Monitoring device, monitoring method, and program

Abstract

The purpose of the present invention is to make it possible to appropriately provide notification of abnormalities in a person being measured, taking into account temporary physical states of the person being measured, such as alcohol drinking and exercise.　A monitoring device (10) comprises: a biological information acquisition unit (11) that acquires biological information of a person being measured; an abnormality determination unit (16) that determines whether or not the person being measured is in an abnormal state on the basis of the biological information; a notification unit (17) that, if it is determined that the person being measured is in an abnormal state, provides notification of the determination; a state determination unit (13) that determines whether or not the person being measured is in a prescribed temporary physical state; and a notification limiting unit (15) that limits the notification by the notification unit (17) if it is determined that the person being measured is in the temporary physical state.

Description

Monitoring device, monitoring method and program

[0001] The present invention relates to a monitoring device, a monitoring method, and a program, and more particularly to a technique for determining physical abnormalities of a subject based on the subject's biological information.

[0002] Systems are known that detect biological information of a subject, such as heart rate and respiratory rate, while the subject is asleep. For example, Patent Document 1 listed below discloses an abnormality evaluation device that acquires the respiratory rate of the subject while sleeping and compares the respiratory rate with a reference respiratory rate to determine whether the subject is in an abnormal physical state. If this device determines that the subject is in an abnormal state, it notifies the subject and a caregiver of this fact.

[0003] Patent No. 6193650

[0004] After drinking alcohol or exercising, the heart rate and respiratory rate of the person being measured change significantly, albeit temporarily. For this reason, with the above-mentioned conventional technology, the person being measured after drinking alcohol or exercising may be judged to be in an abnormal state, and an alert to that effect may be issued.

[0005] The present invention has been made in consideration of the above-mentioned problems, and its purpose is to provide a monitoring device, monitoring method, and program that can appropriately notify of abnormalities in a person being measured, taking into account the person's temporary physical condition, such as drinking or exercise.

[0006] (1) In order to solve the above problem, the monitoring device of the present invention includes a biometric information acquisition means for acquiring biometric information of the person being measured, an abnormality determination means for determining whether the person being measured is in an abnormal state based on the biometric information, a notification means for notifying the person being measured of the abnormal state if it is determined that the person being measured is in an abnormal state, a state determination means for determining whether the person being measured is in a predetermined temporary physical state, and a notification restriction means for restricting the notification by the notification means if it is determined that the person being measured is in the temporary physical state.

[0007] (2) In the monitoring device described in (1) above, the notification limiting means may change the judgment criteria used by the judgment means when it is determined that the person being measured is in the temporary physical condition.

[0008] (3) The monitoring device according to (2) above may further include an influence level determination unit that determines an influence level of the temporary physical condition of the person being measured. The notification limiting unit may change the determination criterion in the determination unit according to the influence level.

[0009] (4) In the monitoring device described in any of (1) to (3) above, the notification restriction means may be configured to prevent the notification means from making a notification when it is determined that the person being measured is in the temporary physical state until it is determined that the person being measured is no longer in the temporary physical state.

[0010] (5) In the monitoring device described in any one of (1) to (4) above, the condition determination means may determine whether the person being measured is in the temporary physical condition based on the biological information.

[0011] (6) In the monitoring device described in (5) above, the state determination means may include a machine learning model trained using the biometric information of a person in the temporary physical state.

[0012] (7) In the monitoring device described in any one of (1) to (6) above, the temporary physical state may be a state after drinking alcohol or a state after exercise.

[0013] (8) Furthermore, the monitoring method of the present invention includes a biometric information acquisition step for acquiring biometric information of the person being measured, an abnormality determination step for determining whether the person being measured is in an abnormal state based on the biometric information, a notification step for notifying the person being measured of the abnormal state if it is determined that the person being measured is in an abnormal state, a condition determination step for determining whether the person being measured is in a predetermined temporary physical state, and a notification restriction step for restricting the notification in the notification step if it is determined that the person being measured is in the temporary physical state.

[0014] (9) A program according to the present invention causes a computer to function as a biological information acquisition means for calculating biological information of a subject, an abnormality determination means for determining whether the subject is in an abnormal state based on the biological information, a notification means for notifying the subject when it is determined that the subject is in an abnormal state, a condition determination means for determining whether the subject is in a predetermined temporary physical state, and a notification limiting means for limiting the notification by the notification means when it is determined that the subject is in the temporary physical state. This program may be stored in a computer-readable information storage medium such as a semiconductor memory or a magneto-optical disk.

[0015] According to the present invention, abnormalities in a subject can be appropriately reported based on the subject's temporary physical condition, such as drinking alcohol or exercise.

