Presence / absence determination system, presence / absence determination method, and program

Abstract

Provided are a presence / absence determination system, a presence / absence determination method, and a program, each of which is capable of accurately determining whether or not a subject is on a bed. When presence / absence information indicates that a subject is not on the bed and the amount of increase in the value indicated by total weight data satisfies a given presence determination condition, a presence / absence change unit (32) changes the retained presence / absence information to reflect that the subject is on the bed. When the presence / absence information indicates that the subject is on the bed and the amount of decrease in the value indicated by the total weight data satisfies a given absence determination condition, the presence / absence change unit (32) changes the retained presence / absence information to reflect that the subject is not on the bed.

Description

Presence / Absence Determination System, Method, and Program

[0001] The present invention relates to a presence / absence determination system, a presence / absence determination method, and a program.

[0002] Various systems for measuring biological information such as the respiratory rate or heart rate of a subject on a bed have been studied. As an example of such a system, Patent Document 1 describes a monitoring device that non-contact determines the respiratory rate and heart rate of a care recipient on a bed using a microwave Doppler sensor.

[0003] Japanese Patent Application Laid-Open No. 2017-134795

[0004] In the technique described in Patent Document 1, when the respiratory rate or heart rate of the subject shows an abnormal value, it cannot be determined whether the cause is that an abnormality has occurred in the subject or that the subject is not on the bed. Therefore, in order to accurately detect that an abnormality has occurred in the subject, it is necessary to accurately determine whether the subject is on the bed or not.

[0005] The present invention has been made in view of the above problems, and an object thereof is to provide a presence / absence determination system, a presence / absence determination method, and a program that can accurately determine whether a subject is on a bed or not.

[0006] (1) The presence / absence determination system according to the present invention includes: Total weight data acquisition means for repeatedly acquiring total weight data indicating the total weight calculated based on the measurement result of the load applied to at least one load sensor located below the position of the person to be measured when the person to be measured is on the bed; presence / absence information holding means for holding presence / absence information indicating whether or not the person to be measured is on the bed; presence / absence information indicating that the person to be measured is not on the bed, and the magnitude of the increase, which is the value obtained by subtracting the value obtained by the total weight data based on the previous measurement result from the value obtained by the total weight data based on the latest measurement result, satisfies a given presence determination condition, and presence / absence information indicating that the person to be measured is on the bed, and the magnitude of the decrease, which is the value obtained by subtracting the value obtained by the total weight data based on the latest measurement result from the value obtained by the total weight data based on the previous measurement result, satisfies a given absence determination condition, and presence / absence information indicating that the person to be measured is on the bed, and absence information indicating that the person to be measured is on the bed, and absence information indicating that the person to be measured is not on the bed

[0007] (2) The presence / absence determination system described in (1) further includes: Doppler data acquisition means for acquiring Doppler data indicating the measurement result by a Doppler sensor provided facing the person to be measured; biometric information generation means for generating biometric information of the person to be measured based on the Doppler data; action execution means for executing a given action in response to the biometric information satisfying a given abnormality determination condition; and suppression means for suppressing the execution of the action in response to the biometric information satisfying the abnormality determination condition when the presence / absence information indicates that the person to be measured is not on the bed.

[0008] (3) The presence / absence determination system described in (2) further includes a body movement determination means that determines whether or not the person being measured is moving based on the magnitude of the fluctuation of the value indicated by the total weight data, wherein the suppression means suppresses the execution of the action corresponding to the fact that the biological information has met the abnormality determination condition when the presence / absence information indicates that the person being measured is on the bed and it is determined that the person being measured is moving.

[0009] (4) The presence / absence determination system according to any one of (1) to (3) further includes a weight estimation means that determines the estimated weight of the person to be measured as a value obtained by subtracting the value of the total weight data when the presence / absence information indicates that the person to be measured is not on the bed from the value of the total weight data when the presence / absence information indicates that the person to be measured is on the bed.

[0010] (5) In the presence / absence determination system described in (4), the presence determination condition is that the magnitude of the increase is greater than a threshold determined based on the estimated weight of the person being measured.

[0011] (6) In the presence / absence determination system described in (4) or (5), the absence determination condition is that the magnitude of the decrease is greater than a threshold determined based on the estimated weight of the person being measured.

[0012] (7) In the presence / absence determination system described in any of (4) to (6), the weight estimation means updates the estimated weight of the person being measured to the increase amount, which is the value obtained by subtracting the value obtained by the total weight data based on the previous measurement result from the value obtained by the total weight data based on the most recent measurement result.

[0013] (8) The presence / absence determination system described in (1), wherein the presence / absence information indicates whether the person to be measured is on the bed, not on the bed, or whether it is unknown whether the person is on the bed, and further includes a person weight data storage means for storing person weight data indicating the weight of the person to be measured, and a total weight data storage means for storing total weight data when the person to be measured is not on the bed, wherein the presence / absence information indicates that the person to be measured is not on the bed, the magnitude of the increase satisfies the presence determination condition, and the sum of the value of the person weight data and the value of the total weight data when the person to be measured If the difference between the value indicated by the total weight data based on the latest measurement results and the value indicated by the total weight data is within a given range, the presence / absence information held is changed to indicate that the person being measured is on the bed. The presence / absence information holds that the presence / absence information indicates that the person being measured is not on the bed, the magnitude of the increase satisfies the presence / absence determination condition, and the difference between the sum of the value of the person's weight data and the value of the total weight data when absent and the value indicated by the total weight data based on the latest measurement results is not within the range, the presence / absence information held is changed to indicate that it is unknown whether the person being measured is on the bed or not.

[0014] (9) The presence / absence determination system described in (8) further includes an updating means that, when the presence / absence information held is changed to indicate that the person to be measured is on the bed, updates the stored person weight data to indicate the increase and updates the stored total weight data when absent to indicate the value indicated by the total weight data based on the most recent measurement result.

