Abnormality determination device

Abstract

To provide an abnormality determination device, etc. that allow a fact that the time of a user's death is approaching to be determined more appropriately.SOLUTION: An abnormality determination device includes: an acquisition part for acquiring a measuring place; a biological signal acquisition part for acquiring a user's biological signals; and a control part. The control part calculates a biological information value of the user on the basis of the biological signals acquired by the biological signal acquisition part and calculates an activity amount of the user from the biological signals, and when the measuring place acquired by the acquisition part is the user's own house, determines that the user is in the state that the time of his or her death is approaching on the basis of the biological information value and the activity amount. When the measuring place acquired by the acquisition part is other than the user's own house, the control part determines that the user is in the state that the time of his or her death is approaching on the basis of the biological information value, the activity amount, and bed-leaving information based on the user's bed-leaving.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an abnormality determination device.

Background Art

[0002] There is known an invention that determines that the end of a user's life is approaching based on the fact that the user's biometric information value or the like indicates an abnormal value.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The present disclosure provides an abnormality determination device and the like that can more appropriately determine that the end of a user's life is approaching.

Means for Solving the Problems

[0005] The abnormality determination device of the present disclosure is an abnormality determination device including an acquisition unit that acquires a measurement location, a biological signal acquisition unit that acquires a user's biological signal, and a control unit. The control unit calculates the user's biological information value based on the biological signal acquired from the biological signal acquisition unit, calculates the user's activity amount from the biological signal, and when the measurement location acquired by the acquisition unit is the user's home, determines that the end of the user's life is approaching based on the biological information value and the activity amount. When the measurement location acquired by the acquisition unit is other than the user's home, it determines that the end of the user's life is approaching based on the biological information value, the activity amount, and the presence / absence information based on the user's getting out of bed.

[0006] The abnormality determination device of the present disclosure is an abnormality determination device including an acquisition unit that acquires a measurement location, a biological signal acquisition unit that acquires a biological signal of a user, and a control unit. The control unit calculates a biological information value of the user based on the biological signal acquired from the biological signal acquisition unit, calculates an activity amount of the user from the biological signal, acquires a time when the user is in bed or out of bed, and when the measurement location acquired by the acquisition unit is the user's home, determines that the user is in a state where death is approaching based on the biological information value and the time when the user is in bed or out of bed, and when the measurement location acquired by the acquisition unit is other than the user's home, determines that the user is in a state where death is approaching based on the biological information value, the activity amount, and the in-bed / out-of-bed information based on the user's getting out of bed.

Effect of the Invention

[0007] According to the present disclosure, it is possible to more appropriately determine that the death of the user is approaching.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Mode for Carrying Out the Invention

[0009] Hereinafter, with reference to the drawings, one mode for carrying out the present invention will be described. Specifically, the case where the abnormality determination device of the present invention is applied will be described, but the scope to which the present invention is applied is not limited to this embodiment.

[0010] There is known a device that determines that the end-of-life of a user is approaching when the activity amount is low when the state in which the biological information value of the user becomes an abnormal value continues for a certain period of time and then the state in which the biological information value becomes within the normal range continues for a certain period of time. The device determines that the user is not out of bed and appropriately determines that the end-of-life of the user is approaching.

[0011] However, as the inventor continued to verify, it was newly invented that the influence of getting out of bed or staying in bed is greatly related based on the location of the user. For example, previously, the state of the person getting out of bed was uniformly estimated regardless of the location where the device was installed. However, the prediction of the time of death in hospitals, nursing facilities, and at home may vary in the time in bed per day (24 hours) before death due to factors such as differences in living environment and treatment policies, so it was discovered that the time of death can be predicted with higher accuracy by considering them separately.

[0012] Based on this discovered content, the following embodiments will be used to explain a system and a device that can more appropriately determine the state of the user, particularly that the time of death is approaching for the user.

[0013] [1. First Embodiment] [1.1 Overall System] FIG. 1 is a diagram for explaining the overall outline of an abnormality determination system 1 to which the abnormality determination device of the present invention is applied. As shown in FIG. 1, the abnormality determination system 1 includes a detection device 3 placed between the floor part of the bed 10 and the mattress 20, and a processing device 5 for processing the value output from the detection device 3. The detection device 3 and the processing device 5 constitute an abnormality determination system (abnormality determination device).

[0014] When a user (hereinafter, for example, referred to as "user P") lies on the mattress 20, the detection device 3 detects body vibration (vibration emitted from the human body) as the biological signal of the user P. Then, based on the detected vibration, the biological information value of the user P is calculated. In the present embodiment, the calculated biological information value (at least the respiratory rate, heart rate, and activity amount) can be output and displayed as the biological information value of the user P. For example, it may be integrally formed by providing a storage unit, a display unit, etc. in the detection device 3. Further, since the processing device 5 may be a general-purpose device, it is not limited to an information processing device such as a computer, and may be configured by a device such as a tablet or a smartphone.

[0015] In addition, the user may be a person who is undergoing convalescence due to illness or in need of care. Also, even a healthy person who does not require care, whether elderly, a child, a person with a disability, or even an animal instead of a human, is acceptable.

[0016] Here, the detection device 3 is configured in a sheet shape so as to be thin. As a result, even when placed between the bed 10 and the mattress 20, it can be used without causing discomfort to the user P, so that the biological information value on the bed can be measured for a long period of time. That is, when the user is lying in bed and at rest, the biological information value and the like are acquired as the state of the user.

[0017] Note that the detection device 3 only needs to be able to acquire the biological signal (body movement, respiratory movement, heart beat, etc.) of the user P. In this embodiment, the heart rate and respiratory rate are calculated based on body vibration. However, for example, it may be detected using an infrared sensor, the biological signal of the user P may be acquired from the acquired video, etc., or a strain gauge-equipped actuator may be used. Also, by using a built-in acceleration sensor or the like, it may be realized by, for example, a smartphone placed on the bed 10 or a tablet.

[0018] Also, the bed 10 is installed in various places. For example, the bed 10 may be installed in the user's own home where the user is, or in a hospital where the user is admitted or a facility where the user resides. Hereinafter, places other than the home including hospitals and facilities are referred to as hospitals, etc. Note that the home is a place where the user lives and may be registered in advance or registered by the user himself / herself.

[0019] [1.2 Functional Configuration] Next, the functional configuration of the abnormality determination system 1 will be described with reference to FIGS. 2 and 3. The abnormality determination system 1 in this embodiment includes a detection device 3 and a processing device 5, and each functional unit (processing) may be realized by either one except for the biological signal acquisition unit 400. That is, by combining these devices, it functions as an abnormality determination device.

[0020] Note that after determining that the abnormal state exists, the abnormality determination system 1 may perform a reporting (notification) operation. At this time, the destination of the report may be a staff member or a family member. Also, as a method of reporting, it may simply be reported (notified) by sound or screen display, or may be reported to the mobile terminal device by email or the like. Further, it may be reported (notified) to other terminal devices or the like.

