing the color of the character string or the graph display and written in the RAM 48. On the other hand, when the instantaneous frequency is less than 0.5 Hz (S60 = No), the process proceeds to the next S62 to determine whether or not the instantaneous frequency is 0.1 to 0.5 Hz or less, which is the normal range. When the instantaneous frequency is 0.1 Hz or less (S62 = Yes), that is, when it is an abnormal value, the process further proceeds to S63, and the instantaneous amplitude is 1 / of the resting average value set in S60: 0.02 kPa. Determine if it is 3 or less. Here, when the instantaneous amplitude exceeds 1/3 of the average value at rest (S63 = No), it is assumed that the body movement has occurred, and in S61, the body movement occurrence is transmitted to, for example, a character string on the monitor 52. After outputting and writing in the RAM 48, the abnormal value detection process (S6) is terminated. On the other hand, when the instantaneous amplitude is 1/3 or less of the average value at rest (S63 = Yes), it is assumed that apnea has occurred, and in S64, the occurrence of apnea is indicated by, for example, a character string on the monitor 52. After outputting and writing in the RAM 48, the abnormal value detection process (S6) is terminated. In addition, when the instantaneous frequency exceeds 0.1 Hz (S62 = No), since the instantaneous frequency is in the normal range (0.1 to 0.5 Hz), the abnormal value detection process (S6) is terminated as it is. ..

After the abnormal value detection process (S6 in FIG. 5) is completed, the data generation means A determines in S7 whether or not to end the respiration measurement. Such a determination is made, for example, by the user operating an operation button (the operation button may be a GUI or the like displayed on the monitor 52 provided on the control device 22) provided on the control device 22 (not shown). Will be done. When the respiration measurement is continuously performed (S7 = No), the acquisition of the pressure fluctuation time series data (S3) to the determination of the end of the respiration measurement (S7) are repeated to continue the acquisition of the respiration data. On the other hand, when the respiration measurement is completed (S7 = Yes), a series of operations shown in FIG. 5 is completed.

Next, the respiration measurement using the respiration measurement device 10 of the present embodiment will be specifically described with reference to the actual measurement data of FIGS. 8 and 9. First, by performing S1 to S3 as the processing of the data generation means A of the control device 22 as shown in FIG. 5, the pressure fluctuation time series data as shown in FIG. 8A is acquired. Then, using the pressure fluctuation time series data, CDM processing is executed for the frequency component related to respiration, that is, the frequency component in the range of 0.1 to 0.5 Hz (see S4 in FIG. 5). , It is possible to acquire the data of the temporal change of the instantaneous frequency and the instantaneous amplitude as shown in FIGS. 8 (b) and 8 (c). Note that FIG. 8 shows data of the sleeping person 12 in a resting state. That is, as shown in FIGS. 8A to 8C, each data is substantially constant with respect to time, and the instantaneous frequency is within the range of 0.1 to 0.5 Hz. You can see that there is.

Subsequently, in FIGS. 9 (a) and 9 (b), the temporal changes in the instantaneous frequency and the instantaneous amplitude when the sleeping person 12 in the resting state becomes apnea (between approximately 125 to 175 seconds in the figure). The data of is shown. In such a case, when the abnormal value detection process shown in FIG. 7 is executed, since the instantaneous frequency is 0.1 Hz or less in approximately 125 to 175 seconds, S60 = No and S62 = Yes. Apnea corresponds to an instantaneous frequency of 0 Hz, but as can be seen from FIG. 9B, it is extremely difficult to accurately measure 0 Hz. Therefore, in the present embodiment, S63 is performed, that is, the instantaneous amplitude between about 125 and 175 seconds is 1/3 or less of that at rest (between about 20 and 120 seconds in FIG. 9). By checking, it can be seen that the occurrence of apnea can be easily detected.

According to the respiration measuring device 10 having a structure according to the present embodiment, it is related to respiration using the pressure fluctuation time series data associated with the respiration movement of the sleeping person 12 detected by the multipoint detection pressure sensitive sensor 20 as a pressure sensor. By performing CDM processing on the frequency component to be performed, time-series data of the instantaneous frequency and instantaneous amplitude of the frequency component related to respiration can be obtained as an instantaneous value instead of the average value obtained by the conventional Fourier conversion process. Can be done. Therefore, it is possible to measure the respiratory state of the sleeping person 12 more accurately, and it is possible to instantly determine the sleep state (REM / non-REM sleep) from the data of the respiratory interval fluctuation obtained in real time. It becomes.

