JP2020185020A - Biological information monitoring system, biological information monitoring method, and bed system - Google Patents

Abstract

To provide a biological information monitoring system capable of acquiring and/or displaying biological information of a subject with high accuracy.SOLUTION: A biological information monitoring system 100 for monitoring biological information of a subject on a bed includes: at least one of load detectors 11-14 for detecting a load of the subject on the bed; an observation waveform acquisition section for acquiring an observation waveform which indicates a temporal change corresponding to breathing or heartbeats of the subject; a filtering section for calculating an estimation waveform which indicates a state of breathing or heartbeats of the subject by a Kalman filter; a body motion determination section for determining presence/absence of a body motion of the subject; an in-phase detection section for detecting an in-phase time when rises of zero crosses of the observation waveform and the estimation waveform or falls of zero crosses of the observation waveform and the estimation waveform simultaneously occur; and a filter control section for controlling an operation of the filtering section based on a determination result of the body motion determination section and a detection result of the in-phase detection section.SELECTED DRAWING: Figure 1

Description

The present invention relates to a biometric information monitoring system, a biometric information monitoring method, and a bed system.

In the fields of medical care and long-term care, it has been proposed to detect the load of a subject on a bed via a load detector and acquire biological information such as the respiratory rate and heart rate of the subject based on the detected load.

Patent Document 1 is a respiratory waveform drawing system that draws a respiratory waveform indicating a respiratory state of a subject on a bed, and includes a drawing compensating unit that compensates for the drawing state of the respiratory waveform, and displays the respiratory waveform in substantially real time. We disclose a respiratory waveform drawing system that can be used. The respiratory waveform drawing system of Patent Document 1 generates a predicted waveform of the subject's respiratory waveform based on the past temporal fluctuation of the position of the center of gravity of the subject, and between the subject's respiratory waveform and the predicted waveform at a predetermined sampling time. Compensates the drawing state of the respiratory waveform based on the distance (correction distance). Specifically, for example, when the correction distance is within a predetermined range, the respiratory waveform is offset to the predicted waveform side by a predetermined distance.

JP-A-2017-205220

An object of the present invention is to provide a biological information monitoring system, a biological information monitoring method, and a bed system capable of acquiring and / or displaying the biological information of a subject with high accuracy.

According to the first aspect of the present invention,
A biometric information monitoring system that monitors the biometric information of subjects on the bed.
With at least one load detector provided under the bed or under the legs of the bed to detect the load of the subject on the bed.
An observation waveform acquisition unit that acquires an observation waveform indicating a temporal fluctuation of the detection value of the load detector according to the respiration or heartbeat of the subject, and an observation waveform acquisition unit.
A filter execution unit that calculates an estimated waveform indicating the respiratory or heartbeat state of the subject by performing a prediction step and a filter step using a Kalman filter.
A body movement determination unit that determines the presence or absence of body movement of the subject based on the detection value of the load detector, and
An in-phase detector that detects an in-phase time at which the rise of the zero cross between the observed waveform and the estimated waveform or the fall of the zero-cross of the observed waveform and the estimated waveform occur at the same time.
It is provided with a filter control unit that controls the operation of the filter execution unit based on the determination result of the body movement determination unit and the detection result of the in-phase detection unit.
The filter execution unit is configured to selectively execute a normal calculation that performs the prediction step and the filter step based on the observed waveform, or a simple calculation that performs only the prediction step.
The filter control unit causes the filter execution unit to perform a simple calculation during the period during which the subject is moving and during the period from the end of the period to the first homeomorphic time after the period. The system is provided.

In the biological information monitoring system of the first aspect, the at least one load detector may be a plurality of load detectors, and the observation waveform acquisition unit may use the observation waveform as the detection value of the plurality of load detectors. The position of the center of gravity of the subject may be obtained based on the above, and the breathing waveform of the subject may be calculated based on the movement of the center of gravity according to the breathing of the subject, and the filter execution unit may calculate the breathing waveform based on the breathing waveform. A filter step may be performed.

The biological information monitoring system of the first aspect may further include a biological information calculation unit that calculates the respiratory rate or heart rate of the subject based on the estimated waveform.

The biological information monitoring system of the first aspect may further include an estimated waveform drawing unit that draws the estimated waveform.

In the biological information monitoring system of the first aspect, the observation waveform acquisition unit uses the observation waveform as the observation waveform, the breath observation waveform showing the temporal variation of the detection value of the load detector according to the breath of the subject, and the load. A heartbeat observation waveform indicating a temporal fluctuation of the detection value of the detector according to the subject's heartbeat may be acquired, and the filter execution unit may obtain a breathing estimation waveform indicating the subject's breathing state as the estimated waveform. , And the heartbeat estimation waveform indicating the state of the heartbeat of the subject may be calculated, and the filter step executed by the filter execution unit is a filter step based on the breathing observation waveform and the heartbeat observation waveform. May include a filter step based on.

According to the second aspect of the present invention,
Bed and
A bed system including the biometric information monitoring system of the first aspect is provided.

According to the third aspect of the present invention,
It is a biological information monitoring method that monitors the biological information of the subject on the bed.
Detecting the load of the subject on the bed by at least one load detector provided on the bed or under the legs of the bed.
Acquiring an observation waveform showing the temporal fluctuation of the detection value of the load detector according to the respiration or heartbeat of the subject, and
Performing a prediction step and a filter step with a Kalman filter to calculate an estimated waveform indicating the respiratory or heartbeat state of the subject, and
Judging the presence or absence of body movement of the subject based on the detection value of the load detector,
A biological information monitoring method including detecting an in-phase time in which a zero cross rise of the observed waveform and the estimated waveform or a zero cross fall of the observed waveform and the estimated waveform occur at the same time.
To calculate the estimated waveform is to selectively execute a normal operation including the prediction step and a filter step based on the observation waveform, or a simple operation including only the prediction step.
Provided is a biological information monitoring method for performing a simple calculation during a period during which the subject is moving, and from the end of the period to the first homeomorphic time after the period.