[0021] [1.2.1 Hardware Configuration] The abnormality determination system 1 (abnormality determination device) includes, as necessary, one or more of a control unit 100, a storage unit (storage 200, ROM 300, and RAM 310), a biological signal acquisition unit 400, a measurement location acquisition unit 500, an input unit 600, an output unit 750, a notification unit 800, and a communication unit 900.

[0022] In the case of FIG. 1, the control unit 100, the biological signal acquisition unit 400, and the storage unit are provided in the detection device 3, and the others may be provided in the processing device 5.

[0023] The control unit 100 controls the entire abnormality determination system 1 (abnormality determination device). The control unit 100 realizes various functions by reading and executing various programs stored in a storage device (for example, storage 200 or ROM 300). The control unit 100 may be realized by one or more control devices / arithmetic devices (CPU (Central Processing Unit), SoC (System on a Chip)). Also, the control unit 100 may be composed of a control circuit.

[0024] The storage 200 is a non-volatile storage device capable of storing programs and data. For example, it may be composed of a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive). Also, the storage 200 may be configured as an externally connectable USB memory, memory card, or the like. Further, the storage 200 may be, for example, a storage area on the cloud.

[0025] The ROM 300 is a non-volatile memory capable of retaining programs and data even when the power is turned off.

[0026] The RAM 310 is a main memory mainly used by the control unit 100 when executing processes. The RAM 310 is a rewritable memory that temporarily holds data including programs read from the storage 200 and the ROM 300, as well as results during execution.

[0027] The biological signal acquisition unit 400 acquires biological signals of the user P. In this embodiment, as an example, body vibrations, which are a type of biological signal, are acquired using a sensor that detects pressure changes. Then, the acquired body vibrations are converted into biological information value data such as respiratory rate, heart rate, and activity level and output by the control unit 100. Furthermore, based on the body vibration data acquired by the biological signal acquisition unit 400, the control unit 100 can also acquire the user's bedridden state (for example, whether the user P is bedridden, in-bed, out-of-bed, semi-reclined, etc.) or the sleep state (sleep, wakefulness) as described later.

[0028] Note that the biological signal acquisition unit 400 in this embodiment acquires the user's body vibrations using, for example, a pressure sensor and acquires respiration and heartbeats from the body vibrations. However, it is also possible to acquire biological signals based on changes in the user's center of gravity position (body movement) using a load sensor, or to acquire biological signals based on displacements of the body surface or bedding using a radar, or to acquire biological signals based on sounds picked up by a microphone by providing a microphone. As long as biological signals of the user can be acquired using any of these sensors.

[0029] That is, the biological signal acquisition unit 400 may be connected to a device such as the detection device 3 or may be configured to receive biological signals from an external device.

[0030] The measurement location acquisition unit 500 acquires the location where the user P measures biological information or the like. Here, the measurement location can mainly acquire the user's home and locations other than home (such as hospitals, etc.).

[0031] Also, for example, the user may manually set the measurement location, and the measurement location acquisition unit 500 may acquire the set measurement location. Further, the measurement location acquisition unit 500 may acquire the measurement location by means of a changeover switch. Additionally, the measurement location acquisition unit 500 may acquire the measurement location by using information such as GPS or a wireless base station of a mobile phone, for example. Moreover, the measurement location acquisition unit 500 may further refer to map information to acquire the measurement location.

[0032] In this embodiment, the measurement location acquired by the measurement location acquisition unit 500 is, for example, acquired as home and locations other than home, but it may be acquired in more detail. For example, the measurement location may be divided into a hospital and a facility. Also, the measurement location acquisition unit 500 may divide the hospital into the hospital where the user is usually admitted and the hospital to which the user has been transferred for specific treatment such as a university hospital, and acquire them separately. Further, the measurement location acquisition unit 500 may acquire the position information, and the control unit 100 may determine and acquire the measurement location.

[0033] The input unit 600 allows the measurer to input various conditions or perform an operation input for starting the measurement. The input unit 600 may be realized by any input means such as a hardware key or a software key, for example.

[0034] The output unit 700 is a functional unit for outputting biological information values such as the sleep state, heart rate, and respiratory rate, or for notifying abnormalities. The output unit 700 may be a display device such as a display, or a notification device (sound output device) for notifying an alarm or the like. It may also be an external storage device for storing data, a transmission device for transmitting data via a communication path, or the like. It may also be a communication device for notifying other devices.

[0035] Further, the input unit 600 and the output unit 700 may be implemented by other devices. For example, they may be implemented using a terminal device (e.g., a smartphone or a tablet used by a user) connected via the communication unit 900. At this time, the terminal device may be capable of executing a program for the control unit 100 to realize the processing described later.

[0036] The notification unit 800 performs notification to users and the like. For example, the notification unit 800 may be a speaker that outputs sound, an LED which is a light-emitting device, etc. Also, the notification unit 800 may perform notification to other devices (e.g., a terminal device such as the user's smartphone, a nurse call, etc.).

[0037] The communication unit 900 communicates with other devices. For example, if the communication unit 900 is a short-range device, it provides communication in a manner such as wireless LAN (or wired LAN), Bluetooth (registered trademark). Also, the communication unit 900 may be a device that provides short-range wireless communication such as NFC. Further, the communication unit 900 may provide communication in a manner capable of mobile communication such as 4G / LTE / 5G / 6G. Also, the communication unit 900 may be an interface (e.g., USB) for communicating with other devices.

[0038] [1.2.2 Software Configuration] With reference to FIG. 3, the software configuration will be described. For example, the control unit 100 realizes each function by executing programs and applications stored in a storage unit (e.g., the storage 200, the ROM 300, the RAM 310).

[0039] The biological information value calculation unit 110 calculates biological information values (such as respiration rate, heart rate, activity level, etc.) of the user P. In this embodiment, the biological information value calculation unit 110 may extract the respiration component and the heart rate component from the body movement acquired from the biological signal acquisition unit 400, and obtain the respiration rate and the heart rate based on the respiration interval and the heart rate interval. Also, the biological information value calculation unit 110 may analyze the periodicity of the body movement (such as Fourier transform, etc.) and calculate the respiration rate and the heart rate from the peak frequency. Further, the biological information value calculation unit 110 may calculate the activity level together. Specifically, the biological information value calculation unit 110 may detect body vibrations per sampling unit time from the biological signal acquisition unit 400 and calculate the activity level based on the number of detected body vibrations. Also, the biological information value calculation unit 110 may calculate the activity level from the change in the user's sleeping posture and movement.

