JP2010063594A - System for predicting getting-out of bed - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To inexpensively provide a less-malfunction and highly-reliable prediction of a bed user's getting out of a bed by acquiring the bed user's posture and weight with few resources. <P>SOLUTION: An apparatus for measuring a bed user's load at a plurality of points on a surface of a bed or the like, estimating the bed user's posture by the acquired load and the center of gravity and predicting the bed user's getting out of a bed, includes: a means for dividing the whole bed surface into 2-4 areas, wherein respective divided areas have panels retained by a plurality of load measuring means for measuring load signals, and computing respective loads of the respective divided areas, and gravity centers of the respective divided areas from the plurality of acquired loads; a means for determining a probability of the user's getting out of the bed based on the acquired loads of the areas and the gravity centers of the areas; and an external output means for a nurse call button or the like. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention predicts the bed user's getting out of bed for elderly people with reduced muscle strength who may fall, and notifies the caregiver etc. at the stage before getting out of bed. Target technical fields that provide the environment necessary to secure a living.

  In recent years, an accident in which elderly people with weak muscles fall or fall around a bed has become a social problem, and countermeasures are required.

At present, the most common countermeasure is to install a floor mat (a sensor that can sound an alarm such as a nurse call when the bed user steps on the floor) on the floor of the bed exit. In this case, when the alarm is issued, the feet are already on the floor, and the caregiver has no time to run.

In addition, there is a measure such as installing a bed prediction sensor (a sensor that can alarm when the bed user gets up and the load on the sensor disappears) under the bedclothes. There are many malfunctions such as reaction, and it is not generally used.

Thus, it cannot be said that the current countermeasures are functioning sufficiently, and there is a need for a bed prediction system with few malfunctions that issues a warning at the stage before bed removal.

It is important to know the posture (position and posture) of the bed user on the bed in order to build a reliable bed prediction and a system with few malfunctions. It is effective to accurately measure the weight of the bed user in order to judge various disturbances such as sitting close to each other and placing objects on the bed.

In order to grasp the posture, there are a method of obtaining data by arranging a large number of load sensors in an array on the bed surface, and a method of measuring the centroid from a four-point load and estimating the posture from the change of the centroid position. .

In the techniques disclosed in Patent Documents 1 and 2, a large number of sensors for measuring the load are arranged in an array on the bed surface, the posture of the bed user is obtained from the output of each sensor, It prevents bed slip.

In the technique disclosed in Patent Document 3, load sensors are arranged at the four corners of the bed surface, the center of gravity position of the bed user and the total load on the bed surface are acquired, and the change of the body position is detected from the change of the center of gravity position. Estimating and detecting various disturbances from changes in the total load makes predictions for getting out of bed with few malfunctions.

In the technique disclosed in Patent Document 4, a plurality of panels having a pair of parallel long sides extending in the widthwise direction of the bed are arranged at positions where the bed can be predicted, the load measurement, The center of gravity is measured only in the hand direction, and the bed is predicted from the load change and the change in the center of gravity.
JP 2001-258957 A JP2003-24392A JP2007-190269 JP 2006-247384 A

We considered that it is necessary for the bed user to analyze the posture until getting out of bed in order to realize the prediction of getting out of bed, and experimented using the same method as in Patent Documents 1 and 2.

Specifically, 94 load sensors are first arranged in a staggered pattern on the entire bed surface, divided into 5 measurements with a 20-channel A / D converter, and a system for loading the load data into the control PC is constructed. did.

An example of the posture obtained by this system is shown in FIG. This is analog data of 256 gradations, where a place where the load is large is dark and a place where the load is small is lightly expressed. Compared with the photograph 2 taken from the same time at the same time, it can be seen that the body position can be accurately grasped.

As a result of collecting the load data at the time of moving up to the bed using this system and analyzing the movement, the common points of interest in the movement up to the floor are the head, buttocks, and bed of the subject. It seems that it is important for the prediction of getting out of bed.

