JP7316038B2 - Event prediction system, sensor signal processing system and program - Google Patents

Description

FIELD OF THE DISCLOSURE The present disclosure relates generally to event prediction systems, sensor signal processing systems and programs, and more particularly to event prediction systems, sensor signal processing systems and programs that use body motion data regarding body motion of a subject.

A non-contact activity sensor (sensor processing system) described in Patent Document 1 includes a Doppler sensor (measuring unit), a distance sensor, and a processor. The processor calculates the amount of activity of the target included in the sensing range (air-conditioned space) of each sensor based on one or both of the amplitude and frequency of the detection signal of the Doppler sensor and the detection signal of the distance sensor. Patent Literature 1 describes that air conditioning control is performed based on the calculated amount of activity.

Moreover, Patent Literature 1 also describes estimating a user's health condition based on the detected amount of activity. For example, based on the change in the amount of activity of the user, the user's health condition such as poor physical condition or injury is grasped.

JP 2017-484 A

However, in Patent Document 1, since air conditioning control is a basic function, it is sufficient to be able to grasp the current state of the user, and it is not assumed that the detected amount of activity is used for other estimations. not

An object of the present disclosure is to provide an event prediction system, a sensor signal processing system, and a program capable of determining even a sign of occurrence of a specific event related to a subject.

An event prediction system according to one aspect of the present disclosure includes an acquisition unit, a sign determination unit, and an existence determination unit. The acquisition unit acquires the body motion data from a measuring device that outputs body motion data relating to body motion of a subject. The sign determination unit determines a sign of occurrence of a specific event related to the subject based on the body motion data acquired during a past reference period. The presence determination unit determines presence/absence of the target person in the target space based on the body motion data. The existence determination unit determines that, when a coefficient of an analysis model of time-series analysis representing the body motion data with a plurality of pieces of body motion data acquired before the acquisition timing of the body motion data exceeds a threshold value, the object It is determined that the target person exists in the space.

A sensor signal processing system according to one aspect of the present disclosure includes an acquisition unit, an acceleration calculation unit, and a presence determination unit. The acquisition unit acquires body motion data from the measuring device. The measuring device outputs the body motion data relating to the subject's body motion. The acceleration calculator obtains an acceleration of the body motion of the subject based on the body motion data. The presence determination unit determines presence/absence of the target person in the target space based on the body motion data. The existence determination unit determines that, when a coefficient of an analysis model of time-series analysis representing the body motion data with a plurality of pieces of body motion data acquired before the acquisition timing of the body motion data exceeds a threshold value, the object It is determined that the target person exists in the space.

A program according to one aspect of the present disclosure acquires the body motion data from a measuring device that outputs body motion data related to the body motion of a subject, and based on the body motion data acquired during a past reference period, the A coefficient of an analysis model for time-series analysis that determines a sign of occurrence of a specific event related to a subject and expresses the body movement data with a plurality of pieces of body movement data acquired prior to the acquisition timing of the body movement data is a program for causing a computer system to execute an event prediction method for determining that the target person exists in the target space when the threshold is exceeded .

According to the present disclosure, there is an advantage that it is possible to provide an event prediction system, a sensor signal processing system, and a program capable of determining even a sign of occurrence of a specific event related to a subject.

FIG. 1 is a block diagram showing configurations of a sensor signal processing system and an event prediction system according to Embodiment 1 of the present disclosure. FIG. 2 is an explanatory diagram of a facility using the same event prediction system. FIG. 3 is a flow chart showing the operation of the event prediction system of the same. FIG. 4 is a flow chart showing a specific example of acceleration calculation processing in the event prediction system. FIG. 5 is a flow chart showing a specific example of a sign determination process in the event prediction system. FIG. 6 is a graph showing an example of body motion data to which the event prediction system is applied.

(Embodiment 1)
(1) Outline An outline of the sensor signal processing system 1 and the event prediction system 10 according to the present embodiment will be described with reference to FIGS. 1 and 2. FIG. The event prediction system 10 has a sensor signal processing system 1 .

The sensor signal processing system 1 is a system that performs signal processing on sensor signals output from a measuring device 2 that monitors a target space 100 (see FIG. 2). The measuring device 2 has a sensor 21 (see FIG. 1) that generates body motion data relating to the body motion of a target person (person) in the target space 100, and outputs a sensor signal including the body motion data. The sensor signal processing system 1 includes an acceleration calculator 111 (see FIG. 1) that obtains the acceleration of the subject's body movement based on the body movement data.

A “target space” as used in the present disclosure is, for example, a specific space within a facility such as a serviced housing for the elderly, a nursing home, or a hospital. If the target space 100 is a private room space in a serviced housing for the elderly or a nursing care facility, the "target person" is a resident of the private room (person receiving care). If the target space 100 is a space in a hospital room, the "subject" is a patient in the hospital room. In addition, the “body movement” referred to in the present disclosure includes not only body movements (rolling over) when a person is sleeping, but also, for example, a person in a standing state, a sitting state and a walking state. including general body movements. The term “acceleration” as used in the present disclosure means the rate of change in speed (moving speed) of at least part of the body of the subject with respect to time when at least part of the body of the subject is in motion. For example, in a subject who is walking, since movement occurs in the entire body, the rate of change in speed of the entire body with respect to time is "acceleration." On the other hand, for example, when a part of the body moves in a sleeping subject, the "acceleration" is the ratio of the speed change of the moving part of the body to the time.

That is, the sensor signal processing system 1 can quantitatively analyze the motion of the subject as acceleration based on the body motion data acquired from the measuring device 2 . The acceleration of the movement of the subject's body can be used, for example, to grasp the behavior of the subject, the state of the subject such as the state of health, and to identify the individual.

The event prediction system 10 according to this embodiment is a system that uses the processing result of the sensor signal processing system 1 to determine a sign of occurrence of a specific event related to a subject. The “specific event” referred to in the present disclosure is a specific event selected from various events that may occur in relation to the subject, such as terminal care (end-of-life care), walking These include falls, illness or injury requiring hospitalization, death, cognitive decline (cognitive impairment), and wandering. Furthermore, individual behaviors in the subject's daily life, such as getting up from a sleeping state (getting out of bed), excretion, going to bed, etc., are also included in the "specific event".

That is, the event prediction system 10 determines a sign of the occurrence of the specific event, that is, a "sign" that appears before the occurrence of the specific event, based on the body motion data acquired from the measuring device 2 . In other words, the event prediction system 10 can make determinations about specific events that are predicted to occur in the future rather than the current state of the subject. As described above, the event prediction system 10 according to the present embodiment has the advantage of being able to determine even a sign of occurrence of a specific event related to the subject.

(2) Details Details of the sensor signal processing system 1 and the event prediction system 10 according to the first embodiment will be described below with reference to the drawings. In the following, as shown in FIG. 2, a case where the target space 100 is set in a private room 50 of a house with services for the elderly will be described as an example. In other words, the "subject" in this case is the occupant of the private room.

(2.1) Configuration The event prediction system 10 includes the sensor signal processing system 1 as described above. In this embodiment, the event prediction system 10 further includes a second arithmetic processing unit 12 as shown in FIG. The sensor signal processing system 1 and the event prediction system 10 including the same are implemented by a computer system installed in, for example, a janitor's room of a facility (here, a serviced elderly house) in which the target space 100 is installed. .

The sensor signal processing system 1 receives sensor signals from the measuring device 2 that monitors the target space 100 . As described above, the measuring device 2 has the sensor 21 that generates body motion data about the body motion of the target person (person) in the target space 100, and sends the sensor signal including the body motion data to the sensor signal processing system 1. Output.