[0040] Specifically, the biological information value calculation unit 110 continuously measures the output value of the sensor at a sampling period of, for example, 16 times per second (960 times per minute). A predetermined measurement threshold value (upper limit value and lower limit value) is set for the measurement value of the sensor. The biological information value calculation unit 110 sequentially outputs the aggregated activity level (0 to 960). The activity level is a numerical quantity (0 to 960) representing how much the user P on the bed has moved the body. The activity level is related to the frequency and intensity of the body movement of the user P. When the activity level is large, it means that the user P has moved the body frequently and greatly on the bed.

[0041] The sleep state determination unit 120 determines the sleep state of the user. For example, the sleep state determination unit 120 determines the sleep state of the user based on the biological signal acquired by the biological signal acquisition unit 400. The sleep state determination unit 120 may determine two states, "awake" and "sleep", as the sleep state. Also, the sleep state determination unit 120 may further determine the "sleep" state as "REM sleep" and "non-REM sleep", or may further perform a determination of multiple levels (depth of sleep) as the "sleep" state.

[0042] Also, the sleep state determination unit 120 is based on the magnitude of the activity level and the state of the time-series change of the activity level It is also possible to determine the sleep state and the awake state. For example, the sleep state determination unit 120 may not determine the awake state even if there is temporary body movement. The sleep state determination unit 120 may determine the awake state when the body movement of the user P continues to a certain extent.

[0043] The user state acquisition unit 130 acquires the state of the user. The state of the user is a general state regarding the user. For example, by using a load sensor or the like provided on the bed 10, the state of whether the user has gotten out of bed or is in bed is acquired. Further, the user state acquisition unit 130 may further acquire the sleeping posture and sleeping position of the user when the user is in bed. The user state acquisition unit 130 may acquire the state of the user based on the biological signal acquired by the biological signal acquisition unit 400 as described above, in addition to a load sensor or the like. Also, the state of the user may include whether the user is sleeping or awake based on the sleep state of the user determined by the sleep state determination unit 120. The user state acquisition unit 130 may acquire the getting out of bed and getting into bed of the user based on the activity amount.

[0044] The user state estimation unit 140 estimates the state of the user from parameters such as biological information values. When the user state estimation unit 140 estimates that the state of the user is abnormal, an alert may be output (notified) by the notification unit 800.

[0045] Also, the user state estimation unit 140 may determine (estimate) that the state in which the state of the user is abnormal continues. Since the state in which the state of the user is abnormal continues, the user state estimation unit 140 may estimate that, for example, the end of life is approaching. Also, the user state estimation unit 140 may notify when the end of life of the user is approaching.

[0046] Here, when it is said that the user's death is approaching, it refers to a state where the time of the user's death is approaching. Here, as one criterion, for example, regarding the timing when the user will die within a predetermined time (for example, 3 hours, 6 hours, 12 hours, 24 hours), it is said to be a state (period) where the death is approaching. In the present embodiment, it is generally assumed to be about 12 to 24 hours, but the predetermined time can also be changed by setting parameters.

[0047] Also, the storage 200 stores the biological information data 210 and the user state data 220.

[0048] The biological information data 210 stores information regarding biological information values calculated by the control unit 100 from the acquired biological signals (body movements), such as the respiratory rate and the heart rate. In the present embodiment, as information regarding biological information values, the respiratory rate, the heart rate, and the body movement are stored, but at least one of them may be stored. Also, other information (for example, a respiratory event index based on fluctuations in respiratory amplitude, a periodic body movement index based on the periodicity of body movement) may be further stored as long as it is a biological information value calculable by the biological information value calculation unit 110. Also, the biological information data 210 is preferably stored in time series at predetermined time intervals.

[0049] The user state data 220 stores the state of the user. As the state of the user acquired by the user state acquisition unit 130, it stores whether the user is "in bed" or "out of bed". Further, the user state data 220 may include the sleep state determined by the sleep state determination unit 120 as the state of the user. For example, when the user is determined to be "in bed" by the user state acquisition unit 130, the sleep state determined by the sleep state determination unit 120 may be stored. Also, the user state data 220 is preferably stored in time series at predetermined time intervals.

[0050] [1.3 Process flow] Also, a method for inferring that the state of the user in the user state inference unit 140 in the present embodiment is abnormal will be described.

[0051] [1.3.1 Overall Flow] Figure 3 is an operation flow chart for explaining the process executed when the control unit 100 estimates the user's state. First, when the control unit 100 attempts to estimate the user's state, it acquires the measurement location (step S12). Here, when the measurement location is outside the home, the control unit 100 executes the first user state estimation process (step S14; Yes → step S16). Also, when the measurement location is at home, the control unit 100 executes the second user state estimation process (step S14; No → step S18). Hereinafter, the first user state estimation process and the second user state estimation process will be described.

[0052] [1.3.2 First User State Estimation Process] Figure 4 is an operation flow chart for explaining the first user state estimation process for estimating the state of a user whose measurement location is outside the home. In the present embodiment, while the first user state estimation process of Figure 4 is being executed, the control unit 100 acquires the following information.

[0053] · The user's biometric information values (e.g., heart rate, respiratory rate, activity level, etc.) acquired by the biometric information value calculation unit 110 · The user's sleep state determined by the sleep state determination unit 120 · The state regarding the user's getting out of bed (being in bed) acquired by the user state acquisition unit 130 is being performed.

[0054] That is, as biometric information values used when the control unit 100 estimates the user's state, the respiratory rate, heart rate, and activity level are important, but the user's state such as sleep / wakefulness (being in bed) / getting out of bed may also be used in combination. By the control unit 100 acquiring information other than biometric information values, it becomes possible to perform a more detailed estimation of the user's state taking into account changes such as the user becoming unable to sleep, an increase in the time spent in bed, an increase in the time not in bed, etc., as well as the continuous in-bed time and continuous out-of-bed time.

[0055] Furthermore, as one of the biometric information values, by obtaining a respiratory disorder index and a periodic body movement index, which are indices (biometric indices) related to the user, it becomes possible to more accurately estimate the user's state from these absolute values, changes in daily average values, changes in the 24-hour time series distribution, and the like. Also, a history of biometric information values may be obtained, and past values, average values, standard deviations, coefficients of variation, and values / ratios of changes in a predetermined recent time may be obtained.

[0056] The control unit 100 may directly obtain biometric information values from, for example, a device capable of obtaining biometric information values as biometric information values, or may calculate them by performing a predetermined operation on the biometric signals output by the biometric signal acquisition unit 400. Also, the control unit 100 may calculate different biometric information values or indices based on biometric information values from a plurality of biometric information values.