However, the sleeping posture varies depending on each subject, and even in the same subject, the sleeping posture varies depending on the time and occasion. For this reason, it has been found difficult to measure the above-mentioned point of interest at a pinpoint on the bed.

As described above, it is difficult to set a point of interest for prediction of getting out of bed by the same method as in Patent Documents 1 and 2, and the number of sensors is large before that, resulting in a very high apparatus cost.

In this method of measuring the load at a large number of “points”, the load on the bed is also applied to the area between the “points” where no sensors are installed, which is necessary for detecting disturbances. There is a deficiency that the exact weight of the bed user cannot be measured.

Next, the use of the center of gravity was tested using the same method as in Patent Document 3.

Specifically, a strain-generating material with a strain gauge attached to four legs that support the bed was installed, and the position of the center of gravity was measured from the four loads measured.

As a result of analyzing the acquired data, it was found that the amount of change in the center of gravity position in the rising action, which is noted in Patent Document 3, is slightly larger than the amount of change caused by other movements of the bed user.

However, even in the same rising operation, if the unit time for measuring the change amount of the center of gravity is lengthened, the change amount of the center of gravity is increased, and if the unit time is shortened, the change amount of the center of gravity is reduced and does not become a constant value. Even if the rising speed is slow and fast, the amount of change in the center of gravity position is different. Therefore, it seems difficult to detect the rising up accurately.

It was also found that when the user is sleeping near the periphery of the bed, the position of the center of gravity is near the periphery of the bed, and it is easy for malfunctions to occur when getting out of bed.

From this fact, it is considered that the prediction of getting out of the bed centering on the information on the position of the center of gravity of Patent Document 3 has many malfunctions.

In Patent Document 4, a plurality of panels that measure the load and the center of gravity in the short direction of the bed are installed at positions where the bed can be predicted, and the optimal panel installation position for the subject must be determined. Therefore, it is difficult to determine the installation position of the panel in advance in order to realize accurate bed leaving prediction and few malfunctions. If this panel is installed on the entire bed surface, it is possible to obtain the posture directly, but the cost becomes high as in Patent Documents 1 and 2.

  In order to solve the problems of Patent Documents 1, 2, 3, and 4, the present invention obtains the bed user's body posture and weight with less resources, and provides a low-cost prediction with low malfunction and high reliability. With the goal.

Therefore, we measured the bed user's load at multiple points on the bed surface such as a bed, estimated the bed user's position from the obtained load and center of gravity, and predicted the bed separation. Each of the divided areas is divided into four areas, and each divided area has a panel held by a plurality of load measuring means for measuring a load signal. We have devised a system that has means to calculate the area center of gravity, means to determine the possibility of getting out of bed from the obtained area load and area center of gravity, and external output means to nurse call etc.

Getting out of bed is when the bed user leaves the bed, and when the bed user puts his feet on the floor and moves his weight from the bed to his feet.

Getting out of bed is to judge getting out of bed at the stage before getting out of bed.

The bed surface is the upper surface of the bed on which the bed user lies.

The load signal is a load obtained by the load measuring means. The area load is the total load applied to each divided area. The area center of gravity is the position of the center of gravity of the total load applied to each divided area.

It is desirable to have four load measuring means in each divided area. The fact that there are four load measuring means means that four independent load signals can be obtained. This is to grasp the area center of gravity two-dimensionally (X-axis direction and Y-axis direction) and to stably fix the bed surface panel.

If there is a divided area where the information on the center of gravity of the area is one-dimensional (only in the X-axis direction or the Y-axis direction), the number of load measuring means may be two if the bed surface panel can be stably fixed.

In order to express the possibility of getting out of the floor numerically, the relationship between the numerical value expressing the possibility of getting out (hereinafter referred to as the level of getting out), the area load, and the area center of gravity is defined in advance as shown in FIG. It is characterized by that.

