JP2015181499A - State detection method, device, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method, a device, and a program for controlling an alarm for actions of a person to be monitored capable of executing control so as to output an alarm according to a state of the person to be monitored, and improving the efficiency of work of a staff member while securing safety of facility users.SOLUTION: A wakening state detection part 101 detects a change from a sleeping state to a wakening state of a person 105 to be monitored on the basis of the output of a sensor 103 installed in a living room that the person 105 to be monitored is in. An alarm control part 102 controls the output of an alarm to a management terminal 104 depending on whether or not a predetermined action of the person 105 to be monitored, such as a rising action and a bed leaving action, is detected by the sensor 103 within a predetermined time from the detection of a change from the sleeping state to the wakening state. After the elapse of the predetermined time, the alarm control part 102 does not perform the detection of the predetermined action or the output of an alarm to the management terminal 104 based on the detection.

Description

  The present invention relates to a method, an apparatus, and a program for detecting a state with respect to a behavior of a monitoring target person.

  In medical facilities and nursing homes, when facilities users with reduced ADL (Activities of Daily Living) are mainly due to partial paralysis of physical functions or decreased physical fitness such as elderly or postoperative It is necessary to take measures to prevent falling.

  For this purpose, conventionally, a sensor that detects the rising or leaving of a facility user in the bed and notifies a staff member away from the bed such as an office through a call device represented by a nurse call is operated. By notifying, the staff visits the facility user and can perform necessary assistance for the facility user, such as walking assistance.

  The staff are busy because they are also taking care of other facility users and other duties, but if they were notified of any kind of bed leaving, they interrupted these work and left the floor. It is necessary to rush to the facility user. Furthermore, in the case of a facility user who frequently gets out of bed, the frequency of business interruptions is frequent, and the impact on other business becomes serious.

  The following technologies are known as conventional technologies for predicting the operation of the monitored person at an early stage when the operation of the monitored person is started, and providing a highly secure monitoring support apparatus that is free from erroneous detection and detection delay. (For example, Patent Document 1). This conventional monitoring support apparatus includes a sensor that detects a plurality of states of a person to be monitored. State identification means for identifying at least one of continuation in one state of the monitored person and state transition from the other state of the monitored person to one state based on a detection signal from the sensor is provided. A selection means for selecting a report content and a report destination based on the continuation or state transition of one state of the monitored person is provided. Transmitting means for transmitting the contents of the report selected by the selecting means to the selected report destination is provided. In this prior art, monitoring of the monitored person is started by the monitoring device when the monitored person transitions from the sleep state to the awake state. In this sleep state where the risk level is low, do not monitor with a monitoring device, prevent unnecessary tension on the monitored person, and reduce the burden on the nurse by starting monitoring from an awake state that requires monitoring While ensuring that accidents can be dealt with.

JP 2011-28583 A

  However, in the above-described conventional technology, the monitoring operation is always performed after the monitored person transitions to the awake state. Therefore, when there is a notification of getting out of bed, the staff must rush to the facility user who left the floor by interrupting other work, and the burden on the staff is not reduced.

  In view of the above, an object of one aspect of the present invention is to perform monitoring in consideration of the state of a monitoring target person.

  In one example, the output of the alarm is controlled depending on whether or not a predetermined behavior of the monitoring subject is detected by the sensor within a predetermined time from the detection of the change from the sleep state to the awake state for the monitoring subject. Cause the computer to execute the process.

  It is possible to perform monitoring in consideration of the status of the person being monitored.

It is a block diagram of the state detection apparatus concerning this embodiment. It is a system configuration figure of a state detection device concerning this embodiment. It is a figure which shows the data structural example of facility user DB. It is a whole flowchart which shows the process example of a monitoring program. It is a figure which shows an example of the state transition of an attitude | position state. It is a figure which shows an example of the state transition between the awakening state and the sleep state in the state of awakening. It is a flowchart which shows the detailed example of the determination process of the state of awakening based on the amount of body movements. It is a flowchart which shows the detailed example of the process which measures the motion amount of the present flame | frame. It is explanatory drawing of the process which measures the motion amount of the present flame | frame. It is a flowchart which shows the example of the alarm determination process which concerns on 1st Embodiment. It is a flowchart which shows the detailed example of the determination process of the alarm mode which concerns on 1st Embodiment. It is a figure which shows the notification screen example of an alarm. It is a flowchart which shows the example of the process which performs the determination of the bed leaving tendency after the awakening of the facility user which concerns on 2nd Embodiment. It is a figure which shows the data structural example of the alarm timing table which concerns on 2nd Embodiment. It is a flowchart which shows the example of the alarm determination process which concerns on 2nd Embodiment. It is a warning characteristic graph which shows the relationship between the elapsed time after awakening and a warning rank concerning a 3rd embodiment. It is a figure which shows the structural example of the alarm rank table which concerns on 3rd Embodiment. It is a flowchart which shows the example of the alarm determination process which concerns on 3rd Embodiment. It is a figure which shows the structural example of the hardware of the server computer which is a state detection apparatus which can execute the monitoring program which implement | achieves the function of a state detection apparatus.

Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings.
FIG. 1 is a block diagram of a state detection device according to the present embodiment.

The state detection device 100 includes a wakefulness detection unit 101 and an alarm control unit 102.
The wakefulness detection unit 101 changes from a sleep state to a wakefulness state for the monitoring target person 105 based on a captured image captured by a sensor 103 (for example, an imaging device) installed in the room where the monitoring target person 105 is located. Is detected.

  The alarm control unit 102 determines whether a predetermined action of the monitoring target person 105, for example, a wake-up action or a get-off action is detected by the sensor 103 within a predetermined time from the detection of the change from the sleep state to the awake state. Controlling the output of alarms to the management terminal 104. After a predetermined time has elapsed, the alarm control unit 102 does not detect a predetermined action and does not output an alarm to the management terminal 104 based on the detection.

  The walk after getting out of bed in the monitoring target 105 with a lowered ADL is a risk of falling when the consciousness is clear after the facility user awakens and when the consciousness is cloudy immediately after awakening. The degree of is very different. More specifically, when the latter consciousness is cloudy, the user may not be able to walk straight, but may be swayed, and the degree of danger of falling dramatically increases. Although it depends on the degree of ADL, there are cases where the assistance of the staff is required for walking when the consciousness is clouded even when the assistance of the staff is not necessary when walking in a state where the consciousness is clear. In other words, in the former case, it is not necessary to notify the staff at the time of getting out of bed, but in the latter case, it is necessary to notify the staff at the time of getting out of the bed.

  With the configuration in FIG. 1, monitoring in consideration of the consciousness state of the monitoring target person 105 becomes possible. More specifically, an alarm is output only when the monitoring subject 105 is apt to become clouded immediately after the sleep state is changed from the sleep state, and after the predetermined time has elapsed, the consciousness will become clear. It is possible to control not to perform alarm output. This makes it possible to ensure the safety of facility users and to increase the efficiency of staff work.

FIG. 2 is a system configuration diagram of the state detection apparatus according to the present embodiment.
The state detection device 100 realized by a server computer executes the monitoring program 201 stored in a memory by a CPU (central processing processor), thereby causing the operations of the functions of the units 101 and 102 of the state detection device 100 in FIG. Execute. A monitoring program 201 and a facility user DB (database) 205 are stored in the memory in the state detection device 100. The monitoring program 201 includes processing units of a facility user detection unit 202, an alarm determination unit 203, and an alarm unit 204.

  The facility user detection unit 202 detects the state of the facility user 210 and the position of the body by image recognition based on captured image data from the imaging device 206 (corresponding to the sensor 103 in FIG. 1).

The alarm determination unit 203 determines the necessity of an alarm using the result of the facility user detection unit 202.
The alarm unit 204 transmits an alarm to the management terminal 104 based on the determination result of the alarm determination unit 203.

  For example, an imaging device 206 equipped with, for example, near-infrared illumination is installed above the bed of a hospital room (living room) of a facility user 210 (corresponding to the monitoring target person 105 in FIG. 1), and the state of the facility user 210 is determined for a certain period of time. Images are taken every time. The captured image data obtained as a result is transmitted to the state detection apparatus 100 via the hospital LAN 220. The imaging device 206 may be installed for each bed in a hospital room (living room). For example, an N: 1 configuration in which a plurality of imaging devices 206 are connected to one state detection device 100 may be used. Further, the imaging device 206 may be connected to the state detection device 100 without going through the hospital LAN 220.

  The monitoring program 201 in the state detection apparatus 100 can output an alarm to the management terminal 104 from the alarm unit 204 via the hospital LAN 220.

  FIG. 3 is a diagram illustrating a data configuration example of the facility user DB 205 of FIG. For each facility user ID of the facility user 210, information on a hospital room (room), a bed number, and a monitoring target flag is registered. The monitoring target flag stores, as a logical value of ON or OFF, whether or not behavior monitoring immediately after the awakening state is performed for each facility user ID.

  FIG. 4 is an overall flowchart showing a processing example of the monitoring program (state detection program) 201 executed by the state detection apparatus 100 of FIG.

  First, captured image data obtained by capturing the facility user 210 (the monitoring target person 105) in the bed with the camera of the imaging device 206 installed above the bed is input (step S401).