In the biological information monitoring system of the third aspect, the at least one load detector may be a plurality of load detectors, and the acquisition of the observed waveform is based on the detection values of the plurality of load detectors. The position of the center of gravity of the subject may be obtained, and the respiratory waveform of the subject may be calculated based on the movement of the center of gravity according to the breathing of the subject. The filter step is a filter step based on the respiratory waveform. You can.

The biological information monitoring system of the third aspect may further include calculating the respiratory rate or heart rate of the subject based on the estimated waveform.

The biological information monitoring system of the third aspect may further include drawing the estimated waveform.

In the biological information monitoring system of the third aspect, acquiring the observation waveform is to acquire a respiration observation waveform showing a temporal variation of the detection value of the load detector according to the respiration of the subject. Often, the calculation of the estimated waveform may be to calculate the respiratory estimation waveform indicating the respiratory state of the subject by the Kalman filter. The biological information monitoring system of the third aspect may further include acquiring a heartbeat observation waveform showing a temporal variation of the detection value of the load detector according to the subject's heartbeat, and the Kalman filter may be used to obtain the subject's heartbeat observation waveform. It may further include calculating a heart rate estimation waveform indicating the state of the heart rate. In the biological information monitoring system of the third aspect, the filter step in the calculation of the respiratory estimation waveform may be a filter step based on the respiratory observation waveform, and the filter step in the calculation of the heartbeat estimation waveform is the heartbeat observation waveform. It may be a filter step based on.

According to the biological information monitoring system, the biological information monitoring method, and the bed system of the present invention, the biological information of the subject can be acquired and / or displayed with high accuracy.

FIG. 1 is a block diagram showing a configuration of a biological information monitoring system according to an embodiment of the present invention. FIG. 2 is an explanatory diagram showing the arrangement of the load detector with respect to the bed. FIG. 3 is a flowchart showing a method of monitoring biological information using a biological information monitoring system. FIG. 4A is an explanatory diagram conceptually showing how the center of gravity of the subject vibrates in the body axis direction in response to the subject's respiration. FIG. 4B is an explanatory diagram for explaining a method of calculating a respiration waveform based on the vibration of the center of gravity according to the respiration of the subject. FIG. 4C is a diagram showing an example of a respiration waveform drawn based on the vibration of the center of gravity according to the respiration of the subject. FIG. 5 is a block diagram showing a specific configuration of the estimated waveform acquisition unit. FIG. 6 is a diagram showing an example of the respiratory waveform obtained by the respiratory waveform acquisition unit and an example of the estimated waveform obtained by the estimated waveform acquisition unit. FIG. 7 is a schematic graph showing the state of fluctuation of the load value detected by the load detector for both the rest period in which the subject is only breathing and the body movement period in which the subject is moving. Is. FIG. 8 is a diagram for explaining control of the Kalman filter execution unit by the Kalman filter control unit. FIG. 9 is a block diagram showing the overall configuration of the bed system according to the modified example.

<Embodiment>
The biometric information monitoring system 100 (FIG. 1) of the embodiment of the present invention will be described as an example in which the biological information monitoring system 100 (FIG. 1) is used together with the bed BD (FIG. 2) to calculate (estimate) the respiratory rate of the subject S on the bed BD. To do.

As shown in FIG. 1, the biological information monitoring system 100 of the present embodiment mainly includes a load detection unit 1, a control unit 3, and a storage unit 4. The load detection unit 1 and the control unit 3 are connected via an A / D conversion unit 2. A display unit 5, a notification unit 6, and an input unit 7 are further connected to the control unit 3.

The load detection unit 1 includes four load detectors 11, 12, 13, and 14. Each of the load detectors 11, 12, 13, and 14 is a load detector that detects a load using, for example, a beam-shaped load cell. Such load detectors are described, for example, in Japanese Patent No. 4829020 and Japanese Patent No. 4002905. The load detectors 11, 12, 13 and 14, respectively, are connected to the A / D converter 2 by wiring or wirelessly.

As shown in FIG. 2, the four load detectors 11 to 14 of the load detection unit ₁ were attached to the lower ends of the legs BL ₁ , BL ₂ , BL ₃ , and BL ₄ at the four corners of the bed BD used by the subject S. It is placed under casters C ₁ , C ₂ , C ₃ , and C ₄ , respectively.

The A / D conversion unit 2 includes an A / D converter that converts an analog signal from the load detection unit 1 into a digital signal, and is connected to the load detection unit 1 and the control unit 3 by wiring or wirelessly, respectively.

The control unit 3 is a dedicated or general-purpose computer, and has a respiratory waveform acquisition unit (observation waveform acquisition unit) 31, an estimated waveform acquisition unit 32, and a respiratory rate calculation unit (biological information calculation unit) 33 built therein.

The storage unit 4 is a storage device that stores data used in the biological information monitoring system 100, and for example, a hard disk (magnetic disk) can be used.

The display unit 5 is a monitor such as a liquid crystal monitor that displays the information output from the control unit 3 to the user of the biological information monitoring system 100.

The notification unit 6 includes a device that aurally performs a predetermined notification based on information from the control unit 3, for example, a speaker.

The input unit 7 is an interface for performing a predetermined input to the control unit 3, and may be a keyboard and a mouse.

An operation of monitoring the biological information (respiratory rate in the present embodiment) of the subject on the bed by using the biological information monitoring system 100 will be described.

As shown in the flowchart of FIG. 3, the subject's biological information monitor using the biological information monitoring system 100 has a load detection step S1 for detecting the subject's load and the subject's respiratory waveform based on the detected load (load value). The respiratory waveform acquisition step S2 for obtaining (details will be described later), the estimated waveform acquisition step S3 for obtaining an estimated waveform (detailed later) indicating the respiratory state of the subject using the respiratory waveform and the Kalman filter, and the subject based on the estimated waveform. The respiratory rate calculation step S4 for calculating the respiratory rate and the display step S5 for displaying the respiratory rate calculated in the respiratory rate calculation step S4 are included.

[Load detection process]
In the load detection step S1, the load of the subject S on the bed BD is detected by using the load detectors 11, 12, 13, and 14. The load of the subject S on the bed BD is distributed and applied to the load detectors 11 to 14 arranged under the legs BL _{1 to} BL ₄ at the four corners of the bed BD, and is dispersed and detected by these.