[0015] The presence / absence determination system described in (10), (8), or (9) further includes: Doppler data acquisition means for acquiring Doppler data indicating the measurement result by a Doppler sensor provided toward the person to be measured; biometric information generation means for generating biometric information of the person to be measured based on the Doppler data; action execution means for executing a given action in response to the biometric information satisfying a given abnormality determination condition; and suppression means for suppressing the execution of the action in response to the biometric information satisfying the abnormality determination condition when the presence / absence information indicates that the person to be measured is not on the bed, wherein the action execution means executes a first action when the presence / absence information indicates that the person to be measured is on the bed, and executes a second action when the presence / absence information indicates that it is unclear whether the person to be measured is on the bed or not.

[0016] (11) In the presence / absence determination system described in any of (1) to (10), the load sensor is provided on each of the four legs of the bed.

[0017] (12) In the presence / absence determination system described in any of (1) to (11), the total weight data acquisition means repeatedly acquires the total weight data which represents the sum of the measured values ​​of the load applied to each of the plurality of load sensors.

[0018] (13) The present or absence determination method according to the present invention includes the steps of repeatedly acquiring total weight data indicating the total weight calculated based on the measurement result of the load applied to at least one load sensor located below the position of the person to be measured when the person to be measured is on the bed; storing presence or absence information indicating whether or not the person to be measured is on the bed; changing the stored presence or absence information to indicate that the person to be measured is on the bed when the presence or absence information indicates that the person to be measured is not on the bed, and the magnitude of the increase, which is the value obtained by subtracting the value obtained by the total weight data based on the previous measurement result from the value obtained by the total weight data based on the most recent measurement result, satisfies a given presence determination condition; and changing the stored presence or absence information to indicate that the person to be measured is not on the bed when the presence or absence information indicates that the person to be measured is on the bed, and the magnitude of the decrease, which is the value obtained by subtracting the value obtained by the total weight data based on the latest measurement result from the value obtained by the total weight data based on the most recent measurement result, satisfies a given absence determination condition.

[0019] (14) The present invention is a program that causes a computer to perform the following steps: repeatedly acquire total weight data indicating the total weight calculated based on the measurement result of the load applied to at least one load sensor located below the position of the person to be measured when the person to be measured is on a bed; maintain presence / absence information indicating whether or not the person to be measured is on the bed; change the maintained presence / absence information to indicate that the person to be measured is on the bed if the presence / absence information indicates that the person to be measured is not on the bed, and the magnitude of the increase, which is the value obtained by subtracting the value obtained by the total weight data based on the previous measurement result from the value obtained by the total weight data based on the most recent measurement result, satisfies a given presence determination condition; and change the maintained presence / absence information to indicate that the person to be measured is not on the bed if the presence / absence information indicates that the person to be measured is on the bed, and the magnitude of the decrease, which is the value obtained by subtracting the value obtained by the total weight data based on the latest measurement result from the value obtained by the total weight data based on the most recent measurement result, satisfies a given absence determination condition. The program may be stored in a computer-readable information storage medium.

[0020] According to the present invention, it is possible to accurately determine whether or not the person being measured is on a bed.

[0021] This is a configuration diagram of a biological information detection system according to an embodiment of the present invention. This is a functional block diagram of a signal processing device according to an embodiment of the present invention. This is a flowchart showing an example of the flow of presence / absence determination processing performed by diagram illustrating an example of processing in the Doppler data acquisition unit. This is a flowchart showing an example of the flow of frequency spectrum selection processing performed by a signal processing device according to an embodiment of the present invention. This is a diagram schematically showing an example of aggregate spectrum generation.

[0022] Embodiments of the present invention will be described in detail below with reference to the drawings.

[0023] Figure 1 is a configuration diagram of a biological information detection system 1 according to an embodiment of the present invention. As shown in the figure, the biological information detection system 1 includes a signal processing device 10, a Doppler sensor unit 12, and a plurality of load sensors 14. The Doppler sensor unit 12 contains a plurality of Doppler sensors. Here, for example, the Doppler sensor unit 12 is assumed to contain six Doppler sensors (first Doppler sensor to sixth Doppler sensor). Also, as shown in Figure 1, the biological information detection system 1 includes four load sensors 14 (14a to 14d).

[0024] As shown in Figure 1, the Doppler sensor unit 12 is positioned facing the person being measured. In the example shown in Figure 1, the Doppler sensor unit 12 is attached to the headboard of the bed 16. Here, multiple Doppler sensors may be positioned symmetrically with respect to the center line L1 of the bed 16 (a line passing through the center of the bed 16 in the width direction and extending in the length direction of the bed 16). Alternatively, multiple Doppler sensors may be arranged perpendicular to the length direction of the bed 16 (the direction along the center line L1 of the bed 16). Furthermore, multiple Doppler sensors may be arranged in a line at equal intervals. Note that the number of Doppler sensors is not limited to six.

[0025] Each Doppler sensor is positioned to face the longitudinal direction of the bed 16 and emits microwaves in the longitudinal direction of the bed 16. The microwaves are reflected by the chest of the person being measured while they are sleeping on the bed 16, and each Doppler sensor receives the reflected waves. Each Doppler sensor generates a Doppler signal from the reflected waves that indicates the movement of the chest associated with breathing, and outputs Doppler data obtained by digitizing this Doppler signal. The microwaves emitted from each Doppler sensor have slightly different frequencies to prevent interference between them.

[0026] Due to the Doppler effect, the reflected wave is frequency-shifted, and by observing this, the respiratory rate of the person being measured can be obtained. The reflected wave is detected by quadrature detection as a Doppler signal containing an I signal, which is in phase with the transmitted wave, and a Q signal, which is an orthogonal component, and is output to the signal processing device 10 in digital format. The Doppler signal input to the signal processing device 10 is time-series data, showing the amplitude (I component and Q component) at each time point.

[0027] The load sensor 14 is positioned below the position of the person being measured when they are on the bed 16. As shown in Figure 1, the load sensor 14 may be provided on each of the four legs of the bed 16. However, the position and number of load sensors 14 are not limited to those shown in Figure 1. For example, the biometric information detection system 1 may include only one load sensor 14. And this single load sensor 14 may be placed on the bed 16 or on the bedding (mattress, etc.) on the bed 16.

[0028] Each load sensor 14 then outputs load data indicating the measurement result of the load applied to that load sensor 14.