For example, if the frequency in the time-series data of the instantaneous frequency is within the range of 0.1 Hz to 0.5 Hz, it is judged that the pressure fluctuation waveform due to respiration at rest is reflected (see FIG. 8 (c)). Based on the findings found by the present inventors that it is possible to detect the occurrence of body movements such as apnea and turning over if the frequency deviates from the above frequency range, for example, the instantaneous frequency range based on pressure fluctuation due to breathing at rest is within the normal range. By setting as, and determining a value that deviates from that range as an abnormal value, it is possible to obtain an accurate instantaneous respiratory rate at rest and detect the occurrence of an abnormality. More specifically, by setting the frequency component for CDM processing to the frequency component related to respiration of 0.1 to 0.5 Hz, the frequency component related to respiration is removed while the frequency component related to respiration is set. It is possible to sample widely. Then, when the instantaneous frequency exceeds the normal range, it can be distinguished as a fluctuation due to body movement, while when the instantaneous frequency falls below the normal range, the instantaneous amplitude at rest is used by using the value of the instantaneous amplitude. Since it can be determined that the apnea is when it becomes 1/3 or less of the average value of, the occurrence of the apnea can be confirmed quickly and easily.

Further, in the present embodiment, by using the sheet-shaped multi-point detection pressure sensor 20 in which a large number of pressure-sensitive points 42 are dispersedly arranged, the sleeping person at bedtime or the like is used without restraining the body of the sleeping person 12. The pressure fluctuation based on the 12 breathing movements can be stably acquired. Moreover, since the upper body 28 extending from at least the chest 24 to the abdomen 26 of the sleeping person 12 that varies depending on the breathing motion is placed on the sheet-shaped multi-point detection pressure sensitive sensor 20, the breathing motion can be performed. the pressure sensing points 42 that can be most advantageously obtain a pressure fluctuation based can be selected, is the pressure variation at-series data including a frequency component related to respiration can advantageously be obtained. In addition, in order to calculate the pressure fluctuation time series data, the pressure sensitive point 42 uses the power spectrum at all the pressure sensitive points 42, and the integrated value in the entire area, that is, the entire frequency band, and the respiratory frequency band (for example, 0.1). ~ 0.5Hz) is calculated, and the ratio of the integrated value in the respiratory frequency band to the integrated value in the entire frequency band is selected in descending order. the average value of the pressure variation at the time-series data of a plurality of pressure sensing points 42, so that it is possible to obtain more reliably the pressure change during-series data including a frequency component related to respiration.

Although the embodiments of the present invention have been described in detail above, the present invention is not limited by the specific description thereof. For example, in the above embodiment, the pressure sensitive point 42 uses the power spectrum at all the pressure sensitive points 42 to calculate the pressure fluctuation time series data, and the integrated value and the respiratory frequency band (for example, 0) in the entire area, that is, the entire frequency band. The integrated value in (1 to 0.5 Hz) was calculated, and it consisted of four selected in descending order of the ratio of the integrated value in the respiratory frequency band to the integrated value in all frequency bands, but the pressure fluctuation time series data was calculated. Therefore, the pressure sensitive points 42 calculate the integrated value in the entire area, that is, the integrated value in the entire frequency band and the integrated value in the respiratory frequency band (for example, 0.1 to 0.5 Hz) using the power spectra at all the pressure sensitive points 42. It may be composed of a plurality of pressure sensitive points 42 selected in descending order of the ratio of integrated values in the respiratory frequency band. As a result, the pressure fluctuation time series data that most favorably represents the pressure fluctuation based on the respiratory movement can be acquired more advantageously.