In this embodiment, the measuring device 2 is not included in the components of the sensor signal processing system 1 and the event prediction system 10 . Therefore, the sensor signal processing system 1 and the event prediction system 10 are applicable in combination with various aspects of the measuring device 2 . However, each of the sensor signal processing system 1 and the event prediction system 10 may include the measurement device 2 as a component.

The measuring device 2 has a sensor 21 and a signal processing circuit 22 . The sensor 21 is a non-contact sensor that detects the body movement of the subject without contacting the subject. In short, the measuring device 2 generates body motion data relating to the body motion of the subject in a non-contact manner. Therefore, the measuring device 2 can generate body motion data without interfering with the movement of the subject. The measuring device 2 is, for example, a radio Doppler sensor.

The sensor 21 is a transducer capable of mutually converting an electric signal and a radio wave, and is, for example, a radio sensor that transmits and receives radio waves in the microwave band. The sensor 21 , for example, transmits radio waves to the target space 100 at predetermined time intervals (one second intervals, for example). The sensor 21 receives a reflected wave (radio wave) reflected by a person (including the target person) or the like existing in the target space 100 .

The signal processing circuit 22 performs signal processing on the output signal (electrical signal corresponding to the reflected wave) of the sensor 21 when the reflected wave is received, and detects the body movement of the subject existing in the target space 100 . generate dynamic data. Specifically, the signal processing circuit 22 uses the Doppler effect to compare the frequency of the received radio wave (reflected wave) with the frequency of the transmitted radio wave to obtain the movement speed of the subject's body. , to generate motion data. Here, the movement speed of the subject's body obtained by the signal processing circuit 22 includes a positive/negative sign to indicate whether the body is moving toward or away from the sensor 21. , contains information about the direction of movement of the body. In other words, the body motion data generated by the signal processing circuit 22 can be treated as a vector value including a "direction" component. As a result, the signal processing circuit 22 can determine whether the body is moving toward or away from the sensor 21, and can detect, for example, the subject getting up (getting out of bed). Furthermore, the acceleration obtained based on the body motion data may also be a vector value containing a "direction" component.

Further, in this embodiment, the signal processing circuit 22 has a function of extracting specific frequency components from the body motion data to generate measurement data indicating the heartbeat and respiratory state of the subject. Specifically, the signal processing circuit 22 filters the body motion data with a heartbeat filter and extracts the frequency component of the body motion caused by the heartbeat to obtain measurement data indicating the state of the heartbeat (hereinafter referred to as “heartbeat measurement data”). Furthermore, the signal processing circuit 22 filters the body motion data with a respiration filter and extracts the frequency component of the body motion caused by respiration to obtain measurement data indicating the state of respiration (hereinafter referred to as “respiration measurement data”). ). Here, the cycle in which the signal processing circuit 22 generates heartbeat measurement data and respiration measurement data is longer than the cycle in which the signal processing circuit 22 generates body motion data. As an example, the signal processing circuit 22 generates body motion data every second, and generates heartbeat measurement data and respiration measurement data every five seconds.

The signal processing circuit 22 also outputs sensor signals including body motion data, heart rate measurement data, and respiration measurement data to the sensor signal processing system 1 . In this embodiment, the signal processing circuit 22 is configured to be able to communicate with the sensor signal processing system 1 by wireless communication conforming to a wireless communication method such as Bluetooth (registered trademark). The measurement device 2 and the sensor signal processing system 1 are not limited to direct communication, and may be configured to communicate via a repeater, for example.

The measuring device 2 is installed in a private room 50 in which a target space 100 is set, as shown in FIG. In the example of FIG. 2, the private room 50 is provided with facilities such as a bed 51, a toilet 52, a washbasin 53, a sliding door 54 at the entrance, and a window 55, for example. Facilities such as a bed 51, a toilet 52, a washbasin 53, a sliding door 54 and a window 55 are not essential in the private room 50, and can be omitted as appropriate. An air conditioner 20 for adjusting the air environment in the private room 50 is installed on the wall of the private room 50 . In this embodiment, as an example, the measurement device 2 is arranged beside (on the side of) the air conditioner 20 .

Here, as an example, the measurement device 2 is installed in such a posture (orientation) that the target space 100 including at least the bed 51 is monitored. In this embodiment, the target space 100 is substantially the entire space inside the private room 50 except for the toilet 52 . The target space 100 is not limited to this example, and may be, for example, the entire space in the private room 50, and can be changed as appropriate.

The sensor signal processing system 1 includes a first arithmetic processing unit 11, an acquisition unit 13, a storage unit 14, and an output unit 15, as shown in FIG.

The acquisition unit 13 acquires body motion data relating to the body motion of the subject. That is, the acquisition unit 13 has a communication function with (the signal processing circuit 22 of) the measuring device 2 . In this embodiment, the acquisition unit 13 is configured to be able to communicate with the measuring device 2 through wireless communication conforming to a wireless communication method such as Bluetooth (registered trademark). The acquiring unit 13 acquires at least body motion data from the measuring device 2 by communicating with the measuring device 2 periodically or irregularly. Specifically, the acquisition unit 13 acquires the body motion data together with the heart rate measurement data and the respiration measurement data by receiving the sensor signal from the measurement device 2 . After acquiring data (body motion data, heartbeat measurement data, and respiration measurement data) from the measuring device 2 , the acquisition unit 13 outputs the acquired data to the first arithmetic processing unit 11 .

The first arithmetic processing unit 11 has at least the functions of the acceleration arithmetic unit 111 and the existence determination unit 112 . The first arithmetic processing unit 11 is implemented by, for example, a computer system including a processor and memory. Functions such as the acceleration calculation unit 111 and the presence determination unit 112 are realized by the processor of the first calculation processing unit 11 executing the program. The program may be recorded in advance in the memory of the first arithmetic processing unit 11 or the storage unit 14, or may be provided through an electric communication line such as the Internet or recorded in a non-temporary recording medium such as a memory card. may

The acceleration calculator 111 obtains the acceleration of the body motion of the subject based on the body motion data. That is, when at least part of the body of the subject moves, the rate of change in speed (moving speed) of at least part of the body of the subject with respect to time is obtained as acceleration by the acceleration calculator 111 . Here, the acceleration calculation unit 111 obtains acceleration according to a predetermined algorithm based on the body motion data obtained by the obtaining unit 13 from the measuring device 2, thereby obtaining data indicating acceleration (hereinafter referred to as “acceleration data”). Execute the acceleration calculation process to be generated. Acceleration calculation processing in the acceleration calculation unit 111 will be described in detail in the section “(2.2.2) Acceleration calculation processing”.

The presence determination unit 112 determines presence/absence of the target person in the target space 100 based on the body motion data. The presence determination unit 112 uses the body motion data acquired by the acquisition unit 13 from the measuring device 2 and follows a predetermined algorithm to perform presence determination processing for generating data indicating the presence or absence of the subject (hereinafter referred to as “presence data”). to run.

In this embodiment, the presence determining unit 112 obtains an analysis model for time-series analysis in which body motion data acquired at a predetermined timing is represented by a plurality of pieces of body motion data acquired before the predetermined timing. Analytical processing. The acquisition unit 13 acquires measurement data from the measurement device 2 every second, for example. As an example, in the time-series analysis process, an analysis model for time-series analysis is obtained in which body motion data acquired at a predetermined timing is represented by a plurality of (for example, 30) pieces of body motion data before the predetermined timing. In the present embodiment, the presence determining unit 112 uses, for example, an auto-regressive (AR) model, and body motion data acquired at a predetermined timing is represented by 30 body motion data for the past 30 seconds. Obtain an analytical model of the correlation function. The analysis model for the time series analysis performed by the presence determination unit 112 is not limited to the autoregression model, and may be another analysis model such as an extended Kalman model, and the analysis model can be appropriately changed in consideration of the amount of calculation.