[0029] The signal processing device 10 may be composed of a known computer, for example, a CPU, memory, input device, and display. The signal processing device 10 generates the respiratory rate of the person being measured based on the Doppler signal output from the Doppler sensor. The signal processing device 10 also calculates the total weight based on the measurement results of the load applied to at least one load sensor 14 and generates total weight data indicating the calculated total weight.

[0030] Figure 2 is a functional block diagram of a signal processing device 10 according to an embodiment of the present invention. As shown in the figure, the signal processing device 10 includes a presence / absence information holding unit 20, a body movement information holding unit 22, a weight data storage unit 24, a load data acquisition unit 26, a total weight data generation unit 28, a presence / absence determination unit 30, a presence / absence modification unit 32, a body movement determination unit 34, a weight data management unit 36, a Doppler data acquisition unit 38, a frequency spectrum generation unit 40, a frequency spectrum selection unit 42, an aggregated spectrum generation unit 44, a biological information generation unit 46, an abnormality determination unit 48, an action execution unit 50, and an operation control unit 52. These functional blocks are realized by the execution of a signal processing program in the signal processing device 10, which is a computer. This signal processing program may be stored in various computer-readable information storage media such as semiconductor memory and loaded from the media to the signal processing device 10. Alternatively, it may be downloaded to the signal processing device 10 via a data communication line such as the Internet.

[0031] The presence / absence information holding unit 20 holds presence / absence information indicating, for example, whether the person being measured is on the bed 16 or not. Here, the presence / absence information may indicate whether the person being measured is on the bed 16, not on the bed 16, or whether it is unknown. In the following explanation, the presence / absence information will be assumed to include a "User 1 Present" flag and an "Unknown" flag. A value of 1 for the "User 1 Present" flag and a value of 0 for the "Unknown" flag will be assumed to indicate that the presence / absence information indicates that the person being measured is on the bed 16. Furthermore, a value of 0 for both the "User 1 Present" flag and a value of 1 for the "Unknown" flag will be assumed to indicate that the presence / absence information indicates that it is unknown whether the person being measured is on the bed 16 or not.

[0032] The body movement information holding unit 22 holds, for example, body movement information indicating whether or not the person being measured is moving. In the following description, a value of 1 for the body movement information corresponds to the body movement information indicating that the person being measured is moving, and a value of 0 for the body movement information corresponds to the body movement information indicating that the person being measured is not moving.

[0033] The weight data storage unit 24 stores, for example, data indicating weight. In the following description, it is assumed that the weight data storage unit 24 stores subject weight data indicating the weight of the person being measured, and absence total weight data indicating the total weight when the person being measured is not on the bed 16. In the initial state, the values ​​of the subject weight data and the absence total weight data are set to given values. For example, a value indicating the weight of the person being measured, which has been measured in advance, may be set as the value of the subject weight data in the initial state. Alternatively, a value indicating the total weight of the bed 16 and bedding, etc., which has been measured in advance, may be set as the value of the absence total weight data in the initial state. Unknown weight data, which will be described later, is also stored in the weight data storage unit 24.

[0034] The load data acquisition unit 26 acquires load data that shows the measurement result of the load applied to at least one load sensor 14 located below the position of the person being measured when the person is on the bed 16. Here, load data output from each of the at least one load sensor 14 may be acquired.

[0035] The total weight data generation unit 28 calculates the total weight based on load data obtained from at least one load sensor 14, and generates total weight data showing the calculated total weight. If the biological information detection system 1 includes one load sensor 14, total weight data showing the measurement result of the load applied to that load sensor 14 may be generated. If the biological information detection system 1 includes multiple load sensors 14, total weight data showing the sum of the measurement results of the load applied to each of the multiple load sensors 14 may be generated.

[0036] The presence / absence determination unit 30, for example, acquires total weight data generated by the total weight data generation unit 28, and determines whether or not the person being measured is on the bed 16 based on the acquired total weight data.

[0037] The presence / absence determination unit 30, for example, if the presence / absence information indicates that the person to be measured is not on the bed 16, determines whether the magnitude of the value obtained by subtracting the value obtained by the total weight data based on the previous measurement result from the value obtained by the total weight data based on the latest measurement result satisfies a given presence determination condition. In the following description, the magnitude of the value obtained by subtracting the value obtained by the total weight data based on the previous measurement result from the value obtained by the total weight data based on the latest measurement result will be referred to as the increase.

[0038] Here, the judgment criterion may be, for example, that the magnitude of the increase is greater than a threshold determined based on the estimated weight of the subject (for example, greater than a threshold that is 80% of the value of the subject's weight data).

[0039] Furthermore, if the presence / absence determination unit 30 indicates, for example, that the person being measured is on the bed 16, it determines whether the magnitude of the difference between the total weight data based on the most recent measurement result and the total weight data based on the most recent measurement result satisfies a given absence determination condition. In the following explanation, the magnitude of the difference between the total weight data based on the most recent measurement result and the total weight data based on the most recent measurement result will be referred to as the decrease amount.

[0040] Here, the absence determination condition may be, for example, that the magnitude of the decrease is greater than a threshold determined based on the estimated weight of the person being measured (for example, greater than a threshold that is 80% of the value of the person's weight data).

[0041] In this embodiment, the following are repeatedly performed (for example, at predetermined time intervals): output of load data indicating the latest measurement result of the load applied to the load sensor 14 from the load sensor 14 to the signal processing device 10, generation of total weight data based on the load data, and determination of whether or not the person being measured is on the bed 16 based on the total weight data.

[0042] The presence / absence change unit 32 changes the held presence / absence information to indicate that the subject is on the bed 16 when the amount of increase described above satisfies a given presence determination condition, for example, when the presence / absence information indicates that the subject is not on the bed 16.

[0043] Also, the presence / absence change unit 32 changes the held presence / absence information to indicate that the subject is not on the bed 16 when the amount of decrease described above satisfies a given absence determination condition, for example, when the presence / absence information indicates that the subject is on the bed 16.

[0044] Note that the threshold value in the presence determination condition and the threshold value in the absence determination condition may be the same or different.