  Next, based on the input captured image data, the body and bed of the facility user 210 (FIG. 2) are detected by image processing (step S402). For example, a feature point such as an edge indicating a discontinuous change in luminance on the captured image data is extracted, and based on the distribution of the extracted feature point, for example, in addition to the head and trunk of the facility user 210, The edge of the bed is detected from the captured image data. Further, the position of the facility user 210 and the position of the bed are also detected.

  Subsequently, the current posture state of the facility user 210 is determined from the past posture state of the facility user 210 and the body detection result in step S402 (step S403).

  Furthermore, the amount of body motion of the facility user 210 is calculated from the captured image data, and the state of awakening of the facility user 210 is determined from the amount of motion (step S404).

  The above steps S401 to S404 correspond to the facility user detection unit 202 of FIG.

  Finally, an alarm determination process is executed (step S405). This process corresponds to the alarm determination unit 203 in FIG. After the process of step S405, the process returns to step S401.

  FIG. 5 is a diagram illustrating an example of state transition of the state of the facility user 210 when the state of the current posture of the facility user 210 is determined in step S403 of FIG. In the state transition example of FIG. 5, the posture state of the facility user 210 is divided into five. The five posture states are, for example, a bedded state, a state where the upper body is raised, a state of sitting in an end position, a state around the bed, and an absent state.

  The bedded state is, for example, a state where the facility user 210 is lying on the bed. For example, in the flooring state, the posture state of the facility user 210 is changed to a state where the upper body is awakened or an end-sitting state due to the wake-up action.

  In the bedded state, for example, the head of the facility user 210 is located on the bed. Therefore, in step S403 in FIG. 4, for example, by detecting that the head of the facility user 210 is present at a height (position) close to the bed, the state of the facility user 210 is changed to the bedded state. Is detected.

  The state where the upper body is raised is, for example, a state where the facility user 210 has raised the upper body on the bed. In the state where the upper body has been raised, the posture state of the facility user 210 transitions to, for example, an end-sitting state, a state around the bed, or a bed state. For example, in a state where the upper body is raised, the posture state of the facility user 210 is changed to a state of being around the bed due to getting out of bed. Alternatively, in a state where the upper body is raised, the posture state of the facility user 210 transitions to the end sitting state due to the behavior before getting out of bed. The end sitting state is, for example, a state where the facility user 210 is sitting on the end of the bed. In the end-sitting position, the posture state of the facility user 210 transitions to, for example, a state around the bed, a state where the upper body has been woken up, or a bed state. For example, in the end-sitting position, the posture state of the facility user 210 is changed to a state of being around the bed by a bed leaving action.

  In the state where the upper body is raised and in the end sitting position, for example, the head of the facility user 210 is located at a height away from the bed. In the state where the upper body is raised, for example, the facility user 210 is located near the center of the bed as compared to the end-sitting position. In other words, in the end-sitting state, for example, the facility user 210 is positioned at the end of the bed as compared to a state where the upper body is raised. Therefore, in step S403 of FIG. 4, for example, the facility user 210 is detected as having the upper body by detecting that the head of the facility user 210 is present at a position away from the bed. It is detected that it is in one of the end sitting positions. Further, in step S403, it is detected whether or not the facility user 210 is in the end sitting position by detecting whether or not the facility user 210 is at the end of the bed.

  The state of being around the bed is, for example, a state where the facility user 210 is beside the bed. In the state around the bed, the state of the posture of the facility user 210 transitions to, for example, any of the absence state, the end-sitting state, and the upper body state. For example, in the state around the bed, the posture state of the facility user 210 is changed to the absence state due to the exiting action.

  In the state around the bed, for example, the facility user 210 is located around the bed. Therefore, in step S403 of FIG. 4, for example, the facility user 210 is in the vicinity of the bed when it is detected that the facility user 210 is present within a predetermined distance from the bed. It is detected.

  The absence state is, for example, a state where the facility user 210 is not around the bed. For example, the state where the facility user 210 has moved to a position away from a predetermined distance from the bed is included in the absence state even when the facility user 210 has not left the room. In the absence state, the posture state of the facility user 210 transitions to a state around the bed, for example.

  In the absence state, for example, the facility user 210 is not around the bed. Therefore, in step S403 in FIG. 4, for example, it is detected that the facility user 210 is not present at a position within a predetermined distance from the bed, thereby detecting that the state of the facility user 210 is absent. Is done.

  FIG. 6 is a diagram illustrating an example of state transition between the awake state and the sleep state when the current awake state of the facility user 210 is determined in step S404 of FIG.