This definition is determined by analyzing the bed leaving motion by a plurality of subjects in advance. By this definition, various stages leading up to bed can be easily specified numerically. In the following, the “numerical value expressing the possibility of getting out of bed” will be referred to as the bed leaving level.

There is a means to set the bed removal level at which the bed user decides to get out as a threshold for external output such as nurse call, and the threshold level set by the system at regular intervals is set. It is characterized by external output when the value is exceeded. This makes it possible to easily adjust the judgment standard for getting out of bed according to the bed user.

For example, assume that the bed user is lying down at level 0 and the level just before the bed level is level 10. The state in the meantime is represented by a numerical value of 1 to 9, and the getting-off operation starts, and the getting-off level judged by the present system increases as the getting-off is approached. Here, it is assumed that the threshold value of the external output is set to 5. When the bed level determined by the system is 5 or more, external output is performed. Also, by increasing or decreasing the threshold by one, it is possible to predict bed leaving at the optimal timing for the user.

By having a function to display the obtained information, the user status and system status can be grasped in real time.

The obtained information is a load signal, a region load, a region center of gravity, a numerical value of a bed leaving level, a presence / absence state, and a parameter representing an internal state of the system.

Furthermore, it has a function of accumulating the obtained information and has means for outputting it to the outside, so that it becomes possible to set a more appropriate external output threshold value with reference to past data. In addition, by analyzing the collected long-term data, it can be used for life management of bed users.

Since this system measures the load not on “points” but on “surfaces”, all the loads on the bed are measured. For this reason, the sum of the area loads in each divided area becomes the sum of the loads on the bed, and the accurate weight of the bed user necessary for detecting the disturbance can be accurately measured, thereby preventing malfunction.

It is also possible to estimate the body posture from the positional relationship between the center of gravity of each region and the ratio of the magnitude of each region load.

In recent years, welfare beds are provided with functions of raising the back and raising the legs as assistance functions for leaving the bed, and the bed surface is often physically divided into 2 to 4 areas. Raising the back is a function to raise the body, and raising the foot is a function to prevent the buttocks from shifting in the foot direction when the back is raised.

Here, taking a bed whose bed surface is divided into four parts as an example, as shown in FIG. 3, regions A, B, C, and D are set in the order of a back raising part, a fixing part, a leg raising part 1, and a leg raising part 2, Let's make the head, buttocks, and bed exits, which are the places of interest for the movement to get out of bed, correspond.

First, when the bed user is sleeping, the head is above the region A (head direction). The focus on the head is to detect rising. By getting up, the head and of course the upper part of the buttocks are not touched. From this, it is possible to monitor the rising up by monitoring the entire area A.

Depending on the position after waking up, a buttock may be on the lower (foot direction) end of region A. Since a certain load is applied, it is not possible to detect the rising with only the area load, and there is a possibility of causing a malfunction. However, since the area center of gravity is extremely below the area A, this can be determined. By combining the area center of gravity and the area load in this way, it is possible to grasp the posture.

Next, regarding the buttocks, when the normal sleeping posture is taken, the buttocks are present in a region B immediately below the back-raising region. When the sleeping posture is slightly below the feet, the region C may be applied. Also, the bed exit spanned both areas B and C.

For this reason, both the regions B and C do not need to be divided because they cover both the buttocks and the bed exit. However, regions B and C need to be divided so as not to hinder the assistance function of the bed.

In addition, regardless of whether the bed exit is on the right side or the left side, the sign of the center of gravity in the X-axis direction is only reversed, so there is an advantage that there is no need to be particularly aware of the left and right sides.

In the region D, no special data is obtained in the normal bed leaving operation. This is because the elderly person who is the subject is not very tall, and when they get up, their feet are also lowered under the bed at the same time, that is, they get up at an angle, and their raised position is already in the end-sitting position (the state of sitting on the edge of the bed) This is because it is often the case. If the sleeping posture is deviated in the foot direction, or if it is necessary to measure the area load and the area center of gravity in order to cope with accurate weight measurement, it is necessary to install a panel.