[0045] The body movement determination unit 34 determines whether or not the subject's body movement has occurred based on, for example, the magnitude of the variation in the value indicated by the total weight data. Here, for example, the body movement determination unit 34 may determine whether or not the subject's body movement has occurred each time a determination timing that arrives at a time interval longer than the time interval at which the total weight data is generated arrives.

[0046] Here, the period between two consecutive determination timings is referred to as a body movement determination period. And the maximum value or minimum value of the value of the total weight data generated in a certain specific body movement determination period may be specified. And the average value of the value of the total weight data generated in the body movement determination period immediately before the specific body movement determination period may be specified. And when the difference between the maximum value specified for the body movement determination period and the average value specified for the body movement determination period immediately before the specific body movement determination period is greater than or equal to a predetermined value, it may be determined that the subject's body movement has occurred at the body movement determination time. Also, when the difference between the minimum value specified for the body movement determination period and the average value specified for the body movement determination period immediately before the specific body movement determination period is greater than or equal to a predetermined value, it may be determined that the subject's body movement has occurred at the body movement determination time. And when neither is the case, it may be determined that the subject's body movement has not occurred at the body movement determination time. Note that the method for determining whether or not the subject's body movement has occurred is not limited to this method.

[0047] And when the body movement determination unit 34 determines that the body movement of the measured person has occurred, it may set 1 to the value of the body movement information held in the body movement information holding unit 22, and when it is determined that the body movement of the measured person has not occurred, 0 may be set to the value of the body movement information held in the body movement information holding unit 22.

[0048] The weight data management unit 36, for example, determines, as an estimated value of the weight of the measured person, a value obtained by subtracting the value of the total weight data when the presence / absence information indicates that the measured person is not on the bed 16 from the value of the total weight data when the presence / absence information indicates that the measured person is on the bed 16. Then, the weight data management unit 36 updates the measured person weight data stored in the weight data storage unit 24 to indicate the determined estimated value.

[0049] When the held presence / absence information is changed from indicating that the measured person is not on the bed 16 to indicating that the measured person is on the bed 16, the weight data management unit 36 may update the estimated value of the weight of the measured person to the above-described increase amount. Here, the weight data management unit 36 may update, for example, the measured person weight data stored in the weight data storage unit 24 to indicate the above-described increase amount. Further, the weight data management unit 36 may update the total weight data when absent stored in the weight data storage unit 24 to indicate the value indicated by the total weight data based on the previous measurement result.

[0050] Here, an example of the processing flow performed by the presence / absence determination unit 30, the presence / absence change unit 32, and the weight data management unit 36 will be described while referring to the flowcharts illustrated in FIGS. 3 to 5. In the following description, it is assumed that the value of the measured person weight data stored in the weight data storage unit 24 is A. Also, it is assumed that the value of the total weight data when absent stored in the weight data storage unit 24 is B.

[0051] FIG. 3 is a diagram showing an example of the processing flow executed when the presence / absence information indicates that the measured person is not on the bed, that is, when the values of the one-person presence flag and the unknown flag held in the presence / absence information holding unit 20 are both 0.

[0052] In this case, the presence / absence determination unit 30 determines whether the above-mentioned increase is greater than or equal to (A × r1) (S101). Here, r1 is a given value, for example, 0.8.

[0053] If the increase is not greater than or equal to (A × r1) (S101: N), the process shown in this example is terminated.

[0054] If the increase is (A × r1) or greater (S101: Y), the presence / absence determination unit 30 determines whether the value of the total weight data based on the latest measurement result is (A + B - d) or greater and (A + B + d) or less (S102). Here, d is a given threshold.

[0055] If the total weight data value based on the latest measurement result is greater than or equal to (A + B - d) and less than or equal to (A + B + d) (S102: Y), the presence / absence change unit 32 sets the value of the user presence flag held in the presence / absence information holding unit 20 to 1 (S103). Then, the weight data management unit 36 ​​updates the measured person weight data stored in the weight data storage unit 24 to reflect the above-mentioned increase, and updates the absence total weight data stored in the weight data storage unit 24 to reflect the value shown in the total weight data based on the most recent measurement result (S104), and the process shown in this example is completed. That is, the above-mentioned increase is set as value A, and the value shown in the total weight data based on the most recent measurement result is set as value B.

[0056] In the process shown in S102, if it is determined that the value of the total weight data based on the latest measurement result is not greater than or equal to (A + B - d) or less than or equal to (A + B + d) (S102: N), the presence / absence change unit 32 sets the value of the unknown flag to 1 (S105). Then, the weight data management unit 36 ​​generates unknown weight data with the value indicating the above-mentioned increase set, and stores the generated unknown weight data in the weight data storage unit 24 (S106). Then, the process shown in this example is terminated. In the following explanation, the value of the unknown weight data is assumed to be C.

[0057] As explained with reference to Figure 3, the presence / absence change unit 32 may change the stored presence / absence information to indicate that the person being measured is on the bed 16 if the presence / absence information indicates that the person being measured is not on the bed 16, the magnitude of the above-mentioned increase satisfies the presence determination condition, and the difference between the sum of the value of the person being measured's weight data and the value of the total weight data when absent and the value indicated by the total weight data based on the latest measurement result is within a given range.

[0058] Furthermore, the presence / absence change unit 32 may change the stored presence / absence information to indicate that it is unclear whether the person being measured is on the bed 16, if the magnitude of the increase described above satisfies the presence determination condition, and the difference between the sum of the measured person's weight data and the total weight data when absent and the total weight data based on the latest measurement result is not within that range.

[0059] Figure 4 shows an example of the processing flow that is executed when the presence / absence information indicates that the person being measured is on the bed, that is, when the value of the "User 1 Present" flag held in the presence / absence information holding unit 20 is 1 and the value of the "Unknown" flag is 0.

[0060] In this case, the presence / absence determination unit 30 determines whether the above-mentioned decrease is greater than or equal to (A × r2) (S201). Here, r2 is a given value, for example, 0.8. The value r2 may be the same as or different from the value r1.