As shown in FIG. 6A, when transition condition 1 is satisfied in the awake state, the awake state transitions to the sleep state. Conversely, when transition condition 2 is satisfied in the sleep state, the sleep state is Transition to the awake state.
Transition conditions 1 and 2 are shown in FIG.

  The total posture (ΣM) within the unit time A of the amount of motion (M) of one frame in the captured image data obtained in step S401 in FIG. 4 is the floor state detected in step S403 in FIG. Due to the transition condition 1 in which ΣM <N1, the awake state transitions to the sleep state. Here, the unit time A is, for example, 15 minutes. N1 is a threshold value of the amount of movement for determining that the body movement is slow for transition from the awake state to the sleep state.

  As part of transition condition 2, if the state of wakefulness is a sleep state and the sum (ΣM) of the amount of movement (M) in one frame within unit time B becomes ΣM> N2, the sleep state transitions to the wake state. . Here, it is desirable that the unit time B is set to an appropriate time for distinguishing between the body movement due to, for example, turning over in the sleeping state and the body movement when transitioning from the sleeping state to the awake state, for example, 3 minutes. It is. N2 is a threshold value of the amount of movement for determining that the body movement is active for transition from the sleep state to the awake state. As another part of the transition condition 2, when the posture state detected in step S403 in FIG. 4 changes from the bed state to another state, the sleep state transitions to the awake state.

  FIG. 7 is a flowchart showing a detailed example of the determination process of the state of awakening based on the amount of movement of the body in step S404 of FIG. 4 described in FIG.

  First, whether the posture state of the facility user 210 determined in step S403 in FIG. 4 with respect to the captured image data one frame before (previous frame) input in step S401 in FIG. It is determined whether or not (step S701).

  If the posture in the previous frame is not in the bedded state (NO in step S701), the flowchart in FIG. 7 is terminated, and the determination process in step S404 in FIG. 4 is terminated. In this case, the current awakening situation is maintained.

  If the posture in the previous frame is a bed resting state (YES in step S701), the state of awakening of the facility user 210 determined in step S404 in FIG. 4 with respect to the captured image data in the previous frame is the awakening state. It is determined whether or not (step S702).

  If the awake state of the previous frame is an awake state (YES in step S702), the amount of motion in the current frame (current frame) input in step S401 of FIG. 4 is measured (step S703), and unit time A The total sum ΣM of the movements is calculated (step S704).

  As a result, it is determined whether or not the transition condition 1 described in FIG. 6B is satisfied (step S705). If transition condition 1 is satisfied (YES in step S705), it is determined that the awakening state corresponding to the captured image data of the current frame is a sleep state (step S706). If transition condition 1 is not satisfied (determination in step S705 is YES), the process in step S706 is skipped. Thereafter, the flowchart of FIG. 7 is terminated, and the determination processing in step S404 of FIG. 4 is terminated.

  If the awakening state in the previous frame is not an awakening state (NO in step S702), the state of the facility user 210 determined in step S403 in FIG. 4 with respect to the captured image data in the current frame is Then, it is determined whether or not the user is in the bedded state (step S707).

  If the posture state in the current frame is a bed resting state (YES in step S707), the amount of motion in the current frame is measured (step S708), and the total motion ΣM in unit time B is calculated (step S708). Step S709).

  In addition, when determination of step S707 is NO, it determines with it being an awake state (step S712), the flow of FIG. 7 is complete | finished, and the determination process of step S404 of FIG. 4 is complete | finished.

  Based on the total motion ΣM within the unit time B calculated in step S709, it is determined whether or not the transition condition 2 described in FIG. 6B is satisfied (step S710). If transition condition 2 is satisfied (YES in step S710), it is determined that the awakening state corresponding to the captured image data of the current frame is the awakening state (step S711). If transition condition 2 is not satisfied (determination in step S705 is YES), the process in step S711 is skipped. Thereafter, the flowchart of FIG. 7 is terminated, and the determination processing in step S404 of FIG. 4 is terminated.

  FIG. 8 is a flowchart showing a detailed example of the process of measuring the motion amount of the current frame executed in step S703 or S708 of FIG. 7, and FIG. 9 is an explanatory diagram thereof.

  Based on the position of the head of the facility user 210 detected in step S402 in FIG. 4 and the center line of the body, a processing range for measuring the amount of movement is set (step S801). For example, in FIG. 9, in both cases (a) and (b), body centerlines 901a and 901b are correctly detected based on the position of the head. The processing ranges 902a and 902b are set based on the center lines 901a and 901b of these bodies. Thereby, even if the facility user 210 is displaced on the bed, for example, it is possible to correctly set the range for measuring the amount of movement.