Even if the number of physical divisions of the bed surface is two or three, the number of divisions of the bed surface may be the same as the number of physical divisions of the bed.

In the case of Patent Document 3, as the information for predicting bed leaving, there is only one center of gravity position of the entire bed surface, so there is a concern about malfunction. However, this method of dividing and measuring the entire bed bed surface into 2 to 4 regions can obtain a larger amount of information than that of Patent Document 3, and can construct a system with few malfunctions.

As is clear from the above explanation, the present method for dividing the bed surface into a plurality of regions and obtaining the region load and the region centroid in each divided region is the size of each region load, the positional relationship between each region centroid, It is possible to obtain a lot of information necessary for bed prediction, such as accurate measurement of bed user's weight, at a low cost. Can be realized at low cost.

Being able to predict getting out of bed with few malfunctions makes it possible to prevent the bed user from leaving the bed alone and contribute to prevention of falling and wrinkles.

A case where four panels held by the load measuring means are attached to a bed whose bed bed surface is physically divided into four will be described with reference to FIG. 4, A, B, C, and D correspond to A, B, C, and D in FIG. 3, A is a back raising portion, B is a fixed portion, C is a foot raising portion 1, and D is a foot raising. Part 2.

The panel 10 is about the same size as the back raising part of the bed body and receives all the load applied to the area of the back raising part. The panel 10 has a rigidity that does not cause distortion due to the load.
The panel 10 is held by load measuring means 11 to 14 at four corners and four points in a mechanism for raising the back of the bed main body.

The load measuring means are installed so that the four load measuring means form a rectangle as large as possible.

The load measuring means uses a strain generating material 11c as shown in FIG. 6 with a strain gauge attached thereto, and the back raising portion of the bed body and the panel 10 are fixed with bolts 11d.

All strain-generating materials used should be the same material and shape. Use the same strain gauges as the strain gauges used, and make all the attachment points to the strain-generating material the same.

The resistance value of the strain gauge to be used is not particularly limited, but using a resistor with a large resistance value can apply a large voltage and improve sensitivity.

The strain gauges 11a and 11b are preferably used in pairs, one on each side of one strain-generating material for temperature compensation.

In order to acquire two pieces of information about the region load and the region center of gravity in the divided regions, the load measuring means 11 to 14 individually assemble the balanced bridges 1 and convert resistance changes into voltage changes. The obtained voltage is amplified by the amplifier 2 and input to the A / D converter 4-1.

When the number of channels of the A / D converter 4-1 is insufficient, A / D conversion is performed in a plurality of times via the photo relay 3 or the like. The load signal obtained by the A / D converter 4-1 calculates the area load and the area center of gravity by the calculator 4-2, determines the bed leaving level by referring to a definition table described later, and the I / O port 4-4. 3 is controlled. An external output, control box 5, storage device 6, and display device 7 are connected to the I / O port 4-3 as required.

As shown in FIG. 5, the balanced bridge 1 is composed of one fixed resistor R1, one variable resistor R2, and two strain gauges 11a and 11b.

The variable resistor R2 is used to adjust the amplified voltage value so as to be within the measurement range of the A / D converter 4-1.

Here, a load signal when the bed user is absent is acquired as an initial value in advance. The value obtained by subtracting the initial value from the load signal acquired thereafter is a value caused only by the object that is added on the bed after the initial value is acquired. For convenience, the value obtained by subtracting the initial value is also called a load signal.

The region load is obtained by processing the obtained load signal in accordance with Equation 1, and the region centroid is obtained by processing in accordance with Equation 2. The numbers in the formulas 1 and 2 are load signals obtained by the load measuring means of the numbers. Two area centroids are obtained: the X-axis direction (bed short direction) and the Y-axis direction (bed long direction).

The area centroid is present at the center portion of the panel 10 when the value is 0, and exists at the portion corresponding to the side of the panel 10 when the absolute value of the area centroid is 1.

