Bed system

Abstract

The present invention provides a posture determination device having at least two vibration sensors and a control unit. The control unit is characterized in that it detects vibrations when a user lies in bed through the vibration sensors, calculates waveforms based on the detected vibrations, determines the user's posture according to the characteristics of the waveforms by identifying the characteristics of the waveforms. Thus, a posture determination device and the like that can appropriately determine the posture as the state of the user can be provided.

Description

[0001] This application is a divisional application of the following patent application:

[0002] Application No.: 201880020409.6

[0003] Filing Date: December 5, 2018

[0004] Title of Invention: Posture Determination Device Technical Field

[0005] The present invention relates to a posture determination device. Background Art

[0006] Conventionally, an invention for determining the state of a user (patient) in bed has been known. For example, the detection device disclosed in Patent Document 1 detects vibrations in the bed and extracts a heartbeat vibration signal originating from a living body's heartbeat. Then, the detection device estimates the posture of the living body in the bed based on the extracted heartbeat vibration signal.

[0007] In addition, the notification device disclosed in Patent Document 2 acquires a user state including at least one of the posture of the user and the position of the user in or out of the bed. Then, the notification device makes a notification based on the acquired user state.

[0008] Prior Art Documents

[0009] Patent Documents

[0010] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2011 - 120667

[0011] Patent Document 2: Japanese Unexamined Patent Application Publication No. 2016 - 192998 Summary of the Invention

[0012] Technical Problem to be Solved by the Invention

[0013] The present invention provides a posture determination device that can appropriately determine the posture of a user based on the waveform of vibrations of the user.

[0014] Technical Solution for Solving the Problem

[0015] The present invention provides a posture determination device, characterized by comprising: at least two vibration sensors capable of detecting vibrations when a user lies in bed; and a control unit that calculates a plurality of waveforms based on the plurality of vibrations and determines the posture of the user when lying in bed based on the characteristics of the plurality of waveforms.

[0016] Advantages of the Invention

[0017] According to the posture determination device of the present invention, it is possible to appropriately determine the posture of a user when lying in bed as one of the states of the user, based on a plurality of waveforms calculated from vibrations. Brief Description of the Drawings

[0018] Figure 1 This is a diagram for explaining the overall first embodiment.

[0019] Figure 2 This is a diagram for explaining the configuration of the first embodiment.

[0020] Figure 3 This is a diagram for explaining the configuration of the first embodiment.

[0021] Figure 4 This is a diagram for explaining the sensor of the first embodiment.

[0022] Figure 5 This is a diagram for explaining the posture determination process of the first embodiment.

[0023] Figure 6 This is a diagram for explaining the posture determination process of the first embodiment.

[0024] Figure 7 This is a diagram showing an example of the waveform in the prone position of the first embodiment.

[0025] Figure 8 This is a diagram showing an example of the waveform in the supine position of the first embodiment.

[0026] Figure 9 This is a diagram showing an example of the waveform in the right lateral position of the first embodiment.

[0027] Figure 10 This is a diagram showing an example of the waveform in the left lateral position of the first embodiment.

[0028] Figure 11 This is a diagram showing an example of the frequency component of the first embodiment.

[0029] Figure 12 This is a diagram for explaining the posture determination process of the second embodiment.

[0030] Figure 13 This is a diagram for explaining the posture determination conditions of the second embodiment.

[0031] Figure 14 This is a diagram for explaining the state of the waveform of the second embodiment.

[0032] Figure 15 This is a diagram for explaining the functional configuration of the state estimation unit of the third embodiment.

[0033] Figure 16 This is a diagram for explaining the neural network of the fourth embodiment.

[0034] Figure 17 This is a diagram for explaining a bed as an application example.

[0035] Figure 18 This is a diagram for explaining the processing in the case of the application example. Detailed implementation manner

[0036] Hereinafter, an embodiment for implementing the present invention will be described with reference to the accompanying drawings. From the viewpoint of preventing a user from falling when the user gets out of bed, a conventional device detects whether the user is getting out of bed (whether the user is in bed). However, the device does not detect what posture the user is in when the user is in bed.

[0037] Regarding the posture of the user, for example, when the device detects a user sitting upright, etc., sometimes the user etc. set many devices (such as sensors) in a wide range of the bed, and the device discriminates between a lying posture and a sitting posture. However, the device cannot determine the direction of the user when the user is sleeping in bed (for example, supine, prone, and side-lying, etc.).

[0038] In order for the device to determine the posture of the user, the user etc. additionally set, for example, a camera device etc., and it is necessary for medical staff, workers, family members, etc. (hereinafter, referred to as staff etc.) to monitor the user, or the device analyzes the images captured by the camera device etc. In this case, it is necessary for the staff etc. to monitor the user all the time, increasing the burden on the staff etc. In addition, in order to monitor the user, it is necessary for the camera device to capture the user frequently. Therefore, from the viewpoint of privacy, sometimes the user will also object to the capture by the camera device.

[0039] Therefore, according to the posture determination device of the present embodiment, the posture of the user can be detected only by detecting the vibration when the user is sleeping in bed.

[0040] It should be noted that in this specification, the user refers to a person using the bed (or mattress), and is not limited to a person receiving treatment due to illness, and as long as it is a person receiving care in a facility or a person sleeping in bed, it can be applied as a user.

[0041] [1. First Embodiment]

[0042] [1.1 Overall System]

[0043] Figure 1 This 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 Figure 1 shown, the system 1 includes a detection device 3 placed between the bed board and the mattress 20 of the bed 10, and a processing device 5 for processing the value output by the detection device 3. The system 1 determines the posture of the user by using the detection device 3 and the processing device 5.

[0044] When the user P is on the bed 10 (or the mattress 20), the detection device 3 detects the body vibration (the vibration emitted by the human body) as the biological signal of the user, that is, the user P. In addition, the processing device 5 calculates the biological information value of the user P based on the vibration detected by the detection device 3.

[0045] The processing device 5 can also output / display the calculated biological information value (for example, respiratory rate, heart rate, activity level) as the biological information value of the user P.

[0046] In addition, by providing a storage unit, a display unit, etc. in the detection device 3, the detection device 3 can also be integrally formed with the processing device 5. In addition, the processing device 5 can also be a general-purpose device, not limited to information processing devices such as computers. For example, it can also be a device such as a tablet computer or a smart phone. In addition, when the detection device 3 has a communication function, the detection device 3 can also be connected (communicate) with the server device instead of the processing device 5.

[0047] In addition, the user can also be a person who is recuperating from illness or a person who needs care. In addition, the user can be either a healthy person who does not need care, or an elderly person, a child, a disabled person, or an animal instead of a human.

[0048] Here, the detection device 3 is, for example, a thin sheet-like device. Thus, even if the detection device 3 is placed between the bed 10 and the mattress 20, the user P can use the detection device 3 without feeling discomfort. Therefore, the detection device 3 can detect the state of the user on the bed for a long time.

[0049] It should be noted that the detection device 3 only needs to detect the vibration of the user P. For example, the detection device 3 can also use an actuator with a strain gauge, a load sensor that measures the load on the legs of the bed 10, etc. In addition, the detection device 3 can also use a smart phone placed on the bed 10, an acceleration sensor built into a tablet computer, etc.