  Next, the difference in the luminance value of the image is calculated between the previous frame and the current frame (step S802).

  Then, the number M of pixels in which the difference exceeding the threshold is detected in the processing range set in step S801 is counted, and the number of pixels M is stored in a memory or the like together with the current frame number (step S803). In this way, for the number of pixels M calculated in step S703 or S708 in FIG. 7, the sum ΣM in step S704 or S709 in FIG. 7 is calculated. In step S705, the transition condition 1 in FIG. 6B is changed to step S710. Thus, the transition condition 2 in FIG. 6B is determined.

  Note that the above-described measurement of the amount of motion may be realized not only by the above-described difference in luminance value of the image but also by a known image processing technique that detects a motion between frames of an image such as an optical flow. For example, when the above-described motion amount is measured using an optical flow, the number of pixels M in which a difference exceeding the threshold value in step S803 is detected can be read as the number M of feature points having a motion vector of a certain amount or more. Good. Furthermore, each movement is specified from the change in the position on the image between the frames of the respective parts of the body such as the head and torso of the facility user 210 detected in step S402 in FIG. The sum may be used as the amount of movement.

  7 and 8 makes it possible to determine whether the facility user 210 is in a sleep state or in an awake state.

  Next, a first embodiment according to the alarm determination process in step S405 of FIG. 4 will be described.

  FIG. 10 is a flowchart illustrating a process example according to the first embodiment of the alarm determination process in step S405 of FIG.

  First, a process for determining whether or not the alarm mode is set is executed (step S1001). FIG. 11 is a flowchart illustrating a detailed example of the process in step S1001. Step S404 in FIG. 4 (FIG. 7 for details) determines whether or not the state of awakening of the facility user 210 has changed from the sleeping state to the awakening state in the current frame (step S1101). If the determination in step S1101 is YES, the current facility user 210 mode is switched to the alarm mode (step S1102).

  As a result of the process of step S1001 of FIG. 10 shown in the flowchart of FIG. 11, it is determined whether the alarm mode is in which a current alarm (alarm) is output or the non-alarm mode in which no alarm is output (step S1002).

  When the current time is the alarm mode (determination in step S1002 is YES), it is determined whether or not it is within a predetermined time (predetermined time: for example, 10 minutes) after switching to the alarm mode (step S1003). If the current alarm mode is not set (NO in step S1002), the process of the flowchart in FIG. 10 is terminated and the alarm determination process in step S405 of FIG. 4 is terminated.

  If the predetermined time has elapsed (NO in step S1003), the current mode is switched to the non-alarm mode (step S1004). Thereby, after the facility user 210 changes from the sleep state to the awake state, when a certain period of time has passed and the consciousness will be clear, the non-alarm mode is set and an unnecessary alarm is transmitted to the management terminal 104. It is possible to control so as not to be performed. After the process of step S1004, the process of the flowchart of FIG. 10 is terminated and the alarm determination process of step S405 of FIG. 4 is terminated.

  If it is within a certain period of time after switching to the alarm mode (YES in step S1003), it is determined whether or not the facility user 210 has left (or woken up) by the determination process in step S403 of FIG. S1005).

  If the facility user 210 leaves (or wakes up) (YES in step S1005), the processing of the alarm unit 204 is called and an alarm is generated and transmitted to the management terminal 104 via the in-hospital LAN 220 of FIG. Is done. Thereafter, the process of the flowchart of FIG. 10 is terminated, and the alarm determination process of step S405 of FIG. 4 is terminated.

  If the facility user 210 is not getting out of bed (or getting up) (NO in step S1005), the process of the flowchart of FIG. 10 is terminated and the alarm determination process of step S405 of FIG. 4 is terminated.

  FIG. 12 is a diagram showing an example of an alarm notification screen transmitted from the state detection apparatus 100 of FIG. 2 to the management terminal 104 in the alarm mode based on the flowcharts of FIGS. 4, 7, 8, 10, and 11. Yes, there is a risk of falling, and an alarm is issued to assist walking. As described above, in the present embodiment, when the facility user 210 is in a state in which the consciousness is easily clouded immediately after the sleep state is changed from the sleep state, the movement of the facility user 210 to get out (or wake up) is detected. An alarm can be issued. Further, when a certain period of time elapses after entering the awake state, it is estimated that the consciousness of the facility user 210 has become clear, and the current mode is changed from the alarm mode to the non-alarm mode (step S1004 in FIG. 10). Accordingly, it is possible to prevent an unnecessary alarm from being issued due to the action of getting out (or getting up) when the consciousness of the facility user 210 will be clear. As a result, it is possible to control so that an alarm (alarm) is output only when the consciousness of the person to be monitored is cloudy, ensuring the safety of the facility user 210 and increasing the efficiency of the staff's work. It becomes possible.