[0057] The control unit 100 determines whether the biometric information value has shown an abnormal value continuously for the first determination time (step S102). Here, the first determination time is preferably 15 minutes or more, and more preferably 30 minutes or more. The control unit 100 determines whether the biometric information value has continuously shown an abnormal value beyond the first determination time. Also, when determining that the biometric information value has continuously shown an abnormal value, preferably, the biometric information value continuously shows an abnormal value within the first determination time, but cases where the average value of the biometric information value at the first determination time shows an abnormal value, or cases where the biometric information value is included in the normal range for only a short period (for example, about several seconds) may also be determined as the abnormal value continuing. Also, the control unit 100 determines whether the biometric information value is an abnormal value by determining abnormal values of the heart rate and / or respiratory rate. Note that as the biometric information value used for determining the abnormal value, other acquirable biometric information values (for example, blood pressure, oxygen saturation, etc.) may be used. Also, the control unit 100 may use both biometric information values of the respiratory rate and heart rate, or may use either biometric information value. Here, as an example of the normal range of each biometric information value, Respiratory rate: 8 - 25 breaths per minute Heart rate: 50 - 110 beats per minute That is, when the biological information value deviates from the normal range, the control unit 100 determines it as an abnormal value. Note that the normal range of the biological information value is an example and may vary depending on, for example, the age and weight of the user. Further, the control unit 100 may set the biological information value when the user is normal (for example, in a healthy state) as the normal range for each user.

[0058] That is, the normal range of the biological information value here may be the range of values shown when the user is normal as the biological information value.

[0059] When the control unit 100 determines that the biological information value has continuously shown an abnormal value for the first determination time, it determines whether the biological information value of the user has stopped showing an abnormal value (step S104). For example, the control unit 100 determines whether the biological information value of the user is again included in the normal range. Here, when the biological information value continuously shows an abnormal value, the control unit 100 repeats the process from step S102 (step S104; No → step S102).

[0060] When the biological information value of the user is again included in the normal range, the control unit 100 determines whether a determination of getting out of bed has been made (step S106). When the control unit 100 determines that the user has gotten out of bed, the process transitions back to step S102 again. This is because the fact that the user has gotten out of bed is not considered to mean that the user's death is imminent.

[0061] Here, for the determination of getting out of bed, the control unit 100 may, for example, determine whether there has been getting out of bed during a predetermined time. Further, the control unit 100 may determine whether the number of times of getting out of bed has exceeded a threshold value during a predetermined time as the determination of getting out of bed.

[0062] Next, the control unit 100 determines whether the time during which the biological information value is within the normal range exceeds the second determination time (step S108). In this way, the control unit 100 detects that the biological information values (respiration rate and heart rate) are within the normal range during the second determination time (step S104; Yes), and that there has been no getting out of bed determination during this second determination time (step S106).

[0063] Here, if the biological information value shows an abnormal value again before exceeding the second determination time (step S108; No → step S104; No), the control unit 100 returns to step S102 and determines the biological information value of the user again. This is because when the biological information value shows an abnormal value again before exceeding the second determination time, even if there is an abnormality in the user, it is considered that the user's death is not imminent.

[0064] Here, the second determination time is preferably 12 hours or more, but may be 6 hours or more. Also, that the biological information value is within the normal range preferably means that the biological information value is continuously within the normal range during the second determination time. However, for example, even if the biological information value is within the abnormal range for less than a predetermined time (for example, 1 minute or for several detection timings), it may be determined as an error and continuously determined to be within the normal range.

[0065] When the time during which the biological information value is within the normal range exceeds the second determination time, the control unit 100 determines whether the activity amount of the user matches the activity amount determination condition (step S110). Here, when the activity amount of the user matches the activity amount determination condition (step S110; Yes), the control unit 100 determines that the user's death is imminent (step S112).

[0066] The activity amount determination condition is a condition for determining whether the user's activity amount or movement indicates that death is imminent. In the case of this embodiment, the activity amount determination condition may be, for example, the following conditions.

[0067] · The average activity amount at the second determination time is 5 times / minute or less · There is a non-physical movement duration exceeding 12 hours (preferably 12 hours, but it may be 6 hours, for example) (for example, a period during which the activity amount is 0 times / minute). When the user's activity amount satisfies the activity amount determination condition, the control unit 100 determines that the user's death is approaching.

[0068] Thus, in a state where the user's death is approaching, after the respiratory rate and heart rate no longer show abnormal values, there is a characteristic that body movement is almost absent for a long time and the user does not get out of bed. That is, when the activity amount is low, it is determined that death is approaching. Whether the activity amount is low is determined by the above-described determination conditions or whether the sleep state continues (the activity amount is so low that it is not determined as sleep).

[0069] Therefore, in the present embodiment, it is possible to determine whether the physical condition has recovered based on the presence or absence of this characteristic, or whether the physical condition has further deteriorated due to the continuation of the abnormality and death is approaching. This makes it possible to prevent misjudging that the state in which death is approaching due to further deterioration of the physical condition as a state where the physical condition has improved.

[0070] Note that when death is approaching, there are some people in whom respiratory events and periodic body movements significantly decrease. Therefore, the control unit 100 may determine whether parameters such as respiratory events and periodic body movements are abnormal values in addition to the biological information values. Respiratory events and periodic body movements are particularly useful when there is body movement of someone other than the user, such as a nurse, caregiver, or accompanying family member.

[0071] [1.3.3 Second User State Estimation Process] Next, the second user state estimation process executed by the control unit 100 in S18 of FIG. 4 will be described with reference to FIG. 6. That is, the second user state estimation process is executed when the user is at home.

[0072] The second user state determination process in FIG. 6 does not have a getting-out-of-bed determination compared to the first user state determination process in FIG. 5. For other processes, since they are the same processes, the same reference numerals are assigned and detailed descriptions are omitted.

[0073] When the user is at home, the user has more opportunities to get out of bed compared to when the user is outside the home. For example, due to household chores (housework, tidying up, etc.), meals, going to the toilet, etc. at home, these may be done outside the bed, and inevitably or according to the user's wish, the user often has to get out of bed. Therefore, when the user is at home, the control unit 100 does not perform a getting-out-of-bed determination, and determines whether the user's death is approaching only based on the biological information value and the activity amount.

[0074] [1.4 Effects] As described above, according to this embodiment, when determining the user's approaching death, by changing the determination method depending on the measurement location, the user's approaching death can be correctly determined. Also, when the control unit 100 determines the user's approaching death, when determining between at home and outside the home, it can more accurately determine the user's approaching death depending on whether getting out of bed is taken into consideration.

[0075] [2. Second Embodiment] Next, the second embodiment will be described. The second embodiment uses the in-bed time when the measurement location of the user is at home. Since medical resources and monitoring systems are often insufficient at home, compared to hospitals, etc., it is more likely to lead to death, and not being able to get out of bed is not only strongly related to the deterioration of the physical condition, but also has an adverse effect on life maintenance such as the inability to live without getting out of bed and the acceleration of the deterioration of the physical condition. Therefore, the inability to get out of bed can also be a strong cause of death. This property becomes more prominent in the case of living alone. Note that the second embodiment has the same hardware configuration and main software configuration as the first embodiment, and only the differences from the first embodiment will be described.