[0050] In addition, in Figure 1 , the direction toward the head side of the bed 10 (mattress 20) is set as the direction H, and the direction toward the foot side is set as the direction F. In addition, in Figure 1 , when the user P is lying on the back on the bed 10, the direction toward the left side of the user P is set as the direction L, and the direction toward the right side of the user P is set as the direction R.

[0051] [1.2 Composition]

[0052] Refer to Figures 2 to 4The configuration of system 1 will be described. System 1 in this embodiment includes a detection device 3 and a processing device 5. In addition, each functional unit other than the detection unit 110 may include either the detection device 3 or the processing device 5. Any one of the detection device 3 and the processing device 5 implements the functions of each functional unit other than the detection unit 110. By combining the detection device 3 and the processing device 5, the detection device 3 and the processing device 5 function as a posture determination device.

[0053] System 1 (posture determination device) includes a control unit 100, a detection unit 110, a first calculation unit 120, a second calculation unit 130, a third calculation unit 135, a determination unit 140, a storage unit 150, an input unit 160, and an output unit 170.

[0054] The control unit 100 controls the operation of system 1. The control unit 100 is, for example, a control device such as a CPU (Central Processing Unit). The control unit 100 realizes various processes by reading various programs stored in the storage unit 150 and executing the various programs. It should be noted that in this embodiment, one control unit 100 is provided as a whole, but it may also be separately provided in the detection device 3 and the processing device 5 as described later Figure 4 like that.

[0055] The detection unit 110 detects vibrations on the detection device 3 and obtains vibration data. As an example, the detection unit 110 in this embodiment uses a sensor that detects pressure changes to detect vibrations (body vibrations) based on the user's actions, etc. In addition, the detection unit 110 obtains vibration data based on the detected vibrations. The detection unit 110 outputs the vibration data to the first calculation unit 120, the second calculation unit 130, and the third calculation unit 135. It should be noted that the vibration data can be analog vibration data or digital vibration data.

[0056] In addition, the detection unit 110 can also, for example, detect the user's vibrations with a pressure sensor and obtain vibration data. In addition, a microphone can be provided instead of the pressure sensor in the detection unit 110. The detection unit 110 can also obtain a biological signal based on the sound picked up by the microphone and obtain vibration data according to the biological signal. In addition, the detection unit 110 can also obtain vibration data based on the output values of an acceleration sensor, a capacitance sensor, and a load sensor. In this way, the detection unit 110 only needs to be able to obtain a biological signal (body vibration indicating the user's body movement) by any method.

[0057] The first calculation unit 120 obtains the user's biological signals based on the vibration data, and calculates biological information values (such as respiratory rate / heart rate / activity level, etc.). In the present embodiment, the respiratory component / heartbeat component is extracted from the vibration (body vibration) data obtained by the detection unit 110. The first calculation unit 120 may also calculate the respiratory rate and heart rate based on the extracted respiratory component / heartbeat component, based on the respiratory interval and heartbeat interval. In addition, the first calculation unit 120 may also analyze the periodicity of the vibration data (such as Fourier transform, etc.), and calculate the respiratory rate and heart rate based on the peak frequency.

[0058] The method for calculating these biological information values of the user can refer to, for example, Japanese Patent Application Laid-Open No. 2010-264193 (invention name: sleep state determination device, program, and sleep state determination system, application date: May 18, 2009), Japanese Patent Application Laid-Open No. 2015-12948 (invention name: sleep evaluation device, sleep evaluation method, and sleep evaluation program, application date: July 4, 2013). The entire content of this patent application is incorporated herein by reference.

[0059] The second calculation unit 130 converts the analog vibration data input from the detection unit 110 into a digital voltage signal at a specified sampling interval, and calculates waveform data representing the waveform of the body vibration (vibration waveform).

[0060] The third calculation unit 135 calculates the frequency distribution based on the waveform data. For example, the third calculation unit 135 calculates the frequency distribution based on the frequency components obtained by performing a fast Fourier transform (FFT: Fast Fourier Transform) on the waveform data calculated by the second calculation unit 130. It should be noted that the third calculation unit 135 may also directly calculate the frequency distribution based on the vibration data output from the detection unit 110.

[0061] The determination unit 140 determines the user's state. For example, the determination unit 140 uses at least one of the vibration data obtained by the detection unit 110, the biological information values calculated by the first calculation unit 120, the values obtained from a load sensor separately provided on the bed 10, the waveform data calculated by the second calculation unit 130, the frequency distribution calculated by the third calculation unit 135, etc. to determine the user's state. The determination unit 140 may also combine multiple data, values, etc. to determine the user's state.

[0062] In the present embodiment, the determination unit 140 determines the user's posture on the bed (which one of supine, prone, and side-lying the user is) as the user's state. In addition, the determination unit 140 may also determine the user's posture such as sitting posture, or states other than the user's posture (such as getting out of bed, being in bed, etc.).

[0063] The storage unit 150 stores various data and programs used for the operation of the system 1. The control unit 100 realizes various functions by reading out and executing these programs. Here, the storage unit 150 is composed of a semiconductor memory (e.g., SSD (Solid State Drive), SD card (registered trademark)), a disk device (e.g., HDD (Hard Disk Drive)), etc. In addition, the storage unit 150 can be an internal storage device or a detachable external storage device. Further, the storage unit 150 can also be a storage area of an external server such as the cloud.

[0064] The storage unit 150 includes a vibration data storage area 152 and a waveform data storage area 154. The vibration data storage area 152 and the waveform data storage area 154 are respectively allocated to each area inside the storage unit 150.

[0065] The vibration data storage area 152 stores the vibration data output from the detection unit 110. Here, the vibration data is stored once every predetermined time. For example, the vibration data is stored at relatively short intervals such as every 1 second or 5 seconds, and at relatively long intervals such as every 30 seconds, 1 minute, or 5 minutes.

[0066] The waveform data storage area 154 stores the waveform data (waveform) of the vibration calculated by the second calculation unit 130 based on the vibration data output from the detection unit 110 or the vibration data stored in the vibration data storage area 152. It should be noted that in this embodiment, the case where the waveform data is set to be stored in the waveform data storage area 154 is described, but the second calculation unit 130 can also calculate as needed each time. In addition, the waveform data can be temporarily stored in the waveform data storage area 154 or can be cumulatively stored.

[0067] The input unit 160 receives operations from the user. For example, the input unit 160 receives inputs of operations indicating the start of obtaining the user's vibration, operations indicating the adjustment of the sensitivity of the detection unit 110, and various operations from the user. The input unit 160 is, for example, a keyboard, a mouse, a touch screen, etc.

[0068] The output unit 170 outputs various information. For example, it is a display device such as a liquid crystal display, a light-emitting component such as an LED, a speaker that outputs sound and voice, an interface that outputs data to other recording media, etc. In addition, the input unit 160 and the output unit 170 can be integrally formed. For example, it can also be a single touch screen that serves as both the input unit 160 and the output unit 170.

[0069] Here, in the above configuration, the first calculation unit 120, the second calculation unit 130, the third calculation unit 135, and the determination unit 140 are mainly implemented by software (program). For example, the control unit 100 reads the software stored in the storage unit 150, and the control unit 100 executes the software. As a result, the control unit 100 realizes the functions of each component.

[0070] That is, by reading and executing the programs that implement the first calculation unit 120, the second calculation unit 130, the third calculation unit 135, and the determination unit 140, the control unit 100 has the functions of each component.