  Next, a second embodiment of the present invention will be described. In the second embodiment, the trend of time until the facility user 210 leaves the awake state from the awake state is accumulated in advance as statistical information. And when the tendency of the time is short, a warning will be output when the action is detected by the sensor at an earlier timing than getting out of the facility user 210 within a predetermined time from the detection of the change from the sleep state to the awake state. The The timing earlier than getting out of bed is, for example, the timing of transition to the awake state or the timing of performing the wake-up action.

  FIG. 13 is a flowchart illustrating an example of processing according to the second embodiment for determining a tendency to get out of the facility user 210 after awakening.

  First, the accumulated time from awakening to getting out of bed (leaving time) in each past sleep of the facility user 210 is referred to (step S1301).

Next, the median value is referenced from the distribution of the bed leaving time (step S1302).
Then, the median bed leaving characteristic value (for example, 6 levels) is determined (step S1303).

  FIG. 14 is a diagram illustrating a data configuration example of an alarm timing table in which the relationship between the bed leaving characteristic value and the alarm timing is registered. This table is referred to in an alarm determination process according to a second embodiment to be described later. For example, if the median of the bed time distribution of the facility user 210 is less than 10 seconds and the bed leaving characteristic value is 1, when the state of awakening of the facility user 210 transitions from a sleep state to a sleep state as an alarm timing, An alarm will be issued. Next, if the median of the distribution of the bedtime of the facility user 210 is 10 seconds or more and less than 20 seconds and the bedtime characteristic value is 2, the warning is issued when the facility user 210 performs the wake-up action as the warning timing. Be emitted. Further, if the median of the distribution of the bedtime of the facility user 210 is 20 seconds or more and less than 45 seconds and the bedout characteristic value is 3, an alarm is issued when the facility user 210 performs the pre-bedout action as the alarm timing. Be emitted. Further, if the median value of the distribution of the bedtime of the facility user 210 is 45 seconds or more and less than 90 seconds and the bedlighting characteristic value is 4, an alarm is issued when the facility user 210 transitions to the end sitting position as the alarm timing. Be emitted. Further, if the median value of the bed time distribution of the facility user 210 is 90 seconds or more and less than 180 seconds and the bed leaving characteristic value is 5, an alarm is issued when the facility user 210 performs the bed leaving action as the alarm timing. It is done. If the median of the bed time distribution of the facility user 210 is 180 seconds or more and the bed leaving characteristic value is 6, an alarm is issued when the facility user 210 transitions to a state around the bed as an alarm timing. It is done.

  FIG. 15 is a flowchart illustrating a process example according to the second embodiment of the alarm determination process in step S405 of FIG. In FIG. 15, the same step numbers are assigned to the same processing steps as the alarm determination processing of FIG. 10 according to the first embodiment. The process of FIG. 15 differs from the process of FIG. 10 in the following points.

  In FIG. 10, it is determined in step S1005 whether or not the facility user 210 has just left the floor (wakes up), and an alarm is issued in step S1006.

  On the other hand, in step S1501 in FIG. 15 replacing step S1005 in FIG. 10, it is determined whether or not the movement of the facility user 210 matches the operation at a predetermined alarm timing. The predetermined alarm timing is an alarm timing corresponding to the bed leaving characteristic value on the alarm timing table in FIG. 14 determined in step S1303 in FIG. 13 or a bed leaving characteristic value higher than the bed leaving characteristic value associated with the alarm timing. The associated alarm timing.

  If the movement of the facility user 210 matches the operation of the alarm timing (determination in step S1501 is YES), an alarm is issued in step S1006.

  If the movement of the facility user 210 does not match the alarm timing operation (NO in step S1501), the process of the flowchart in FIG. 15 is terminated and the alarm determination process in step S405 of FIG. 4 is terminated.

  In this manner, in the second embodiment, it is possible to issue an alarm at an appropriate action timing of the facility user 210 in accordance with the time trend of the facility user 210 until he leaves the bed from the awake state.

  Next, a third embodiment will be described. In 3rd Embodiment, the action used as the trigger for outputting an alarm is controlled according to the elapsed time from the detection of the change from the sleep state to the awake state within a predetermined time. For example, immediately after awakening, for example, an alarm is output when a transition is made to an awakening state or when a wake-up action is performed. On the other hand, when time has passed since awakening, no warning is issued when transitioning to the above-mentioned awakening state, and after that, for example, when leaving the bed, transitioning to a state around the bed An alarm is output.

