JP2023153132A - Electrically-driven furniture - Google Patents

Abstract

To provide a posture determining device or the like which can appropriately determine a posture as a state of a patient.SOLUTION: A posture determining device includes at least two oscillation sensors and a control unit. The control unit detects oscillations in a recumbent state of an object person by the oscillation sensor, calculates a wave form from the detected oscillations, and determines a posture of the object person from characteristics by recognizing the characteristics of the wave form.SELECTED DRAWING: Figure 1

Description

The present invention relates to electric furniture.

Conventionally, inventions for determining the condition of a patient on a patient bed apparatus are known. For example, in Patent Document 1, a bed state detection device is disposed on a bed and separately detects a component of vibration applied to the bed along the longitudinal direction of the bed and a component along the vertical direction with respect to the floor surface of the bed. a vibration detection unit that detects a vibration; a signal extraction unit that extracts a heartbeat vibration signal derived from the heartbeat of the living body existing on the bed from the vibration detected by the vibration detection unit; An invention is disclosed that includes a state estimating section that estimates the posture of the living body on a bed.

Further, Patent Document 2 discloses a user status acquisition unit that acquires a user status including at least one of the user's posture and the user's position on the bed or outside the bed, and a user status acquisition unit in which the user status corresponds to a dangerous state. a risk determination unit that determines whether or not to do so; a safety determination unit that determines whether the user state corresponds to a safe state; and a state setting unit that sets either a valid state or an invalid state. and a notification unit that performs notification processing when the user state is determined to be in a dangerous state and is in the valid state, and the status setting unit switches from the valid state to the invalid state when the notification process is performed. , discloses a notification device characterized in that when it is determined that the user state corresponds to a safe state in the disabled state, the notification device switches from the disabled state to the enabled state.

Japanese Patent Application Publication No. 2011-120667 Japanese Patent Application Publication No. 2016-192998

An object of the present invention is to provide a posture determination device that can appropriately determine the posture of a patient from a body vibration waveform.

In order to solve the above-mentioned problems, the patient posture determining device of the present invention is a posture determining device having at least two vibration sensors and a control section, the control section being configured to detect the subject's posture from the vibration sensor. The present invention is characterized by detecting vibrations while lying down, calculating a waveform from the detected vibrations, recognizing characteristics of the waveform, and determining the posture of the subject from the characteristics.

According to the posture determination device of the present invention, the posture of the subject is appropriately determined as one of the conditions of the subject based on the body vibration waveform, along with other biological information such as heart rate, breathing rate, amount of activity, and sleep state. can be determined.

FIG. 2 is a diagram for explaining the entire first embodiment. FIG. 2 is a diagram for explaining the configuration in the first embodiment. FIG. 2 is a diagram for explaining the configuration in the first embodiment. It is a figure for explaining a sensor in a 1st embodiment. FIG. 3 is a diagram for explaining patient posture determination processing in the first embodiment. FIG. 3 is a diagram for explaining patient posture determination processing in the first embodiment. It is a figure showing an example of a waveform in a prone position in a 1st embodiment. It is a figure showing an example of a waveform in a supine position in a 1st embodiment. FIG. 3 is a diagram showing an example of a waveform in the left lateral decubitus position in the first embodiment. FIG. 3 is a diagram showing an example of a waveform in the right lateral decubitus position in the first embodiment. It is a figure showing an example of a frequency component in a 1st embodiment. FIG. 7 is a diagram for explaining patient posture determination processing in the second embodiment. FIG. 7 is a diagram for explaining posture determination conditions in a second embodiment. FIG. 7 is a diagram for explaining the state of waveforms in the second embodiment. It is a figure for explaining the functional composition of the patient condition estimating part in a 3rd embodiment. FIG. 7 is a diagram for explaining a neural network in a fourth embodiment. FIG. 3 is a diagram for explaining a bed device as an application example. FIG. 7 is a diagram for explaining processing in an application example.

EMBODIMENT OF THE INVENTION Hereinafter, one form for implementing this invention is demonstrated with reference to drawings. From the viewpoint of preventing falls when a patient leaves the bed, in comparative examples, detection of the patient leaving the bed (staying in bed) was often performed. However, detection of the patient's posture while in bed has not been done.

Regarding the patient's posture, for example, if the patient's sitting position is detected as an action before getting out of bed, the patient's lying position or sitting position may be determined by installing a large number of sensors over a wide area of the bed. It was not possible to determine the sleeping position of the patient (eg, supine, prone, lateral, etc.).

In order to determine the patient's posture, it has been necessary, for example, to provide a separate camera device or the like to monitor the patient or to perform image analysis. In this case, the burden on the person monitoring the patient was heavy. In addition, since it requires a camera to constantly take pictures, some patients have objected to this from the perspective of protecting their privacy.

Therefore, according to the posture determination device of this embodiment, it is possible to detect even the patient's posture by simply detecting vibrations when the patient is sleeping on the bed device.

In addition, in this specification, a patient refers to a person who uses a bed device (mattress), and is not limited to those who receive treatment for illness, but also includes those who receive care at a facility or those who sleep on a bed device. Applicable as a person.

[1. First embodiment]
[1.1 Entire system]
FIG. 1 is a diagram for explaining the overall outline of a system 1 to which the posture determination device of the present invention is applied. As shown in FIG. 1, the system 1 includes a detection device 3 placed between the floor of the bed device 10 and the mattress 20, and a processing device 5 for processing the values output from the detection device 3. It is composed of The detection device 3 and the processing device 5 constitute a system for determining the patient's posture.

When a subject (hereinafter referred to as "patient P" as an example) is on the mattress 20, the detection device 3 detects body vibrations (vibrations emitted from the human body) as biological signals of the patient P, who is the subject. Furthermore, the biological information value of the patient P can be calculated based on the detected vibrations.

The calculated biological information value (for example, breathing rate, heart rate, activity amount) may be output/displayed as the patient P's biological information value.