The existence determination unit 112 determines whether or not the subject is present at a predetermined timing based on the determination conditions including the conditions regarding the coefficients of the analysis model obtained in the time-series analysis process. The presence determination unit 112 satisfies the determination condition that, for example, the coefficient of the analysis model obtained by the time-series analysis process exceeds a predetermined threshold value, or the size of the acquired body motion data exceeds a predetermined determination value. Based on this, it is determined whether or not the target person exists at a predetermined timing. That is, when the first-order coefficient of the autoregressive model exceeds the threshold value or the size of the body motion data exceeds the determination value, the presence determining unit 112 determines that the target person exists in the target space 100 (in a room-occupied state). ). The presence determining unit 112 determines that the target person does not exist in the target space 100 (absent state) when the primary coefficient of the autoregressive model is equal to or less than the threshold and the size of the measurement data is equal to or less than the determination value. judge.

As described above, in the event prediction system 10 according to the present embodiment, the presence determination unit 112 determines the presence or absence of the subject at a predetermined timing based on the determination condition regarding the coefficient of the analysis model obtained in the time series analysis process. Judging. Therefore, the presence determining unit 112 is less likely to be affected by temporary fluctuations in the measurement data, and the accuracy of determination by the presence determining unit 112 can be improved.

Furthermore, the presence determination unit 112 determines whether or not the subject is present in the target space 100 using at least one of the heartbeat measurement data, respiration measurement data, and acceleration data instead of or together with the body motion data. may In other words, the existence determination unit 112 may perform processing for determining the presence or absence of the target person in the target space 100 based on at least one of body motion data, heartbeat measurement data, respiration measurement data, and acceleration data. Heartbeat measurement data, respiration measurement data, and acceleration data are all based on body motion data. Therefore, for example, even when determining the presence or absence of the target person based on the acceleration data, the presence determination unit 112 indirectly determines the presence or absence of the target person in the target space 100 based on the body motion data. In short, the presence determination unit 112 may directly or indirectly determine the presence or absence of the target person in the target space 100 based on the body motion data.

Here, when determining the presence or absence of the target person based on the acceleration data, the presence determination unit 112 can also identify the individual and determine the presence or absence of the target person. That is, the presence determination unit 112 can identify whether or not the person is the target person (the resident of the private room 50) from the acceleration data. As a result, when a person other than the target person (for example, facility staff) is in the target space 100, the presence determination unit 112 determines that the target person does not exist in the target space 100 or exists.

The second arithmetic processing unit 12 has at least the functions of the sign determination unit 121 and the state determination unit 122 . Like the first arithmetic processing unit 11, the second arithmetic processing unit 12 is realized by, for example, a computer system including a processor and memory. The functions of the symptom determination unit 121, the state determination unit 122, and the like are realized by the processor of the second arithmetic processing unit 12 executing the programs. The program may be recorded in advance in the memory or storage unit 14 of the second arithmetic processing unit 12, or may be provided by being recorded in a non-temporary recording medium such as a memory card or through a telecommunication line such as the Internet. may In this embodiment, the second arithmetic processing unit 12 is configured by a computer system different from the first arithmetic processing unit 11 .

Output data of the first arithmetic processing unit 11 is input to the second arithmetic processing unit 12 . In this embodiment, the output data of the first arithmetic processing unit 11 includes body motion data, heart rate measurement data, and respiration measurement data acquired by the acquiring unit 13 from the measuring device 2 . Furthermore, in the present embodiment, the output data of the first arithmetic processing unit 11 includes acceleration data (acceleration data) regarding the movement of the subject's body, which is obtained by the acceleration arithmetic unit 111 based on the body motion data. I'm in. Moreover, the output data of the first arithmetic processing unit 11 includes data (presence/absence data) indicating the determination result of the presence determination unit 112 . Therefore, in addition to the body motion data, heartbeat measurement data, respiration measurement data, acceleration data, and presence/absence data are input to the second arithmetic processing unit 12 . Heartbeat measurement data, respiration measurement data, and acceleration data are all obtained based on body motion data, in other words, data based on body motion data.

The portent determination unit 121 determines a portent of occurrence of a specific event related to the subject based on the body motion data. That is, the sign determination unit 121 executes sign determination processing for determining a sign of occurrence of a specific event such as terminal care according to a predetermined algorithm based on the body motion data acquired by the acquisition unit 13 from the measuring device 2. do. Here, the portent determination unit 121 determines a portent based on body motion data acquired during the past reference period. The “reference period” referred to in the present disclosure is a period in the past (before the reference time point) when the time point at which the portent determination unit 121 performs determination is set as the reference time point. In this embodiment, the reference period is a period of a certain period of time (eg, 30 minutes, 1 hour, 1 day, etc.) ending at the reference time. In other words, the portent determination unit 121 obtains a set of body motion data acquired by the acquisition unit 13 during a reference period starting from a time point a certain time before the reference time point and ending at the reference time point. A sign determination process is executed based on the data.

Furthermore, similarly to the existence determination unit 112, the sign determination unit 121 also uses at least one data of the heartbeat measurement data, the respiration measurement data, and the acceleration data instead of or together with the body movement data to detect the specific event. may be determined. That is, the portent determination unit 121 detects a portent of the occurrence of the specific event based on at least one of the body motion data, the heart rate measurement data, the respiration measurement data, and the acceleration data included in the output data of the first arithmetic processing unit 11. You may perform the symptom determination process to determine. Heartbeat measurement data, respiration measurement data, and acceleration data are all based on body motion data. In short, the portent determination unit 121 may directly or indirectly determine a portent of occurrence of a specific event based on body motion data.

The determination result of the portent determination unit 121 is output to the output unit 15 as portent data. In this embodiment, the portent determination unit 121 determines only the presence or absence of a portent of occurrence of a specific event. In other words, the portent determination unit 121 determines whether or not there is a portent of the occurrence of the specific event without distinguishing between the types of specific events for which the portent is determined. Therefore, the portent data generated by the portent determination unit 121 indicates whether or not there is a portent of the occurrence of the specific event.

In addition, the portent determination unit 121 determines a portent based on changes in the amount of physical activity of the subject based on the body motion data. The “amount of physical activity” referred to in the present disclosure is represented by the product of the strength (intensity) of physical activity and the duration of the physical activity. "Physical activity" in the present disclosure is an activity performed by a subject, not limited to exercise for the purpose of maintaining or improving physical strength, but all activities (actions) that consume more energy than a resting state means Details of the sign determination processing in the sign determination unit 121 will be described in the section “(2.2.3) Sign determination processing”.

The state determination unit 122 determines the subject's current state, such as the subject's health condition, sleep state, and mental state, based on the body motion data. That is, the state determination unit 122 determines the current state of the subject, such as good physical condition or poor physical condition, according to a predetermined algorithm based on the body motion data acquired by the acquisition unit 13 from the measurement device 2. Execute the process.

Furthermore, similarly to the sign determination unit 121, the state determination unit 122 uses at least one data of the heartbeat measurement data, the respiration measurement data, and the acceleration data instead of or together with the body movement data to may determine the current state of In other words, the state determination unit 122 determines the current state of the subject based on at least one of the body motion data, the heart rate measurement data, the respiration measurement data, and the acceleration data included in the output data of the first arithmetic processing unit 11. You may perform the state determination processing to determine. In short, the state determination unit 122 may directly or indirectly determine the subject's current state based on the body motion data.