Each of the load detectors 11 to 14 detects the load (load change) and outputs it as an analog signal to the A / D converter 2. The A / D conversion unit 2 converts an analog signal into a digital signal with a sampling period of, for example, 5 milliseconds, and outputs the digital signal (hereinafter, “load signal”) to the control unit 3. In the following, the load signals obtained by digitally converting the analog signals output from the load detectors 11, 12, 13, and 14 by the A / D converter 2 are the load signals s ₁ , s ₂ , s ₃ , and s, respectively. Call it ₄ .

[Respiratory waveform acquisition process]
In the respiratory waveform acquisition step S2, the respiratory waveform acquisition unit 31 obtains a respiratory waveform (an example of the observed waveform) indicating the respiratory state of the subject S based on the observation of the load fluctuation of the subject S.

Human breathing is performed by moving the thorax and diaphragm to inflate and contract the lungs. Here, during inspiration, that is, when the lungs expand, the diaphragm moves downward and the internal organs also move downward. On the other hand, when exhaling, that is, when the lungs contract, the diaphragm rises upward and the internal organs also move upward. As described in the specification of Japanese Patent No. 6105703 granted to the applicant of this case, the center of gravity of the human being slightly moves with the movement of the internal organs, and the movement direction is the extending direction of the spine (body axis direction). ) Is almost along.

In the present invention and the present specification, the “breathing waveform” refers to the vibration of the center of gravity of the subject (hereinafter, appropriately referred to as “breathing vibration”) that vibrates in the body axis direction of the subject in response to the breathing of the subject. It means the waveform shown by expanding to. One cycle of the respiratory waveform corresponds to one breath (expiration and inspiration) of the subject. The amplitude of the respiratory waveform is affected by the subject's physique and breathing depth. Specifically, for example, when the subject is large or the subject takes a deep breath, the amplitude becomes large, and when the subject is small or the subject takes a shallow breath, the amplitude becomes small.

As a specific example, the respiratory waveform acquisition unit 31 obtains data indicating the respiratory waveform by the following procedure.

First, the respiratory waveform acquisition unit 31 calculates the position of the center of gravity G of the subject S at each sampling time based on the load signals s _{1 to} s ₄ from the load detection unit 1. As shown in FIG. 4A, the center of gravity G of the subject S vibrates substantially along the direction of the body axis SA of the subject S in response to the respiration of the subject S.

Next, as shown in FIG. 4B, the respiration waveform acquisition unit 31 projects the calculated position of the center of gravity G at each sampling time onto the body axis SA to obtain the position GA, and the vibration of the vibration corresponding to the respiration of the center of gravity G. The distance X between the center O and the position GA is calculated. When the calculated distance X at each sampling time is sequentially plotted with the horizontal axis as the time axis, the respiratory waveform RW shown in FIG. 4C is obtained. Instead of the distance X, the linear distance X'between the center of gravity G and the vibration center O may be used. There are various methods for setting the vibration center O. Specifically, for example, two extreme points (as an example, points E1 and E2 in FIG. 4A) that occur continuously and in close proximity to the locus of the center of gravity G. ), And its midpoint can be set as the center of vibration O.

The respiratory waveform acquisition unit 31 calculates the distance X at each sampling time as data indicating the respiratory waveform RW, and outputs this to the estimated waveform acquisition unit 32. Further, the respiratory waveform acquisition unit 31 may include a respiratory waveform drawing unit (not shown) that draws the respiratory waveform RW based on the data indicating the respiratory waveform RW. In this case, the respiratory waveform drawing unit may draw the respiratory waveform RW as shown in FIG. 4C and display it on the display unit 5.

[Estimated waveform acquisition step S3]
In the estimated waveform acquisition step S3, the estimated waveform acquisition unit 32 obtains an estimated waveform indicating the respiratory state of the subject S by using the Kalman filter and the data indicating the respiratory waveform RW obtained in the respiratory waveform acquisition step S2.

As shown in FIG. 5, the estimated waveform acquisition unit 32 shows a Kalman filter execution unit (filter execution unit) 321 that calculates data indicating the estimated waveform, a Kalman filter control unit 322 that controls the Kalman filter execution unit 321 and an estimated waveform. It includes an estimated waveform drawing unit 323 that draws an estimated waveform based on the data. Further, the Kalman filter control unit 322 includes a body movement determination unit 322a, an in-phase detection unit 322b, and a mode switching unit (filter control unit) 322c.

The Kalman filter execution unit 321 executes the Kalman filter and calculates data indicating an estimated waveform EW indicating the respiratory state of the subject S.

In general, the Kalman filter is a calculation method that efficiently estimates the state of the analysis target, and analyzes at the next time point based on the state estimation value indicating the state of the analysis target at a certain time point and the state equation created appropriately. A prediction step for predicting a state estimate indicating the state of the target and a filter step for correcting the state estimate at the next time point predicted in the prediction step using the observed value at the time point are alternately performed.

The Kalman filter execution unit 321 performs the calculation under the following two operation modes.

(1) Normal Calculation Mode In the normal calculation mode, the Kalman filter execution unit 321 performs a general Kalman filter processing including a prediction step and a filter step. That is, the Kalman filter execution unit 321 uses the prediction step of predicting a new state estimation value using the past state estimation value and the predetermined equation of state, and the observation value of the new state estimation value predicted by the prediction step. Perform a filter step to correct (update).

Specifically, the Kalman filter execution unit 321 in the prediction step _(here, the estimated value of the distance X at the sampling time t _n) state estimate EV n at the predetermined sampling time t _n using a with a predetermined state equation Then, the predicted estimated value PEV _{n + 1 at} the next sampling time t _{n + 1} (here, the predicted estimated value of the distance X at the sampling time t _{n + 1} ) is calculated.

Then, the prediction Kalman filter execution unit 321, the filter step, observations at sampling time t _{n + 1} (in this case, the distance value of X in the respiratory waveform RW at sampling time t _{n + 1)} based on and Kalman gain was calculated in the prediction step The estimated value PEV _{n + 1} is corrected, and the state estimated value EV _{n + 1} at a predetermined sampling time t _{n + 1} is obtained. If the Kalman gain is made larger, the state estimation value EV _{n + 1} becomes a value closer to the observed value.