Further, the detection device 3 may be formed integrally with the processing device 5 by providing a storage section, a display section, etc. Furthermore, since the processing device 5 may be a general-purpose device, it is not limited to an information processing device such as a computer, and may be configured by a device such as a tablet or a smartphone. Furthermore, if the detection device 3 has a communication function, it may be connected to a server device instead of the processing device 5.

Further, the target person may be a person undergoing medical treatment for an illness or a person who requires nursing care. In addition, it may be a healthy person who does not require care, an elderly person, a child, a disabled person, or an animal other than a person.

Here, the detection device 3 is configured in a sheet shape so that the thickness is small. Thereby, even if it is placed between the bed device 10 and the mattress 20, it can be used without making the patient P feel uncomfortable, so the condition of the patient in bed can be detected for a long period of time.

Note that the detection device 3 only needs to be able to detect vibrations of the patient P. For example, an actuator with a strain gauge or a load cell placed on a bed leg or the like that measures the load may be used. Further, by using a built-in acceleration sensor or the like, it may be realized, for example, by a smartphone, a tablet, etc. placed on the bed device 10.

Further, in FIG. 1, the head side of the bed device 10 (mattress 20) is designated as direction H, the foot side is designated as direction F, the left side is designated as direction L, and the right side is designated as direction R when the patient P is in the supine position.

[1.2 Configuration]
Next, the configuration of the system 1 will be explained with reference to FIGS. 2 to 4. The system 1 in this embodiment has a configuration including a detection device 3 and a processing device 5, and each functional unit (processing) may be realized by either of them except for the vibration detection unit 110. That is, by combining these devices, they function as an attitude determination device.

The system 1 (posture determination device) includes a control section 100, a vibration detection section 110, a biological information calculation section 120, a waveform calculation section 130, a frequency distribution calculation section 135, a patient condition determination section 140, and a storage section 150. , an input section 160 , and an output section 170 .

The control unit 100 controls the operation of the system 1. For example, it is a control device such as a CPU (Central Processing Unit). The control unit 100 implements various processes by reading and executing various programs stored in the storage unit 150. In this embodiment, the control section 100 operates as a whole, but it can also be provided in each of the detection device 3 and the processing device 5 as shown in FIG. 4, which will be described later.

The vibration detection unit 110 detects vibrations on the detection device 3 and acquires vibration data. In the present embodiment, as an example, vibrations (body vibrations) based on patient movement etc. are detected using a sensor that detects pressure changes. Vibration data is acquired based on the detected vibrations, and is output to the biological information calculation section 120, the waveform calculation section 130, and the frequency distribution calculation section 135 for processing. Note that the vibration data to be output may be analog vibration data or digital vibration data.

Further, the vibration detection unit 110 may obtain vibration data by detecting vibrations of the patient using a pressure sensor, for example, or may provide a microphone instead of the pressure sensor to generate biological signals based on sounds picked up by the microphone. may be obtained, and vibration data may be obtained from the biological signal. Further, vibration data may be acquired from output values of an acceleration sensor, a capacitance sensor, or a load sensor. In this way, any method may be used as long as the biological signal (body vibration indicating the patient's body movement) can be acquired.

The biological information calculation unit 120 acquires biological signals of the patient P from the vibration data, and calculates biological information values (respiration rate, heart rate, activity amount, etc.). In this embodiment, a breathing component and a heartbeat component may be extracted from the vibration (body vibration) data acquired by the vibration detection unit 110, and the breathing rate and heartbeat rate may be determined based on the breathing interval and heartbeat interval. Alternatively, the periodicity of body movements may be analyzed (Fourier transform, etc.) and the respiration rate and heart rate may be calculated from the peak frequency.

The waveform calculation unit 130 converts analog vibration data input from the vibration detection unit 110 into a digital voltage signal at predetermined sampling intervals, and calculates a waveform of body vibration (vibration waveform).

In this embodiment, for convenience of explanation, the waveform of vibration data (vibration waveform) is calculated by the waveform calculation unit 130 from the vibration data output by the vibration detection unit 110, and processing is performed based on the calculated waveform. This will be explained as execution. However, the waveform calculation unit 130 may not be provided and various processes may be executed based on the vibration data.

Further, the frequency distribution calculation unit 135 calculates a frequency distribution from the vibration (waveform). For example, the frequency distribution calculation unit 135 calculates frequency components by performing Fast Fourier Transform (FFT) on the waveform calculated by the waveform calculation unit 130.

The patient condition determining section 140 determines the condition of the patient. For example, the patient's condition is determined based on vibration data acquired by the vibration detection unit 110, biological information values calculated by the biological information calculation unit 120, a load sensor separately provided in the bed apparatus 10, and the like.

In this embodiment, the patient condition determination unit 140 mainly determines the posture of the patient (supine position, prone position, lateral position). In addition, the patient condition determination unit 140 may determine other patient postures such as an edge-sitting position, or other postures (for example, out of bed, in bed, etc.).

The storage unit 150 stores various data and programs for operating the system 1. The control unit 100 realizes various functions by reading and executing programs stored in the storage unit 150. Here, the storage unit 150 is configured with a semiconductor memory (eg, SSD (Solid State Drive) or SD card (registered trademark)), a magnetic disk device (eg, HDD (Hard Disk Drive)), and the like. Furthermore, the storage unit 150 may be a built-in storage device or a removable external storage device. Alternatively, it may be a storage area of an external server such as a cloud.

The storage unit 150 has a vibration data storage area 152 and a waveform data storage area 154 secured therein.

The vibration data storage area 152 stores vibration data output from the vibration detection section 110. Here, the interval to be stored as vibration data may be every predetermined time, for example, it may be a short interval such as every 1 second or every 5 seconds, or it may be a relatively short interval such as every 30 seconds, 1 minute, or 5 minutes. It may be a long interval.

The waveform data storage area 154 stores vibration waveform data (waveform data) calculated by the waveform calculation unit 130 based on the vibration data output from the vibration detection unit 110 or the vibration data stored in the vibration data storage area 152. do. In this embodiment, the description will be made assuming that the waveform data is stored in the waveform data storage area 154, but the waveform calculation unit 130 may calculate the waveform data as necessary each time.