The determination result of the state determination unit 122 is output to the output unit 15 as state data. Furthermore, output data (body motion data, heart rate measurement data, respiration measurement data, acceleration data, and presence/absence data) of the first arithmetic processing unit 11 is also output from the second arithmetic processing unit 12 to the output unit 15 .

The storage unit 14 includes, for example, electrically rewritable nonvolatile memory such as EEPROM (Electrically Erasable Programmable Read Only Memory) and volatile memory such as RAM (Random Access Memory). The storage unit 14 stores body motion data, heartbeat measurement data, respiration measurement data, and the like acquired by the acquisition unit 13 . Here, the storage unit 14 stores at least body motion data over the reference period. Furthermore, the storage unit 14 also stores the calculation results of the first calculation processing unit 11 (including acceleration data and presence/absence data), the calculation results of the second calculation processing unit 12 (including predictor data and state data), and the like.

The output unit 15 outputs the judgment result of the symptom judgment unit 121 . Furthermore, the output unit 15 outputs the determination results of the existence determination unit 112 and the state determination unit 122 . Here, the output unit 15 has a communication function with devices such as the display 3 and the information terminal 4, for example. The information terminal 4 is, for example, a smart phone, a tablet terminal, or a personal computer. The output unit 15 outputs data input from the second arithmetic processing unit 12 to these devices.

That is, the presence/absence data generated by the presence determination unit 112, the sign data generated by the sign determination unit 121, and the state data generated by the state determination unit 122 are output from the output unit 15 to the display 3, the information terminal 4, and the like. be done. The output unit 15 may output the determination result when there is a change in the determination result of the presence determination unit 112, the sign determination unit 121, or the state determination unit 122, or, for example, when requested by the information terminal 4 You may output the judgment result to .

As a result, the determination result of the presence determination unit 112, the determination result of the sign determination unit 121, and the determination result of the state determination unit 122 are presented to the manager on the display 3, the information terminal 4, and the like. The manager can check the determination results of the presence determination unit 112, the symptom determination unit 121, and the state determination unit 122 on the display 3, the information terminal 4, and the like. The mode of presentation on the display 3 and the information terminal 4 may be, for example, display, voice output, printout (printing), writing to a non-temporary recording medium, transmission to another information terminal, etc. good.

Moreover, the output unit 15 may output the calculation result (acceleration data) of the acceleration calculation unit 111 . Furthermore, the configuration of the output unit 15 is not limited to outputting the determination result to devices such as the display 3 and the information terminal 4 through communication. For example, the output unit 15 itself may output the determination result by display, voice output, printout (printing), writing to a non-temporary recording medium, transmission to an information terminal, or the like.

(2.2) Operation (2.2.1) Overall operation The overall operation of the sensor signal processing system 1 according to the present embodiment and the event prediction system 10 including the same will be described with reference to the flowchart of FIG. do.

The acquiring unit 13 regularly or irregularly performs acquisition processing for acquiring body motion data from the measuring device 2 (S1). In this embodiment, as an example, the acquiring unit 13 acquires body motion data, heart rate measurement data, and respiration measurement data from the measuring device 2 every second. The acquisition unit 13 outputs data (body motion data, heartbeat measurement data, and respiration measurement data) acquired from the measurement device 2 to the first arithmetic processing unit 11 . Here, it is assumed that the measuring device 2 updates body motion data every second, and updates heart rate measurement data and respiration measurement data every five seconds. Therefore, the heartbeat measurement data and respiration measurement data acquired by the acquisition unit 13 from the measurement device 2 are updated every five seconds.

When the data (body movement data, heart rate measurement data, and respiratory measurement data) are input from the acquisition unit 13, the first arithmetic processing unit 11 performs preprocessing such as noise reduction and moving average calculation on these data. (S2). The first arithmetic processing unit 11 stores the pre-processed data in the storage unit 14 .

The first arithmetic processing unit 11 performs acceleration arithmetic processing in the acceleration arithmetic processing unit 111 based on the body motion data after the pre-processing, and obtains acceleration data indicating the acceleration (S3). Details of the acceleration calculation process will be described in the section "(2.2.2) Acceleration calculation process".

Next, the first arithmetic processing unit 11 executes presence determination processing in the presence determination unit 112 based on the body motion data after the preprocessing, and determines the presence or absence of the target person in the target space 100 by time-series analysis. Determine (S4).

As a result of the existence determination process, if it is determined that the target person exists in the target space 100 (S5: Yes), the second arithmetic processing unit 12 performs the sign determination unit 121 based on the body movement data after the preprocessing. , to determine a sign of occurrence of a specific event (S6). Details of the portent determination processing will be described in the section “(2.2.3) portent determination processing”.

Next, the second arithmetic processing unit 12 executes state determination processing in the state determination unit 122 based on the body motion data after the pre-processing, and determines the current state of the subject (S7).

Then, the output unit 15 executes an output process of outputting the determination result of the presence determination unit 112, the determination result of the sign determination unit 121, and the determination result of the state determination unit 122 (S8). Further, if it is determined that the target person does not exist in the target space 100 as a result of the existence determination process (S5: No), the sign determination process and the state determination process are skipped, and the process proceeds to the output process. In this case, the output unit 15 outputs only the determination result of the existence determination unit 112 in the output process.

The sensor signal processing system 1 and the event prediction system 10 provided with it perform the processing from step S1 to step S8, for example, at intervals of one second.

However, the order of processing shown in FIG. 3 is merely an example, and the order of processing can be changed as appropriate.

(2.2.2) Acceleration Calculation Processing The acceleration calculation processing (S3 in FIG. 3) performed by the acceleration calculation unit 111 will be described.

In the acceleration calculation process, the acceleration calculation unit 111 obtains acceleration by performing differentiation processing on a plurality of pieces of body motion data arranged in time series. Specifically, the acceleration calculation unit 111 performs a differential operation on a plurality of pieces of body motion data arranged in time series and stored in the storage unit 14, so that the subject's body motion occurring in a predetermined period is Find the acceleration of

That is, the acceleration is the rate of change in speed (moving speed) of at least part of the subject's body with respect to time when at least part of the subject's body moves. On the other hand, the body motion data is data reflecting the speed of motion of the subject. Therefore, in the acceleration calculation process, the acceleration calculation unit 111 obtains the differential value of the body motion data as the acceleration and generates the acceleration data. As a result, when there is no time for the movement speed of the subject reflected in the body movement data, that is, when the movement speed is constant, the acceleration obtained by the acceleration calculation process becomes zero. The greater the change over time in the movement speed of the subject reflected in the body motion data, the greater the acceleration obtained by the acceleration calculation process.

FIG. 4 is a flowchart showing a specific example of acceleration calculation processing.

As shown in FIG. 4, when the acceleration calculation process starts, the acceleration calculator 111 calculates the time difference dt between two adjacent pieces of body motion data among the plurality of pieces of body motion data arranged in time series (S31). Here, attention is focused on the body motion data acquired at the predetermined timing "t" and the body motion data acquired at the timing "t-1" one before the predetermined timing. In this case, the acceleration calculation unit 111 calculates the difference between the “time (t)” corresponding to the predetermined timing “t” and the “time (t−1)” corresponding to the timing “t−1” one before that. Let the difference be the time difference dt.