In this way, the Kalman filter execution unit 321 in the normal calculation mode corrects the predicted estimated value calculated in the predicted step and the predicted estimated value calculated in the predicted step by the observed value in the prediction step for calculating the predicted estimated value using the past state estimated value and the state equation. By alternately repeating the filter step for obtaining the new state estimated value EV _{n + 1} , the state estimated values EV _n , EV _{n + 1} , EV _{n + 2,} ... At each sampling time are sequentially calculated. The calculated state estimated values EV _n , EV _{n + 1} , EV _{n + 2,} ... Are data indicating the estimated waveform EW.

In the following description, the estimated waveform EW drawn based on the state estimated values EV _n , EV _{n + 1} , EV _{n + 2,} ... Calculated by the normal calculation mode is referred to as a normal estimated waveform EW1.

A schematic example of the normal estimated waveform EW1 is as shown in the period up to the time T1 in FIG. 6 and the period after the time T3. The normal estimated waveform EW1 shown in FIG. 6 has non-constant amplitude and period, which reflects the actual respiratory condition of subject S. That is, since the normal estimated waveform EW1 is obtained through a filter step of correcting the predicted estimated value by the respiratory waveform RW acquired based on the actual breathing of the subject S, it reflects the actual respiratory state of the subject S. It becomes a shape.

The respiratory waveform RW is also drawn in FIG. 6, and the normal estimated waveform EW1 and the respiratory waveform RW are substantially the same (note that in FIG. 6, both the respiratory waveform RW and the normal estimated waveform EW1 can be visually recognized. Both waveforms are drawn slightly offset). As described above, it is desirable but not essential to design the Kalman filter so that the normal estimated waveform EW1 substantially matches the respiratory waveform RW, and how close the normal estimated waveform EW1 is to the respiratory waveform RW is appropriately determined by the design of the Kalman filter. Can be changed.

(2) Simple calculation mode In the simple calculation mode, the Kalman filter execution unit 321 executes only the prediction step, and does not correct the predicted estimated value by the filter step. In addition, in the Kalman filter of the present specification and the present invention, "performing only the prediction step" simply means that the prediction step is performed without performing the filter step, and even if other auxiliary calculation steps are not performed. It doesn't mean anything.

Specifically, the Kalman filter execution unit 321 _(here, the predicted estimates of the distance X at the sampling time t _n) predicted estimate PEV n at the predetermined sampling time t _n by the a predetermined state equation, the next sampling time t _{n +} predicted estimate at ₁ PEV n + _{1 (in} this case, the predicted estimate of the distance X at the sampling time t _{n + 1)} is calculated, and similarly thereafter, by with a predetermined state equation newly obtained predicted estimate, The predicted estimated value PEV _{n + 1} , the predicted estimated value PEV _{n + 2,} ... Are sequentially calculated.

In this way, the Kalman filter execution unit 321 in the simple calculation mode continuously executes the prediction steps without performing the filter steps to obtain the predicted estimated values PEV _n , PEV _{n + 1} , PEV _{n + 2,} ... At each sampling time. Calculate sequentially. The calculated predicted estimated values PEV _n , PEV _{n + 1} , PEV _{n + 2,} ... Are data indicating the estimated waveform EW.

In the following description, the estimated waveform EW drawn based on the predicted estimated values PEV _n , PEV _{n + 1} , PEV _{n + 2,} ... Calculated by the simple calculation mode is referred to as a simple estimated waveform EW2.

A schematic example of the simple estimated waveform EW2 is as shown in the period of time T1 to time T3 in FIG. The simple estimated waveform EW2 shown in FIG. 6 has substantially constant amplitude and period. This is because the simple estimated waveform EW2 is a waveform drawn based only on the prediction step, and is not affected by the filter step, in other words, the respiration waveform RW based on the actual respiration of the subject S. In general, the simple estimated waveform EW2 exhibits an idealized and relatively monotonous change as compared to the normal estimated waveform EW1.

In the period of time T1 to time T3 in FIG. 6, the respiratory waveform RW fluctuates greatly and randomly in the period of time T1 to time T2. This means that the subject S has body movements (movement of the subject's head, body (trunk), limbs, etc. Movements of organs, blood vessels, etc. associated with breathing, heartbeat, etc. are not included in body movements). This is because the center of gravity G of the subject S is larger than the respiratory vibration and moves aperiodically. The simple estimated waveform EW2, which is not affected by the filter step based on the respiratory waveform RW, shows a substantially constant amplitude and period even during the period in which the subject S is moving and the respiratory waveform RW is disturbed.

The Kalman filter control unit 322 controls the Kalman filter execution unit 321 to calculate the normal estimated waveform EW1 in the normal mode during the period when the subject S does not have body movement, and the period during which the subject S has body movement. , And the simple estimated waveform EW2 is calculated in the simple mode during a predetermined period after the body movement of the subject S is completed.

Specifically, the control of the Kalman filter execution unit 321 by the Kalman filter control unit 322 is performed as follows, for example.

First, the body movement determination unit 322a of the Kalman filter control unit 322 detects the presence or absence of body movement of the subject S. Specifically, for example, the detection of the presence or absence of body movement of the subject S is performed by the following procedure.

The body movement determination unit 322a has a standard deviation (moving standard deviation) of sampling values included in a predetermined sampling period (for example, the past 5 seconds) for each of the load signals s _{1 to} s ₄ , σ ₁ , σ ₂ , σ. _3. Calculate σ ₄ . The calculation can be done at each sampling time.

Since the standard deviation represents the magnitude of the variation in the sampling value, as shown in FIG. 7, the subject S on the bed BD is at rest, and the amount of fluctuation of the load signals s _{1 to} s ₄ is small during the period P ₁ The standard deviations σ _{1 to} σ ₄ also become smaller. On the other hand and moving the subject S is the body in a (which occurs motion in the subject S), and larger standard deviation σ ₁ ~σ ₄ during the period P ₂ the amount of fluctuation is large in the load signal s ₁ ~s ₄ ..