More specifically, in FIG. 7, when the area centroid X = 0 and the area centroid Y = 0, the position indicated by the area centroid is point O. When the area centroid X = 0 and the area centroid Y = −1, the area centroid The position indicated by is the point P. When the area centroid X = 1 and the area centroid Y = −1, the position indicated by the area centroid is the point Q.

The panels 20, 30 and 40 are also installed in the same manner, and the area load and the area center of gravity are obtained.

Since each panel is installed on the bed surface of the bed body, it is installed so that adjacent panels do not come into contact with each other by the movement of the back raising and foot raising functions.

The system obtains load signals at regular intervals, calculates the area load and area center of gravity, and determines the bed leaving level in light of the definition shown in FIG.
The definition of FIG. 8 is defined based on the result of analyzing the bed leaving motion of a plurality of subjects in advance.

FIG. 8 will be described. Classification No. Is a sequential number assigned for the sake of convenience.

A, B, C, and D in the “classification by area load” column are symbols corresponding to areas A, B, C, and D in FIG. 4. Value. “Α” in the “classification by area load” column is the sum of area loads A, B, C, and D acquired first when the bed user is lying on the bed, that is, the weight of the bed user. The column “A classification by region load” is a column indicating the ratio of the sum of the calculated region loads A, B, C, and D to α acquired first.

Similarly, column A is a column that indicates how much the calculated area load A value is relative to the initially acquired α, and column C is the sum of the calculated area loads B and C. This is a column indicating the ratio of the first acquired α, and column D is a column indicating the ratio of the calculated area load D to the first acquired α.

In the column A, if the value is 90% or more and less than 110%, it means that the bed user is present on the bed. 1 to 21 events are defined. The width of 90% or more and less than 110% of 20% is provided to absorb the fluctuation of the load signal due to the movement of the bed user. Classification No. No. 22 defines the state immediately before leaving the bed when the total load on the bed decreases to less than 90%. Classification No. 23 defines a state in which there is a disturbance such as a person other than the bed user sitting on the bed when the total load on the bed increases to 110% or more.

A, B, C, and D in the “classification by area centroid” column are also symbols corresponding to areas A, B, C, and D in FIG. 4. Value. Columns “o” and “ki” of “classification based on area centroid” are fields indicating the values of the calculated area centroids A and D, and the column “C” indicates the larger of the calculated area centroid B and area centroid C, whichever has the larger absolute value. It is a column which shows the value of. X is the X-axis direction, and Y is the Y-axis direction, indicating the direction on the bed. The minus part in the table means that the value can be anything.

The “leaving level” column is a value defined as the leaving level when all conditions of the area load and area center of gravity of an arbitrary row are satisfied.

For example, classification No. Describing 1, when the total area load A + B + C + D is 90% or more and less than 110% of α and the area load A is 30% or more of α, the bed leaving level is defined as 0.

For example, classification No. 19, the total area load A + B + C + D is 90% to less than 110% of α, the area load A is less than 10% of α, the area load D is less than 10% of α, and the area When the center of gravity B and C and the larger absolute value in the X-axis direction is 0.5 or more and less than 0.7, the bed leaving level is defined as 7.

The bed leaving level is defined so that the value becomes larger as the bed gets closer to the bed. The numerical values in the definition table shown in FIG. 8 show an example, and the contents are not limited.

For example, a control box 5 is provided with a switch for setting a threshold value for external output to a nurse call or the like, and a value corresponding to the bed user is set.

In addition to the buttons and switches described above, the control box 5 may be provided with other switches, buttons, and LEDs as necessary. For example, an LED indicating that the system is in operation, an LED indicating that the initial value of the load signal has been acquired, an LED indicating that the weight has been acquired, and an LED indicating that an external output has been performed Etc.

As shown in FIG. 4, a device 6 for storing the load signal obtained by this system, the area load, the area center of gravity, the numerical value of the bed leaving level, the presence / absence state, the parameters indicating the internal state of the system, etc. is required. Provide accordingly. A display function 7 is also provided as necessary.