  FIG. 16 is an alarm characteristic graph showing the relationship between the elapsed time after awakening and the alarm rank according to the third embodiment. Relational data corresponding to this graph is stored in, for example, a memory or the like in the state detection device 100, and the alarm rank is determined according to the elapsed time after waking up. Immediately after awakening, the value of the alarm rank takes a maximum value of 6, and the alarm rank value decreases from the maximum value of 6 toward 0 as the elapsed time after awakening becomes longer.

  FIG. 17 is a diagram illustrating a data configuration example of an alarm rank table according to the third embodiment. This table is stored in, for example, a memory in the state detection device 100. This table is referred to in an alarm determination process according to a third embodiment to be described later. For example, if the alarm rank of the facility user 210 is 6, an alarm is issued when the alert state of the facility user 210 transitions from a sleep state to an awake state as the alarm timing. If the alarm rank of the facility user 210 is 5, an alarm is issued when the facility user 210 performs a wake-up action as an alarm timing. If the alarm rank of the facility user 210 is 4, an alarm is issued as the alarm timing when the facility user 210 performs the action before leaving the bed. If the alarm rank of the facility user 210 is 3, an alarm is issued when the facility user 210 transitions to the end-sitting position as the alarm timing. Furthermore, if the facility user 210 has an alarm rank of 2, the alarm is issued when the facility user 210 performs a bed leaving action as the alarm timing. If the alarm rank of the facility user 210 is 1, an alarm is issued when the facility user 210 transitions to a state around the bed as an alarm timing. If the facility user 210 has a warning rank of 0, no warning is issued.

  FIG. 18 is a flowchart illustrating a process example according to the third embodiment of the alarm determination process in step S405 of FIG.

  First, it is determined whether or not the facility user 210 is within a certain period of time after entering the awake state (step S1801).

  If more than a certain time has elapsed since the facility user 210 entered the awake state (NO in step S1801), the process of the flowchart in FIG. 18 is terminated and the alarm in step S405 in FIG. The determination process ends.

  If the facility user 210 enters the awake state within a predetermined time (determination in step S1802 is YES), the alarm rank is determined by referring to the graph data of FIG. 16 according to the elapsed time after the awakening. . Further, the alarm timing is determined by referring to the alarm rank table having the data configuration example of FIG. 17 according to the determined alarm rank (step S1802).

  Next, it is determined whether or not the alarm rank determined in step S1802 is not 0 (step S1803).

  If the alarm rank is 0 (NO in step S1803), no alarm is issued, so the process of the flowchart of FIG. 18 is terminated and the alarm determination process of step S405 of FIG. 4 is terminated.

  If the alarm rank is not 0 (determination in step S1803 is YES), step S1804 is executed. In step S1804, it is determined whether the movement of the facility user 210 matches the operation at a predetermined alarm timing. The predetermined alarm timing is the alarm timing determined in step S1802 on the alarm timing table of FIG. 17 or the alarm timing associated with an alarm rank lower than the alarm rank with which the alarm timing is associated.

  If the movement of the facility user 210 coincides with the operation at a predetermined alarm timing (determination in step S1804 is YES), the function of the alarm unit is called in step S1805 to generate an alarm.

  If the movement of the facility user 210 does not match the alarm timing operation (NO in step S1804), the process of the flowchart of FIG. 18 is terminated and the alarm determination process of step S405 of FIG. 4 is terminated.

  In this way, in the third embodiment, an alarm is issued at an appropriate action timing of the facility user 210 according to the elapsed time after awakening when the consciousness of the facility user 210 becomes clear from the cloudy state. It becomes possible to emit.

  According to the first to third embodiments described above, it is efficient by making the alarm notification trigger different from the normal time immediately after awakening of the facility user 210 that is most likely to cause a fall accident and needs the assistance of the staff. The staff can be notified and the work of the staff is improved. In other words, it can be expected to be able to use time for other work such as care, which leads to improvement of the quality of care.

  FIG. 19 is a diagram illustrating a hardware configuration example of a server computer that is a state detection device 100 capable of executing the monitoring program 201 of FIG. 8 that realizes the function of the alarm output device 100 of FIG.

  The computer shown in FIG. 19 includes a CPU 1901, a memory 1902, an input device 1903, an output device 1904, an external storage device 1905, a portable recording medium driving device 1906 in which a portable recording medium 1909 is inserted, and a communication interface 1907. These have a configuration in which they are connected to each other by a bus 1908. The configuration shown in the figure is an example of a server computer that can implement the state detection apparatus 100, and such a server computer is not limited to this configuration.

  The CPU 1901 controls the entire computer. The memory 1902 stores various data such as the database of the monitoring program 201 and the facility user DB 205, the alarm timing table of FIG. 14, the alarm characteristic graph data of FIG. 16, the alarm rank table of FIG. Stores control data. The CUP 1901 performs overall control by reading the monitoring program 201 into the memory 1902 and executing it.