Next, the acceleration calculation unit 111 calculates the amount of change dv in the value (magnitude) of body motion data between two adjacent body motion data (S32). At this time, the acceleration calculation unit 111 calculates the "body motion (t)" acquired at a predetermined timing "t" and the "body motion (t-1)" acquired at the timing "t-1" )” is the amount of change dv.

Next, the acceleration calculator 111 calculates acceleration data by dividing the variation dv by the time difference dt (S33). As a result, the acceleration calculator 111 can specify the absolute value of the acceleration (S34), and can also specify the positive/negative (sign) of the acceleration (S35). The absolute value of acceleration corresponds to the magnitude of motion of the subject. The positive/negative of the acceleration depends on the moving direction of the subject (the direction in which the body approaches or leaves the sensor 21) and the acceleration/deceleration of the subject.

However, the order of processing shown in FIG. 4 is merely an example, and the order of processing can be changed as appropriate.

Here, the body motion data to be differentiated may be body motion data before preprocessing, or body motion data after preprocessing such as noise reduction and moving average calculation. good too. By executing the acceleration calculation process using the body motion data after the preprocessing, it is possible to suppress the influence of noise.

(2.2.3) Omen Judgment Processing The omen judgment processing (S6 in FIG. 3) performed by the omen judgment unit 121 will be described.

In the sign determination process, the sign determination unit 121 determines a sign of occurrence of a specific event based on body motion data acquired during the past reference period. Specifically, the portent determination unit 121 analyzes the magnitude, the amount of change, the frequency of change, the time period of change, etc. for a plurality of pieces of body motion data arranged in time series stored in the storage unit 14, and analyzes the data. A sign of occurrence of a specific event is determined from the result.

That is, before a specific event occurs, a characteristic tendency often appears in the subject's body motion data. In particular, characteristic tendencies tend to appear in time-series data of body motion data. Therefore, in the portent determination process, the portent determination unit 121 analyzes the time-series data of the body motion data, such as the average value (magnitude), the amount of variation, the frequency of change, the time period of change, and the like, and determines whether the specific event has occurred. predictor and generate predictor data. As a result, for example, when a characteristic tendency such as the average value of the body motion data continuously and significantly decreasing appears, it is determined in the sign determination process that there is a sign of the occurrence of terminal care.

FIG. 5 is a flow chart showing a specific example of the sign determination process.

As shown in FIG. 5, when the sign determination process starts, the sign determination unit 121 first sets various variables (S61). The variables set at this time include a threshold value V1 to be compared with statistical values such as body motion data, the start date of predictor determination, an aggregation period, a specified value N1, which will be described later, and the like.

Next, the portent determination unit 121 calculates statistical values such as body motion data for the aggregation period after the start date (S62). The "statistical value" referred to in the present disclosure is a value statistically obtained from time-series data such as body movement data (including acceleration data) during the aggregation period, for example, the average value of body movement data (moving average value ), median, mode, minimum and variance.

The portent determination unit 121 compares the calculated statistical value with the threshold value V1 (S63). If the statistical value is less than the threshold value V1 (S63: Yes), the sign determination unit 121 changes the value of the alert flag from "0" to "1" (S64). Further, the portent determination unit 121 calculates the cumulative value of alert flags with a value of "1" within the counting period (S65).

The sign determination unit 121 compares the calculated cumulative value of the alert flags with the specified value N1 (S66). If the cumulative value is equal to the prescribed value N1 (S66: Yes), the portent determination unit 121 determines that there is a “prediction” of the specific event (S67). On the other hand, if the statistical value is equal to or greater than the threshold value V1 (S63: No) or if the cumulative value is less than the specified value N1 (S66: No), the sign determination unit 121 determines that there is no sign of the specific event ( S68).

However, the order of processing shown in FIG. 5 is merely an example, and the order of processing can be changed as appropriate.

The flowchart shown in FIG. 5 shows a sign determination process for a specific event that occurs over a long period of time (for example, several days to several weeks or longer) such as a disease requiring hospitalization. However, it is not limited to this type of specific event. For example, it is also possible to implement predictive judgment processing for specific events that occur in a short period of time (for example, several minutes to several hours or less) such as getting up from a sleeping state (getting out of bed). is. For specific events that occur in a short period of time (for example, several minutes to several hours or less), for example, the direction of the inequality sign in the process of comparing the statistical value with the threshold value V1 (S63) is opposite to the example in FIG. . Further, which of the long-term specific event and the short-term specific event is to be used as the specific event in the sign determination process can be switched by, for example, changing the length of the reference period set in step S61. be. As an example, if the reference period is set to 60 days, the portent determination unit 121 can determine whether or not there is a portent of the occurrence of a long-term specific event from the change point of the 60-day cycle. On the other hand, if the reference period is set to 10 seconds, the portent determination unit 121 can determine whether or not there is a portent of the occurrence of a short-term specific event from the change point of the 10-second cycle.

Next, as an example, the symptom determination processing in the case illustrated in FIG. 6 will be described. FIG. 6 is a graph showing changes in body motion data, with the horizontal axis being the time axis and the vertical axis being the average value (moving average value) of the body motion data. In FIG. 6, one scale of the horizontal axis corresponds to "5 days".

In the example of FIG. 6, before time t1, the average value of the body motion data maintains the threshold value V1 or more, but after the time point t1, the average value of the body motion data becomes less than the threshold value V1. In other words, at time t1, the average value of the body motion data falls below the threshold V1, so that a different tendency from before time t1 appears. In this example, the test subject (subject) has an illness requiring hospitalization several days after time t1. In other words, a sign of a specific event of illness requiring hospitalization appears as a characteristic trend of the subject's body motion data at time t1.

In particular, in this embodiment, as described above, the portent determination unit 121 can determine a portent based on changes in the amount of physical activity of the subject based on body motion data. Here, according to the equation of motion (F=ma), the product of acceleration and mass is force (F), and in physical activity using the whole body of the subject, the weight of the subject corresponds to mass. Assuming that the weight gain and loss of the subject is negligible, the acceleration is considered to correspond to the muscle force exerted by the subject during physical activity. Therefore, the acceleration obtained by the acceleration calculator 111 corresponds to the intensity of physical activity, and the amount of physical activity over a certain period of time is represented by the cumulative value of acceleration over a certain period of time. Therefore, in the present embodiment, in the sign determination process, the sign determination unit 121 determines a sign of the occurrence of the specific event based on the acceleration data obtained by the acceleration calculation process.

Judgment conditions in the sign judgment process include, for example, the following first to fourth conditions. The first condition is that the state in which the amount of physical activity is equal to or less than the first threshold continues for a first period of time (for example, several days or several weeks). The second condition is that the amount of decrease in the amount of physical activity during a second period of time (for example, several tens of minutes or several hours) is greater than or equal to the second threshold. The third condition is that the amount of physical activity is equal to or less than the third threshold (smaller than the first threshold). The fourth condition is that the difference between the average value of the amount of physical activity at rest during sleep and the average value of the amount of physical activity at non-rest is equal to or less than the fourth threshold. Conditional expressions and parameters (including thresholds and the like) for realizing these determination conditions are stored in the storage unit 14, for example.

As another example, in the portent determination process, body motion data may be used as vector values to determine a portent of the occurrence of a specific event. In this case, the portent determination unit 121 analyzes acceleration data as a vector value including a “direction” component in addition to a scalar quantity such as an average value of body motion data. As a result, for example, when the movement of the subject in the private room 50 in the direction of leaving the private room 50 is detected, the portent determination unit 121 determines that there is a portent of the occurrence of a specific event such as loitering or a decline in cognitive function. do.