[0071] In addition, Figure 2 is a diagram showing the configuration of the posture determination device of the system 1. These components can also be implemented by, for example, a single device capable of detecting vibration, as Figure 1 shown, the posture determination device can also be composed of a detection device 3 and a processing device 5. In addition, an external server capable of providing the same service can be used instead of the processing device 5 to implement the posture determination device.

[0072] Regarding Figure 2 the case where the system 1 of Figure 1 is composed of the detection device 3 and the processing device 5 of Figure 3 the posture determination device of the system 1 will be described. The detection device 3 includes a control unit 300, a detection unit 320 as a sensor, a storage unit 330, and a communication unit 390.

[0073] In addition, the control unit 300 executes the software (program) stored in the storage unit 330, whereby the control unit 300 functions as the first calculation unit 310. The detection unit 320 outputs vibration data based on the detected vibration.

[0074] The first calculation unit 310 calculates a biological information value based on the vibration data. Then, the detection device 3 stores the biological information value in the biological information value data 340, or sends it to the processing device 5 via the communication unit 390. In addition, at the same time, the detection device 3 also sends the vibration data detected by the detection unit 320 to the processing device 5 via the communication unit 390.

[0075] The timing of sending the vibration data from the detection device 3 to the processing device 5 and the timing of storing the biological information value (biological information) in the biological information value data 340 can be real-time transmission or transmission at regular intervals.

[0076] It should be noted that the detection unit 320 is Figure 2 the detection unit 110 of Figure 4 Here, the detection unit 320 will be described using

[0077] Figure 4This is a view of the bed 10 (mattress 20) observed from above. Figure 4 In it, the direction towards the upper side is Figure 1 the direction H, and the direction towards the lower side is Figure 1 the direction F. Additionally, the direction towards Figure 4 the right side of Figure 1 is the direction L, and the direction towards Figure 4 the left side of Figure 1 is the direction R.

[0078] The detection device 3 is placed between the bed 10 and the mattress 20 or on the mattress 20. As the part where the detection device 3 is placed, the vicinity of the user's spine is preferred, so it is at least in the direction closer to the H side (the user's head side) than the center of the bed 10 (mattress 20).

[0079] Moreover, the detection device 3 incorporates sensors (detection units 110 / 320). This sensor is, for example, a vibration sensor that can detect the vibration of the user (body vibration). And at least two sensors are provided in the detection device 3. For example, in Figure 4 two sensors (vibration sensors 320a, 320b) are provided on the left and right of the detection device 3.

[0080] The vibration sensor 320a is arranged away from the vibration sensor 320b. The interval between the vibration sensor 320a and the vibration sensor 320b can be, for example, the interval when the user is located on the sensors. Preferably, the interval between the two sensors is about 15 - 60 cm.

[0081] In addition, Figure 3 the first calculation unit 310 is Figure 2 the first calculation unit 120. Moreover, the communication unit 390 is, for example, an interface capable of connecting (communicating) with a network (e.g., LAN / WAN).

[0082] The processing device 5 includes a control unit 500, a storage unit 530, an input unit 540, an output unit 550, and a communication unit 590. The processing device 5 receives vibration data from the detection device 3 via the communication unit 590. The processing device 5 stores the received vibration data in the vibration data storage area 532.

[0083] The control unit 500 executes the software (program) stored in the storage unit 530. Thus, the control unit 500 functions as a second calculation unit 502, a third calculation unit 504, and a determination unit 506. Additionally, the waveform data calculated by the second calculation unit 502 is stored in the waveform data storage area 534.

[0084] It should be noted that the second calculation unit 502 is Figure 2 the second calculation unit 130. The third calculation unit 504 isFigure 2 The third calculation unit 135. The determination unit 506 is Figure 2 The determination unit 140. The input unit 540 is Figure 2 The input unit 160. The output unit 550 is Figure 2 The output unit 170. The storage unit 530 is Figure 2 The storage unit 150.

[0085] [1.3 Processing Flow]

[0086] Refer to Figure 5 The posture determination process in this embodiment will be described. The posture determination process is a process executed by the control unit 100 (determination unit 140).

[0087] First, the control unit 100 (determination unit 140) acquires vibration data (step S102). Specifically, the determination unit 140 reads vibration data from the vibration data storage area 152 to acquire vibration data, or receives it from the detection unit 110 to acquire vibration data.

[0088] Next, the determination unit 140 determines the user's posture based on the vibration data. At this time, the determination unit 140 determines the user's posture based on the correlation between sensors and the correlation within a sensor. Here, the correlation between sensors refers to the correlation relationship of multiple data output by multiple sensors. In addition, the correlation within a sensor refers to the correlation relationship of multiple data output by one sensor.

[0089] First, the second calculation unit 130 calculates a waveform based on the vibration data and outputs it as waveform data (step S104). The second calculation unit 130 outputs the waveform data to each vibration sensor. For example, in Figure 4 There are two vibration sensors (vibration sensor 320a and vibration sensor 320b). Therefore, the second calculation unit 130 also outputs the waveform data to each vibration sensor. The second calculation unit 130 outputs two pieces of waveform data.

[0090] In addition, the second calculation unit 130 can also store the waveform data in the waveform data storage area 154, or output it to the output unit 170. When the output unit 170 is a display device, the output unit 170 displays the waveform data.

[0091] Next, the determination unit 140 determines whether there is a correlation between sensors based on each waveform data (step S106). For example, the determination unit 140 determines whether there is similarity in the waveform data output to each of the two sensors (for example, Figure 4 Vibration sensor 320a and vibration sensor 320b of

[0092] The determination unit 140 obtains the correlation between sensors for two waveform data by using the cross-correlation function. The determination unit 140 can output a value normalized to "0" to "1" based on the similarity of the two waveform data by using the cross-correlation function. The value output by using this cross-correlation function changes according to the similarity of the two waveform data. For example, when the value of the cross-correlation function is "1", the two waveform data are exactly the same and the similarity is the maximum. In addition, when the value of the cross-correlation function is "0", the two waveforms are completely different and the similarity is the minimum.

[0093] Moreover, when the determination unit 140 determines whether there is a correlation in the two waveform data, the determination unit 140 determines whether the output value of the cross-correlation function exceeds a threshold value. For example, when the threshold value is set to "0.7", if the output value of the cross-correlation function is "0.7" or less, the determination unit 140 determines that there is no correlation in the two waveform data. If the output value of the cross-correlation function exceeds "0.7", the determination unit 140 determines that there is a correlation in the two waveform data. That is, when there is a correlation in the two waveform data, the determination unit 140 determines that there is a correlation between sensors.

[0094] Next, when there is a correlation between sensors (step S106; Yes), the determination unit 140 determines that the user's posture is "supine or prone" (steps S116 to S120). It should be noted that usually, since the time when the user is prone during sleep is very short, the determination unit 140 can also determine the user's posture as only "supine or prone". However, there are also reports that the risk of prone asphyxia is high and it is related to sudden infant death syndrome. In addition, the purpose is to prohibit the automatic operation of the electric bed when the user is prone. Therefore, in order to distinguish between supine and prone, the determination unit 140 can also determine whether the user's posture is "supine" and whether the user's posture is "prone".