[0016] It is an overall configuration diagram of a watching system according to an embodiment of the present invention. It is a functional block diagram of a watching device according to an embodiment of the present invention. It is a flow diagram showing an operation example of the watching device. It is a flow diagram showing a modified operation example of the watching device.

[0017] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

[0018] FIG. 1 is a diagram illustrating the overall configuration of a monitoring system according to an embodiment of the present invention. The monitoring system 1 shown in the figure is centered around a bed 40 installed in a home. A speaker microphone 43 and a Doppler sensor 45 are attached to the headboard of the bed 40. The Doppler sensor 45 irradiates microwaves toward the chest of a person sleeping in the bed 40 and receives the reflected waves. The reflected waves generate a Doppler signal indicating chest movement associated with heartbeat and breathing, and this Doppler signal is digitized and output as Doppler data. The speaker microphone 43 is also installed facing the person sleeping in the bed 40. A weighing scale 41 is attached to the floorboard or the like of the bed 40 to measure the weight of the person, bedding, etc. The weighing scale 41, speaker microphone 43, and Doppler sensor 45 are connected to a monitoring device 10 installed in the same home.

[0019] The monitoring device 10 acquires biometric information (heart rate and pulse rate in this case) of a person being measured (not shown) sleeping in bed 40 based on Doppler data measured by the Doppler sensor 45, and detects abnormalities in the person being measured based on this biometric information. If an abnormality is detected, a voice message such as "An abnormality has been detected. Are you okay?" is output from the speaker of the speaker-microphone 43. If the person being measured does not respond to this message with a message such as "I'm okay," the monitoring device 10 transmits a message of abnormality to the monitoring server 20 connected via a communication network 30 such as the Internet. The monitoring server 20 is a computer installed in a remote monitoring center. When the monitoring server 20 receives the message of abnormality, a staff member at the monitoring center again calls the person being measured through the speaker of the speaker-microphone 43 and, if necessary, requests the dispatch of a medical professional such as a doctor or an ambulance to the residence.

[0020] FIG. 2 is a functional block diagram of a monitoring device 10 according to an embodiment of the present invention. As shown in the figure, the monitoring device 10 functionally includes a biological information acquisition unit 11, an admission determination unit 12, a status determination unit 13, an impact level determination unit 14, a notification restriction unit 15, an abnormality determination unit 16, and a notification unit 17. The monitoring device 10 includes a general-purpose computer including a CPU and memory, and the functions shown in FIG. 2 are realized by executing a program according to an embodiment of the present invention on this computer. The program may be supplied to the computer from a computer-readable information storage medium such as a semiconductor memory, or may be supplied to the computer by being downloaded from another computer via a communication network 30 such as the Internet.

[0021] The biological information acquisition unit 11 acquires biological information of the subject based on Doppler data detected by the Doppler sensor 45. Here, the acquired biological information is the heart rate and respiratory rate. For example, peaks corresponding to the heart rate and respiratory rate are identified by Fourier analysis of the Doppler data, and the heart rate and respiratory rate are acquired from the positions (frequencies) of these peaks.

[0022] The bed entry determination unit 12 determines whether the person being measured is sleeping in bed 40 based on the weight detected by the weighing scale 41. For example, the weight of the person being measured is stored in advance, and the bed entry determination unit 12 determines that the person being measured has entered bed when the weight detected by the weighing scale 41 increases by the weight of the person being measured. The bed entry determination unit 12 also determines that the person being measured has left bed when the weight detected by the weighing scale 41 decreases by the weight stored in advance.

[0023] The state determination unit 13 determines whether the subject is in a predetermined temporary physical state. Here, the "predetermined temporary physical state" refers to a state after drinking alcohol and a state after exercise. Based on the Doppler data detected by the Doppler sensor 45, it is determined whether the subject is in a post-drinking state (a state in which the effects of drinking alcohol remain), a post-exercise state (a state in which the effects of exercise remain), a normal state, or another state. As an example, this determination can be made using a machine learning model. Specifically, Doppler data or its feature values ​​for a certain period (e.g., five minutes) of a person in a post-drinking state are labeled with a label indicating the post-drinking state to create training data. Furthermore, Doppler data or its feature values ​​for a certain period of a person in a post-exercise state are labeled with a label indicating the post-exercise state to create training data. Furthermore, Doppler data or its feature values ​​for a certain period of a person in a normal state (a state that is neither a post-drinking state nor a post-exercise state) are labeled with a label indicating the normal state to create training data. Then, using this training data, a machine learning model that classifies physical states from the Doppler data or its feature values ​​is trained. When a person is in a post-alcohol state, the heart rate and breathing rate tend to increase and the intensity also tends to be high. Furthermore, there tends to be little change in the heart rate and breathing rate even after falling asleep. Furthermore, the ratio of inhalation to exhalation tends to change in breathing. On the other hand, when a person is in a post-exercise state, the heart rate and breathing rate tend to increase and the intensity also tends to be high. Furthermore, after falling asleep, the heart rate and breathing rate tend to gradually decrease. Regarding breathing, the ratio of inhalation to exhalation tends to fall within a certain range. By having a machine learning model learn these characteristics, it is possible to determine whether the person being measured is in a post-alcohol state, a post-exercise state, a normal state, or some other state.