Also, the determination conditions (including the magnitude of the threshold value, the length of the period, etc.) used in the sign determination process may be set for each subject. Therefore, in a serviced housing for the elderly having a plurality of private rooms 50 , a determination condition used in the sign determination process may be set for each private room 50 .

(3) Modifications Embodiment 1 is merely one of various embodiments of the present disclosure. Embodiment 1 can be modified in various ways according to design and the like as long as the object of the present disclosure can be achieved. A function similar to that of the sensor signal processing system 1 may be embodied by a sensor signal processing method, a (computer) program, a non-temporary recording medium recording the program, or the like. A sensor signal processing method according to one aspect acquires body motion data from a measuring device 2 that outputs body motion data related to body motion of a subject (step S1 in FIG. 3), and based on the body motion data, Acceleration of body motion is obtained (step S3 in FIG. 3). A program according to one aspect is a program for causing a computer system to execute the sensor signal processing method. Also, functions similar to those of the event prediction system 10 may be embodied by an event prediction method, a (computer) program, or a non-temporary recording medium recording the program. An event prediction method according to one aspect acquires body motion data relating to body motion of a subject (step S1 in FIG. 3), and determines a sign of occurrence of a specific event related to the subject based on the body motion data. (Step S6 in FIG. 3). A program according to one aspect is a program for causing a computer system to execute the event prediction method.

Modifications of the first embodiment are listed below. Modifications described below can be applied in combination as appropriate.

The execution subject of the sensor signal processing system 1, the event prediction system 10, the sensor signal processing method, or the event prediction method in the present disclosure includes a computer system. A computer system is mainly composed of a processor and a memory as hardware. The processor executes the program recorded in the memory of the computer system, thereby realizing the function as the execution subject of the sensor signal processing system 1, the event prediction system 10, and the sensor signal processing method or event prediction method in the present disclosure. be. The program may be recorded in advance in the memory of the computer system, may be provided through an electric communication line, or may be recorded in a non-temporary recording medium such as a computer system-readable memory card, optical disk, or hard disk drive. may be provided. A processor in a computer system consists of one or more electronic circuits, including semiconductor integrated circuits (ICs) or large scale integrated circuits (LSIs). A plurality of electronic circuits may be integrated into one chip, or may be distributed over a plurality of chips. A plurality of chips may be integrated in one device, or may be distributed in a plurality of devices.

In addition, it is not an essential configuration for the sensor signal processing system 1 that a plurality of functions in the sensor signal processing system 1 are aggregated in one housing, and the components of the sensor signal processing system 1 include a plurality of housings may be provided dispersedly. Furthermore, at least part of the functions of the sensor signal processing system 1 may be implemented by, for example, a server device and a cloud (cloud computing). Similarly, it is not an essential configuration of the event prediction system 10 that a plurality of functions of the event prediction system 10 are integrated in one housing, and the components of the event prediction system 10 are distributed over a plurality of housings. may be provided as Furthermore, at least part of the functions of the event prediction system 10 may be realized by, for example, a server device and cloud (cloud computing). Conversely, in Embodiment 1, functions distributed across multiple devices, such as the measurement device 2 and the sensor signal processing system 1, may be integrated into one housing.

In addition, the facility in which the target space 100 is set is not limited to facilities such as serviced housing for the elderly, nursing care facilities, and hospitals where employees reside, but may be childcare facilities such as daycare centers. In this case, the "subject" is the infant or toddler being cared for. Furthermore, the facility in which the target space 100 is set may be a general residential (single-family house or collective housing) facility. In this case, the "subject" is a resident (resident) of the housing facility. In a situation where the subject lives alone in the facility where the target space 100 is set, the determination result of the event prediction system 10 is, for example, the subject's family living away from the subject, or Local community supporters (care managers, life counselors, etc.) are notified.

In addition, the measuring device 2 may be configured to output at least body motion data to the sensor signal processing system 1, and outputting data other than body motion data (heartbeat measurement data and respiratory measurement data) is necessary for sensor signal processing. This is not an essential configuration for system 1.

Moreover, the measuring device 2 is not limited to the radio wave Doppler sensor, and may be, for example, an ultrasonic Doppler sensor that transmits ultrasonic waves. Furthermore, the measuring device 2 may generate body motion data relating to the body motion of the subject in the target space 100 . Therefore, the measuring device 2 may be a sensor other than the Doppler sensor, such as a radio wave sensor using a frequency modulated continuous wave radar system, a TOF (Time Of Flight) system, or a sensor using an image sensor. Furthermore, the measuring device 2 is not limited to a non-contact type sensor that generates body motion data related to the body motion of the subject in a non-contact manner. sensor.

Further, the portent determination unit 121 may determine a portent of occurrence of a specific event based on at least body motion data. It is not an essential configuration for the prediction system 10 . If the sign determination unit 121 does not use acceleration data, the acceleration calculation unit 111 can be omitted as appropriate.

In addition, the provision of the acquisition unit 13, the storage unit 14, and the output unit 15 in the sensor signal processing system 1 is not an essential configuration of the event prediction system 10, and the event prediction system 10 is provided separately from the sensor signal processing system 1. , an acquisition unit 13 , a storage unit 14 and an output unit 15 . The storage unit 14 and the output unit 15 are not essential components of the event prediction system 10, and at least one of the storage unit 14 and the output unit 15 may be omitted as appropriate.

Further, the acceleration data generated by the acceleration calculation unit 111 does not necessarily have to be used by the sign determination unit 121, the state determination unit 122, etc., and may be used only for determination by the existence determination unit 112, for example. Furthermore, the calculation result of the acceleration calculation unit 111 may be output by the output unit 15 only. In this case, the sign determination unit 121 and the state determination unit 122 are not essential components of the event prediction system 10, and at least one of the sign determination unit 121 and the state determination unit 122 may be omitted as appropriate.

Moreover, the communication method between the measuring device 2 and the sensor signal processing system 1 is not limited to wireless communication, and may be wired communication (including power line carrier communication).

Also, the first arithmetic processing unit 11 and the second arithmetic processing unit 12 may be configured by one computer system. Conversely, the first arithmetic processing unit 11 and the second arithmetic processing unit 12 may be configured by three or more computer systems.

Also, a technique such as machine learning may be used for part of the processing of the sensor signal processing system 1 and the event prediction system 10, such as acceleration calculation processing and sign determination processing. For example, it is preferable to use a technique such as machine learning as a criterion for judging a sign based on body motion data acquired in the past reference period.

As an example, when determining the threshold V1, in addition to the basal metabolic rate that can be estimated from the age, height, weight, etc. of the subject (subject), using machine learning from the measured respiratory rate or heart rate, etc., the threshold V1 may be determined. Also, the threshold V1 is not limited to a constant value, and the threshold V1 may have a gradient with respect to the time axis.

Further, machine learning may be used when setting the time range for calculating the specified value N1 or the statistical value based on age, height, weight, amount of meals, or the like.

In Embodiment 1, in the binary comparison, "greater than or equal to" includes both the case where the two values are equal and the case where one of the two values exceeds the other, but is not limited to this. It may be "greater than". In other words, whether or not two values are equal can be arbitrarily changed depending on the setting of the reference value, etc., so there is no technical difference between "greater than" and "greater than or equal to". Similarly, "less than" may be "less than".

(Embodiment 2)
The event prediction system 10 according to the present embodiment differs from the event prediction system 10 according to the first embodiment in the content of the symptom determination processing in the symptom determination unit 121 . In the following, configurations similar to those of the first embodiment are denoted by common reference numerals, and descriptions thereof are omitted as appropriate.