[0076] FIG. 7 shows the second user state estimation process in the second embodiment. The process in FIG. 7 replaces the second user state estimation process in FIG. 6 of the first embodiment.

[0077] When the user's biometric information value has been an abnormal value continuously for the first determination time (step S102; Yes), the control unit 100 determines whether the user's in-bed time or out-of-bed time (number of times of getting out of bed) matches the determination condition (step S202). When the user's in-bed time or out-of-bed time (number of times of getting out of bed) matches the determination condition (step S202; Yes), the control unit 100 determines that the user's death is approaching (step S112).

[0078] Here, as the user's in-bed time, the control unit 100 can use any of the following times.

[0079] · The time when the user is in bed (the time when the user is not out of bed) · The time when the person out of bed is in a lying position (for example, the time excluding the time in a long sitting position, an upright sitting position, etc.) · The time when the user's sleep state is in sleep

[0080] Also, as the user's out-of-bed time, the control unit 100 can use the time other than the in-bed time (for example, when the user's being in bed is not detected).

[0081] Also, as the determination condition, the control unit 100 can use any of the following conditions.

[0082] · The in-bed time or out-of-bed time in the immediately preceding predetermined time exceeds the threshold value (for example, among the immediately preceding 24 hours, the in-bed time exceeds 1300 minutes, and among the immediately preceding 12 hours, the ratio of the out-of-bed time is less than 10%) · The in-bed time of the previous day exceeds the threshold value (for example, the in-bed time of the previous day exceeds 20 hours) · The in-bed time continuously exceeds the threshold value (for example, the immediately preceding in-bed time exceeds 15 hours) · The out-of-bed time in the immediately preceding predetermined time is below the threshold value (for example, the out-of-bed time in the immediately preceding 6 hours is 5 minutes or less) · The number of times of getting out of bed in the immediately preceding predetermined time is below the threshold value (for example, the number of times of getting out of bed in the immediately preceding 6 hours is 0 times)

[0083] Here, the previous day's in-bed time refers to, for example, the in-bed time of the day before the day of measurement. For example, if it is around noon on April 10th, it refers to the in-bed time on April 9th. Also, the previous day's in-bed time may be the in-bed time for the time period from 24 hours before the current measurement time (the immediate previous 24 hours).

[0084] In this way, when the biological information value continues to be an abnormal value for a predetermined time or more and the user's in-bed time has become long, the control unit 100 can determine that the user's death is approaching. Also, when the user has not gotten out of bed for a certain period of time, the control unit 100 can determine that the user's death is approaching.

[0085] [3. Third Embodiment] Next, the third embodiment will be described. In the first embodiment, in the user state estimation unit 140, it was described that the input biological information is determined based on all, any one, or a combination of the biological information value and "user movement (activity level)", "variation in respiratory amplitude (respiratory event index)", and "periodic user movement (periodic body movement index)" to determine the abnormal state of the user, particularly the user's approaching death.

[0086] This embodiment will describe the case where the user state estimation unit 140 estimates the user's state using artificial intelligence (machine learning).

[0087] In this embodiment, instead of the first user state estimation process in FIG. 5 and the second user state estimation process in FIG. 6, the user state, particularly whether the user's death is approaching, is estimated based on the user state estimation unit 140A in FIG. 8.

[0088] Here, the operation of the user state estimation unit 140A in this embodiment will be described. The user state estimation unit 140A uses biological information and the user's state as input values (input data), and utilizes artificial intelligence and various statistical indicators to estimate the user's state.

[0089] As shown in FIG. 8, the user state estimation unit 140A includes a feature extraction unit 142, an identification unit 144, an identification dictionary 146, and a user state output unit 148. Further, the identification dictionary 146 has a first identification dictionary 146A used when the measurement location is outside the home and a second identification dictionary 146B used when the measurement location is at home.

[0090] First, as input data input to the user state estimation unit 140A, various parameters are input and used. For example, in the present embodiment, based on the body vibration data acquired by the biological signal acquisition unit 400, the "respiration rate", "heart rate", "activity amount" calculated by the biological information calculation unit, and the "sleep state" determined by the sleep state determination unit 120 are used. "Variation in respiration rate" and "variation in heart rate" calculated from these biological information values, and "respiratory disorder index" and "periodic limb movement index" calculated from the same body vibration data can also be used.

[0091] In addition, when the measurement location is outside the home, the user state estimation unit 140A also inputs, as parameters, parameters related to the user's "getting out of bed" and "being in bed". Therefore, the first identification dictionary 146A is a generated identification dictionary including parameters related to "getting out of bed" and "being in bed" as parameters.

[0092] In addition, when the measurement location is at home, the user state estimation unit 140A does not include parameters related to the user's "getting out of bed" and "being in bed" in the parameters. Therefore, the second identification dictionary 146B is a generated identification dictionary that does not include parameters related to "getting out of bed" and "being in bed" as parameters.

[0093] Then, each feature point is extracted by the feature extraction unit 142 and output as a feature vector. Here, examples of what the feature extraction unit 142 extracts as feature points include the following.

[0094] (1) The respiration rate is 30 [times / minute] or more or 8 [times / minute] or less and continues for a certain period of time or more (2) The heart rate is 120 [times / minute] or more or 40 [times / minute] or less and continues for a certain period of time or more (3) The trend of the heart rate or respiratory rate increases (by 10% or more) from the start to the end of nighttime sleep (4) The variation (standard deviation, coefficient of variation) of the respiratory rate or heart rate during the night (21:00 - 6:59) is equal to or greater than a certain value (5) The apnea - hypopnea index or periodic limb movement index significantly decreases (6) The apnea - hypopnea index or periodic limb movement index significantly increases or is equal to or greater than a certain value (during the night) (7) The activity level significantly increases or decreases (8) The sleep determination continues for a certain period of time or more, and the nighttime wakefulness determination is 95% or more

[0095] The feature extraction unit 142 outputs a feature vector by combining one or more of these feature points. Note that the ones described as feature points are just examples and are not limited to those values. For example, taking (1) as an example, the respiratory rate may be 25 [breaths / min] or more, or 10 [breaths / min] or less. Thus, the values of the parameters are for the convenience of explanation. And the feature extraction unit 142 may output "1" for the corresponding feature points and "0" for the non - corresponding feature points, or may output a random variable

[0096] Also, the feature extraction unit 142 may further include the following parameters according to the measurement location of the user

[0097] (9) Information regarding the user's getting out of bed and being in bed (in - bed / out - of - bed information) The in - bed / out - of - bed information is information regarding the user's getting out of bed and being in bed. For example, the control unit 100 may output in a time series whether the user is "out of bed" or "in bed". Also, the control unit 100 may include information such as the number of times the user gets out of bed, the frequency of getting out of bed, and the time of getting out of bed

[0098] When the measurement location of the user is outside the home and all of the above-described feature points are included, the feature space is nine-dimensional, and the feature extraction unit 142 outputs a nine-dimensional feature vector to the identification unit 144. Also, when the measurement location of the user is at home, the feature space is eight-dimensional, and the feature extraction unit 142 outputs an eight-dimensional feature vector to the identification unit 144.