[0024] Although the temporary physical state of the subject is determined using a machine learning model here, other methods may also be used. For example, a sensor may be installed on a dining table or refrigerator, and the detection results may be used to determine whether the subject is in a state after drinking alcohol. A breath sensor may also be installed, and the alcohol concentration in the breath may be used to determine whether the subject is in a state after drinking alcohol. A video camera may be installed near the dining table in the home, and the captured image may be used to determine whether the subject has drunk alcohol. A video camera may also be installed in the living room, and the captured image may be used to determine whether the subject has exercised at home. The subject may also input whether they have drunk alcohol or exercised into the monitoring device 10.

[0025] The influence level determination unit 14 determines the influence level (the extent of the influence of alcohol) of the subject when the subject is in a post-drinking state. Also, the influence level (the extent of the influence of exercise) of the subject when the subject is in a post-exercise state. For example, the greater the influence of alcohol remains after drinking, the higher the subject's heart rate and respiratory rate. Therefore, a range of heart rate and / or respiratory rate is set for each influence level (for example, three levels), and when the state determination unit 13 determines that the subject is in a post-drinking state, the influence level determination unit 14 checks to which influence level range the subject's heart rate and respiratory rate acquired by the biological information acquisition unit 11 belong, and thereby determines the influence level of alcohol.

[0026] Furthermore, the greater the effects of exercise remaining after exercise, the higher the heart rate and respiratory rate of the person being measured. Therefore, a range of heart rate and / or respiratory rate may be set in advance for each influence level (for example, in three levels), and when the state determination unit 13 determines that the person being measured is in a post-exercise state, the influence level determination unit 14 may check to which influence level range the heart rate and respiratory rate of the person being measured acquired by the biological information acquisition unit 11 belong, and thereby determine the influence level of exercise.

[0027] The abnormality determination unit 16 determines whether the subject is in an abnormal state based on the subject's heart rate and respiratory rate acquired by the biological information acquisition unit 11. As an example, multiple numerical ranges for the heart rate are provided, and a risk value is assigned to each of the ranges. The risk value for the heart rate is determined by checking which range the heart rate falls into. Similarly, multiple numerical ranges for the respiratory rate are provided, and a risk value is assigned to each of the ranges. The risk value for the respiratory rate is determined by checking which range the respiratory rate falls into. The abnormality determination unit 16 then calculates a total risk value by adding the risk value for the heart rate and the risk value for the respiratory rate. If this total risk value is equal to or greater than a given threshold, the subject is determined to be in an abnormal state. The given thresholds include a normal threshold used when the subject is in a normal state and a post-drinking threshold used when the subject is in a post-drinking state. As described below, multiple post-drinking thresholds may be provided depending on the degree to which the effects of drinking remain (impact level). Furthermore, whether the subject is in an abnormal state may also be determined when the subject is in a post-exercise state, in which case a post-exercise threshold may also be prepared. Multiple post-exercise thresholds may also be prepared depending on the degree to which the effects of exercise remain (influence level).

[0028] When it is determined that the person being measured is in an abnormal state, the notification unit 17 notifies the person to that effect. Specifically, it outputs a voice message from the speaker unit of the speaker microphone 43. Furthermore, if there is no response from the person being measured to the speaker microphone 43, it transmits a message to the monitoring server 20 via the communication network 30 indicating that the person being measured is in an abnormal state.

[0029] The notification limiting unit 15 limits notification by the notification unit 17 when the subject is determined to be in a post-drinking state or a post-exercise state. For example, when the subject is in a post-drinking state, the abnormality determination unit 16 changes its abnormality determination criteria. Specifically, the threshold value compared with the total risk value is changed to a post-drinking threshold value that is increased by a predetermined value from the normal threshold. This reduces the chance of the total risk value exceeding the threshold and being determined to be an abnormal state. In this way, notification by the notification unit 17 can be suppressed. In this case, the degree to which the threshold is increased may be changed depending on the level of influence of alcohol. Specifically, the higher the level of influence of alcohol, the larger the post-drinking threshold value that can be used. In this way, the determination criteria can be set appropriately depending on the level of influence of alcohol. The notification limiting unit 15 may also suspend abnormality determination by the abnormality determination unit 16 or suspension of notification by the notification unit 17 until the subject is determined not to be in a post-drinking state.