Note that the waveform data storage area 154 is reserved as a storage area for convenience of explanation, but it is sufficient if it is calculated from vibration data as appropriate and temporarily stored.

The input unit 160 receives operation input from the user. For example, the user performs various operation inputs such as starting vibration acquisition and adjusting the sensitivity of the vibration detection unit 110.

The output unit 170 outputs various information. For example, it is composed of a display device such as a liquid crystal display, a light emitting member such as an LED, a speaker that outputs sound or audio, an interface that outputs data to another recording medium, and the like. Furthermore, the input section 160 and the output section 170 may be configured as a touch panel by integrally configuring the input section 160 and the output section 170.

Here, among the configurations described above, the biological information calculation section 120, the waveform calculation section 130, the frequency distribution calculation section 135, and the patient condition determination section 140 are mainly realized by software. For example, the control unit 100 reads and executes software stored in the storage unit 150. When the software is executed, the control unit 100 will be realized as each component.

That is, the control unit 100 reads and executes a program that implements the biological information calculation unit 120, the waveform calculation unit 130, the frequency distribution calculation unit 135, and the patient condition determination unit 140, so that the control unit 100 can perform the functions of each component. It will be held.

Further, FIG. 2 conceptually explains the configuration of the posture determination device as the system 1. These configurations may be realized, for example, by one device capable of detecting vibrations, or may be configured separately into a detection device 3 and a processing device 5, as shown in FIG. Further, instead of the processing device 5, an external server capable of providing the same service may be implemented.

A case where the system 1 of FIG. 2 is implemented by the detection device 3 and the processing device 5 of FIG. 1 will be described with reference to FIG. 3. The detection device 3 includes a control section 300, a vibration detection section 320 that is a sensor, a storage section 330, and a communication section 390.

Further, the control unit 300 functions as a biological information calculation unit 310 by executing software (program) stored in the storage unit 330. The vibration detection unit 320 outputs vibration data based on the detected vibrations.

The biological information calculation unit 310 calculates biological information based on the vibration data. The calculated biometric information (value) is then stored in the biometric information data 340 or transmitted to the processing device 5 via the communication unit 390. In addition, vibration data detected by the vibration detection section 320 is also transmitted to the processing device 5 via the communication section 390.

The timing of transmitting vibration data from the detection device 3 to the processing device 5 and the timing of storing biological information values (biological information) in the biological information data 340 may be in real time or at predetermined time intervals. good.

Note that the vibration detection section 320 is the vibration detection section 110 in FIG. 2 . Here, the vibration detection section 320 will be explained using FIG. 4.

FIG. 4 is a top view of the bed device 10 (mattress 20). In FIG. 4, the upper side is the direction H in FIG. 1, and the lower side is the direction F in FIG. Further, the right side of FIG. 4 is the direction F of FIG. 1, and the left side of FIG. 4 is the direction R of FIG.

The detection device 3 is placed between the bed device 10 and the mattress 20 or on the mattress 20. The place where it is placed is preferably near the patient's back, so it is at least in a direction closer to the H side than the center of the bed device 10 (mattress 20).

Furthermore, the detection device 3 has a built-in sensor (vibration detection section 110/320). The sensor is, for example, a vibration sensor, and can detect vibrations (body vibrations) of the patient. At least two sensors are provided; for example, in FIG. 4, two (vibration sensor 320a, vibration sensor 320b) are provided on the left and right sides of the detection device 3.

The vibration sensor 320a and the vibration sensor 320b are provided at a predetermined interval. The distance between the two sensors may be, for example, the width of the patient in the lateral direction, and preferably the distance between the two sensors is about 15 to 60 cm.

Furthermore, the biometric information calculation unit 310 in FIG. 3 is the biometric information calculation unit 120 in FIG. 2 . Further, the communication unit 390 is, for example, a communication interface connectable to a network (for example, LAN/WAN).

The processing device 5 includes a control section 500, a storage section 530, an input section 540, an output section 550, and a communication section 590. The processing device 5 receives vibration data from the detection device 3 via the communication unit 590. The received vibration data is stored in the vibration data storage area 532.

The control unit 500 functions as a waveform calculation unit 502, a frequency distribution calculation unit 504, and a patient condition determination unit 506 by executing software (programs) stored in the storage unit 530. Further, the waveform data calculated by the waveform calculation unit 502 is stored in the waveform data storage area 534.

Note that the waveform calculation section 502 is the waveform calculation section 130 in FIG. The frequency distribution calculation unit 504 is the frequency distribution calculation unit 135 in FIG. The patient condition determining section 506 is the patient condition determining section 140 in FIG. 2 . Input section 540 is input section 160 in FIG. The output section 550 is the output section 170 in FIG. The storage unit 530 is the storage unit 150 in FIG.

[1.3 Process flow]
Next, posture determination processing in this embodiment will be described with reference to FIG. 5. The posture determination process is a process executed by the patient condition determination unit 140.

First, the control unit 100 (patient condition determination unit 140) acquires vibration data (step S102). The patient condition determination section 140 acquires vibration data by reading it from the vibration data storage area 152 or receiving it from the vibration detection section 110 .

Subsequently, the posture of the patient is determined based on the vibration data, the correlation between the sensors, and the correlation within the sensor. In this embodiment, for example, the waveform calculation unit 130 calculates a waveform and outputs it as waveform data (step S104). Waveform data is calculated for each vibration sensor. For example, as shown in FIG. 4, since two vibration sensors (vibration sensor 320a and vibration sensor 320b) are provided in this embodiment, waveform data are also calculated for each.

Furthermore, the waveform data may be stored in the waveform data storage area 154 or output by the output unit 170. For example, if the output unit 170 is a display device, the waveform data is displayed on the display device.

Subsequently, the patient condition determination unit 140 determines whether there is a correlation between the sensors from the waveform data (step S106). Here, the correlation between sensors is determined by determining whether there is similarity in waveforms calculated based on vibration data acquired from two sensors (for example, vibration sensor 320a and vibration sensor 320b in FIG. 4). This is one method.