[0061] If the decrease is not greater than or equal to (A × r²) (S201: N), the process shown in this example is terminated.

[0062] If the decrease is greater than or equal to (A × r²) (S201: Y), the presence / absence change unit 32 sets the value of the user presence flag held in the presence / absence information holding unit 20 to 0, and the process shown in this example is terminated.

[0063] Figure 5 shows an example of the processing flow that is executed when the presence / absence information indicates that it is unclear whether the person being measured is on the bed, that is, when the value of the "User 1 Present" flag held in the presence / absence information holding unit 20 is 0 and the value of the "Unknown" flag is 1.

[0064] In this case, the presence / absence determination unit 30 determines whether the above-mentioned decrease is greater than or equal to (C × r3) (S301). Here, r3 is a given value, for example, 0.8. The value r3 may be the same as or different from the value r1. Also, the value r3 may be the same as or different from the value r2.

[0065] If the decrease is not greater than or equal to (C × r³) (S301: N), the process shown in this example is terminated.

[0066] If the decrease is (C × r3) or greater (S301: Y), the presence / absence change unit 32 sets the value of the unknown flag held in the presence / absence information holding unit 20 to 0 (S302). Then, the weight data management unit 36 ​​deletes the unknown weight data stored in the weight data storage unit 24 (S303), and the process shown in this example is completed.

[0067] The Doppler data acquisition unit 38 acquires Doppler data for each of the multiple Doppler sensors provided facing the person being measured, showing the measurement results from that Doppler sensor over a certain period of time.

[0068] The Doppler data acquisition unit 38, for example as shown in Figure 6, extracts a portion of the Doppler data representing the measurement results for each of the multiple Doppler sensors

[0069] For example, from the first Doppler data acquired from the first Doppler sensor, Doppler data D(1,1), D(2,1), D(3,1), D(4,1), ... showing the measurement results in the first time window, second time window, third time window, fourth time window, ... are extracted. Similarly, from the second Doppler data acquired from the second Doppler sensor, Doppler data D(1,2), D(2,2), D(3,2), D(4,2), ... showing the measurement results in the first time window, second time window, third time window, fourth time window, ... are extracted.

[0070] Similarly, for the third to sixth Doppler sensors, Doppler data showing the measurement results for each time window is extracted. For example, from the third Doppler data acquired from the third Doppler sensor, Doppler data D(1,3), D(2,3), D(3,3), D(4,3), ... showing the measurement results for the first, second, third, fourth, ... time windows are extracted. Also, from the fourth Doppler data acquired from the fourth Doppler sensor, Doppler data D(1,4), D(2,4), D(3,4), D(4,4), ... showing the measurement results for the first, second, third, fourth, ... time windows are extracted. Furthermore, Doppler data D(1,5), D(2,5), D(3,5), D(4,5), ... showing the measurement results in the first time window, second time window, third time window, fourth time window, ... are extracted from the fifth Doppler data acquired from the fifth Doppler sensor. Similarly, Doppler data D(1,6), D(2,6), D(3,6), D(4,6), ... showing the measurement results in the first time window, second time window, third time window, fourth time window, ... are extracted from the sixth Doppler data acquired from the sixth Doppler sensor.

[0071] Each time window has a constant length (for example, 60 seconds), and the start time of each time window is shifted by a predetermined amount of time (for example, 2 seconds). Furthermore, the time windows applied to each of the first through sixth Doppler data are the same. That is, for each of the first through sixth Doppler data, for example, the period corresponding to the first time window is the same, and the period corresponding to the second time window is also the same.

[0072] The frequency spectrum generation unit 40 generates a frequency spectrum for each of the multiple Doppler sensors, for example, based on the Doppler data indicating the measurement results from the Doppler sensor.

[0073] The frequency spectrum generation unit 40 converts the input Doppler data into a frequency spectrum by, for example, performing a Fast Fourier Transform (FFT) on the Doppler data. Here, for example, the data for the I signal and the data for the Q signal may each be converted into a frequency spectrum.

[0074] The frequency spectrum selection unit 42 selects, for example, a plurality of frequency spectra from among those generated for each of the plurality of Doppler sensors, based on the peak frequencies in the frequency spectra generated for each of the plurality of Doppler sensors.

[0075] Here, an example of the flow of the frequency spectrum selection process performed by the frequency spectrum selection unit 42 will be explained with reference to the flowchart illustrated in Figure 7. Here, for example, we will explain the process of selecting multiple frequency spectra from the I signal data and Q signal data of six Doppler data D(1,1), D(1,2), D(1,3), D(1,4), D(1,5), and D(1,6) extracted for the first time window.

[0076] First, the frequency spectrum selection unit 42 identifies the peak frequency, which is the frequency of the maximum amplitude in the frequency spectrum generated for each of the six Doppler data D(1,1), D(1,2), D(1,3), D(1,4), D(1,5), and D(1,6), as the respiratory rate corresponding to that frequency spectrum (S401).

[0077] Then, the frequency spectrum selection unit 42 calculates a representative value (in this case, an average value) of the respiratory rate identified in the process shown in S401 (S402).

[0078] Then, the frequency spectrum selection unit 42 checks whether there is a frequency spectrum corresponding to a respiratory rate (i.e., an outlier respiratory rate) whose difference from the representative value calculated in the process shown in S402 is greater than or equal to a predetermined value (for example, 5) (S403).

[0079] If an outlier respiratory rate exists (S403: Y), the frequency spectrum selection unit 42 excludes the frequency spectrum corresponding to the outlier respiratory rate (S404) and returns to the process shown in S402.

[0080] If there are no outlier respiratory rates (S403: N), the process shown in this example is terminated. Note that in the process shown in this example, all frequency spectra may be selected without any frequencies being excluded.

[0081] The frequency spectra that remain after the above process are selected from the frequency spectra generated for the I signal data and Q signal data of each of the six Doppler data extracted for the first time window. The process described above is performed for each of the multiple time windows.