[0030] Similarly, when the subject is in a post-exercise state, the abnormality determination unit 16 may change its abnormality determination criteria. Specifically, the threshold value compared with the overall risk value is changed to a post-exercise threshold value that is increased by a predetermined value from the normal threshold value. In this case, the degree to which the threshold value is increased may be changed depending on the level of influence of exercise. Specifically, the higher the level of influence of exercise, the larger the post-exercise threshold value that may be used. Alternatively, the notification limiting unit 15 may stop the abnormality determination unit 16 from making an abnormality determination or the notification unit 17 from issuing a notification until it is determined that the subject is not in a post-exercise state.

[0031] 3 is a flow diagram showing an example of the operation of the monitoring device 10. As shown in the figure, the bed entry determination unit 12 of the monitoring device 10 first monitors whether the subject has entered bed 40 (S101). If the subject has entered bed, the biological information acquisition unit 11 then acquires Doppler data transmitted from the Doppler sensor 45 (S102). The monitoring device 10 repeats the processes of S101 and S102 until one minute has elapsed (S103), and then determines whether five minutes or more have elapsed since the subject entered bed (S104). If five minutes or more have not elapsed, the abnormality determination unit 16 sets the normal threshold as the determination criterion (S109).

[0032] If it is determined in S104 that five minutes or more have passed since the subject went to bed, the state determination unit 13 determines the subject's state, and the influence level determination unit 14 determines the influence level of alcohol, etc. (S105). Specifically, the state determination unit 13 inputs Doppler data from the most recent five minutes into a machine learning model to determine whether the subject is in a state after drinking alcohol or exercise. If the state determination unit 13 determines that the subject is in a normal state (S106), the abnormality determination unit 16 sets a normal threshold as the determination criterion (S109). If the state determination unit 13 determines that the subject is in a post-drinking state (S107), the abnormality determination unit 16 sets a post-drinking threshold corresponding to the influence level of alcohol as the determination criterion (S109). If the state determination unit 13 determines that the subject is in a post-exercise state (S108), the notification unit 17 outputs a voice from the speaker unit of the speaker-microphone 43 indicating that the abnormality determination is temporarily suspended (suspended) (S112), and the process returns to S101. Furthermore, if it is determined that the person being measured is not in a post-exercise state (S108), the alarm unit 17 outputs a voice from the speaker unit of the speaker microphone 43 indicating that the abnormality determination will be stopped (S111), and sends a message to that effect to the monitoring server 20.

[0033] Thereafter, the abnormality determination unit 16 calculates a total risk value (S113) based on the heart rate and respiratory rate acquired by the biological information acquisition unit 11. Then, the abnormality determination unit 16 performs an abnormality determination by comparing the total risk value with the threshold set in S109 or S110 (S114).

[0034] When the abnormality determination unit 16 determines that an abnormal state has occurred in the most recent five consecutive abnormality determinations (S115), the notification unit 17 issues a call through the speaker unit of the speaker microphone 43 (S116). If the microphone unit of the speaker microphone 43 picks up a sound of the subject's response to this call (S117), the process returns to S101. If the microphone unit of the speaker microphone 43 does not pick up a sound of the subject's response (S117), the notification unit 17 transmits a message to the monitoring server 20 indicating that an abnormality has occurred in the subject (S118), and the process returns to S101.

[0035] Also, if the abnormality determination unit 16 determines in S115 that the last five consecutive abnormality determinations have not resulted in an abnormal state, the process returns to S101. After one minute has elapsed, the state determination unit 13 again determines the state of the person being measured based on the Doppler data from the last five minutes, and the influence level determination unit 14 determines the influence level of drinking, etc. (S105). If the monitoring device 10 determines that the person being measured has moved from the post-drinking state (S107) to a normal state (S106), it sets a normal threshold as the determination criterion (S109) and continues the subsequent processing.

[0036] According to the monitoring system 1 described above, the heart rate and respiratory rate of the person being measured after going to bed or while asleep can be obtained based on the Doppler data acquired by the Doppler sensor 45, and a total risk value can be calculated from these values. The total risk value is compared with a given threshold, and if it is equal to or greater than the threshold, it is determined that an abnormality has occurred. The notification unit 17 notifies the person being measured and the staff at the monitoring center of this. In this embodiment, the Doppler data is used to determine whether the person being measured is in a temporary physical state due to drinking or exercise, and the notification by the notification unit 17 is limited accordingly. This makes it possible to prevent excessive notifications from being made to the person being measured and the staff at the monitoring center.