[0095] The determination unit 140 determines whether there is a correlation within the sensor (step S106; Yes → step S116). Here, the determination unit 140 determines whether there is a correlation within the sensor. For example, the determination unit 140 evaluates the periodic intensity in the waveform data to determine whether there is a correlation within the sensor. As an example, for the waveform data of one sensor, the determination unit 140 determines whether there is a correlation by using the autocorrelation function. The autocorrelation function outputs a value normalized to "0" to "1" based on the periodic intensity output of the waveform within the same sensor. For example, when the value of the autocorrelation function is "1", the waveform data is output completely periodically, and the determination unit 140 determines that there is a complete correlation within the sensor. In addition, when the value of the autocorrelation function is "0", the determination unit 140 determines that there is no correlation within the sensor.

[0096] In addition, the determination unit 140 may also calculate the periodicity intensity using Fourier transform or chi-square periodogram, etc., as a value normalized to "0" to "1", and determine the correlation within the sensor based on this value.

[0097] Moreover, when the determination unit 140 determines whether there is a correlation in one waveform data, it may also determine whether the output value of the autocorrelation function exceeds a threshold value. For example, when the threshold value is set to "0.7", if the output value of the autocorrelation function is "0.7" or less, the determination unit 140 determines that the calculated waveform data (one detected vibration data) has no correlation. If the output value of the autocorrelation function exceeds "0.7", the determination unit 140 determines that there is a correlation in this waveform data.

[0098] When the determination unit 140 determines that there is a correlation within the sensor, the determination unit 140 determines the user's posture as "supine" (step S116; Yes → step S118). On the other hand, when the determination unit 140 determines that there is no correlation within the sensor, the determination unit 140 determines the user's posture as "prone" (step S116; No → step S120).

[0099] In addition, when the determination unit 140 determines in step S106 that there is no correlation between the sensors (step S106; No), the determination unit 140 determines the user's posture as "side lying" (step S110). It should be noted that the determination unit 140 may only determine the user's posture as "side lying", but in order to distinguish between "right side lying" and "left side lying", the determination unit 140 may further determine whether the user's posture is "right side lying" and whether the user's posture is "left side lying" (step S114). Staff, etc., in order to confirm whether the user has changed body position, or when a part of the user is numb, etc., staff, etc. must pay attention to whether the user sleeps on the side with the numb side down, so the determination unit 140 may also determine whether the user's posture is "right side lying" and whether the user's posture is "left side lying".

[0100] In this case, the heartbeat signal when the user is lying on the left side becomes greater than the heartbeat signal when the user is lying on the right side. Therefore, when the magnitude of the heartbeat signal is set to be above a specified threshold value (in the case of extracting a high-frequency signal), the determination unit 140 determines that the user's posture is "left side lying".

[0101] As a method for the determination unit 140 to determine the magnitude of the heartbeat signal, there are various methods. For example, the determination unit 140 may use the ratio of the frequency component corresponding to the heartbeat signal to the frequency component corresponding to the breathing signal, the signal intensity of the data after high-pass filter processing, etc., to determine the magnitude of the heartbeat signal.

[0102] In this way, the posture determination device of the present embodiment can determine the user's posture (sleeping posture) based on the vibration data.

[0103] In addition, for the sake of description, the determination unit 140 of the present embodiment determines the presence or absence of correlation between sensors and the presence or absence of correlation within the sensor based on the waveform data calculated by the second calculation unit 130. However, it is not limited thereto. For example, the determination unit 140 may simply determine the correlation between sensors and the correlation within the sensor based on the vibration data. In this case, the determination unit 140 may not perform the process of step S104.

[0104] In addition, the determination unit 140 calculates waveform data based on the vibration data of the user to determine the user's posture. However, it is not limited thereto. For example, the determination unit 140 may also evaluate the shape of the frequency distribution to determine the posture. In this case, the determination unit 140 comprehensively evaluating the shape of the frequency distribution of the vibrations obtained by two or more sensors will be more accurate, but the sensor for obtaining the vibration may also be one.

[0105] For example, with reference to Figure 6 , the process in which the determination unit 140 uses frequency analysis and determines the user's posture will be described. Similar to Figure 5 , the determination unit 140 obtains vibration data from the detection unit 110 (step S152), and the second calculation unit 130 calculates (outputs) waveform data (step S154).

[0106] Next, the third calculation unit 135 calculates the frequency distribution of the waveform data output by the second calculation unit 130 (step S156). For example, when the high-frequency components in the frequency distribution are not calculated, the determination unit 140 designates the user's posture as "lying on the side".

[0107] Due to the relationship between the position of the heart and the position of the sensor varying depending on the user's posture, the body tissues (muscles, fat, bones, internal organs, etc.) in between are also different. Therefore, since the vibration transmission method from the user to the sensor also changes, differences appear in the frequency components measured by the system 1.

[0108] Determine the movement of the heart and the directions of the chest and abdominal movements caused by breathing. For example, regarding the chest and abdominal movements caused by breathing, when the user is lying on their back, the movement in the direction perpendicular to the sensor (detection unit 110) increases. In addition, when the user is lying on their side, the movement in the direction parallel to the sensor (detection unit 110) increases. Thus, the frequency distribution is different for each user posture. Therefore, a frequency distribution is stored for each posture, and the determination unit 140 can determine the posture by comparing the actually extracted frequency distribution with the frequency distributions stored for each posture.

[0109] For example, when the user is lying on their side, it is difficult to detect components with a frequency higher than the heartbeat component frequency (except for integer multiples of the frequency equivalent to the heartbeat component frequency). Therefore, when the high-frequency components are fewer than the heartbeat components, the determination unit 140 may determine the user's posture as "lying on the side".

[0110] It should be noted that in the above description, the third calculation unit 135 calculates the frequency distribution based on the waveform data, but it may also directly calculate the frequency distribution based on the vibration data. In this case, step S154 is not executed.

[0111] [1.4 Conditions for Posture Determination]

[0112] Here, the above-mentioned posture determination process may also be executed by the determination unit 140 when the user is in bed. In addition, the posture determination process may also be executed by the determination unit 140 when the states of the user and the bed reach specified conditions. Hereinafter, the posture determination process uses the user's body movement as the condition for the determination unit 140 to execute the posture determination process. Here, the user's body movement refers to a situation where the user's posture changes, such as when the user turns over.

[0113] (1) The determination unit 140 determines the user's posture for each interval.

[0114] The determination unit 140 divides the time for determining the user's posture into multiple intervals. For example, in an interval without body movement, the determination unit 140 determines that the user's posture continues in the same posture. Moreover, the determination unit 140 outputs the user as being in the same posture.

[0115] In this case, the determination unit 140 may also analyze the intervals without the user's body movement uniformly (applying the results of the posture determination process). In addition, the determination unit 140 may also summarize the results of the posture determination process in a certain fixed interval (for example, every 3 minutes, etc.). When there is an interval without body movement in this fixed interval, the determination unit 140 outputs the determination result of the posture with the largest number in the interval without body movement as the result of the posture determination process for this interval.

[0116] (2) The determination unit 140 determines the user's posture only in intervals without body movement

[0117] The determination unit 140 determines the user's posture only in intervals without the user's body movement. For example, when the detection unit 110 detects the user's body movement, the determination unit 140 suspends the execution of the posture determination process. In addition, after a specified time (for example, 10 seconds after the body movement is no longer detected) from when the detection unit 110 detects the user's body movement, the determination unit 140 executes the posture determination process again.