As an example, a cross-correlation function is used for two waveforms. The cross-correlation function is output as a value normalized to "0" to "1" based on the degree of similarity between the two waveforms. This value changes depending on the degree of similarity between the two waveforms. For example, when the value of the cross-correlation function is "1", it can be seen that the two waveforms completely match and the degree of similarity is maximum. Further, when the value of the cross-correlation function is "0", it can be seen that the two waveforms do not match at all and the degree of similarity is the minimum.

When determining whether there is a correlation between the two waveforms, the patient condition determination unit 140 determines whether the output value of the cross-correlation function exceeds the cross-correlation function threshold. For example, when the cross-correlation function threshold is set to "0.7", it is determined that two waveforms have no correlation if the output value of the cross-correlation function is "0.7" or less. On the other hand, if the output value of the cross-correlation function exceeds "0.7", it is determined that the two waveforms are correlated.

Next, if there is a correlation between the sensors (step S106; Yes), the patient condition determination unit 140 determines that the patient's posture is “supine or prone” (steps S116 to S120). Since the amount of time the patient is in the prone position during sleep is extremely short, the patient's position may simply be determined as ``supine or prone,'' but prone positioning has a high risk of suffocation and is associated with Sudden Infant Death Syndrome. The supine position and prone position may be further determined for purposes such as prohibiting automatic operation of the electric bed when the patient is in the prone position.

That is, it is determined whether there is a correlation within the sensor (step S106; Yes→step S116). Here, whether or not there is a correlation within the sensor is determined by evaluating, for example, the strength of periodicity in the waveform. As an example, it is determined whether or not there is a correlation between the waveforms of one sensor by using an autocorrelation function. The autocorrelation function outputs a value normalized to "0" to "1" based on the strength of the periodicity of the waveform within the same sensor. For example, when the value of the autocorrelation function is "1", it can be seen that the waveform data is output completely periodically, and there is a perfect correlation within the sensor.

Furthermore, the strength of periodicity may be calculated as a value standardized to "0" to "1" using Fourier transform, chi-square periodogram, or the like.

When determining whether one waveform has a correlation, the patient condition determination unit 140 may determine whether the output value of the autocorrelation function exceeds an autocorrelation function threshold. For example, if the autocorrelation function threshold is "0.7", if the output value of the autocorrelation function is "0.7" or less, there is no correlation in the calculated waveform (one detected vibration data). It is determined that On the other hand, if the output value of the autocorrelation function exceeds 0.7, it is determined that the waveforms are correlated.

If it is determined that there is a correlation within the sensor, the patient condition determining unit 140 determines that the patient's posture is “supine” (step S116; Yes→step S118). On the other hand, if it is determined that there is no correlation within the sensor, the patient condition determining unit 140 determines that the patient's posture is prone (step S116; Yes→step S118).

Further, when the patient condition determination unit 140 determines in step S106 that there is no correlation between the sensors (step S106; No), the patient's posture is determined to be “lateral decubitus” (step S110). Note that it is sufficient to determine that the patient's position is ``side-lying''; however, in order to check whether the patient has changed their position, if the patient has a paralyzed side, it is necessary to check that the patient is not sleeping on the paralyzed side. Therefore, the right lateral decubitus position and the left lateral decubitus position may be further determined (step S114).

In this case, when the patient is in the left lateral decubitus position, the input of the heartbeat signal is greater than in the right lateral decubitus position. Therefore, when a predetermined number of heartbeat signals or more are extracted (when a high frequency signal is extracted), it is determined that the patient is in the left lateral decubitus position.

There are various methods for determining the magnitude of the input heartbeat signal, but for example, the ratio of the frequency component corresponding to the heartbeat signal to the frequency component corresponding to the breathing signal, the signal of data subjected to high-pass filter processing, etc. Strength etc. may be used.

In this way, according to this embodiment, the patient's posture (sleeping posture) can be determined from the vibration data.

Furthermore, for convenience of explanation, the correlation between sensors and the presence or absence of correlation within a sensor have been described based on the waveform data calculated by the waveform calculation unit 130, but the correlation between sensors is simply determined based on vibration data. The correlation within the sensor or the correlation within the sensor may be determined. In this case, the process of step S104 does not need to be executed.

Further, in the above-described embodiment, the patient's posture was determined by calculating the waveform from the patient's vibration data, but the posture may be determined by evaluating the shape of the frequency distribution. In this case, the accuracy will be higher if the shape of the frequency distribution of vibrations acquired by two or more sensors is comprehensively evaluated, but the number of sensors that acquire vibrations may be one.

For example, a process of determining a patient's posture using frequency analysis will be described with reference to FIG. Similar to FIG. 5, vibration data is acquired from the vibration detection section 110 (step S152), and waveform data is calculated (output) by the waveform calculation section 130 (step S154).

Subsequently, the frequency distribution calculation unit 135 calculates the frequency distribution of the waveform data output by the waveform calculation unit 130 (step S156). For example, if a high frequency component is not calculated in the frequency distribution, the patient's posture is identified as “lateral decubitus”.

Since the relationship between the heart position and the sensor position differs depending on the posture, the body tissues (muscles, fat, bones, internal organs, etc.) between them differ. Therefore, since the way vibrations are transmitted also changes, differences appear in the measured frequency components.

The movement of the heart and the direction of movement of the chest and abdomen due to breathing are fixed. For example, when the patient is in the supine position, the movement of the thorax and abdomen due to breathing will be large in the direction perpendicular to the sensor (vibration detection unit 110). Furthermore, when the patient is in the lateral position, the movement in the direction parallel to the sensor (vibration detection unit 110) becomes large. Therefore, the frequency distribution differs depending on the patient's posture. Therefore, the posture can be determined by storing a frequency distribution for each posture and comparing the actually extracted frequency distribution with the frequency distribution stored for each posture.

For example, in the lateral position, components having a higher frequency than the heartbeat component are difficult to detect (except for integral multiples of the frequency corresponding to the heartbeat component). Therefore, if the high frequency component is smaller than the heartbeat component, the patient's posture can be determined to be in the "lateral decubitus position".