  The input device 1903 detects an input operation by the administrator of the state detection device 100 using a keyboard, a mouse, or the like, and notifies the CPU 1901 of the detection result.

  The output device 1904 outputs data sent under the control of the CPU 1901 to a display device or a printing device.

  The external storage device 1905 is, for example, a hard disk storage device. It is mainly used for storing various data and the monitoring program 201.

  The portable recording medium driving device 1906 accommodates a portable recording medium 1909 such as an optical disk, SDRAM, or CompactFlash (registered trademark), and has an auxiliary role for the external storage device 1905.

  The communication interface 1907 is a device for connecting the communication line of the hospital LAN 810 in FIG.

  The system according to the present embodiment is realized by the CPU 1901 executing the monitoring program 201 equipped with the function of the state detection apparatus 100 of FIG. The program may be distributed by being recorded in, for example, the external storage device 1905 or the portable recording medium 1909, or may be acquired from the network by the network connection device 1907.

Regarding the above embodiment, the following additional notes are disclosed.
(Appendix 1)
Controlling the output of the alarm in accordance with whether or not the predetermined behavior of the monitoring subject is detected by the sensor within a predetermined time from the detection of the change from the sleep state to the awake state for the monitoring subject.
A state detection method characterized by causing a computer to execute processing.
(Appendix 2)
The predetermined action of the monitoring subject is a wake-up action or a get-off action,
The state detection method according to supplementary note 1, wherein:
(Appendix 3)
After the predetermined time has elapsed since the detection of the change from the sleep state to the awake state, the detection of the predetermined action and the output of the alarm based on the detection are not performed.
The state detection method according to any one of appendix 1 or 2, characterized in that:
(Appendix 4)
Accumulating in advance a trend of the time to get out of the awake state for the monitoring subject,
When the tendency of the time is short, the behavior at the timing earlier than the getting out of the monitoring subject is detected by the sensor within the predetermined time from the detection of the change from the sleep state to the awake state. Output an alarm,
The state detection method according to any one of appendices 1 to 3, wherein:
(Appendix 5)
Controlling the predetermined action for outputting the alarm according to an elapsed time from detection of a change from the sleep state to the awake state within the predetermined time;
The state detection method according to any one of appendices 1 to 4, characterized in that:
(Appendix 6)
An alarm output that controls the output of an alarm according to whether or not a predetermined behavior of the monitoring target is detected by a sensor within a predetermined time from detection of a change from the sleep state to the awakening state of the monitoring target Control unit,
A state detection device comprising:
(Appendix 7)
Controlling the output of the alarm according to whether or not the predetermined behavior of the monitoring subject is detected by the sensor within a predetermined time from the detection of the change from the sleep state to the awake state for the monitoring subject,
A status detection program that causes a computer to execute processing.

1901 CPU
1902 Memory 1903 Input Device 1904 Output Device 1905 External Storage Device 1906 Portable Recording Medium Drive Device 1907 Communication Interface 1908 Bus 1909 Portable Recording Medium

Claims (7)

Controlling the output of the alarm in accordance with whether or not the predetermined behavior of the monitoring subject is detected by the sensor within a predetermined time from the detection of the change from the sleep state to the awake state for the monitoring subject.
A state detection method characterized by causing a computer to execute processing. The predetermined action of the monitoring subject is a wake-up action or a get-off action,
The state detection method according to claim 1. After the predetermined time has elapsed since the detection of the change from the sleep state to the awake state, the detection of the predetermined action and the output of the alarm based on the detection are not performed.
The state detection method according to claim 1, wherein: Accumulating in advance a trend of the time to get out of the awake state for the monitoring subject,
When the tendency of the time is short, the behavior at the timing earlier than the getting out of the monitoring subject is detected by the sensor within the predetermined time from the detection of the change from the sleep state to the awake state. Output an alarm,
The state detection method according to any one of claims 1 to 3. Controlling the predetermined action for outputting the alarm according to an elapsed time from detection of a change from the sleep state to the awake state within the predetermined time;
The state detection method according to claim 1, wherein: Alarm output control for controlling the output of an alarm according to whether or not a predetermined action of the monitoring target is detected by a sensor within a predetermined time from detection of a change from a sleep state to a wakefulness state for the monitoring target Part,
A state detection device comprising: Controlling the output of the alarm in accordance with whether or not the predetermined behavior of the monitoring subject is detected by the sensor within a predetermined time from the detection of the change from the sleep state to the awake state for the monitoring subject.
A status detection program that causes a computer to execute processing.