[0118] [1.5 Explanation Based on Waveform]

[0119] Here, a method for the posture determination unit 140 to determine the user's posture based on waveforms will be described. Figures 7 to 10 Each of the waveforms shown is a waveform based on vibrations detected by two vibration sensors. In Figures 7 to 10 the waveform, the horizontal axis represents time, and the vertical axis represents the voltage value, respectively representing the waveforms based on the vibrations detected by the detection unit 110. It should be noted that, for ease of description, a method for the posture determination unit 140 to determine the user's posture based on waveforms will be described. As a method for the posture determination unit 140 to determine the posture, for example, the trend may also be detected from the time-series vibration data to determine the user's posture.

[0120] Figure 7 The two waveforms shown have no correlation with the waveforms inside the sensor and do not repeat the same shape. That is, this waveform is a waveform indicating that the user is in a prone posture ( Figure 5 step S120).

[0121] Figure 8 The two waveforms shown have a correlation with the waveforms between the sensors and also have a correlation with the waveforms inside the sensor. That is, this waveform is a waveform indicating that the user is in a supine posture ( Figure 5 step S118).

[0122] Figure 9 and Figure 10 The two waveforms shown are waveforms that have no correlation between the sensors. That is, this waveform is a waveform indicating that the user is in a side-lying posture ( Figure 5 step S110). Comparing Figure 9 and Figure 10 , in the waveform of the left side-lying position, i.e., the waveform shown in Figure 10 , a high-frequency signal generated by the heartbeat can be clearly seen.

[0123] Figure 11 is a chart for obtaining the frequency distribution according to the waveform. Figure 11 is a chart showing the frequency distribution obtained in Figure 6 step S156, Figure 11 (a) of Figure 11 is a diagram of the chart showing the supine position,

[0124] [2. Second Embodiment]

[0125] The second embodiment will be described. The second embodiment is an embodiment for determining the user's posture based on multiple posture determination conditions.

[0126] It should be noted that theFigure 5 The operation process is replaced with Figure 12 The implementation manner of the operation process. In this implementation manner, the description of the parts that are the same as those in the first implementation manner in terms of configuration, etc. is omitted.

[0127] First, the determination unit 140 acquires vibration data (step S202). Then, the second calculation unit 130 calculates a waveform (step S204). Next, the determination unit 140 calculates an index (value) for each condition (steps S206 to step S214). The determination unit 140 calculates a total value based on the calculated index values (step S216).

[0128] The determination unit 140 calculates the total values corresponding to "supine", "prone", and "(left / right) side lying" as the total value. Then, the determination unit 140 determines the posture that becomes the maximum total value as the user's posture.

[0129] Here, in steps S206 to step S214, the determination unit 140 calculates index values for each condition. Refer to Figure 13 The calculation method of the index value is described.

[0130] (1) First index calculation (step S206)

[0131] The determination unit 140 calculates the value of the first index calculated based on the correlation of the waveform in the sensor. First, the determination unit 140 calculates the value of the autocorrelation function based on the waveform in the sensor by using the method described in the first implementation manner.

[0132] Then, the determination unit 140 weights the output value of the autocorrelation function and calculates the value of the first index. Here, the method of weighting performed by the determination unit 140 is described.

[0133] Figure 13 The table of... shows the waveform characteristics of the sensor for each posture of the user. For example, the correlation based on the waveform in the sensor is "present" when the user's posture is "supine", "present" when the user's posture is "side lying", and "absent" when the user's posture is "prone".

[0134] Here, the determination unit 140 executes the autocorrelation function and outputs a value between "0 and 1". Here, in Figure 13 , when the correlation based on the waveform in the sensor is "present" ("supine", "side lying"), the determination unit 140 directly uses the output value as the value of the first index. In addition, in Figure 13 , when the correlation based on the waveform in the sensor is "absent", the determination unit 140 uses the value obtained by subtracting the output value of the autocorrelation function from the maximum value as the value of the first index.

[0135] A specific example will be described. When the determination unit 140 determines that the output value of the autocorrelation function is "0.8", the value of the first index is 0.8 when in the "supine" position, 0.2 when in the "prone" position, and 0.8 when in the "lateral" position.

[0136] (2) Second index calculation (step S208)

[0137] The determination unit 140 calculates the value of the second index calculated based on the correlation between the waveforms of the sensors. First, the determination unit 140 calculates the output value of the cross-correlation function using the above method with two waveform data.

[0138] Then, the determination unit 140 weights the output value and calculates the value of the second index. Here, the method of weighting performed by the determination unit 140 will be described.

[0139] For example, referring to Figure 13 , the correlation based on the waveforms between the sensors is, for example, "present" when the user's posture is "supine", "present" or "absent" when the user's posture is "prone", and "absent" when the user's posture is "lateral".

[0140] The determination unit 140 executes the cross-correlation function and outputs a value between "0" and "1". Here, in Figure 13 , when the correlation based on the waveforms between the sensors is "present" ("supine"), the determination unit 140 directly uses the output value as the value of the second index. In addition, in Figure 13 , when the correlation based on the waveforms between the sensors is "absent" ("lateral"), the determination unit 140 uses the value obtained by subtracting the output value from the maximum value as the value of the second index. In addition, in Figure 13 , when the correlation based on the waveforms between the sensors is "present" or "absent" ("prone"), the determination unit 140 uses the value obtained by halving the output value as the value of the second index.

[0141] A specific example will be described. When the determination unit 140 determines that the output value of the cross-correlation function is "0.9", the value of the second index is 0.9 when in the "supine" position, 0.45 when in the "prone" position, and 0.1 when in the "lateral" position.

[0142] (3) Third index calculation (step S210)

[0143] The determination unit 140 calculates the value of the third index based on the waveform output from the second calculation unit 130. For example, the determination unit 140 determines, by the third calculation unit 135, whether the ratio of the power spectral density of the frequency of the heartbeat component and its integer multiple frequencies in the high-frequency component above the frequency of the respiratory component is above a fixed value. When the ratio is above the fixed value, the determination unit 140 determines that the waveform output from the second calculation unit 130 strongly carries the waveform of the heartbeat (the waveform containing a large amount of heartbeat components).

[0144] Then, when the waveform output from the second calculation unit 130 carries the waveform of the heartbeat, the determination unit 140 outputs "1", and when the waveform output from the second calculation unit 130 does not carry the waveform of the heartbeat (the waveform not containing a large amount of heartbeat components), it outputs "0". Moreover, the determination unit 140 outputs the value obtained by weighting the output value as the value of the third index. A method of weighting performed by the determination unit 140 will be described.

[0145] For example, it will be described with reference to Figure 13 the table. When the user's posture is "supine", the carrier frequency of the heartbeat waveform in the waveform output from the second calculation unit 130 becomes smaller (the waveform output from the second calculation unit 130 contains relatively few heartbeat components). In addition, when the user's posture is "prone", the carrier frequency of the heartbeat waveform in the waveform output from the second calculation unit 130 becomes smaller. In addition, when the user's posture is "side-lying", the carrier frequency of the heartbeat waveform in the waveform output from the second calculation unit 130 becomes larger (the waveform output from the second calculation unit 130 contains relatively many heartbeat components).

[0146] In addition, regarding the user's posture of "side-lying", especially when the user's posture is "left side-lying", in the waveform output from the second calculation unit 130, the carrier frequency of the heartbeat waveform becomes larger.

[0147] In addition, when the user's posture is "right side-lying", when compared with the user's postures of "supine" and "prone", the carrier frequency of the heartbeat waveform in the waveform output from the second calculation unit 130 becomes larger. However, when the user's posture is "right side-lying", when compared with the user's posture of "left side-lying", the carrier frequency of the heartbeat waveform in the waveform output from the second calculation unit 130 becomes smaller.