The actual operation will be described next.

Place a bed with a panel on the floor, and place a bed, mattress, pillow, comforter, etc. to obtain the initial value of the load signal. Thereby, the state when the bed user is absent can be set. If the absence can be automatically determined, it may be acquired automatically.

Thereafter, the weight of the bed user is acquired while the bed user enters the bed and takes a stable sleeping posture. If a stable sleeping posture can be automatically determined, it may be acquired automatically. After getting the weight, it is ready for external output.

Thereafter, the system acquires load signals at regular intervals, calculates the area load and area center of gravity of the four areas, and determines the bed leaving level by referring to the definition of the bed leaving level.

When the bed leaving level determined by this system exceeds the external output threshold for a certain period of time, external output is performed. Once external output is performed, external output cannot be performed until the bed user is in a stable sleeping position again. This is because if the state in which external output can be continued continues, even if the nurse call is stopped once, the nurse call sounds again and cannot be stopped indefinitely.

  Even when the bed surface is divided into two or three, it can be realized by the same method as in the case of dividing into four, except that the bed leaving level is newly defined.

  Although load measurement using a strain gauge has been described, a load cell or a pressure sensor may be used, and the processing method of the obtained data can be realized in the same way as when using a strain gauge.

A case where four panels held by the load measuring means are attached to a bed whose bed bed surface is physically divided into four will be described with reference to FIG.

The panel 10 was made of a steel φ25.4 mm pipe with an outer peripheral frame and reinforced with an L angle. Inside the frame, a steel mesh was attached to hold the mattress.
The size of the panel 10 is the same as the component of the back-up part of the bed body in the bed short direction, and the bed length direction is 2 cm shorter.

The panels 20, 30, and 40 were also made the same size as the constituent members of the bed body in each region in the bed short direction. In the longitudinal direction of the bed, the panel 20 was 4 cm shorter, the panel 30 was 2 cm shorter, and the panel 40 was the same.

The reason for shortening each panel is to prevent adjacent panels from coming into contact with each other due to the movement of the back raising portion and the foot raising portion.

Sixteen load measuring means are prepared, each having a structure in which 120 Ω strain gauges 11a and 11b are attached to the central portion of the strain-generating material 11c having the shape shown in FIG. The panels 10, 20, 30, and 40 were fixed to the constituent members of the fixing portion, the foot raising portion 1, and the foot raising portion 2 by bolts 11d at four corners and four points.

The strain gauges were connected as shown in FIG. The voltage applied to the balanced bridge 1 was 1.5V. The fixed resistor R1 is 120Ω, and the variable resistor R2 is 120 ± 10Ω by combining a multi-rotation trimmer 20Ω and a 110Ω fixed resistor.

The amplification factor of the amplifier 2 was 2000 times.

A SH7125 microcomputer 4 manufactured by Renesas Corp. was used for the A / D converter 4-1, the arithmetic unit 4-2, and the I / O port 4-3.

SH7125 has 8 channels for A / D conversion. Therefore, using the photo relay 3, 16 load signals were A / D-converted by dividing each of the eight load signals into eight.

The measurement cycle was 10 Hz, that is, the speed at which the entire bed surface was measured 10 times per second.

The control box 5 includes a button for collecting an initial value of the load signal, a button for collecting the weight of the bed user, a switch for setting a bed leaving level as a threshold for external output, and an initial value of the area load. An LED indicating that the value is being collected, an LED indicating that the bed user's weight is being collected, an LED indicating that the external output function is disabled, and an indication that the system is operating LED.

A personal computer is used for the storage device 6 and the display device 7 and is connected to the microcomputer 4 via an RS-232C interface.

The situation of bed leaving prediction will be described using the system of the first embodiment.

The threshold for external output was set to 7. When the bed user was absent, the initial value of the load signal was acquired and zero adjustment was performed.

Body weight was obtained when the bed user was in bed.

The bed leaving level was defined as shown in FIG.