[0099] The identification unit 144 identifies the class corresponding to the user state from the input feature vector. At this time, as the identification dictionary 146, the class is identified by comparing with a plurality of prototypes prepared in advance. The prototype may be stored as a feature vector corresponding to each class, or may store a feature vector representing the class.

[0100] When the feature vector representing the class is stored, the class to which the nearest prototype belongs is determined. At this time, it may be determined by the nearest neighbor rule, or may be identified by the k-nearest neighbor method.

[0101] Note that the identification dictionary 730 used by the identification unit 144 may store the prototype in advance, or may store it using machine learning.

[0102] Also, when the user is measured outside the home, the identification unit 144 identifies the class using the first identification dictionary 146A since, for example, a nine-dimensional feature vector is input. Also, when the user is measured at home, the identification unit 144 identifies the class using the second identification dictionary 146B since, for example, an eight-dimensional feature vector is input.

[0103] Then, corresponding to the class identified by the identification unit 144, the user state output unit 148 outputs the user state. The user state output unit 148 may output "normal" or "abnormal" as the state of the user. Also, the user state output unit 148 may identify "imminent death", "recovered", etc. after the user is determined to be abnormal. Also, the user state output unit 148 may output the probability variable as it is.

[0104] Accordingly, according to the present embodiment, it is possible to obtain information including "getting out of bed" and "being in bed" based on biological information such as "respiration rate", "heart rate", and "activity level", and the measurement location, and to infer whether the user's death is approaching from these biological information.

[0105] Note that the above-described third embodiment has been described for the case of applying to the first embodiment, but it may also be applied to the second embodiment. In this case, when the measurement location of the user is at home, the control unit 100 may further use the in-bed time as a parameter.

[0106] [4. Fourth Embodiment] Next, the fourth embodiment will be described. The fourth embodiment replaces the functional configuration of the software in FIG. 3 of the first embodiment with FIG. 9. In addition to the functional configuration of the first embodiment, this embodiment further includes a user diary output unit 150.

[0107] The user diary output unit 150 outputs the acquired biological information values and the sleep state (0: getting out of bed, 1: being in bed and awake, 2: sleeping) as image data (image data of "1440 pixels × number of days of pixels") with the value of the pixel value per minute with one row being 24 hours. The user diary output unit 150 can output a respiration diary representing the respiration rate of the user, a heart rate diary representing the heart rate of the user, a sleep diary representing the sleep state of the user, an activity level diary representing the body movement of the user, a respiration disorder diary representing the number of respiration disorder events, a periodic body movement diary representing the number of periodic body movement events, etc. as the user diary. Note that the user diary output unit may output a combination of various parameters as one user diary. Further, the user diary output unit 150 can output graphs of these user diaries as diary data which is image data.

[0108] [5. Fifth Embodiment] Next, a fifth embodiment will be described. The fifth embodiment includes a user state estimation unit 140B that estimates the state of a user using a neural network, instead of the user state estimation unit 140 of the first embodiment and the user state estimation unit 140A of the third embodiment.

[0109] The user state estimation unit 140B estimates the user state from the input log data. Here, as a process for estimating the user state, recently, deep learning (deep neural network) has shown high accuracy particularly in image recognition, and this method is used as an example in this embodiment. The process in this deep learning will be briefly described with reference to FIG. 10.

[0110] First, the user state estimation unit 140B inputs the parameters when the user state estimation unit 140 estimated the user state in the first and second embodiments into a neural network composed of a plurality of layers and neurons included in each layer. Also, the user state estimation unit 140B inputs the signal of the log data (image data) output by the user log output unit 150 output in the fourth embodiment into a neural network composed of a plurality of layers and neurons included in each layer. Each neuron receives signals from a plurality of other neurons, performs calculations, and outputs the calculated signals to a plurality of other neurons. When the neural network has a multi-layer structure, in the order in which the signals flow, it is called an input layer, an intermediate layer (hidden layer), and an output layer.

[0111] A neural network having a plurality of intermediate layers is called a deep neural network (for example, a Convolutional Neural Network (convolutional neural network) having a convolution operation), and a machine learning method using this is called deep learning.

[0112] The input parameters and log data are subjected to various operations (such as convolution operations, pooling operations, normalization operations, matrix operations, etc.) on the neurons of each layer of the neural network, flow while changing their forms, and a plurality of signals are output from the output layer.

[0113] For each of the plurality of output values from the neural network, a process such as associating them with the user's state and inferring the user's state associated with the output value with the largest value is performed. Alternatively, even if the user's state is not directly output, one or more output values may be passed through a classifier, and the user's state may be inferred from the output of the classifier.

[0114] The parameters, which are coefficients used in various operations of the neural network, are determined by inputting a large number of parameters into the neural network in advance, the user's state in the parameters, the log data, and the user's state of the log data, and propagating the error between the output value and the correct value in the reverse direction of the neural network by the error backpropagation method, and updating the parameters of the neurons in each layer many times. In this way, the process of updating and determining the parameters is called learning.

[0115] The structure of the neural network and the individual operations are well-known techniques described in books and papers, and any of these techniques may be used.

[0116] The user state inference unit 140B can output whether the user is approaching death as the user's state from input data such as the user's biometric information and log data.

[0117] Note that, in the case of log data, the user state estimation unit 140B inputs log data with one line representing 24 hours and uses a neural network. However, log data with one line representing 7 days considering the weekly rhythm, log data with one line representing 28 days considering the roughly monthly rhythm, log data with one line representing 365 days considering the yearly rhythm, etc. may also be used. Further, the user state estimation unit 140B may input biometric information values that do not consider the rhythm in advance and use a neural network. That is, information such as "heart rate", "respiratory rate", "activity level", "getting out of bed", "being in bed", "number of respiratory events", and "number of periodic body movements" may be input into the neural network with their respective time axes synchronized and learned to estimate the user state.

[0118] [6. Example] An example of determining the state of the user (that death is approaching) will be described by using the above-described embodiment. FIG. 11 is a diagram showing the relationship between the biometric information, activity level, and sleep of the user, and FIGS. 12 to 15 are diagrams for explaining an example of the user log.

[0119] FIG. 11 shows the state of the user over a predetermined period from the top. From the top of FIG. 11, it is a diagram plotting the heart rate, respiratory rate, and activity level every minute. And the control unit 100 determines whether the user is out of bed or in bed, and it is a graph showing out of bed as 0, in bed (the user is awake) as 1, and in bed (the user is sleeping) as 2.