In the present embodiment, the portent determination unit 121 distinguishes between types of specific events for determining portents.
In other words, the portent determination unit 121 not only determines whether or not there is a portent of the occurrence of the specific event, but also discriminates and determines the type of the specific event. That is, when the sign determination unit 121 determines that there is a sign of a specific event, the sign determination unit 121 determines which type of specific event is determined to be a sign of occurrence among various events that may occur in relation to the subject. Identify.

Examples of specific event types include terminal care, falls while walking, illness or injury requiring hospitalization, death, cognitive decline, wandering, getting up from a sleeping state (getting out of bed), excretion, and going to bed. be. However, specific events are not limited to specific classifications such as terminal care and falls while walking. It may be an abstract classification such as whether the event occurs within a certain period of time). Therefore, the portent data generated by the portent determination unit 121 indicates whether or not there is a portent of occurrence of a specific event, and if there is a portent of occurrence, the type of the specific event.

Specifically, the portent determination unit 121 determines a portent of the occurrence of the specific event by distinguishing even the type of the specific event using determination conditions associated with each type of the specific event. The determination conditions are stored in the storage unit 14, for example. Here, in the symptom determination process, a score is calculated for each determination condition, and when a plurality of determination conditions are satisfied, for example, the specific event corresponding to the determination condition with the highest score may be adopted.

In addition, the portent determination unit 121 is not limited to the configuration for identifying which type of specific event is determined to be “predicted” after determining that the specific event “has a portent”. After that, it may be determined whether or not there is a sign of occurrence of the specific event. In this case, it is possible to narrow down the specific events to be subjected to the determination of the presence or absence of a sign. For example, when a specific event such as terminal care is to be determined as to whether or not there is an omen, the omen determination unit 121 does not determine whether or not there is an omen of occurrence other than the target specific event, such as deterioration of cognitive function.

In addition, the determination conditions used in the sign determination process may include conditions related to external information input to the event prediction system 10 from outside the event prediction system 10, such as the subject's care level, past history, and care records. good. As a result, the portent determination unit 121 can improve the determination accuracy of the portent determination process based on information such as the level of care, past history, and care record.

Further, based on the external information as described above, for example, the output unit 15 may change the order and mode of presenting the determination result of the symptom determination unit 121 to the manager or the like. As an example, when it is determined that a plurality of subjects are “presence of a sign” of a specific event at approximately the same time, the output unit 15 outputs the determination results in descending order of priority for the subjects based on the external information about each subject. may be presented.

Further, the portent determination unit 121 may change the determination conditions used in the portent determination process for each subject according to past body motion data, determination results, or the like. Thereby, the portent determination unit 121 can determine a portent of occurrence of a specific event in accordance with a change in life rhythm and a change in the physical condition of each subject, thereby improving the determination accuracy of the portent determination process.

Further, in the event prediction system 10, based on the determination result of the predictive determination process, the content of the care performed by the caregiver (the person who takes care of the subject) and the change in the state of the subject after the care is performed are detected. An event caused by the determination result of the determination process may be used as feedback information. As a result, based on the feedback information, the event prediction system 10 can, for example, propose care to be performed in response to the determination result, or confirm whether the determination result of the symptom determination process is correct or not.

Further, the event prediction system 10 may output the determination result of the symptom determination process and the like to an external nursing care work system or the like. Furthermore, the output unit 15 may change the output destination of the determination result of the symptom determination process according to the determination result of the symptom determination process. As a result, depending on which type of specific event is determined to be “predicted”, the output unit 15 can determine the predictor determination process to an appropriate person, such as a doctor, nurse, or caregiver, according to the specific event. I can present the results.

The configuration described in the second embodiment can be applied in appropriate combination with the various configurations (including modifications) described in the first embodiment.

(summary)
As described above, the event prediction system (10) according to the first aspect includes an acquisition section (13) and a sign determination section (121). An acquisition unit (13) acquires body motion data relating to body motion of a subject. A portent determination unit (121) determines a portent of occurrence of a specific event related to the subject based on the body motion data.

According to this aspect, it is possible to determine a sign of occurrence of a specific event that can occur in relation to the subject, that is, a "sign" that appears before the occurrence of the specific event, based on the body motion data. Therefore, according to the event prediction system (10), there is an advantage that it is possible to determine even a sign of occurrence of a specific event related to the subject.

In the event prediction system (10) according to the second aspect, in the first aspect, the portent determination unit (121) determines a portent based on body motion data acquired during the past reference period.

According to this aspect, by using the body motion data acquired in the past reference period, it is possible to relatively determine a sign. Decrease can be suppressed.

In the event prediction system (10) according to the third aspect, in the first or second aspect, the sign determination unit (121) determines a sign judge.

According to this aspect, it is possible to improve the accuracy of predictor determination. That is, since the sign of the occurrence of the specific event is likely to appear in the change in the amount of physical activity of the subject, the sign determination unit (121) makes a determination based on such a change in the amount of physical activity, thereby increasing the determination accuracy. can be improved.

An event prediction system (10) according to a fourth aspect, in any one of the first to third aspects, further comprises a presence determining section (112). A presence determination unit (112) determines presence/absence of a target person in the target space (100) based on the body motion data.

According to this aspect, the body motion data can be used both for determination of a sign of the occurrence of the specific event and determination of the presence or absence of the subject in the subject space (100).

An event prediction system (10) according to a fifth aspect, in any one of the first to fourth aspects, further comprises an output section (15). The output unit (15) outputs the determination result of the sign determination unit (121).

According to this aspect, it is possible to notify the manager or the like of the determination result of the symptom determination unit (121), for example.

In the event prediction system (10) according to the sixth aspect, in any one of the first to fifth aspects, the portent determination section (121) distinguishes the types of specific events for which portents are determined.

According to this aspect, not only the presence or absence of a sign of the occurrence of the specific event but also the type of the specific event for which the sign is determined is distinguished, so the judgment result of the sign judgment unit (121) can be easily analyzed.

An event prediction method according to a seventh aspect acquires body motion data relating to body motion of a subject, and determines a sign of occurrence of a specific event related to the subject based on the body motion data.

According to this aspect, it is possible to determine a sign of occurrence of a specific event that can occur in relation to the subject, that is, a "sign" that appears before the occurrence of the specific event, based on the body motion data. Therefore, according to the event prediction method, there is an advantage that it is possible to determine even a sign of occurrence of a specific event related to the subject.

A program according to an eighth aspect is a program for causing a computer system to execute the event prediction method according to the seventh aspect.

According to this aspect, it is possible to determine a sign of occurrence of a specific event that can occur in relation to the subject, that is, a "sign" that appears before the occurrence of the specific event, based on the body motion data. Therefore, according to the above program, there is an advantage that it is possible to determine even a sign of occurrence of a specific event related to the subject.

Various configurations (including modifications) of the event prediction system (10) according to Embodiments 1 and 2, without being limited to the above aspects, can be implemented by an event prediction method, a program, and a non-temporary recording medium recording the program. Realization is possible.

The configurations according to the second to sixth aspects are not essential configurations for the event prediction system (10) according to the first aspect, and can be omitted as appropriate.

A sensor signal processing system (1) according to a ninth aspect includes an acquisition section (13) and an acceleration calculation section (111). An acquisition unit (13) acquires body motion data from the measurement device (2). The measuring device (2) outputs body motion data relating to the subject's body motion. An acceleration calculator (111) obtains the acceleration of the subject's body motion based on the body motion data.