FIG. 9 shows a part of data acquired when the bed user performs the bed leaving operation. FIG. 9 describes three pieces of information, that is, the area center of gravity, the area load, and the bed leaving level from the top. The horizontal axis shows the passage of time.

The graph of the area centroid shows a total of 8 calculation results, 2 each for XY in 4 areas, but only Ay and Bx are shown. Ay is a value in the Y-axis direction (bed longitudinal direction) of the region A, and Bx is a value in the X-axis direction (bed short direction) of the region B. The area load graph shows the ratio of each area load to the total of the 4 areas, and A, B, C, and D correspond to the areas A, B, C, and D shown in FIG. The bed leaving level is a value determined by the system according to the definition of FIG. 8 from the values of the area load and the area centroid.

Looking at the graph in chronological order, the region load D first decreased to around 0%. Next, the area load A decreased to near 0%, and the area loads B and C increased accordingly. This indicates that the bed user has woken up.

At the same timing when the area load A decreased to near 0%, the area gravity center Ay changed from near 0 to near -1. This indicates that the position of the center of gravity of the region A has changed in the foot direction. The graph is cut off halfway, because the area centroid cannot be calculated because the load on the area A is lost.

Thereafter, the area centroid Bx gradually changed from near 0 to near 1. This indicates that the center of gravity has moved in the direction of the bed exit.

The bed level determined by the system gradually changed to a larger number due to the bed user's bed leaving operation, and when the set threshold value 7 was reached, external output was performed.

It was predicted to get out of bed 20 seconds faster than the actual bed leaving time.

The data used in this example was created based on information stored in the storage device 6.

  Similar results were obtained when the same system as in Example 1 was constructed using a bed whose bed surface was divided into three.

  Similar results were obtained when the same system as in Example 1 was constructed using a bed whose bed surface was divided into two.

Two-dimensional display image of sleeping posture obtained by installing a number of load sensors in a staggered pattern on the bed Photo taken at the same time as Fig. 1 Figure of bed where bed surface is divided into 4 areas Configuration diagram of bed leaving prediction system Diagram of balanced bridge that converts resistance change of strain gauge into voltage change Image of load measuring means using strain gauge and strain generating material Position explanatory diagram of the area center of gravity Bed level definition table Graph of data accumulated during bed removal

Explanation of symbols

10, 20, 30, 40 Panel 11-14 Load measuring means 11a, 11b Strain gauge 11c Strain generating material 11d Bolt 1 Balance bridge 2 Amplifier 3 Photo relay 4 Microcomputer 4-1 A / D converter in microcomputer 4-2 Microcomputer 4-3 I / O port in microcomputer 5 Control box 6 Storage device 7 Display device

Claims (5)

  In a device that measures the bed user's load at multiple points on the bed surface, such as a bed, and estimates the bed user's position from the resulting load and center of gravity to predict bed leaving, the entire bed surface is divided into 2 to 4 areas. Each of the divided areas has a panel held by a plurality of load measuring means for measuring the load signal, and the area load of each divided area and the center of gravity of each divided area are obtained from the obtained plurality of load signals. A bed leaving prediction system comprising: means for calculating the above, means for judging the possibility of getting out of bed from the obtained area load and area center of gravity, and means for external output to a nurse call or the like.   The bed leaving prediction system according to claim 1, wherein each of the divided areas has four load measuring means.   In order to express the possibility of getting out of the floor numerically, the relationship between the numerical value expressing the possibility of getting out (hereinafter referred to as the level of getting out), the area load, and the area center of gravity is defined in advance. The bed leaving prediction system according to claim 1 or 2.   There is a means to set the bed removal level at which the bed user decides to get out as a threshold for external output such as nurse call, and the threshold level set by the system at regular intervals is set. The bed leaving prediction system according to claim 1, wherein external output is performed when the value is exceeded.   5. The bed leaving prediction system according to claim 1, further comprising means for displaying the obtained information and a function for accumulating, and means for outputting the accumulated information to the outside.

Images