Further, in the above embodiment, the pressure sensitive points 42 are composed of four in order to calculate the pressure fluctuation time series data, but may be five or more. Further, in the above embodiment, the sheet-shaped multi-point detection pressure sensor 20 as the pressure sensor is arranged on the support surface 18 of the floor portion 16 of the bed 14 that supports the sleeping person 12, but the floor portion It may be arranged inside the mattress or the like constituting 16 or on the lower surface side. In addition, the pressure sensor is not limited to the example multi-point detection pressure sensor 20, and any pressure sensor can be used as long as it can detect the pressure fluctuation accompanying the breathing motion of the sleeping person 12. In the above embodiment, the work content and the measurement result of the control device 22 are output to the monitor 52, but can be output to any known output means such as a printer or voice.
In addition, the present invention originally includes the inventions described below, and the configuration and action / effect thereof will be added.
The present invention
(I) CDM processing is performed on the pressure sensor that detects the pressure fluctuation accompanying the breathing motion of the sleeping person and the frequency component related to breathing in the pressure fluctuation time series data detected by the pressure sensor, and the time of the instantaneous frequency. A respiration measuring device comprising a data generating means for generating series data and an output means for outputting the data obtained by the data generating means.
(Ii) The data generation means includes a normal range setting unit for setting a normal range of an instantaneous frequency and an abnormal value determination unit for determining an abnormal value when the value of the instantaneous frequency deviates from the normal range. The respiration measuring device according to (i).
(Iii) When the value of the instantaneous frequency exceeds the upper limit value of the normal range, the abnormal value determination unit determines that the abnormal value is due to a body movement other than the breathing motion of the sleeping person (iii). The respiration measuring device according to ii),
(Iv) The data generation means further generates time-series data of instantaneous amplitude by the CDM processing, and the normal range setting unit further sets the average value of the instantaneous amplitude at rest. When the abnormal value determination unit is 1/3 or less of the resting average value of the instantaneous amplitude, it is determined to be an abnormal value due to apnea (iii) or (iii). Respiratory measuring device described,
(V) The respiration measuring device according to any one of (i) to (iv), wherein the frequency component related to respiration in the pressure fluctuation time series data is 0.1 to 0.5 Hz.
(Vi) The pressure sensor has a sheet shape in which a large number of pressure sensitive points are dispersedly arranged, is arranged on a support surface supporting the sleeping person, and is an upper body extending from at least the chest to the abdomen of the sleeping person. The respiration measuring device according to any one of (i) to (v), which is composed of a multi-point detection pressure sensor on which the device is mounted.
(Vii) The data generation means includes a pressure fluctuation calculation unit, and the pressure fluctuation calculation unit obtains from a plurality of pressure sensitive points selected in descending order of output values in the respiratory frequency band from among the large number of pressure sensitive points. The respiration measuring device according to (vi), which calculates the pressure fluctuation time series data from the average value of the output values.
(Viii) The data generation means includes a pressure fluctuation calculation unit, and the pressure fluctuation calculation unit selects in descending order of the ratio of the respiratory frequency band to the total area of the power spectrum of the acquired pressure among the large number of pressure sensitive points. The respiratory measuring apparatus according to (vi), wherein the pressure fluctuation time series data is calculated from the average value of the output values obtained from the plurality of pressure sensitive points.
Including inventions related to.
In the invention described in (i) above, the frequency component related to breathing in the pressure fluctuation time series data associated with the breathing motion of the sleeping person detected by the pressure sensor is subjected to CDM (Complex Demodulation) processing in the data generation means. As a result, the time-series data of the instantaneous frequency can be obtained, and the time-series data of the obtained instantaneous frequency can be output via the output means by an arbitrary method. As a result, it is possible to measure the instantaneous frequency related to respiration, which was not obtained by the conventional Fourier transform process, and it is possible to obtain the respiratory rate and respiratory vibration of the sleeping person as instantaneous values instead of the average value in a certain section. .. As a result, the breathing state of the sleeping person can be measured more accurately, and the sleep state is judged (REM / non-REM sleep) and the apnea state is detected instantly from the data of the breathing interval fluctuation obtained in real time. It becomes possible.
In the invention described in (ii) above, if the output frequency in the time-series data of the instantaneous frequency is within the range of 0.1 Hz to 0.5 Hz, the pressure fluctuation waveform due to respiration at rest is reflected and the range is deviated. Based on the findings found by the present inventors that it is possible to detect the occurrence of body movements such as apnea and turning over, for example, the range of the instantaneous frequency based on the pressure fluctuation due to breathing at rest is set as the normal range, and the range is set as the normal range. By determining a value that deviates from the range as an abnormal value, it is possible to obtain an accurate instantaneous respiratory rate at rest and detect the occurrence of an abnormality.
In the invention described in (iii) above, when the instantaneous frequency exceeds the normal range, it can be distinguished as a fluctuation due to body movement, and the instantaneous respiratory rate at rest can be obtained by removing such noise.
In the invention described in (iv) above, in the data generation means, by further generating time-series data of instantaneous amplitude by CDM processing, the value of instantaneous amplitude is used and such a value is preset in the normal range setting unit. When it becomes 1/3 or less of the average value at rest, it can be determined that the patient has apnea, and the occurrence of apnea can be promptly confirmed.
In the invention described in (v) above, by setting the frequency component to one of 0.1 to 0.5 Hz, the frequency component related to respiration is widely sampled while removing the frequency component related to heartbeat. Can be done. This makes it possible to easily confirm the apnea state in addition to the instantaneous respiratory rate at rest.
In the invention described in the above (vi), by using a sheet-shaped multi-point detection pressure-sensitive sensor in which a large number of pressure-sensitive points are dispersedly arranged, the sleeping person at bedtime or the like is used without restraining the sleeping person's body. It is possible to stably acquire pressure fluctuations based on the breathing motion of. Moreover, since the seat-shaped multi-point detection pressure-sensitive sensor is designed to mount at least the upper body of the sleeping person, which fluctuates due to the respiration movement, from the chest to the abdomen, the pressure fluctuation based on the respiration movement is the most. It is possible to select a pressure sensitive point that can be advantageously obtained, and it is possible to advantageously obtain pressure fluctuation time-series data including a frequency component related to respiration.
In the invention described in the above (vii), the pressure fluctuation time series data is calculated based on the average value of the output values obtained from the plurality of pressure sensitive points selected in descending order of the output value of the respiratory frequency band. It is possible to advantageously acquire pressure fluctuation time series data that most favorably represents pressure fluctuations based on respiratory movements.
In the invention described in the above (viii), pressure fluctuation time series data is obtained from the average value of output values obtained from a plurality of pressure sensitive points selected in descending order of the ratio of the respiratory frequency band to the total area of the power spectrum of the acquired pressure. By calculating, it is possible to more reliably obtain pressure fluctuation time series data including frequency components related to respiration. The number of pressure sensitive points to be selected is preferably 2 to 5, more preferably about 4 because the parts where the body movement corresponding to breathing appears may be dispersed. It is desirable to do.