According to this aspect, the acceleration of the subject's body movement can be obtained based on the body movement data acquired from the measuring device (2). According to the equation of motion, the product of acceleration and mass is force, and in physical activity using the whole body of the subject, the weight of the subject corresponds to mass. Assuming that the weight gain and loss of the subject is negligible, the acceleration is considered to correspond to the muscle force exerted by the subject during physical activity. Therefore, the acceleration obtained by the acceleration calculator (111) corresponds to the intensity of physical activity, and the amount of physical activity over a certain period of time is represented by the cumulative value of acceleration over a certain period of time. Therefore, in the sensor signal processing system (1), the amount of physical activity can be quantitatively evaluated by obtaining the acceleration from the body motion data, and the detection accuracy of the amount of physical activity can be improved.

In the sensor signal processing system (1) according to the tenth aspect, in the ninth aspect, the acceleration calculator (111) obtains acceleration by performing differentiation processing on a plurality of pieces of body motion data arranged in time series.

According to this aspect, the acceleration can be obtained by a relatively simple calculation called differential processing, and the processing load of the acceleration calculator (111) can be reduced.

In the sensor signal processing system (1) according to the eleventh aspect, in the ninth or tenth aspect, the measuring device (2) is a non-contact sensor (21) for detecting body movement data without contacting the subject. have

According to this aspect, the acceleration can be obtained without interfering with the movement of the subject for obtaining the body motion data.

In the sensor signal processing system (1) according to the twelfth aspect, in the eleventh aspect, the sensor (21) is a radio sensor that transmits and receives radio waves.

According to this aspect, even a minute movement of the subject can be detected by the sensor (21), so that the detection accuracy of the amount of physical activity can be further improved.

In the sensor signal processing system (1) according to the thirteenth aspect, in the twelfth aspect, the measuring device (2) has the frequency of the radio wave transmitted by the sensor (21) and the frequency of the radio wave received by the sensor (21) The moving speed of the subject is obtained by comparing .

According to this aspect, since the moving speed of the subject can be determined by the measuring device (2), the sensor signal processing system (1) can determine the acceleration by relatively simple processing.

The sensor signal processing system (1) according to the fourteenth aspect, in any one of the ninth to thirteenth aspects, further comprises an output section (15). An output unit (15) outputs the calculation result of the acceleration calculation unit (111).

According to this aspect, it is possible to notify the manager or the like of the calculation result of the acceleration calculator (111), for example.

A sensor signal processing method according to a fifteenth aspect acquires body motion data from a measuring device (2) that outputs body motion data relating to the body motion of a subject, and based on the body motion data, detects the body motion of the subject. Find the acceleration for

According to this aspect, the acceleration of the subject's body movement can be obtained based on the body movement data acquired from the measuring device (2). According to the equation of motion, the product of acceleration and mass is force, and in physical activity using the whole body of the subject, the weight of the subject corresponds to mass. Assuming that the weight gain and loss of the subject is negligible, the acceleration is considered to correspond to the muscle force exerted by the subject during physical activity. Therefore, the acceleration obtained by the acceleration calculation process corresponds to the strength of the physical activity, and the amount of physical activity over a certain period of time is represented by the accumulated value of the acceleration over the certain period of time. Therefore, in the sensor signal processing method, the amount of physical activity can be quantitatively evaluated by obtaining the acceleration from the body motion data, and the detection accuracy of the amount of physical activity can be improved.

A program according to a sixteenth aspect is a program for causing a computer system to execute the sensor signal processing method according to the fifteenth aspect.

According to this aspect, the acceleration of the subject's body movement can be obtained based on the body movement data acquired from the measuring device (2). According to the equation of motion, the product of acceleration and mass is force, and in physical activity using the whole body of the subject, the weight of the subject corresponds to mass. Assuming that the weight gain and loss of the subject is negligible, the acceleration is considered to correspond to the muscle force exerted by the subject during physical activity. Therefore, the acceleration obtained by the acceleration calculation process corresponds to the strength of the physical activity, and the amount of physical activity over a certain period of time is represented by the accumulated value of the acceleration over the certain period of time. Therefore, in the above program, the amount of physical activity can be quantitatively evaluated by obtaining the acceleration from the body motion data, and the detection accuracy of the amount of physical activity can be improved.

Various configurations (including modifications) of the sensor signal processing system (1) according to Embodiments 1 and 2, without being limited to the above aspects, include a sensor signal processing method, a program, and a non-temporary recording medium recording the program. It can be embodied in

The configurations according to the tenth to fourteenth aspects are not essential configurations for the sensor signal processing system (1) according to the ninth aspect, and can be omitted as appropriate.

REFERENCE SIGNS LIST 10 Event prediction system 2 Measuring device 13 Acquisition unit 15 Output unit 21 Sensor 111 Acceleration calculation unit 112 Presence determination unit 121 Sign determination unit

Claims (11)

an acquisition unit that acquires the body movement data from a measuring device that outputs body movement data related to body movement of a subject;
a sign determination unit that determines a sign of occurrence of a specific event related to the subject based on the body motion data acquired in the past reference period;
a presence determination unit that determines the presence or absence of the target person in the target space based on the body movement data;
The existence determination unit determines that, when a coefficient of an analysis model of time-series analysis representing the body motion data with a plurality of pieces of body motion data acquired before the acquisition timing of the body motion data exceeds a threshold value, the object An event prediction system that determines that the subject exists in space. further comprising an acceleration calculation unit that obtains an acceleration of the movement of the subject's body based on the body movement data;
The event prediction system according to claim 1, wherein the portent determination unit determines the portent based on a change in the amount of physical activity of the subject based on the body motion data. 3. The event prediction system according to claim 2, wherein the acceleration calculation unit obtains the acceleration by performing differentiation processing on the plurality of pieces of body motion data arranged in time series. The event prediction system according to claim 2 or 3, further comprising an output section that outputs a calculation result of the acceleration calculation section. 4. The event prediction system according to any one of claims 1 to 3, further comprising an output section for outputting a determination result of said sign determination section. The event prediction system according to any one of claims 1 to 5, wherein the measuring device has a non-contact sensor that detects the body motion data without contacting the subject. The event prediction system according to claim 6, wherein the sensor is a radio sensor that transmits and receives radio waves. 8. The event prediction system according to claim 7, wherein the measuring device obtains the moving speed of the subject by comparing the frequency of the radio wave transmitted by the sensor and the frequency of the radio wave received by the sensor. The event prediction system according to any one of claims 1 to 8, wherein the portent determination unit distinguishes between types of the specific events for which the portent is determined. an acquisition unit that acquires the body movement data from a measuring device that outputs body movement data related to body movement of a subject;
an acceleration calculation unit that obtains an acceleration of the movement of the subject's body based on the body movement data;
a presence determination unit that determines the presence or absence of the target person in the target space based on the body movement data;
The existence determination unit determines that, when a coefficient of an analysis model of time-series analysis representing the body motion data with a plurality of pieces of body motion data acquired before the acquisition timing of the body motion data exceeds a threshold value, the object A sensor signal processing system that determines that the subject exists in space. Acquiring the body motion data from a measuring device that outputs body motion data relating to the body motion of the subject, and determining the occurrence of a specific event related to the subject based on the body motion data acquired during a past reference period. When a sign is determined and the coefficient of the analysis model of the time-series analysis representing the body motion data represented by a plurality of pieces of body motion data acquired before the acquisition timing of the body motion data exceeds a threshold value, the object space A program for causing a computer system to execute an event prediction method for determining that the subject exists.

Images