The body movement determination unit 322a calculates a simple mean value (arithmetic mean value) σ _AV with standard deviations σ _{1 to} σ ₄ , and compares the simple mean value σ _AV with a predetermined threshold σ _th to move the subject S. Is determined whether or not is occurring. That is, when the simple mean value σ _AV is larger than the threshold value σ _th , it is determined that the subject S is moving, and when the simple mean value σ _AV is equal to or less than the threshold value σ _th , the subject S is moving. Judge that it is not.

Based on the above determination, the body movement determination unit 322a outputs a body movement pulse BMP indicating a rise during the period in which the subject S is performing body movement to the mode switching unit 322c.

An example of the body motion pulse BMP is as shown in the second stage of FIG. The first stage of FIG. 8 is the respiratory waveform RW and the estimated waveform EW shown in FIG. The body movement pulse BMP in the second stage of FIG. 8 is on during the period from time T1 to time T2 when the subject S is moving, and is off during the other periods.

In parallel with the detection of the presence or absence of body movement of the subject S by the body movement determination unit 322a, the in-phase detection unit 322b of the Kalman filter control unit 322 has the respiratory waveform RW rising from the zero cross (that is, from the negative side to the positive side of the waveform). shows the rise), and the estimated waveform EW is time indicating the rise of the zero crossing (hereinafter, detects the called "in-phase time t _p").

Specifically, the common mode detection unit 322b specifies the time t _{R0 at} which the respiratory waveform RW indicates the rise of zero cross and the time t _{E0 at} which the estimated waveform EW indicates the rise of zero cross (third and fourth stages in FIG. 8). ), The in-phase pulse SPP indicating the rise and fall is output to the mode switching unit 322c at the in-phase time t _p where the time t _R0 and the time t _E0 coincide. It should be noted that the coincidence between the time t _R0 and the time t _E0 does not necessarily have to be an exact match, and a certain degree of timing deviation is acceptable within the range necessary for achieving the desired accuracy. Specifically, for example, a phase shift of about ± 5% in the cycle of respiration (heartbeat in the following modification) may be regarded as coincident in timing. In this case, the in-phase time t _p has a width of period about ± 5% of the respiratory (heart in the modification example below). In the present specification and the present invention, when it is said that "the rising (or falling) of the zero cross of the observed waveform and the estimated waveform occur at the same time", both of them are within a period having a range of about ± 5% of the respiratory (or heartbeat) cycle. May mean that both occur together, and the "in-phase time" can be a period having a time width of about ± 5% of the respiration (or heartbeat) cycle.

An example of the common mode pulse SPP is as shown in the fifth stage of FIG. Phase pulse SPP has risen at time t _p the rise occurs simultaneously zero crossing of the respiratory waveform RW and estimated waveform EW, after lost motion of the subject S, the first time t _p, is equal to the time T3.

The mode switching unit 322c simplifies the calculation mode of the Kalman filter execution unit 321 to the normal calculation mode based on the body movement pulse BMP output from the body movement determination unit 322a and the in-phase pulse SPP output from the in-phase detection unit 322b. Switch between the calculation mode.

Specifically, the mode switching unit 322c rises at the same time as the rise of the body motion pulse BMP, and falls at the same time as the fall of the first in-phase pulse SPP that occurs after the fall of the body motion pulse BMP (FIG. 8). 6th stage) is generated. Then, during the period when the timing pulse TMP is on (time T1 to time T3 in FIG. 8), the calculation mode of the Kalman filter execution unit 321 is set to the simple calculation mode, and during the period when the timing pulse TMP is off, the Kalman filter is set. The calculation mode of the execution unit 321 is set to the normal calculation mode.

That is, the mode switching unit 322c is based on the body movement pulse BMP and the in-phase pulse SPP, and the period during which the body movement occurs in the subject S and after the body movement of the subject S disappears, the respiratory waveform RW and the estimated waveform EW The operation mode of the Kalman filter execution unit 321 is set to the simple calculation mode during the period until the rise of the zero cross occurs, and the operation mode of the Kalman filter execution unit 321 is set to the normal calculation mode during the other period.

The timing pulse TMP may be generated so as to fall at the same time as the rise and fall of the first in-phase pulse SPP that occurs after the fall of the body motion pulse BMP. Further, the mode switching unit 322c does not have to obtain the timing pulse TMP. That is, without relying on the timing pulse TMP, the Kalman filter execution unit 321 is switched to the simple calculation mode at the time when the body motion pulse BMP shows a rise, and from the rise of the in-phase pulse SPP that occurs first after the fall of the body motion pulse BMP. It is also necessary to switch the Kalman filter execution 321 to the normal calculation mode at any time until the falling edge.

The estimation waveform drawing unit 323 draws the estimated waveform EW based on the calculation result of the Kalman filter execution unit 321 and outputs it to the control unit 3. Specifically, the normal estimated waveform EW1 is drawn based on the state estimated values EV _n , EV _{n + 1} , EV _{n + 2,} ... Calculated by the normal calculation mode, and the predicted estimated value PEV _n , calculated by the simple calculation mode. A simple estimated waveform EW2 is drawn based on PEV _{n + 1} , PEV _{n + 2,} ....

Here, the mode switching unit 322c of the Kalman filter control unit 322 of the present embodiment does not start at the time when the body movement of the subject S stops (time T2), but after the body movement of the subject S stops, the rise of the zero cross of the respiratory waveform RW. The reason for switching the Kalman filter execution unit 321 from the simple calculation mode to the normal calculation mode at the time when the rising edge of the zero cross of the estimated waveform EW coincides with each other (time T3) will be described.