[0082] The aggregated spectrum generation unit 44 generates an aggregated spectrum based, for example, on some or all of the frequency spectra generated for each of the multiple Doppler sensors. As shown in Figure 5, with respect to the first time window, the aggregated spectrum (1) may be generated by adding the amplitudes (intensities) of the frequency spectra of the data of the I signal of D(1,1), the Q signal of D(1,1), the I signal of D(1,2), the Q signal of D(1,2), the I signal of D(1,3), the Q signal of D(1,3), ..., the Q signal of D(1,6) for each frequency. The aggregated spectrum (1) corresponds to the aggregated spectrum of the first time window. In this way, the aggregated spectrum generation unit 44 may generate an aggregated spectrum which is the sum of some or all of the frequency spectra generated for each of the multiple Doppler sensors.

[0083] Furthermore, the aggregated spectrum generation unit 44 may generate an aggregated spectrum, which is a frequency spectrum obtained by averaging some or all of the frequency spectra generated for each of the multiple Doppler sensors.

[0084] Furthermore, the aggregated spectrum generation unit 44 may generate an aggregated spectrum based on a plurality of frequency spectra selected by the frequency spectrum selection unit 42. For example, suppose the frequency spectra of the I signal data of D(1,5), the Q signal data of D(1,5), the I signal data of D(1,6), and the Q signal data of D(1,6) are excluded, and the other frequency spectra are selected. In this case, an aggregated spectrum (1) may be generated, which is the sum of the frequency spectra of the I signal data of D(1,1), the Q signal data of D(1,1), the I signal data of D(1,2), the Q signal data of D(1,2), the I signal data of D(1,3), the Q signal data of D(1,3), the I signal data of D(1,4), and the Q signal data of D(1,4). By doing so, the influence of outliers in the generated aggregated spectrum can be suppressed.

[0085] Figure 8 shows the aggregated spectrum (1), which is the aggregated spectrum of the first time window. Similarly, the aggregated spectrum (2), which is the aggregated spectrum of the second time window, the aggregated spectrum (3), which is the aggregated spectrum of the third time window, and so on are generated.

[0086] The biological information generation unit 46 generates biological information of the subject based on, for example, Doppler data. The biological information generation unit 46 may also generate biological information of the subject for a given period based on an aggregated spectrum. Here, the biological information generation unit 46 may generate the respiratory rate of the subject for a given period based on the peak frequency in the aggregated spectrum. For example, the frequency of the maximum amplitude in the aggregated spectrum may be identified as the peak frequency. The peak frequency in the aggregated spectrum may then be generated as the respiratory rate.

[0087] Furthermore, the biological information generation unit 46 may identify the maximum amplitude and the second largest amplitude in the aggregated spectrum. The biological information generation unit 46 may then generate an amplitude ratio obtained by dividing the maximum amplitude by the second largest amplitude.

[0088] Furthermore, the biological information generation unit 46 may generate an amplitude ratio for each of the aggregated spectra calculated for multiple time windows (for example, the first to the 30th time windows). The biological information generation unit 46 may then extract aggregated spectra in which the amplitude ratio is equal to or greater than a predetermined threshold (for example, 1.5). The biological information generation unit 46 may then generate the average value of the respiratory rate generated for each of the aggregated spectra in which the amplitude ratio is equal to or greater than a predetermined threshold (for example, 1.5) as the respiratory rate for a period (for example, 1 minute) associated with the multiple time windows.

[0089] Furthermore, the biological information generation unit 46 may generate a confidence score as the ratio of the number of aggregated spectra whose amplitude ratio is equal to or greater than a predetermined threshold (e.g., 1.5) to the number of aggregated spectra calculated for multiple time windows.

[0090] In this embodiment, the frequency spectrum selection unit 42 does not necessarily have to select a frequency spectrum. The aggregated spectrum generation unit 44 may use all the frequency spectra generated by the frequency spectrum generation unit 40 for a certain period to generate an aggregated spectrum for that period.

[0091] Furthermore, the frequency spectrum selection unit 42 may select frequency spectra on a Doppler sensor basis. That is, the frequency spectrum selection unit 42 may select multiple Doppler sensors from among the multiple Doppler sensors based on the peak frequencies in the frequency spectra generated for each of the multiple Doppler sensors. The aggregated spectrum generation unit 44 may then generate an aggregated spectrum based on the frequency spectra generated for each of the selected multiple Doppler sensors.

[0092] Furthermore, it is not necessary for the data of the I signal and the data of the Q signal to be converted into frequency spectra. For example, the data of a complex signal with the I component as the real part and the Q component as the imaginary part may be converted into a frequency spectrum. Then, an aggregate spectrum may be generated based on that frequency spectrum.

[0093] When measuring a subject's biological information using multiple Doppler sensors, the accuracy of the measurement results obtained from each of these sensors may be low depending on the subject's position and posture.

[0094] Therefore, if biological information is measured using each of these multiple Doppler sensors, and a representative value such as the average of the measured biological information is used as the biological information value of the person being measured, the biological information value obtained in this way may not be a valid value.

[0095] In this embodiment, as described above, biological information is generated based on the aggregated spectrum. Therefore, according to this embodiment, it is possible to measure the biological information of a subject with greater accuracy compared to the case where a representative value such as the average value of the measured biological information is used as the value of the subject's biological information.

[0096] Furthermore, for example, each Doppler sensor has a different angle relative to the person being measured, so the peak frequencies of the frequency spectra based on the measurement results may differ slightly. However, when multiple Doppler sensors are arranged perpendicular to the length of the bed 16, symmetrically with respect to the center line of the bed 16, or in a line at equal intervals, an aggregate spectrum is generated by appropriately averaging the frequency spectra corresponding to each Doppler sensor. As a result, it becomes possible to measure the biological information of the person being measured with greater accuracy.

[0097] In the following explanation, we will assume that the above respiratory rate is generated at one-minute intervals.

[0098] The abnormality determination unit 48 determines, for example, whether the biological information of the person being measured satisfies a given abnormality determination condition. Here, the abnormality determination condition may be, for example, that the respiratory rate being 8 or less or 25 or more applies to any of the five most recently generated respiratory rates.

[0099] The action execution unit 50 executes a given action, for example, when the biological information of the person being measured meets the above-mentioned abnormality determination conditions.