[0148] Therefore, when the user's posture is "side-lying", the determination unit 140 directly outputs the output value as the value of the third index. In addition, when the user's posture is "supine" or "prone", the determination unit 140 outputs a smaller value of the third index (for example, multiplied by "0.1", or "0", etc.).

[0149] (4) Fourth index calculation (step S212)

[0150] The determination unit 140 calculates a fourth index related to the shapes of peaks and valleys in the waveform of the heartbeat. For example, the portion of the waveform from a valley to a peak is the portion from time t1→t2 of Figure 14 Another example is that the portion from a peak to a valley is the portion from time t2→t3 of Figure 14 These portions of waveform changes represent transitions of vibrations (transitions of pressure), which usually correspond to the user's exhalation / inhalation.

[0151] The determination unit 140 evaluates how well the first time from a valley to a peak and the second time from a peak to a valley match between two sensors (vibration sensor 320a and vibration sensor 320b) for each cycle of breathing (from valley to valley or from peak to peak). For example, if the ratio of the first time in one cycle of vibration sensor 320a and the ratio of the first time in one cycle of vibration sensor 320b are within a desired range, the determination unit 140 evaluates that vibration sensors 320a and 320b match (also referred to as the same pair).

[0152] The determination unit 140 calculates the value of the fourth index based on the ratio of cycles that become the same pair in two sensors (vibration sensor 320a and vibration sensor 320b) in each cycle included in the posture determination interval. When the number of cycles that become the same pair in the cycles included in the posture determination interval exceeds a desired threshold, the determination unit 140 evaluates that they match between the two sensors (vibration sensor 320a and vibration sensor 320b).

[0153] Refer to Figure 13 , when the user's posture is "supine", the waveform changes between the two sensors are the same. Also, when the user's posture is "prone", the waveform changes between the two sensors are the same. Therefore, the determination unit 140 directly outputs the ratio of the time correspondence relationship between the two sensors that becomes the same pair within the posture determination interval as the value of the fourth index. Additionally, when the user's posture is "side lying", sometimes the waveform changes between the two sensors are the same and sometimes they are opposite. Thus, the determination unit 140 outputs the value obtained by multiplying the ratio of the same pairs included in the posture determination interval by "0.5" as the value of the fourth index.

[0154] (5) Fifth index calculation (step S214)

[0155] The determination unit 140 calculates the comparison result of the first time from a valley to a peak and the second time from a peak to a valley in the waveform as the fifth index.

[0156] At this time, the determination unit 140 compares the length of the first time from the valley to the peak and the length of the second time from the peak to the valley for each cycle of breathing (from valley to valley or from peak to peak) included in the posture determination section based on the waveform, and determines the portion where the length of the first time is "short" and the length of the second time is "long" ("short → long"). Then, the determination unit 140 calculates the value of the fifth index based on the ratio of the pair in which the length of time becomes the "short → long" relationship in both of the two sensors (vibration sensor 320a, vibration sensor 320b) within the posture determination section.

[0157] For example, the determination unit 140 counts the "breathing in which the first time is shorter than the second time" included in the posture determination section, and outputs the consistent ratio as the value of the fifth index.

[0158] Refer to Figure 13 , when the user's posture is "supine", a large number of portions having the "short → long" relationship can be seen in the waveform. Therefore, the determination unit 140 outputs the value obtained by subtracting "0.5" from the ratio of the pair of the two sensors within the posture determination section as the value of the fifth index. In addition, when the user's posture is "prone" and when the user's posture is "side-lying", it is difficult for the determination unit 140 to show characteristics in the waveform. Therefore, the determination unit 140 does not output the value of the fifth index. That is, the determination unit 140 outputs "0" as the value of the fifth index.

[0159] In this way, the determination unit 140 can determine the user's posture using each index value, and can more appropriately determine the user's posture.

[0160] It should be noted that in the present embodiment, the determination unit 140 also determines the posture using the waveform calculated by the second calculation unit 130 based on the vibration data. However, the determination unit 140 can also simply determine the user's posture based on the vibration data. That is, in Figure 12 processing, the determination unit 140 may not execute step S204, and may transition the processing from step S202 to step S206.

[0161] For example, the determination unit 140 can output the third index by performing frequency analysis on the vibration data. In addition, when outputting the fifth index, as long as the maximum value (nearby) and the minimum value (nearby) of the vibration data can be extracted, the peaks and valleys of the waveform can be determined in the same manner as the waveform.

[0162] [3. Third Embodiment]

[0163] Next, the third embodiment will be described. This embodiment describes the case where the determination unit 140 uses artificial intelligence (machine learning) to determine the user's posture.

[0164] This embodiment is based on Figure 15The estimation unit 700 estimates the user's posture, which is one of the user's states, instead of Figure 5 processing.

[0165] Here, the operation of the estimation unit 700 in the present embodiment will be described. The estimation unit 700 estimates the user's posture by using vibration data and the user's state as input values (input data) and applying artificial intelligence and various statistical metrics.

[0166] As Figure 15 shown, the estimation unit 700 includes an extraction unit 710, an identification unit 720, an identification dictionary 730 (learning model), and an output unit 740.

[0167] First, the estimation unit 700 is input with and utilizes various parameters. For example, in the present embodiment, vibration data and waveform data calculated based on the vibration data are input to the estimation unit 700.

[0168] Moreover, the extraction unit 710 extracts each feature point of the input data and outputs it as a feature vector. Here, as the content extracted as feature points by the extraction unit 710, for example, the following content can be considered.

[0169] (1) Whether there is a correlation within the sensor

[0170] (2) Whether there is a correlation between sensors

[0171] (3) Whether the waveform data carries a heartbeat waveform

[0172] (4) Whether the time from the valley to the peak of the respiration waveform is shorter or longer than the time from the peak to the valley

[0173] (5) Whether the appearance rate of the double-peak waveform is high or low

[0174] (6) Whether there is a difference in the area from the center line to the upper and lower in the waveform data

[0175] (7) Whether there is a difference in the heartbeat waveform between sensors / whether there is a difference in the carrier frequency

[0176] The extraction unit 710 outputs a feature vector by combining one or more of these feature points. It should be noted that the content described as feature points above is an example, and other content extracted as feature points can also be combined. In addition, for ease of explanation, the values output as feature points by the extraction unit 710 are convenient values. Moreover, the extraction unit 710 outputs the value of the corresponding feature point as "1" and the value of the non-corresponding feature point as "0". It should be noted that the extraction unit 710 can also output random variables as the values of feature points.

[0177] Moreover, in the case of including all the above-mentioned feature points, the feature space is seven-dimensional, and the extraction unit 710 outputs a seven-dimensional feature vector to the recognition unit 720.

[0178] The recognition unit 720 recognizes the class corresponding to the user's state based on the input feature vector. At this time, the recognition unit 720 recognizes the class by comparing with a plurality of prototypes prepared in advance as a recognition dictionary 730. The prototypes can be stored as feature vectors corresponding to each class, or the feature vectors representing the classes can be stored.

[0179] When the feature vectors representing the classes are stored in the recognition dictionary 730, the recognition unit 720 determines the class to which the closest prototype belongs. At this time, the recognition unit 720 can determine the class according to the nearest neighbor determination rule, or can recognize and determine the class according to the k-nearest neighbor method.