[0120] And the sleep logs showing the biometric information values, the state of the user being in bed or out of bed in a highly visible state are FIGS. 12 to 15. The log is an example of a user log representing the sleep state of the user every day, and a graph showing the sleep state every day is displayed in the vertical direction.

[0121] All user logs have one line representing 24 hours, with the center of the graph indicating 0:00 am. Also, for example, regarding the respiration rate and heart rate as biometric information values, the normal range is shown in green (light color in a grayscale graph). Here, as the value approaches an abnormal value, it changes to red or blue. Also, the hatched area is a non-measured interval (power off).

[0122] That is, if the value becomes higher than the normal range, it changes from green to yellow to red. Also, if the value becomes lower than the normal range, it changes from green to light blue to blue. In the case of grayscale, it changes from a light color to a dark color.

[0123] Also, regarding the sleep state, it can be displayed with various color-codings. For example, getting out of bed may be represented by white, awake (in bed) by orange, and sleep (in bed) by blue. In each figure, getting out of bed, awake (in bed), and sleep (in bed) are shown with darker colors.

[0124] Also, when showing the sleep state, it is also possible to display the activity level. For example, by superimposing black vertical bars that show the graph of the activity level, it is possible to confirm the sleep state and the activity level together.

[0125] Since biometric information values usually show periodic fluctuations over 24 hours, one week, etc., by using the user log in this way, the long-term fluctuations of the user become easier to see, and there is an advantage that one can quickly notice a physical condition deterioration. With these multiple log data, it becomes possible to accurately estimate the user's state. For example, if there is no change in other biometric information values such as the respiration rate during a time period when the heart rate shows an abnormal value, or if it shows an abnormal value at a certain time period every day, it is possible not to determine it as abnormal, etc.

[0126] Multiple user logs can be used. Usually, the user's abnormality is visually judged manually using the user log, but by using the system of this embodiment, it is possible to automatically estimate and notify the user's abnormality based on certain criteria that do not depend on individual differences or capabilities in judgment, without the labor of visual judgment.

[0127] Figures 12 to 15 are examples of the user log of a user who died during measurement at home, and the time of death was around 9:00 on October 31. Figure 12 is a sleep log showing the timing of getting in and out of bed. Figure 13 is a sleep log showing the combined activity level. Figure 14 is a log showing the user's respiratory rate. Figure 15 is a log showing the user's heart rate.

[0128] The user's heart rate started to rise at around 18:00 on October 29, returned to the normal range at around 6:00 in the morning on the following 30th, and multiple getting-out-of-bed determinations have been observed from the graph of the state thereafter. Then, from around 19:00 on October 30, the user got back into bed again, the activity level decreased, and the user died. However, with the conventional method, the process is reset when the user gets out of bed, whereas by applying this algorithm, it becomes possible to observe the decrease in the activity level without stopping the process.

[0129] [7. Other methods for determining the time of death] The following describes other methods for determining the time of death. The method for determining the time of death is, for example, a method in which the control unit 100 determines the time of death of the user in the second user state estimation process. For example, in the second user state estimation process, the following method for determining the time of death may be used. Also, the following method for determining the time of death uses one week (7 days) as the reference number of days for a predetermined past period, but other reference numbers of days (for example, 10 days, 5 days) may also be used.

[0130] (1) The first method for determining the time of death The control unit 100 calculates the total getting-out-of-bed time for the past 24 hours every day. Then, the control unit 100 calculates the minimum value for the reference number of days (for example, the past one week (7 days)). And the control unit 100 may determine that the time of death is approaching when the getting-out-of-bed time on the current day is lower than the minimum value for the past 7 days.

[0131] In addition, in the first end-of-life determination method, the control unit 100 may calculate the average value instead of the minimum value. In this case, the control unit 100 determines that the end of life is approaching when the out-of-bed time on the current day is less than the average value, or when the average value has been continuously exceeded for a predetermined number of days. Further, the control unit 100 may multiply the minimum value or the average value by a predetermined coefficient.

[0132] (2) Second end-of-life determination method The control unit 100 calculates the maximum out-of-bed time for one out-of-bed per day. Then, the control unit 100 calculates the average of the reference days (for example, the past 7 days) of the maximum out-of-bed time. Then, the control unit 100 determines that the end of life is approaching when the maximum out-of-bed time per day is less than the average value of the past out-of-bed times.

[0133] (3) Third end-of-life determination method The control unit 100 calculates the time zone when the user is always out of bed within the reference days (for example, the past 7 days). Then, when the user is in bed at the time calculated as always being out of bed, the control unit 100 determines that the end of life is approaching. For example, in the sleep log of FIG. 16, the area of the dotted line is the time zone when the user is always out of bed. In this time zone, when the control unit 100 determines that the user is in bed, it may determine that the end of life is approaching.

[0134] Here, the control unit 100 may calculate the out-of-bed time zone for each day of the week in the past month, for example, and use it for the determination that the end of life is approaching. For example, by calculating the out-of-bed time zone for each day of the week, the control unit 100 can accommodate users with different lifestyles depending on the day of the week.

[0135] (4) Fourth end-of-life determination method The control unit 100 quantifies the sleep state of the user and determines that the end of life of the user is approaching. For example, the control unit sets out-of-bed as "0", in-bed (awake) as "1", and in-bed (sleep) as "2", and calculates the average value of these numerical values every minute. Here, paying attention only to the change in the out-of-bed time, without distinguishing between in-bed (awake) and in-bed (sleep), out-of-bed may be set as "0", and in-bed (awake) or in-bed (sleep) as "1".

[0136] The control unit 100 obtains the absolute value of the difference between the state of the day (the digitized sleep state every minute) and the average value of the number of reference days from the day (for example, the past seven days). Then, the control unit 100 uses the average value of the absolute values of one day as the state change coefficient. When the state change coefficient exceeds a threshold value (for example, 0.7 to 0.8), it is determined that death is approaching.

[0137] For example, in FIG. 17, the solid line represents the state of the day, the thick line represents the average state of the past seven days, and the state obtained by subtracting the average of the past seven days from the day is shown by the dotted line. The average value of the absolute values of the data shown by the dotted line corresponds to the state change coefficient. FIG. 17 shows the state on May 1 in FIG. 16.

[0138] Also, FIG. 18 is a graph in which the state change coefficient is calculated and plotted for the daily data of the sleep diary in FIG. 16. In this embodiment, since the state data of the past seven days is used, the state change coefficient is calculated from April 28. In the graph of the state change coefficient, the state change coefficient starts to increase from April 30 when the time in bed starts to become long. The state change coefficient on April 30 is 0.88, and the state change coefficient on May 1 is 1.27, exceeding the threshold value. Therefore, in the case of the user in FIG. 16, the control unit 100 can determine that death is approaching after April 30.