10: Respiration measuring device, 12: Sleeper, 18: Support surface, 20: Multipoint detection pressure sensor (pressure sensor), 24: Chest, 26: Abdomen, 28: Upper body, 42: Pressure sensitive point, 52: Monitor (Output means), A: Data generation means A, a ₁ : Pressure fluctuation calculation unit, a ₂ : Normal range setting unit, a ₃ : Abnormal value determination unit

Claims (7)

A pressure sensor that is placed on the floor that supports the sleeping person and detects pressure fluctuations associated with the sleeping person's breathing movement.
A data generation means for generating instantaneous frequency time-series data by performing CDM processing on the frequency components related to respiration in the pressure fluctuation time-series data detected by the pressure sensor.
It is provided with an output means for outputting the data obtained by the data generation means.
The data generation means includes a normal range setting unit for setting a normal range of the instantaneous frequency and an abnormal value determination unit for determining an abnormal value when the value of the instantaneous frequency deviates from the preset normal range. respiratory measuring device, characterized in that out. According to claim 1 , when the value of the instantaneous frequency exceeds the upper limit value of the normal range, the abnormal value determination unit determines that the value is an abnormal value due to a body movement other than the breathing motion of the sleeping person. The described respiration measuring device. The data generation means further generates time-series data of instantaneous amplitude by the CDM processing, and the normal range setting unit further sets the average value of the instantaneous amplitude at rest. When the value of the instantaneous frequency is equal to or less than the lower limit of the normal range and the value of the instantaneous amplitude is equal to or less than 1/3 of the average value at rest, the abnormal value determination unit causes an abnormality due to aspiration. The respiration measuring device according to claim 1 or 2 , wherein the value is determined. The respiration measuring device according to any one of claims 1 to 3 , wherein the frequency component related to respiration in the pressure fluctuation time series data is 0.1 to 0.5 Hz. The pressure sensor has a sheet shape in which a large number of pressure sensitive points are dispersedly arranged, is arranged on a support surface supporting the sleeping person, and the upper body extending from at least the chest to the abdomen of the sleeping person is placed. The respiration measuring device according to any one of claims 1 to 4 , which is composed of a multi-point detection pressure sensitive sensor. The data generation means includes a pressure fluctuation calculation unit, and the pressure fluctuation calculation unit selects output values from a plurality of pressure sensitive points selected in descending order of output values in the respiratory frequency band from among the large number of pressure sensitive points. The respiration measuring device according to claim 5 , wherein the pressure fluctuation time series data is calculated from the average value of. The data generation means includes a pressure fluctuation calculation unit, and the pressure fluctuation calculation unit selects a plurality of pressure sensitive points in descending order of the ratio of the respiratory frequency band to the total area of the power spectrum of the acquired pressure. The respiratory measurement device according to claim 5 , wherein the pressure fluctuation time series data is calculated from the average value of the output values obtained from the pressure sensitive points.

Images