[0180] It should be noted that the recognition dictionary 730 used by the recognition unit 720 can store prototypes in advance, or can be re-stored or updated at any time using machine learning.

[0181] Moreover, corresponding to the class recognized by the recognition unit 720, the output unit 740 outputs the (sleeping) posture as one of the user's states. The states of the user output by the output unit 740 can recognize and output "supine", "prone", "side lying", etc., or can directly output random variables.

[0182] Thus, according to the present embodiment, the determination unit 140 can obtain the vibration data output from the sensor, and based on this information, use machine learning to infer the user's posture.

[0183] [4. Fourth Embodiment]

[0184] The fourth embodiment will be described. The fourth embodiment is an embodiment in which the inference unit 700 of the third embodiment uses deep learning using a neural network to infer the user's posture.

[0185] In this embodiment, the waveform data of the user is input to the inference unit 700. The inference unit 700 infers the user state (posture) based on the input waveform data. The inference unit 700 uses deep learning as the process for inference. It should be noted that the process using deep learning (deep neural network) can perform inference with high accuracy especially in image recognition. Figure 16 It is a diagram for explaining the process of the inference unit 700 performing inference using deep learning.

[0186] First, the estimation unit 700 inputs the signal of the waveform data (image data) output by the second calculation unit 130 into a neural network composed of multiple layers and neurons included in each layer. Each neuron receives signals from multiple other neurons and outputs the signal after performing an operation to multiple other neurons. When the neural network is composed of multiple layers, in the order of signal flow, it is called an input layer, an intermediate layer (hidden layer), and an output layer.

[0187] A network in which the intermediate layer of a neural network is composed of multiple layers is called a deep neural network (for example, a Convolutional Neural Network (CNN) with a convolutional operation), and the machine learning method using this network is called deep learning.

[0188] The waveform data performs various operations (convolutional operation, pooling operation, normalization operation, matrix operation, etc.) on the neurons of each layer of the neural network, flows while changing its shape, and outputs multiple signals from the output layer.

[0189] Each of the multiple output values from the neural network is associated with the user's posture. The estimation unit 700 estimates the user's posture as the user's posture associated with the output value with the largest value. In addition, the estimation unit 700 may pass one or more output values through a classifier and estimate the user's posture based on the output of the classifier, rather than directly outputting the posture as the user's state.

[0190] A large amount of waveform data and the corresponding user's posture in the waveform data are input to the neural network in advance as parameters, which are coefficients for various operations of the neural network. In addition, the error between the output value of the neural network and the correct value is transmitted in the opposite direction in the neural network by the error backpropagation method. Thus, the parameters of the neurons in each layer are updated and determined multiple times. The process of updating and determining such parameters is called learning.

[0191] Regarding the structure and each operation of the neural network, they are well-known technologies explained in books or papers, and any one of these technologies can be used.

[0192] In this way, by using the estimation unit 700, referring to the vibration wave data (waveform data) calculated based on the vibration data output from the sensor, the user's posture is output accordingly.

[0193] It should be noted that in the present embodiment, an example in which the estimation unit 700 uses a neural network to estimate the image data of the waveform is described. In addition to this, the estimation unit 700 may also input only the vibration data (time-series voltage output values) and estimate the user's posture through learning. Alternatively, it may be that the estimation unit 700 inputs the data of the signal transformed into the frequency domain in Fourier transform or discrete cosine transform and estimates the user's posture through learning.

[0194] [5. Application Example]

[0195] Regarding the above-mentioned posture determination device, by incorporating it into other devices, the following application examples can be considered.

[0196] [5.1 Bed]

[0197] In Figure 17 the configuration of the bed is shown. The bed 10 has a back bottom plate 12, a waist bottom plate 14, a knee joint bottom plate 16, and a foot bottom plate 18. The user P is supported on the upper body by the back bottom plate 12 and on the waist by the waist bottom plate 14.

[0198] The drive control unit 1000 controls the drive of the bed. Here, the drive control unit 1000 includes the function of the bottom plate control unit 1100 for controlling functions such as back elevation and knee joint elevation (foot lowering) by causing the bottom plate to move.

[0199] To achieve the back elevation function, the bottom plate control unit 1100 is connected to the back bottom plate drive unit 1110 and the knee joint bottom plate drive unit 1120. The back bottom plate drive unit 1110 is, for example, an actuator and is connected to the back elevation link via a link mechanism. Moreover, the back bottom plate drive unit 1110 controls the movement of the back bottom plate 12 placed on the link. The back bottom plate drive unit 1110 performs back elevation / back lowering control.

[0200] In addition, the knee joint bottom plate drive unit 1120 is, for example, an actuator. The knee joint bottom plate drive unit 1120 is connected to the knee joint elevation link via a link mechanism. Moreover, the knee joint bottom plate drive unit 1120 controls the movement of the foot bottom plate 18 further connected to the knee joint bottom plate 16 placed on the link. The knee joint bottom plate drive unit 1120 performs knee joint elevation / knee joint lowering (foot lowering / foot elevation) control.

[0201] Moreover, when the bed performs a back elevation operation, the bottom plate control unit 1100 does not perform a back elevation operation when the determination unit 140 (or the estimation unit 700) determines the user's posture as "lying face down". That is, even if the user selects a back elevation operation, the bottom plate control unit 1100 does not drive the back bottom plate drive unit 1110 and does not perform a back elevation operation. In addition, even when the bed is operating automatically, when the determination unit 140 (or the estimation unit 700) determines the user's posture as "lying face down", the bottom plate control unit 1100 does not perform a back elevation operation.

[0202] Refer to Figure 18 The operation in this case will be described. First, the control unit 100 determines whether an operation has been selected by the user (step S302). For example, the user selects the back elevation button through the input unit 160 (operating the remote control). Thereby, the control unit 100 determines that the back elevation operation has been selected.

[0203] Next, the control unit 100 (determination unit 140) performs a posture determination process (step S304). The posture determination process is for the determination unit 140 to perform any of the above posture determination processes to determine the posture of the user on the bed. Additionally, the posture determination process can also be for the estimation unit 700 to determine the user's posture.

[0204] Here, the control unit 100 determines whether the user's posture has become a specific posture (step S306). In this application example, if the user's posture is "prone", the control unit 100 does not perform the back elevation action (step S306; Yes). If it is any other posture, the control unit 100 performs the back elevation action by issuing an instruction to the bottom plate control unit 1100 (back bottom plate drive unit 1110) (step S306; No → step S308).

[0205] Moreover, when the user cancels the back elevation action (for example, by the user performing an abort operation or releasing the back elevation button, etc.), the control unit 100 stops the back elevation action (step S310; Yes → step S312).

[0206] Additionally, during the back elevation action, even when the user's posture is in a specific posture (for example, being prone during the back elevation action), the control unit 100 also stops the back elevation action (step S306; Yes → step S312).

[0207] [5.2 Body Position Conversion Device]

[0208] In a body position conversion device for converting the user's body position, a posture determination device is applied. For example, the control unit 100 automatically stores the frequency of body position conversion (posture conversion) and the ratio of each posture in order to grasp the risk of pressure ulcers. That is, the control unit 100 can apply the user's posture determined by the determination unit 140 to nursing and treatment by automatic storage.