[0139] [8. Modification Example] The present disclosure is not limited to the above-described embodiments, and various modifications are possible. That is, embodiments obtained by appropriately combining technical means modified within the scope not departing from the gist of the present disclosure are also included in the technical scope.

[0140] Also, the above-described embodiments are described separately for convenience of explanation, but they can be executed in combination within the possible range. Also, any technology described in the specification has the intention of obtaining rights in corrections or divisional applications, etc.

[0141] In each embodiment, the program operating in each device is a program (a program that functions a computer) that controls a CPU or the like so as to realize the functions of the above-described embodiments. And the information handled by these devices is temporarily stored in a temporary storage device (e.g., RAM) during its processing, and then stored in a storage device such as various ROMs and HDDs, and read by the CPU as needed for correction and writing.

[0142] Here, as the recording medium for storing the program, any of a semiconductor medium (e.g., ROM, non-volatile memory card, etc.), an optical recording medium or magneto-optical recording medium (e.g., DVD (Digital Versatile Disc), CD (Compact Disc), BD (Blu-ray (registered trademark) Disc), etc.), a magnetic recording medium (e.g., magnetic tape, flexible disk, etc.) may be used.

[0143] Also, when distributing to the market, the program can be stored in a portable recording medium for distribution, or transferred to a server computer connected via a network such as the Internet. In this case, the storage device of the server device is of course also included in the present disclosure.

[0144] Also, the above-described data may not be stored in the device, but stored in an external device and appropriately called. For example, the data may be stored in a NAS (Network Attached Storage) or stored on the cloud.

[0145] Note that the scope of the present disclosure is not limited to the configurations explicitly described in the specification, and combinations of the technologies disclosed in this specification are also included in the scope. Among the present disclosure, the configuration for which a patent is sought is described in the appended claims, but it is not intended to exclude it from the technical scope for the reason that it is not described in the claims.

[0146] In addition, in the above-mentioned specification, the descriptions such as "in the case of ~" and "when ~" are illustrative examples and do not limit the configuration to the described content. For configurations other than these cases and times, the disclosure also includes what is obvious to those skilled in the art and has the intention of obtaining rights.

[0147] Also, for the descriptions of the processes and data flows described in the specification that involve an order, they are not limited to the described order. For example, configurations in which part of the process is deleted or the order is changed are also disclosed and have the intention of obtaining rights.

[0148] In addition, although the functions described in the embodiments are described as being executed by respective devices, they may be realized by one device or by further using an external server.

[0149] Each functional block or various features of the devices used in the above-described embodiments can be implemented or executed by an electric circuit, for example, an integrated circuit or a plurality of integrated circuits. The electric circuit designed to execute the functions described in this specification may include a general-purpose use processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic, discrete hardware components, or a combination thereof. The general-purpose use processor may be a microprocessor or a conventional processor, controller, microcontroller, or state machine. The above-mentioned electric circuit may be composed of a digital circuit or an analog circuit. Also, when an integrated circuit technology that replaces the current integrated circuit appears due to the progress of semiconductor technology, one or more aspects of the present disclosure can also use the new integrated circuit by such technology.

[0150] In addition, in this embodiment, although the processing device 5 outputs biological information based on the result output by the detection device 3, all of the calculations may be performed by the detection device 3. Further, it may be realized not only by installing an application in a terminal device (for example, a smartphone, a tablet, a computer), but also by performing processing on the server side and returning the processing result to the terminal device.

[0151] For example, by uploading biological information from the detection device 3 to the server, the above-described processing may be realized on the server side. This detection device 3 may be realized by a device such as a smartphone incorporating an acceleration sensor and a vibration sensor.

Explanation of Reference Numerals

[0152] 1 Abnormality determination system 3 Detection device 5 Processing device 100 Control unit 200 Storage 300 ROM 310 RAM 400 Biological signal acquisition unit 500 Measurement location acquisition unit 600 Input unit 700 Output unit 800 Notification unit 900 Communication unit 10 Bed 20 Mattress

Claims

1. An abnormality determination device comprising an acquisition unit that acquires a measurement location, a biological signal acquisition unit that acquires a user's biological signal, and a control unit, wherein the control unit calculates a biological information value of the user based on the biological signal acquired from the biological signal acquisition unit, calculates the activity amount of the user from the biological signal, when the measurement location acquired by the acquisition unit is the user's home, determines that the user is in a state where death is approaching based on the biological information value and the activity amount, when the measurement location acquired by the acquisition unit is other than the user's home, determines that the user is in a state where death is approaching based on the biological information value, the activity amount, and the presence / absence of bed information based on the user's getting out of bed, An abnormality determination device characterized by the above.

2. The control unit when the measurement location is the user's home, determines that the user's abnormality is continuing based on the biological information value and the activity amount, and when the state where the biological information value becomes an abnormal value continues for a first determination time and then the state where the biological information value is within the normal range continues for a second determination time, and the activity amount is low at the second determination time, determines that the user is in a state where death is approaching, when the measurement location is other than the user's home, determines that the user's abnormality is continuing based on the biological information value and the activity amount, and when the state where the biological information value becomes an abnormal value continues for the first determination time and then the state where the biological information value is within the normal range continues for the second determination time, and the activity amount is low and the user is not out of bed at the second determination time, determines that the user is in a state where death is approaching The abnormality determination device according to Claim 1.

3. When the measurement location acquired by the control unit from the acquisition unit is other than the user's home and it is determined that the user has gotten out of bed as the presence / absence of bed information based on the getting out of bed, the control unit determines that the user is not in a state where death is approaching The abnormality determination device according to Claim 1.

4. When the measurement location is other than the user's home and it is determined that the user has a duration of immobility for a predetermined time, the control unit determines that the user's activity amount is low The abnormality determination device according to Claim 2.

5. An abnormality determination device comprising an acquisition unit that acquires a measurement location, a biological signal acquisition unit that acquires a biological signal of a user, and a control unit, wherein the control unit calculates a biological information value of the user based on the biological signal acquired from the biological signal acquisition unit, calculates an activity amount of the user from the biological signal, acquires the in-bed time or out-of-bed time of the user, when the measurement location acquired by the acquisition unit is the user's home, determines that the user is in a state where death is approaching based on the biological information value and the in-bed time or out-of-bed time of the user, when the measurement location acquired by the acquisition unit is other than the user's home, determines that the user is in a state where death is approaching based on the biological information value, the activity amount, and the in-bed / out-of-bed information based on the user's getting out of bed, characterized by the above-mentioned abnormality determination device. **Claim 6** When the measurement location acquired by the acquisition unit is the user's home, the control unit determines that death is approaching when the in-bed time of the previous day exceeds a predetermined threshold or when the in-bed time continuously exceeds the predetermined threshold. The abnormality determination device according to claim 5.

Images