Note that in the above description, the frequency distribution calculation unit 135 calculates the frequency distribution from the waveform data, but the frequency distribution may be calculated directly from the vibration data. In this case, step S154 will not be executed.

[1.4 Conditions for posture determination]
Here, the posture determination process described above may be executed normally, or may be executed when a predetermined condition is met. Hereinafter, the patient's body movement will be used as a condition for posture determination. Here, the patient's body movement refers to movement in the patient's posture, such as the patient turning over in bed.

(1) Determine the posture for each section For example, in a continuous section that does not include body movement, it is determined that the same posture continues, and the same posture is output.

In this case, sections that do not include body movements may be analyzed together (the results of the posture determination process may be applied). In addition, the results of the posture determination process are totaled in a certain period (for example, every 3 minutes, etc.). If there is a section that does not include body movement within the certain section, the determination result of the most common posture in the section that does not include body movement is output as the result of posture determination processing for that section.

(2) Perform posture determination only in sections where no body movement occurs Posture determination is performed only in sections where no body movement occurs. For example, when the vibration detection unit 110 detects a patient's body movement, execution of the posture determination process is stopped. Furthermore, the posture determination process may be executed again after a predetermined period of time has elapsed since body movement was detected and no longer detected (for example, 10 seconds after body movement was no longer detected).

[1.5 Explanation based on waveform]
Here, determination of the patient's posture based on the waveform will be explained. Each waveform is a waveform based on vibrations detected based on two vibration sensors. The horizontal axis represents time and the vertical axis represents voltage value, each representing a waveform based on the detected vibration. Although the determination of the patient's posture will be explained based on waveforms for convenience of explanation, the patient's posture may also be determined by detecting trends from time-series vibration data.

In FIG. 7, there is no correlation between the waveforms within the sensor, and the same shape is not repeatedly represented. That is, this waveform is a waveform indicating a prone position (step S120 in FIG. 5).

In FIG. 8, there is a correlation between the waveforms between the sensors, and there is also a correlation between the waveforms within the sensor. That is, this waveform is a waveform indicating a supine posture (step S118 in FIG. 5).

9 and 10 show waveforms with no correlation between sensors. That is, this waveform is a waveform indicating a lateral posture (step S110 in FIG. 5). Comparing FIG. 9 and FIG. 10, in FIG. 10, which shows the waveform in the left lateral position, the high-frequency signal due to the heartbeat is more noticeable.

FIG. 11 is a graph showing the frequency distribution obtained from the waveform. FIG. 11 is a graph showing the frequency distribution obtained in step S156 of FIG. 6, with FIG. 11(a) showing the supine position, and FIG. 11(b) showing the lateral position. . In this way, the patient's posture can be determined by further extracting the frequency distribution from the waveform (vibration data).

[2. Second embodiment]
A second embodiment will be described. The second embodiment is an embodiment in which the patient's posture is determined based on a plurality of posture determination conditions.

Note that this is an embodiment in which the operational flow of FIG. 5 of the first embodiment is replaced with the operational flow of FIG. 12, and descriptions of the same parts such as the configuration will be omitted.

First, the patient condition determination unit 140 acquires vibration data (step S202). Then, the waveform calculation unit 130 calculates the waveform (step S204). Subsequently, the patient condition determination unit 140 calculates an index (value) for each condition (steps S206 to S214), and calculates a total value from the calculated index values (step S216).

The total value is calculated as the total value for each of the "supine position", "prone position", and "(left and right) lateral decubitus position". Then, the patient condition determining unit 140 outputs the posture with the largest value as the patient's posture.

Here, the patient condition determination unit 140 calculates an index value for each condition in steps S206 to S214, and a method for calculating the index value will be described with reference to FIG. 13.

(1) Intra-sensor correlation index calculation (step S206)
A value of an intra-sensor correlation index calculated by correlation based on a waveform within the sensor is calculated. First, the patient condition determination unit 140 calculates the value of the autocorrelation function based on the waveform within the sensor using the method described in the first embodiment.

Then, the patient condition determination unit 140 weights the output value of the autocorrelation function and outputs it as the value of the intra-sensor correlation index. Here, the weighting method will be explained.

For example, referring to FIG. 13, the intra-sensor correlation index is "present" for "supine position" and "lateral decubitus position", and "absent" for "prone position".

The output value of the autocorrelation function is between "0 and 1". Here, in FIG. 13, the output value is directly used as the value of the intra-sensor correlation index for the locations marked as "present" ("supine position" and "prone position"). In addition, in FIG. 13, the value of the intra-sensor correlation index is determined by subtracting the output value of the autocorrelation function from the maximum value in the “none” portion.

To explain a specific example, if the output value of the autocorrelation function is "0.8", the patient condition determination unit 140 determines "supine position = 0.8" and "prone position = 0.2" and "lateral decubitus position = 0.8" are output.

(2) Inter-sensor mutual index calculation (step S208)
A value of an inter-sensor correlation index calculated by correlation based on waveforms between sensors is calculated. First, the patient condition determination unit 140 calculates an output value using the cross-correlation function of two waveform data using the method described above.

Then, the patient condition determination unit 140 weights the output value and outputs it as an intra-sensor correlation index value. Here, the weighting method will be explained.

For example, referring to FIG. 13, the inter-sensor correlation index is "present" for "supine position", "present/absent" for "prone position", and "absent" for "lateral position".

The output value of the cross-correlation function is between "0 and 1". Here, in FIG. 13, the output value is directly used as the value of the intra-sensor correlation index for the "present" ("supine position") location. In addition, in FIG. 13, for the "none" location ("lateral supine position"), the value of the intra-sensor correlation index is the value obtained by subtracting the output value from the maximum value. Further, in FIG. 13, for the "present/absent" ("prone position") location, the value of the inter-sensor mutual index is set by halving the output value.

To explain a specific example, if the output value of the cross-correlation function is "0.9", the patient condition determination unit 140 determines "supine position = 0.9" and "prone position = 0.45" and "lateral decubitus position = 0.1" are output.