[0209] Additionally, when the user's posture determined by the determination unit 140 has been in the same state for a specified time, the body position conversion device gives a notification or automatically performs a body position conversion. For example, as Figure 17 shown, the control unit 100 and the drive control unit 1000 control the body position conversion drive unit 1200. The body position conversion drive unit 1200 converts the user's body position, for example, by inflating and deflating airbags provided on the left and right of the user or by undulating the left and right bottom plates.

[0210] The drive control unit 1000 controls the body position conversion drive unit 1200 according to the posture determined by the determination unit 140 to perform control for changing the body position of the user P.

[0211] For example,Figure 18 This will be described by taking the processing of

[0212] as an example. When the user selects the action of body position conversion (step S302; Yes), the determination unit 140 determines the user's posture by any of the above methods (step S304).

[0213] Here, the body position conversion drive unit 1200 performs processing according to the posture determined by the determination unit 140. For example, if the user is in the right lateral lying position, the body position conversion drive unit 1200 controls the airbag set on the right side. The user performs body position conversion. In addition, if the user is in the left lateral lying position, the body position conversion drive unit 1200 controls the airbag set on the left side. The user performs body position conversion. In addition, if the user is lying prone, the body position conversion drive unit 1200 does not control either the left or right airbag. The user does not perform body position conversion.

[0214] In addition, when the determination unit 140 determines that the user's posture has continued in the same posture for a specified time, body position conversion can also be performed by the body position conversion drive unit 1200. For example, when the user's posture determined by the determination unit 140 has continued in the same posture for more than 10 minutes, the body position conversion drive unit 1200 performs body position conversion.

[0215] [5.3 Notification Device]

[0216] The notification device makes a notification according to the user's posture determined by the determination unit 140. As a method of notification by the notification device, it can also be a notification by voice output from a voice output device or a notification by display (light) of a display device. In addition, the notification device can also notify other terminal devices (for example, a mobile terminal device owned by medical staff).

[0217] The following timings can be considered as the timings for the notification device to make a notification. For example, when the user has numbness, if the numb side is downward, the risk of bedsore increases. Therefore, when the posture determined by the determination unit 140 is a lateral lying position with the numb side downward, the notification device makes a notification.

[0218] In addition, since infants and young children are at an increased risk of suffocation death when sleeping prone. Therefore, when the posture of the user (infant or young child) determined by the determination unit 140 is a prone position, in order to prevent suffocation death, the notification device makes a notification.

[0219] [6. Modification Example]

[0220] As described above, the embodiments of the present invention have been described in detail with reference to the accompanying drawings. However, the specific configuration is not limited to this embodiment, and designs and the like within the scope not departing from the gist of the present invention are also included in the scope of the claims.

[0221] In addition, in the present embodiment, based on the result output from the detection device 3, the posture of the user is determined in the processing device 5, but it may also be determined entirely by one device. In addition, not only is the application installed and implemented on the terminal device (e.g., smartphone, tablet computer, computer), but for example, processing may also be performed on the server side and the processing result may be returned to the terminal device.

[0222] For example, the above-described processing may also be implemented on the server side by uploading vibration data from the detection device 3. The detection device 3 may be implemented by a device such as a smartphone having an acceleration sensor and a vibration sensor built therein.

[0223] In addition, in the above embodiment, it has been described that two vibration sensors are provided, but two or more may also be provided. In addition, in the method of calculating the frequency distribution in the first embodiment to determine the posture, it can also be achieved even with one sensor.

[0224] In addition, in the embodiment, the program running in each device is a program for controlling a CPU or the like to implement the functions of the above-described embodiment (a program for causing a computer to function). Moreover, the information processed by these devices is temporarily stored in a temporary storage device (e.g., RAM) during the processing, and then is stored in a storage device such as various ROMs, HDDs, or SSDs, and is read out by the CPU as needed for correction / writing.

[0225] In addition, in the case of being distributed on the market, the program can be stored in a portable recording medium for distribution, or transmitted to a server computer connected via a network such as the Internet. Of course, in this case, the storage device of the server computer is also included in the present invention.

[0226] Reference Signs

[0227] 1 System

[0228] 3 Detection Device

[0229] 5 Processing Device

[0230] 10 Bed

[0231] 12 Back Bottom Plate

[0232] 14 Waist Bottom Plate

[0233] 16 Knee Joint Bottom Plate

[0234] 18 foot sole

[0235] 20 mattress

[0236] 100 control unit

[0237] 110 detection unit

[0238] 120 first calculation unit

[0239] 130 second calculation unit

[0240] 135 third calculation unit

[0241] 140 determination unit

[0242] 150 storage unit

[0243] 152 vibration data storage area

[0244] 154 waveform data storage area

[0245] 160 input unit

[0246] 170 output unit

[0247] 700 speculation unit

[0248] 710 feature extraction unit

[0249] 720 recognition unit

[0250] 730 recognition dictionary

[0251] 740 output unit

Claims

1. A bed system, characterized in that, Comprising: A bed, including a bed frame, a bed board provided on the bed frame, and a driving part for driving the bed board; At least two sensors capable of detecting biological information when a user lies on the bed; And A control unit that determines the posture of the user when the user lies on the bed based on the characteristics of the biological information. The control unit is configured to perform control so that when the posture of the user is a prone posture, the bed board is not driven. The posture is any one of a supine posture, a prone posture, a left lateral lying posture, and a right lateral lying posture. The sensors include a first sensor and a second sensor. The control unit determines the posture of the user when the user lies on the bed based on the correlation in the first waveform of the biological information obtained by the first sensor and the correlation between the first waveform and the second waveform. The second waveform is the waveform of the biological information detected by the second sensor. When there is a correlation between the first waveform and the second waveform, the control unit judges the correlation in the first waveform.

2. The bed system according to claim 1, characterized in that The bed board can support the user's back. The control unit performs control so that when the posture of the user is a prone posture, even if the control unit receives an instruction from the user, the bed board is not driven. The instruction is a back elevation command.

3. The bed system according to claim 2, characterized in that The control unit performs control to drive the bed board when the posture of the user is not a prone posture.

4. The bed system according to claim 1, characterized in that When the control unit determines that there is no correlation between the first waveform and the second waveform, the control unit determines that the posture of the user is a lateral lying posture; when the control unit determines that there is a correlation between the first waveform and the second waveform, the control unit determines that the posture of the user is a supine posture or a prone posture.

5. The bed system according to claim 4, characterized in that When the control unit determines that there is a correlation in the first waveform, the control unit determines that the posture of the user is a supine posture; when the control unit determines that there is no correlation in the first waveform, the control unit determines that the posture of the user is a prone posture.

6. The bed system according to claim 1, characterized in that When the control unit determines that there is a correlation in the first waveform, the control unit determines that the posture of the user is a supine posture; when the control unit determines that there is no correlation in the first waveform, the control unit determines that the posture of the user is a prone posture.

7. The bed system according to claim 1, characterized in that The control unit calculates the frequency distribution of the waveform of the biological information detected by the sensors and determines the posture of the user when the user lies on the bed based on the frequency distribution.

8. The bed system according to claim 1, characterized in that The control unit determines the posture of the user based on at least one of the following: (1) whether the first waveform and the second waveform contain a large amount of heartbeat components; (2) the degree of consistency between the first time and the second time between the two sensors, the waveform of the heartbeat component changes from valley to peak at the first time, and the waveform of the heartbeat component changes from peak to valley at the second time; (3) whether the ratio of the first time being shorter than the second time exceeds a threshold value.

Images