[0037] The present invention is not limited to the above-described embodiment, and various modifications are possible, and such modifications also fall within the technical scope of the present invention.

[0038] For example, in the operation example of FIG. 3 , the state of drinking, etc., is not assessed until at least five minutes have passed since going to bed. However, the state of drinking, etc., may be assessed one minute after going to bed. FIG. 4 is a flow chart showing the operation of the monitoring device in this case. As shown in the figure, one minute after going to bed, the state assessment unit 13 assesses the state of the subject, and the influence level assessment unit 14 assesses the influence level of drinking, etc. (S105a). At this time, if less than five minutes have passed since going to bed, the state assessment unit 13 assesses the state of the subject based on the Doppler data for the previous minute, for example, using a first machine learning model. Furthermore, if five minutes or more have passed since going to bed, the state of the subject is assessed based on the Doppler data for the previous five minutes, for example, using a second machine learning model. When creating the first machine learning model, labels indicating the state after drinking, etc. are assigned to one minute of Doppler data or its features of a person in each of the following states: after drinking, after exercise, and normal, to create learning data. Then, a first machine learning model that classifies physical conditions from one minute of Doppler data or its feature values ​​is trained using the training data thus created. Similarly, when creating a second machine learning model, five minutes of Doppler data or its feature values ​​of a person in each of the following states are assigned a label indicating the state, such as "after drinking," to create training data. Then, a second machine learning model that classifies physical conditions from five minutes of Doppler data or its feature values ​​is trained using the training data thus created. According to this modification, an abnormality can be appropriately determined based on the subject's temporary physical condition one minute after going to bed.

[0039] 1 Monitoring system, 10 Monitoring device, 11 Biometric information acquisition unit, 12 Bed admission determination unit, 13 Status determination unit, 14 Impact level determination unit, 15 Notification restriction unit, 16 Abnormality determination unit, 17 Notification unit, 20 Monitoring center server, 30 Communication network, 40 Bed, 41 Weight scale, 43 Speaker microphone, 45 Doppler sensor.

Claims

1. A monitoring device comprising: a biometric information acquisition means for acquiring biometric information of a person being measured; an abnormality determination means for determining whether or not the person being measured is in an abnormal state based on the biometric information; a notification means for notifying the person being measured that he or she is in an abnormal state when it is determined that he or she is in an abnormal state; a state determination means for determining whether or not the person being measured is in a predetermined temporary physical state; and a notification restriction means for restricting the notification by the notification means when it is determined that the person being measured is in the temporary physical state.

2. A monitoring device as claimed in claim 1, characterized in that the notification limiting means changes the judgment criteria in the judging means when the person being measured is judged to be in the temporary physical condition.

3. A monitoring device as described in claim 2, further comprising an impact level determination means for determining an impact level of the temporary physical condition of the person being measured, and wherein the notification limiting means changes the determination criteria in the determination means according to the impact level.

4. A monitoring device as described in claim 1, characterized in that the notification restriction means, when it is determined that the person being measured is in the temporary physical state, does not issue a notification by the notification means until it is determined that the person being measured is not in the temporary physical state.

5. A monitoring device according to claim 1, wherein the condition determining means determines whether or not the person being measured is in the temporary physical condition based on the biological information.

6. A monitoring device as described in claim 5, characterized in that the state determination means includes a machine learning model trained using the biometric information of a human being in the temporary physical state.

7. A monitoring device according to claim 6, characterized in that the temporary physical state is a state after drinking alcohol or a state after exercise.

8. A monitoring method comprising: a bio-information acquisition step for acquiring bio-information of a person being measured; an abnormality judgment step for judging whether or not the person being measured is in an abnormal state based on the bio-information; a notification step for notifying the person being measured of the abnormal state if it is judged that the person being measured is in an abnormal state; a condition judgment step for judging whether or not the person being measured is in a predetermined temporary physical state; and a notification restriction step for restricting the notification in the notification step if it is judged that the person being measured is in the temporary physical state.

9. A program for causing a computer to function as: a biometric information acquisition means for calculating the biometric information of the person being measured; an abnormality judgment means for judging whether the person being measured is in an abnormal state based on the biometric information; a notification means for notifying the person being measured that he or she is in an abnormal state when it is judged that the person being measured is in an abnormal state; a condition judgment means for judging whether the person being measured is in a predetermined temporary physical state; and a notification restriction means for restricting the notification by the notification means when it is judged that the person being measured is in the temporary physical state.

Images