(3) Heartbeat waveform index value calculation (step S210)
A heartbeat waveform index value indicating the extent to which the heartbeat waveform is included in the output waveform is calculated. For example, if the frequency distribution calculation unit 135 determines that the ratio of the power spectrum density of the frequency of the heartbeat component in the high frequency component higher than the frequency of the respiratory component and the frequency that is an integer multiple of the frequency is a certain value or more, the heartbeat waveform is strongly overlapped. It is determined that the

Then, the patient condition determination unit 140 outputs "1" if the heartbeat waveform is included in the waveform, and "0" if it is not included. Furthermore, the patient condition determination unit 140 outputs the weighted output value as the value of the heartbeat waveform index. Here, the weighting method will be explained.

For example, referring to FIG. 13, the heartbeat waveform is smaller in the "supine position" and the "prone position". Also, if you are in the ``lateral position'', it will appear larger. Also, regarding "lateral decubitus position", "left decubitus position" has a large value, but "right decubitus position" has a large value compared to "supine position" and "prone position", but "left decubitus position" It appears small compared to .

Therefore, in the case of the "lateral supine position", the output value is directly output as the heartbeat waveform index value. Further, "supine position" and "prone position" are outputted in a smaller size (for example, multiplied by "0.1", "0", etc.).

(4) Inter-sensor respiration index value (step S212)
The patient condition determination unit 140 calculates an index regarding the shape of peaks and valleys in the waveform. The portion where the waveform changes from a valley to a peak is, for example, the time t1→t2 portion in FIG. 14, and the portion where the peak becomes a valley is, for example, the time t2→t3 portion in FIG. These parts show vibrational transitions (pressure transitions), which typically correspond to exhalation/inhalation. The patient condition determination unit 140 determines whether the time from a valley to a peak or the time from a peak to a valley is the same between sensors, and pairs of sensors with the same correspondence of times included in the posture determination section. The ratio is calculated as the inter-sensor respiration index value. Referring to FIG. 13, the "supine position" and the "prone position" are the same between the sensors.

Therefore, the value of the inter-sensor respiration index is output as is. In addition, in the case of "side-lying position", the ratio of the same pair described above is multiplied by "0.5" and is output as the value of the inter-sensor respiration index, since the ratio may be the same or the opposite in some cases. .

(5) Calculation of respiratory index value (step S214)
The patient condition determination unit 140 calculates an index regarding the comparison result between the time from trough to peak in the waveform and the time from peak to trough.

At this time, the patient condition determination unit 140 compares the length of the time from the valley to the peak included in the posture determination section and the length of the time from the peak to the valley from the waveform, and the length of the two sections is changed from "short to long." ” is detected. Then, the patient condition determination unit 140 outputs the ratio of pairs having a "short→long" relationship as an output value.

Further, referring to FIG. 13, in the case of the "supine position", since there are many "short → long" changes, the patient condition determination unit 140 uses the value obtained by subtracting 0.5 from the calculated output value as the respiratory index. Output as is as the value. In addition, the patient condition determination unit 140 does not output a respiratory index value (outputs "0") since characteristics are difficult to emerge between "prone position" and "lateral decubitus position".

In this way, it becomes possible to determine the patient's posture using each index value, and it becomes possible to determine a more appropriate patient's posture.

Note that in this embodiment as well, the waveform is calculated based on the vibration data, but the patient's posture may be determined simply based on the vibration data. That is, step S204 in FIG. 12 may not be executed, and the process may transition from step S202 to step S206.

For example, the heartbeat waveform index can be determined by frequency analyzing vibration data. Furthermore, in the respiratory index, the peaks and troughs of the waveform can be captured in the same way as the waveform if the maximum value (nearby) and minimum value (nearby) of the vibration data can be extracted.

[3. Third embodiment]
Next, a third embodiment will be described. In this embodiment, a case will be described in which the patient condition determination unit 140 determines the patient's posture using artificial intelligence (machine learning).

In this embodiment, instead of the process in FIG. 5, the patient's posture, which is one of the patient's conditions, is estimated based on the patient condition estimation unit 700 in FIG. 15.

Here, the operation of the patient condition estimating section 700 in this embodiment will be explained. The patient condition estimating unit 700 uses vibration data and the patient's condition as input values (input data), and estimates the patient's posture by using artificial intelligence and various statistical indicators.

As shown in FIG. 15, the patient condition estimation section 700 includes a feature extraction section 710, an identification section 720, an identification dictionary 730, and a patient condition output section 740.

First, various parameters are input as input data to the patient condition estimating section 700 and used. For example, in this embodiment, vibration data as vibration data and waveform data calculated from the vibration data are used.

Then, the feature extraction unit 710 extracts each feature point and outputs it as a feature vector. Here, what the feature extraction unit 710 extracts as feature points may include, for example, the following.

(1) Is there a correlation within the sensor? (2) Is there a correlation between the sensors? (3) Is the heartbeat waveform included in the waveform data? (4) The time from the valley to the peak of the respiratory waveform is from the peak to the peak. Is it short or long compared to the trough time? (5) Is there a high or low incidence of double peak waveforms? (6) Is there a difference in the area above and below the center line in the waveform data? (7) Heartbeat between sensors Are there any differences in waveforms and placement?

The feature extraction unit 710 outputs a feature vector by combining one or more of these feature points. Note that the feature points described are just one example, and are not limited to these values. Moreover, in this way, each value is a value for convenience of explanation. Then, "1" may be output for the applicable feature point, "0" may be output for the non-applicable feature point, or a random variable may be output.

If all of the feature points described above are included, the feature space is seven-dimensional, and is output to the identification unit 720 as a seven-dimensional feature vector.

The identification unit 720 identifies a class corresponding to the patient condition from the input feature vector. At this time, the class is identified by comparing it with a plurality of prototypes prepared in advance as the identification dictionary 730. The prototype may be stored as a feature vector corresponding to each class, or may be stored as a feature vector representing the class.