At the time when the body movement of the subject S is stopped, a phase difference is usually generated between the estimated waveform EW and the respiratory waveform RW (as an example, the simple estimated waveform EW2 and the respiratory waveform RW at the time T2 in FIG. 6). See the phase difference in). If the Kalman filter execution unit 321 is returned from the simple calculation mode to the normal calculation mode at such a timing, the simple estimation waveform EW2 calculated in the simple calculation mode is applied by applying the filter step using the respiratory waveform RW. A discontinuity (that is, a phase shift, a waveform shift in the X-axis direction, etc.) occurs with the normal estimated waveform EW1 calculated by the normal calculation mode. Such a discontinuity also appears in the estimated waveform EW drawn by the estimated waveform drawing unit 323, which is not preferable for observing the waveform. It also causes an error in calculating the respiratory rate using the estimated waveform EW (details will be described later).

On the other hand, as in the Kalman filter control unit 322 of the present embodiment, the time point at which the rise of the zero cross of the respiratory wave RW coincides with the rise of the zero cross of the estimated waveform EW, that is, the phase of the respiratory waveform RW is the simple estimated waveform EW2. If the operation mode of the Kalman filter execution unit 321 is switched from the simple mode to the normal mode at the time when the phases match, the influence of the filter step at the time of switching can be eliminated or reduced. As a result, the simple estimated waveform EW2 and the normal estimated waveform EW1 can be smoothly continued (for example, refer to the connection portion between the simple estimated waveform EW2 and the normal estimated waveform EW1 at the time T3 in the fourth stage of FIG. ).

[Respiratory rate calculation step S4]
In the respiratory rate calculation step S4, the respiratory rate calculation unit 33 calculates (estimates) the respiratory rate of the subject S based on the estimated waveform EW calculated by the estimated waveform acquisition unit 32. The respiratory rate calculation by the respiratory rate calculation unit 33 is specifically performed as follows, for example, using the peak detection of the estimated waveform EW.

First, the respiratory rate calculation unit 33 sequentially detects the positive peak pp (hereinafter, simply referred to as “peak pp”) of the estimated waveform EW (that is, the normal estimated waveform EW1 and the simple estimated waveform EW2). FIG. 6 shows the peaks pp _n-2 , pp _n-1 , and pp _n of the estimated waveform EW specified by the respiratory rate calculation unit 33 at the times T _n-2 , T _n-1 , and T _n .

Next, the respiratory rate calculation unit 33 calculates the respiratory rate R, which is an estimated value of the respiratory rate per minute of the subject S, as follows, based on the time when the peak of the estimated waveform EW is specified.

The respiratory rate calculation unit 33 sets the time T _n-1 at the time when the peak pp _n-1 is specified at the time T _n-1, and the time T _n-2 when the peak pp _n-2 is specified immediately before that. The elapsed time TT _n-1 during the period is calculated, and the respiratory rate R is calculated using the following (Equation 1).
(Equation 1)
R = 60 / TT _m [times / minute] (m = 1, 2, ..., n-1, n)

That is, by dividing 60 seconds by the latest respiratory cycle, an estimated value of the respiratory rate per minute that reflects the latest respiratory condition is calculated. Respiration rate calculation unit 33, likewise, when the peak pp _n at time _{T n} is identified, and the time _{T n,} between time _{T n-1} peak _{pp n-1} is identified immediately before The elapsed time TT _n is calculated, and the respiratory rate R is calculated using (Equation 1). The calculated respiratory rate R is output to the control unit 3.

[Display process]
In the display step S5, the control unit 3 displays the respiratory rate of the subject S on the display unit 5 based on the output of the respiratory rate calculation unit 33, and the estimated waveform EW of the subject S based on the output of the estimated waveform drawing unit 323. Is displayed on the display unit 5.

Further, in the display step S5, in addition to or instead of the display using the display unit 5, the notification unit 6 may be used for notification. In this case, for example, when the respiratory rate of the subject S becomes smaller than a predetermined threshold value, a notification sound is emitted to notify a nurse, a caregiver, or the like who is a user of the biological condition monitoring system 100 of the occurrence of body movement. ..

The effects of the biological information monitoring system 100 of this embodiment are summarized below.

The biological condition monitoring system 100 of the present embodiment includes a Kalman filter control unit 322 that controls the operation mode of the Kalman filter execution unit 321. The Kalman filter control unit 322 does not stop the body movement of the subject S, but the body movement of the subject S. The Kalman filter execution unit 321 is switched from the simple calculation mode to the normal calculation mode when the rise of the zero cross of the respiratory waveform RW and the rise of the zero cross of the estimated waveform EW coincide with each other. In this way, switching from the simple calculation mode to the normal calculation mode, in other words, restarting the filter step, the phases of the respiratory waveform RW and the estimated waveform EW are aligned, and the influence of the filter step on the Kalman filter calculation is small. By performing the timing, it is possible to suppress the deviation between the simple estimated waveform EW2 and the normal estimated waveform EW1 due to the restart of the filter step.

Therefore, when the estimated waveform EW is drawn and displayed on the display unit 5, a smooth estimated waveform EW can be displayed, and the respiratory rate of the subject S can be calculated with high accuracy based on the estimated waveform EW. Calculations can be made. That is, the waveform indicating the respiratory state of the subject S can be displayed with high accuracy, and the respiratory rate of the subject S can be acquired with high accuracy.

The biological condition monitoring system 100 of the present embodiment monitors the biological condition of the subject S by using load detectors 11 to 14 arranged under the legs BL _{1 to} BL ₄ of the bed BD. Therefore, it is not necessary to attach the measuring device to the body of the subject S, and the subject S does not feel uncomfortable or uncomfortable.

[Modification example]
In the biological information monitoring system 100 of the above embodiment, the following modified modes can also be adopted.

In the biological information monitoring system 100 of the above embodiment, the respiratory waveform acquisition unit 31 may be configured as a waveform acquisition unit that obtains a waveform showing load fluctuation according to the heartbeat of the subject S in addition to the respiratory waveform RW. In this case, the estimated waveform acquisition unit 32 may be configured to obtain an estimated waveform indicating the heart rate state of the subject S in addition to the estimated waveform indicating the respiratory state of the subject S, and may be configured to obtain a respiratory rate calculation unit (biological information calculation unit). ) 33 may be configured to calculate the heart rate of the subject S in addition to the respiratory rate of the subject S.