[0100] Here, the action execution unit 50 may perform a first action if the presence / absence information indicates that the person being measured is on the bed 16, and a second action if the presence / absence information indicates that it is unclear whether the person being measured is on the bed 16 or not. For example, if the value of the "User 1 Present" flag held in the presence / absence information holding unit 20 is 1, an emergency notification action (e.g., requesting an ambulance) may be performed in response to the fact that the person being measured's biometric information has met the abnormality judgment conditions. If the value of the "Unknown" flag held in the presence / absence information holding unit 20 is 1, an abnormality notification action (e.g., outputting an alarm sound) may be performed in response to the fact that the person being measured's biometric information has met the abnormality judgment conditions.

[0101] For example, if the presence / absence information indicates that the person being measured is not on the bed 16, the motion control unit 52 suppresses the execution of an action corresponding to the biological information meeting the abnormality determination conditions.

[0102] Here, the operation control unit 52 may monitor the value of the user presence flag and the unknown flag held in the presence / absence information holding unit 20.

[0103] Furthermore, the operation control unit 52 may stop the Doppler sensor in response to detecting a change in the situation from when the value of the "One User Present" flag or the "Unknown" flag is 1 to when both the "One User Present" flag and the "Unknown" flag are 0.

[0104] Furthermore, the operation control unit 52 may activate the Doppler sensor in response to detection that the value of either the "One User Present" flag or the "Unknown" flag has changed from 0 to 1. The generation of respiratory rate data may then be resumed.

[0105] In this way, by stopping the operation of the Doppler sensor, the execution of actions corresponding to the abnormality detection conditions of the biological information may be suppressed.

[0106] Furthermore, the motion control unit 52 may suppress the execution of actions corresponding to the abnormality judgment condition of the biological information if it determines that the person being measured is on the bed 16, and that the person being measured is moving.

[0107] For example, suppose the value of the "One User Present" flag is 1. In this case, the motion control unit 52 may monitor the value of the body movement information held in the body movement information holding unit 22.

[0108] Furthermore, the motion control unit 52 may stop the Doppler sensor in response to detecting a change in the value of the body movement information from 1 to 0.

[0109] Furthermore, the motion control unit 52 may activate the Doppler sensor in response to detecting a change in the value of the body movement information from 0 to 1. The generation of respiratory rate may then be resumed.

[0110] Furthermore, the action execution unit 50, rather than the motion control unit 52, may suppress the execution of an action corresponding to the abnormality judgment condition of the biological information if the presence / absence information indicates that the person being measured is not on the bed 16. For example, suppose the Doppler sensor is operating and the above-mentioned respiratory rate is being generated at one-minute intervals. In this situation, even if the biological information of the person being measured meets the given abnormality judgment condition, the action execution unit 50 does not have to execute the given action if both the value of the "1 user present" flag and the value of the "unknown" flag held in the presence / absence information holding unit 20 are 0. Also, even if the biological information of the person being measured meets the abnormality judgment condition, the action execution unit 50 does not have to execute the given action if the value of the body movement information held is 1.

[0111] When the respiratory rate of a person being measured shows an abnormal value, it is not possible to determine whether the cause is that something is wrong with the person being measured or that the person is not on the bed 16. In this embodiment, as described above, it is possible to accurately determine whether or not the person being measured is on the bed. Therefore, when the respiratory rate of a person being measured shows an abnormal value, it becomes possible to determine whether the cause is that something is wrong with the person being measured or that the person is not on the bed 16.

[0112] Furthermore, as explained above, if the presence / absence information indicates that the person being measured is not on the bed 16, the execution of actions corresponding to the abnormality judgment condition of the biometric information may be suppressed. By doing so, for example, it is possible to prevent actions such as calling an ambulance or outputting an alarm sound from being mistakenly executed when the person being measured is not on the bed 16.

[0113] Furthermore, if body movement occurs, there is a high probability that the respiratory rate cannot be accurately detected. Therefore, as explained above, if it is determined that the subject is moving, the execution of actions corresponding to the abnormality of the biological information may be suppressed. This prevents actions from being erroneously executed based on a respiratory rate that is likely to be inaccurate.

[0114] Furthermore, as explained above, the Doppler sensor may be stopped if the presence / absence information indicates that the person being measured is not on the bed 16. This makes it possible to reduce the power consumption of the Doppler sensor.

[0115] Furthermore, as explained above, if a person or animal (such as a pet or the subject's child) on bed 16 is significantly heavier or lighter than the subject, it may be determined that it is unclear whether the subject is on bed 16 or not. In this case, although the biometric information meets the abnormality detection criteria, an emergency call action (e.g., calling an ambulance) will not be performed, but an abnormality notification action (e.g., outputting an alarm sound) will be performed. This reduces false alarms and prevents unnecessary execution of emergency call actions.

[0116] Furthermore, in this embodiment, the estimated weight of the person being measured (value A mentioned above) may be output (for example, as a display output or audio output). In this way, the person being measured can have their weight, which fluctuates slightly from day to day, measured without any burden simply by lying down.

[0117] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are possible. For example, in the above description, respiratory rate was generated as the subject's biological information, but heart rate can be generated in the same way. In this case, a given action may be executed in response to whether the heart rate satisfies a given abnormality determination condition.

Claims

1. A presence / absence determination system comprising: Total weight data acquisition means for repeatedly acquiring total weight data indicating the total weight calculated based on the measurement result of the load applied to at least one load sensor located below the position of the person being measured while the person being measured is on a bed; Presence / absence information holding means for holding presence / absence information indicating whether or not the person being measured is on the bed; Presence / absence information changing means for changing the held presence / absence information to indicate that the person being measured is on the bed when the presence / absence information indicates that the person being measured is not on the bed, and the magnitude of the increase, which is the value obtained by subtracting the value obtained by the total weight data based on the previous measurement result from the value obtained by the total weight data based on the most recent measurement result, satisfies a given presence determination condition; and Absence / Absence information changing means for changing the held presence / absence information to indicate that the person being measured is not on the bed when the presence / absence information indicates that the person being measured is on the bed, and the magnitude of the decrease, which is the value obtained by subtracting the value obtained by the total weight data based on the latest measurement result from the value obtained by the total weight data based on the most recent measurement result, satisfies a given absence determination condition.