If a feature vector representing a class is stored, the class to which the closest prototype belongs is determined. At this time, the identification unit 720 may perform the determination using the nearest neighbor decision rule or may perform the identification using the k-nearest neighbor method.

Note that the identification dictionary 730 used by the identification unit 720 may store prototypes in advance, or may store them using machine learning.

Then, in accordance with the class identified by the identification unit 720, the patient status output unit 740 outputs (lying) posture as one of the patient statuses. As the patient's state to be output, "supine position", "prone position", "lateral decubitus position", etc. may be identified, or a random variable may be output as is.

As a result, according to this embodiment, it is possible to acquire vibration data output from the sensor and estimate the patient's posture from this information.

[4. Fourth embodiment]
A fourth embodiment will be described. The fourth embodiment is an embodiment in which the patient condition estimation unit 700 of the third embodiment estimates the patient's posture based on waveform data using deep learning using a neural network.

In this embodiment, the patient's waveform data is input to the patient condition estimation section 700. The patient condition estimation unit 700 estimates the patient condition (posture) from the input waveform data, and recently, deep learning (deep neural network) has been used for this estimation process, which has achieved high accuracy especially in image recognition. . This embodiment also utilizes this method as an example. This deep learning process will be briefly explained using FIG. 16.

First, the patient condition estimation section 700 inputs a signal of waveform data (image data) outputted by the waveform calculation section 130 to a neural network constituted by a plurality of layers and neurons included in each layer. Each neuron receives signals from other neurons and outputs the processed signals to the other neurons. When a neural network has a multilayer structure, the layers are called an input layer, an intermediate layer (hidden layer), and an output layer in the order in which signals flow.

A neural network in which the middle layer consists of multiple layers is called a deep neural network (for example, a Convolutional Neural Network with convolutional operations), and the machine learning method using this is called deep learning. call.

The waveform data is subjected to various operations (convolution operation, pooling operation, normalization operation, matrix operation, etc.) to neurons in each layer of the neural network, and flows while changing its shape, and multiple signals are output from the output layer.

Each of the plurality of output values from the neural network is associated with a patient's posture, and processing is performed such that the patient's posture is inferred to be associated with the output value having the largest value. Alternatively, one or more output values may be passed through a classifier, and the patient's posture may be inferred from the output of the classifier, without directly outputting the posture, which is the patient's condition.

Parameters, which are coefficients used in various calculations of the neural network, are determined by inputting a large amount of waveform data and the patient's posture corresponding to the waveform data into the neural network in advance, and calculating the error between the output value and the correct value by inverting the error. The propagation method is determined by propagating backward through the neural network and updating the parameters of neurons in each layer many times. The process of updating and determining parameters in this way is called learning.

The structure of the neural network and the individual operations are known techniques explained in books and papers, and any of these techniques may be used.

In this way, by using the patient condition estimating unit 700, the posture of the patient is output by referring to the vibration wave data calculated from the vibration data output from the sensor.

In addition, in this embodiment, an example was explained in which a neural network is used to estimate the waveform image data, but it is also possible to estimate the patient's posture by simply inputting vibration data (time-series voltage output values) and letting it learn. It is also possible to do so. Alternatively, the patient's posture may be estimated by inputting data converted into a frequency domain signal using Fourier transform or discrete cosine transform and performing learning.

[5. Application example]
The following application examples can be considered by incorporating the above-mentioned attitude determination device into other devices.

[5.1 Bed device]
FIG. 17 shows the configuration of the bed device. The bed device 10 has a back bottom 12, a waist bottom 14, a knee bottom 16, and a leg bottom 18. The upper body of the user P is supported by the back bottom 12, and the lower back is supported by the waist bottom 14.

The drive control unit 1000 controls the drive of the bed device. Here, the drive control section 1000 realizes the functions of the bottom control section 1100 for controlling functions such as back raising, knee raising (leg lowering), etc. by operating the bottom.

In order to realize the back raising function, a back bottom drive section 1110 and a knee bottom drive section 1120 are connected to the bottom control section 1100. The back bottom drive unit 1110 is, for example, an actuator, and is connected to a link for raising the back via a link mechanism. Then, under the control of the back bottom drive unit 1110, the back bottom 12 placed on the link is operated, and back raising/back lowering control is performed.

Further, the knee bottom drive unit 1120 is, for example, an actuator, and is connected to a knee-raising link via a link mechanism. Then, under the control of the knee bottom drive unit 1120, the knee bottom 16 placed on the link and the connected foot bottom 18 are operated, and knee raising/lowering (leg lowering/leg raising) control is performed. .

When the bed apparatus attempts to raise the back, the bottom control section 1100 does not perform the back raising operation if the patient condition determining section 140 determines that the patient's posture is "prone position." That is, even if the user selects the back-raising operation, the bottom control section 1100 does not drive the back-bottom driving section 1110, and the back-raising operation is not performed. Further, even when the bed apparatus is in automatic operation, if the patient condition determination unit 140 determines that the patient's posture is “prone position”, the bottom control unit 1100 does not perform the back-raising operation.

The operation in this case will be explained with reference to FIG. First, it is determined whether an action has been selected by the user (step S302). For example, the user selects a back-up button using the input unit 160 (operation remote control). Thereby, the control unit 100 determines that the back-raising motion has been selected.

Subsequently, the control unit 100 (patient condition determination unit 140) executes posture determination processing (step S304). In the posture determination process, the patient condition determination unit 140 executes any of the posture determination processes described above to determine the posture of the patient on the bed apparatus.

Here, the control unit 100 determines whether the patient is in a specific posture (step S306). In this application example, the control unit 100 does not perform the back raising operation if the patient's posture is “prone position” (step S306; Yes). If the posture is other than that, the control unit 100 issues an instruction to the bottom control unit 1100 (back bottom drive unit 1110) to perform a back-raising operation (step S306; Yes→step S308).

Then, when the back-raising operation is canceled by the user (for example, the user/patient performs a cancel operation or the back-raising button is released), the control unit 100 stops the back-raising operation ( Step S310; Yes→Step S312).