In this modification, the waveform acquisition unit uses at least one of the load signals s _{1 to} s _{4 to} obtain components contained in the heartbeat frequency band (about 0.5 Hz to about 3.3 Hz) by a bandpass filter or the like. It is separated and sent to the estimation waveform acquisition unit 32 as a waveform showing the load fluctuation according to the heartbeat of the subject S.

The estimated waveform acquisition unit 32 obtains an estimated waveform indicating the heartbeat state of the subject S by the same step as the estimated waveform acquisition step S3 of the above embodiment. Since the waveform showing the load fluctuation according to the heartbeat of the subject S is also a waveform in the same time domain as the respiratory waveform RW, the estimated waveform acquisition unit 32 of the modified example is the respiratory waveform of the estimated waveform acquisition unit 32 of the above embodiment. An estimated waveform indicating the state of the heartbeat of the subject S can be obtained by a process equivalent to the process performed on the RW. At this time, the Kalman filter execution unit 321 calculates the predicted estimated value using the equation of state for obtaining the waveform indicating the heartbeat state of the subject S in the prediction step, and receives it from the waveform acquisition unit in the filter step. The predicted estimated value is corrected using the waveform showing the load fluctuation according to the heartbeat of.

The waveform calculation unit may be configured as a waveform calculation unit that calculates only the waveform showing the load fluctuation according to the heartbeat of the subject S. In this case, the estimated waveform acquisition unit 32 may be configured to obtain only the estimated waveform indicating the heart rate state of the subject S, and the respiratory rate calculation unit (biological information calculation unit) 33 obtains only the heart rate of the subject S. It may be a configuration.

In the biological information monitoring system 100 of the above embodiment, the respiratory waveform acquisition unit 31 is a component included in the respiratory frequency band (about 0.2 Hz to about 0.33 Hz) from at least one of the load signals s _{1 to} s _4. The waveform obtained by separating the above may be calculated, and the waveform may be sent to the estimation waveform acquisition unit 32 instead of the respiratory waveform RW. Since the waveform obtained by separating the components included in the respiration frequency band from at least one of the load signals s _{1 to} s ₄ is also a waveform in the same time domain as the respiration waveform RW, the estimated waveform acquisition unit 32 is described above. An estimated waveform indicating the respiratory state of the subject S can be obtained by a process equivalent to the process performed on the respiratory waveform RW in the embodiment.

The breath waveform and the waveform obtained by separating the components contained in the frequency band of breath from at least one of the load signals s _{1 to} s ₄ are described in the scope of the patent claim as "subject of the detection value of the load detector". An example of an "observed waveform showing temporal fluctuations according to breathing", and a waveform obtained by separating a component contained in the frequency band of the heartbeat from at least one of the load signals s _{1 to} s ₄ is a patent claim. This is an example of the “observed waveform showing the temporal fluctuation of the detection value of the load detector according to the subject's heartbeat” described in the range. Further, the respiratory waveform acquisition unit 31 of the above embodiment and the waveform acquisition unit of the modified example are examples of the “observation waveform acquisition unit” described in the claims.

In the biological information monitoring system 100 of the above embodiment, the body movement determination unit 322a of the Kalman filter control unit 322 determines the presence or absence of body movement of the subject S by comparing the simple mean values σ _AV of the standard deviations σ _{1 to} σ ₄ with the threshold value. It was judged, but it is not limited to this. The body movement determination unit 322a compares any one of the standard deviations σ _{1 to} σ ₄ with the threshold value, and compares the simple average value or total value of any two or more of the standard deviations σ _{1 to} σ ₄ with the threshold value, etc. The presence or absence of body movement of the subject S may be determined according to the above, and a variance that is the square of the standard deviation may be used instead of the standard deviations σ _{1 to} σ ₄ .

In the biological information monitoring system 100 of the above embodiments, phase detector 322b of the Kalman filter control unit 322 as an in-phase time t _p, the but respiratory waveform RW and the estimated waveform EW has detected the time indicating the rise of the zero crossing this Is not limited. Phase detecting unit 322b, as the in-phase time t _p, the respiratory waveform RW and the estimated waveform EW may detect a time indicating the fall of the zero crossing.

In the biological information monitoring system 100 of the above embodiment, the Kalman filter control unit 322 does not have to include the estimation waveform drawing unit 323. Further, the biological information monitoring system 100 of the above embodiment does not have to include the respiratory rate calculation unit 33.

In the biological information monitoring system 100 of the above embodiment, the respiratory rate calculation unit 33 may calculate the respiratory rate R based on the identification of the negative peak np instead of the positive peak pp. The specific steps in this case are the same as those in the above embodiment.

The biological information monitoring system 100 of the above embodiment does not necessarily include all of the load detectors 11 to 14, and may only include any one of them. Further, the load detectors do not necessarily have to be arranged at the four corners of the bed, and may be arranged at arbitrary positions so that the load of the subject on the bed and its fluctuation can be detected. Further, the load detectors 11 to 14 are not limited to the load sensor using the beam type load cell, and for example, a force sensor can also be used.

In the biological information monitoring system 100 of the above embodiment, each of the load detectors 11 to 14 is arranged under the casters C attached to the lower ends of the legs of the bed BD, but the present invention is not limited to this. Each of the load detectors 11-14 may be provided between the four legs of the bed BD and the floorboard of the bed BD, or the upper part if the four legs of the bed BD can be divided into upper and lower parts. It may be provided between the leg and the lower leg. Further, the load detectors 11 to 14 may be integrally or detachably combined with the bed BD to form a bed system BDS including the bed BD and the biological information monitoring system 100 of the present embodiment (FIG. 9). In the present specification, the "load detector provided on the bed" refers to the load detector provided between the four legs of the bed BD and the floor plate of the bed BD as described above, or the upper leg. It means a load detector provided between the lower leg and the lower leg.

In the biological information monitoring system 100 of the above embodiment, between the load detection unit 1 and the A / D conversion unit 2, a signal amplification unit that amplifies the load signal from the load detection unit 1 and filtering that removes noise from the load signal. A part may be provided.

In the biological condition monitoring system 100 of the above embodiment, the display unit 5 is a simple device such as a printer that prints and outputs information representing biological information, a lamp that displays biological information, or the like, in place of or in addition to the monitor. Visual display means may be provided. The notification unit 6 may include a vibration generation unit that notifies by vibration in place of or in addition to the speaker.