2. A presence / absence determination system according to claim 1, further comprising: Doppler data acquisition means for acquiring Doppler data indicating the measurement result by a Doppler sensor provided facing the person to be measured; biometric information generation means for generating biometric information of the person to be measured based on the Doppler data; action execution means for executing a given action in response to the biometric information satisfying a given abnormality determination condition; and suppression means for suppressing the execution of the action in response to the biometric information satisfying the abnormality determination condition when the presence / absence information indicates that the person to be measured is not on the bed.

3. The presence / absence determination system according to claim 2, further comprising: a body movement determination means for determining whether or not body movement of the person being measured has occurred based on the magnitude of the fluctuation of the value indicated by the total weight data, wherein the suppression means suppresses the execution of the action corresponding to the fact that the biological information has met the abnormality determination condition when the presence / absence information indicates that the person being measured is on the bed and it is determined that body movement of the person being measured has occurred.

4. A presence / absence determination system according to claim 1, further comprising: weight estimation means for determining an estimated weight of the person to be measured by subtracting the value of the total weight data when the presence / absence information indicates that the person to be measured is not on the bed from the value of the total weight data when the presence / absence information indicates that the person to be measured is on the bed.

5. The presence / absence determination system according to claim 4, wherein the presence determination condition is that the magnitude of the increase is greater than a threshold determined based on the estimated weight of the person being measured.

6. The presence / absence determination system according to claim 4, wherein the absence determination condition is that the magnitude of the decrease is greater than a threshold determined based on the estimated weight of the person being measured.

7. The presence / absence determination system according to claim 4, wherein the weight estimation means updates the estimated weight of the person to be measured by the increase amount, which is the value obtained by subtracting the value obtained by the total weight data based on the previous measurement result from the value obtained by the total weight data based on the most recent measurement result.

8. The presence / absence determination system according to claim 1, wherein the presence / absence information indicates whether the person to be measured is on the bed, is not on the bed, or is it unknown whether the person is on the bed or not; further comprising: a person weight data storage means for storing person weight data indicating the weight of the person to be measured; and a total weight data storage means for storing total weight data when the person to be measured is not on the bed; the presence / absence change means changes the stored presence / absence information to indicate that the person to be measured is on the bed when the presence / absence information indicates that the person to be measured is not on the bed, the magnitude of the increase satisfies the presence determination condition, and the difference between the sum of the value of the person weight data and the value of the total weight data when the person to be measured is within a given range; The presence / absence determination system includes a presence / absence information change, which indicates that the person being measured is not on the bed, the magnitude of the increase satisfies the presence / absence determination condition, and the difference between the sum of the value of the person being measured's weight data and the value of the total weight data when absent and the value indicated by the total weight data based on the latest measurement result is not within the range, thereby changing the retained presence / absence information to indicate that it is unknown whether the person being measured is on the bed or not.

9. The presence / absence determination system according to claim 8, further comprising an updating means for updating the stored person weight data to indicate the increase when the stored presence / absence information is changed to indicate that the person being measured is on the bed, and updating the stored total weight data when absent to indicate the value indicated by the total weight data based on the most recent measurement result.

10. The presence / absence determination system according to claim 8, further comprising: Doppler data acquisition means for acquiring Doppler data indicating the measurement result by a Doppler sensor provided toward the person to be measured; biometric information generation means for generating biometric information of the person to be measured based on the Doppler data; action execution means for executing a given action in response to the biometric information satisfying a given abnormality determination condition; and suppression means for suppressing the execution of the action in response to the biometric information satisfying the abnormality determination condition when the presence / absence information indicates that the person to be measured is not on the bed, wherein the action execution means executes a first action when the presence / absence information indicates that the person to be measured is on the bed, and executes a second action when the presence / absence information indicates that it is unclear whether the person to be measured is on the bed or not.

11. The presence / absence determination system according to claim 1, wherein the load sensor is provided on each of the four legs of the bed.

12. The presence / absence determination system according to claim 1, wherein the total weight data acquisition means repeatedly acquires the total weight data which indicates the sum of the measured values ​​of the load applied to each of the plurality of load sensors.

13. A method for determining presence or absence of a person, comprising: repeatedly acquiring total weight data indicating the total weight calculated based on the measurement results of the load applied to at least one load sensor located below the position of the person being measured while the person being measured is on a bed; maintaining presence / absence information indicating whether or not the person being measured is on the bed; changing the maintained presence / absence information to indicate that the person being measured is on the bed if the presence / absence information indicates that the person being measured is not on the bed, and the magnitude of the increase, which is the value obtained by subtracting the value obtained by the total weight data based on the previous measurement result from the value obtained by the total weight data based on the most recent measurement result, satisfies a given presence / absence condition; and changing the maintained presence / absence information to indicate that the person being measured is not on the bed if the presence / absence information indicates that the person being measured is on the bed, and the magnitude of the decrease, which is the value obtained by subtracting the value obtained by the total weight data based on the latest measurement result from the value obtained by the total weight data based on the most recent measurement result, satisfies a given absence / absence condition.

14. A program to cause a computer to perform the following steps: repeatedly acquire total weight data indicating the total weight calculated based on the measurement results of the load applied to at least one load sensor located below the position of the person being measured while the person being measured is on a bed; maintain presence / absence information indicating whether or not the person being measured is on the bed; if the presence / absence information indicates that the person being measured is not on the bed, and the magnitude of the increase, which is the value obtained by subtracting the value obtained by the total weight data based on the previous measurement result from the value obtained by the total weight data based on the most recent measurement result, satisfies a given presence determination condition, change the maintained presence / absence information to indicate that the person being measured is on the bed; and if the presence / absence information indicates that the person being measured is on the bed, and the magnitude of the decrease, which is the value obtained by subtracting the value obtained by the total weight data based on the latest measurement result from the value obtained by the total weight data based on the most recent measurement result, satisfies a given absence determination condition, change the maintained presence / absence information to indicate that the person being measured is not on the bed.

Images