Furthermore, the control unit 100 also stops the back-raising operation when the patient's posture becomes a specific posture (for example, when the patient becomes prone during the back-raising operation). Step S306; Yes→Step S312).

[5.2 Body position change]
In order to understand the risk of pressure ulcers, the frequency of body position changes and the ratio of each position are automatically memorized. That is, by automatically storing the patient's posture determined by the patient condition determining unit 140, it is utilized for nursing care and treatment.

Further, when a predetermined period of time has elapsed while the posture determined by the patient condition determination unit 140 remains the same, a notification is given or the body position is automatically changed. For example, as shown in FIG. 17, the drive control section 1000 controls the body position change drive section 1200. The body position changing drive unit 1200 changes the patient's body position, for example, by inflating and deflating air cells provided on the left and right sides of the patient, and by raising and lowering the left and right bottoms.

The drive control unit 1000 controls the body position change drive unit 1200 according to the posture determined by the patient condition determination unit 140, and performs control to change the body position of the patient P.

For example, the process in FIG. 18 will be described as an example. When the action of changing the body position is selected (step S302; Yes), the patient condition determination unit 140 determines the patient's posture using one of the methods described above (step S304).

Here, it is determined whether the determined posture is a specific posture. For example, if the patient is in the right lateral position, the body position change drive unit 1200 changes the body position by controlling the air cell provided on the right side. Furthermore, if the patient is in the left lateral position, the body position change drive unit 1200 changes the body position by controlling the air cell provided on the left side. Also, if the patient is in the prone position, no position change is performed.

Further, the body position may be changed when the posture continues for a predetermined period of time. For example, if the posture determined by the posture stabilization process continues to be the same for 10 minutes or more, a process of changing the body position may be executed.

[5.3 Notification device]
A configuration is provided that provides notification in accordance with the patient's posture determined by the patient condition determination unit 140. The notification method may be an audio output device, a display device, or a method of notifying another terminal (for example, a mobile terminal device owned by a medical worker).

As for the timing of notification, for example, if the patient is paralyzed, placing the patient on the paralyzed side will increase the risk of bedsores. Therefore, when the posture determined by the patient condition determination unit 140 is a lateral position with the paralyzed side facing down, notification is made.

Furthermore, in order to prevent infants from suffocating when they sleep on their stomachs, a notification is issued when the patient condition determination unit 140 determines that the infant is in the prone position.

[6. Modified example]
Although the embodiments of this invention have been described above in detail with reference to the drawings, the specific configuration is not limited to these embodiments, and designs etc. within the scope of the gist of this invention are also within the scope of the claims. include.

Furthermore, in this embodiment, the patient's posture is determined in the processing device 5 based on the results output by the detection device 3, but all determinations may be made by one device. In addition to implementing this by installing an application on a terminal device (for example, a smartphone, tablet, or computer), it is also possible to perform processing on the server side and return the processing results to the terminal device.

For example, the above-described processing may be implemented on the server side by uploading vibration data from the detection device 3 to the server. This detection device 3 may be realized by a device such as a smartphone having a built-in acceleration sensor or vibration sensor, for example.

Further, in the embodiment described above, it has been explained that there are two vibration sensors, but more than two vibration sensors may be provided. Further, the method of calculating the frequency distribution and determining the posture according to the first embodiment can be implemented using only one sensor.

Further, in the embodiment, the program that runs on each device is a program that controls the CPU, etc. (a program that causes the computer to function) so as to realize the functions of the embodiment described above. The information handled by these devices is temporarily stored in a temporary storage device (for example, RAM) during processing, and then stored in various ROM, HDD, and SSD storage devices, and read out by the CPU as necessary. , corrections and writing are performed.

Furthermore, when distributing the program on the market, the program can be stored in a portable recording medium and distributed, or it can be transferred to a server computer connected via a network such as the Internet. In this case, it goes without saying that the storage device of the server computer is also included in the present invention.

1 System 3 Detection device 5 Processing device 10 Bed device 12 Back bottom 14 Waist bottom 16 Knee bottom 18 Foot bottom 20 Mattress 100 Control section 110 Vibration detection section 120 Biological information calculation section 130 Waveform calculation section 135 Frequency distribution calculation section 140 Patient condition determination Section 150 Storage section 152 Vibration data storage area 154 Waveform data storage area 160 Input section 170 Output section 700 Patient condition estimation section 710 Feature extraction section 720 Identification section 730 Identification dictionary 740 Patient condition output section

Claims (4)

at least two vibration sensors;
a control unit that detects vibrations of the subject while lying on the bed using the vibration sensor, and determines the posture of the subject from the waveform of the detected vibrations;
When the subject's posture is a specific posture, the button does not operate even if the button on the input section is selected, and when the subject's posture is not in the specific posture, when the button on the input section is selected. a drive unit that operates the button;
Equipped with
In the electric furniture, the control unit determines that the subject is in a lateral position if there is no correlation between the two vibration sensors. at least two vibration sensors;
a control unit that detects vibrations of the subject while lying on the bed using the vibration sensor, and determines the posture of the subject from the waveform of the detected vibrations;
When the subject's posture is a specific posture, the button does not operate even if the button on the input section is selected, and when the subject's posture is not in the specific posture, when the button on the input section is selected. a drive unit that operates the button;
Equipped with
In the electric furniture, the control unit determines that the subject is in a supine position if there is a correlation between the two vibration sensors and a correlation within the vibration sensor. at least two vibration sensors;
a control unit that detects vibrations of the subject while lying on the bed using the vibration sensor, and determines the posture of the subject from the waveform of the detected vibrations;
When the subject's posture is a specific posture, the button does not operate even if the button on the input section is selected, and when the subject's posture is not in the specific posture, when the button on the input section is selected. a drive unit that operates the button;
Equipped with
In the electric furniture, the control unit determines that the posture of the subject is prone when there is a correlation between the two vibration sensors and there is no correlation within the vibration sensor. The electric furniture according to any one of claims 1 to 3, wherein the specific posture is a prone position, and the button is a back-up button.

Images