As long as the features of the present invention are maintained, the present invention is not limited to the above-described embodiment, and other modes considered within the scope of the technical idea of the present invention are also included within the scope of the present invention. ..

According to the biological information monitoring system of the present invention, the biological information of the subject can be acquired and / or displayed more accurately, which can contribute to the improvement of the quality of medical care, long-term care, and the like.

1 Load detector, 11, 12, 13, 14 Load detector, 2 A / D conversion unit, 3 Control unit, 31 Respiratory waveform acquisition unit, 32 Estimated waveform acquisition unit, 321 Kalman filter execution unit, 322 Kalman filter control unit, 322a Body movement determination unit, 322b in-phase detection unit, 322c mode switching unit, 323 estimated waveform drawing unit, 33 respiratory rate calculation unit, 4 storage unit, 5 display unit, 6 notification unit, 7 input unit, 100 biometric information monitoring system, BD Bed, BDS bed system, S subject

Claims (11)

A biometric information monitoring system that monitors the biometric information of subjects on the bed.
With at least one load detector provided under the bed or under the legs of the bed to detect the load of the subject on the bed.
An observation waveform acquisition unit that acquires an observation waveform indicating a temporal fluctuation of the detection value of the load detector according to the respiration or heartbeat of the subject, and an observation waveform acquisition unit.
A filter execution unit that calculates an estimated waveform indicating the respiratory or heartbeat state of the subject by performing a prediction step and a filter step using a Kalman filter.
A body movement determination unit that determines the presence or absence of body movement of the subject based on the detection value of the load detector, and
An common mode detection unit that detects a common mode time at which the rise of the zero cross of the observed waveform and the estimated waveform or the fall of the zero cross of the observed waveform and the estimated waveform occur at the same time.
It is provided with a filter control unit that controls the operation of the filter execution unit based on the determination result of the body movement determination unit and the detection result of the in-phase detection unit.
The filter execution unit is configured to selectively execute a normal calculation that performs the prediction step and the filter step based on the observed waveform, or a simple calculation that performs only the prediction step.
The filter control unit causes the filter execution unit to perform a simple calculation during the period during which the subject is moving and during the period from the end of the period to the first homeomorphic time after the period. system. The at least one load detector is a plurality of load detectors.
The observation waveform acquisition unit obtains the position of the center of gravity of the subject based on the detection values of the plurality of load detectors as the observation waveform, and the respiration of the subject based on the movement of the center of gravity position according to the respiration of the subject. Calculate the waveform,
The biometric information monitoring system according to claim 1, wherein the filter execution unit performs the filter step based on the respiratory waveform. The biometric information monitoring system according to claim 1 or 2, further comprising a biometric information calculation unit that calculates the respiratory rate or heart rate of the subject based on the estimated waveform. The biometric information monitoring system according to any one of claims 1 to 3, further comprising an estimated waveform drawing unit for drawing the estimated waveform. The observation waveform acquisition unit uses the observation waveform as the observation waveform, the breath observation waveform showing the temporal fluctuation of the detection value of the load detector according to the breath of the subject, and the heartbeat of the subject of the detection value of the load detector. Acquire the heartbeat observation waveform showing the time fluctuation according to the time
The filter execution unit is configured to calculate, as the estimated waveform, a breathing estimation waveform indicating the respiratory state of the subject and a heartbeat estimation waveform indicating the heartbeat state of the subject.
The biometric information monitoring system according to any one of claims 1 to 4, wherein the filter step executed by the filter execution unit includes a filter step based on the respiration observation waveform and a filter step based on the heartbeat observation waveform. Bed and
A bed system including the biometric information monitoring system according to any one of claims 1 to 5. It is a biological information monitoring method that monitors the biological information of the subject on the bed.
Detecting the load of the subject on the bed by at least one load detector provided on the bed or under the legs of the bed.
Acquiring an observation waveform showing the temporal fluctuation of the detection value of the load detector according to the respiration or heartbeat of the subject, and
Performing a prediction step and a filter step with a Kalman filter to calculate an estimated waveform indicating the respiratory or heartbeat state of the subject, and
Judging the presence or absence of body movement of the subject based on the detection value of the load detector,
A biological information monitoring method including detecting an in-phase time in which a zero cross rise of the observed waveform and the estimated waveform or a zero cross fall of the observed waveform and the estimated waveform occur at the same time.
To calculate the estimated waveform is to selectively execute a normal operation including the prediction step and a filter step based on the observation waveform, or a simple operation including only the prediction step.
A biological information monitoring method in which a simple calculation is performed during a period during which the subject is moving, and during a period from the end of the period to the first homeomorphic time after the period. The at least one load detector is a plurality of load detectors.
To acquire the observed waveform, the position of the center of gravity of the subject is obtained based on the detection values of the plurality of load detectors, and the respiratory waveform of the subject is calculated based on the movement of the center of gravity position according to the respiration of the subject. Is to do
The biometric information monitoring method according to claim 7, wherein the filter step is a filter step based on the respiratory waveform. The biological information monitoring method according to claim 7 or 8, further comprising calculating the respiratory rate or heart rate of the subject based on the estimated waveform. The biological information monitoring method according to any one of claims 7 to 9, further comprising drawing the estimated waveform. Acquiring the observation waveform is to acquire a respiration observation waveform showing a temporal variation of the detection value of the load detector according to the respiration of the subject.
Further including acquiring a heartbeat observation waveform showing a temporal fluctuation of the detection value of the load detector according to the heartbeat of the subject.
To calculate the estimated waveform is to calculate a respiratory estimated waveform indicating the respiratory state of the subject by the Kalman filter.
Further including calculating a heartbeat estimation waveform indicating the heartbeat state of the subject by the Kalman filter.
The filter step in the calculation of the respiratory estimation waveform is a filter step based on the respiratory observation waveform, and the filter step in the calculation of the heartbeat estimation waveform is a filter step based on the heartbeat observation waveform. The biometric information monitoring method described in item 1.

Images