Monitoring system and program

Abstract

To make it easy for caregivers to obtain information they need about the living situation of the person being cared for. [Solution] The monitoring system includes an acquisition unit 412 that acquires multiple monitoring data indicating the living situation of the person being monitored, a display unit that displays questions regarding the living situation of the person being monitored, a reception unit that accepts the selection of questions from the person being monitored, and an answer generation unit 416 that generates answers to the selected questions based on the multiple monitoring data.

Description

[Technical Field]

[0001] The present invention relates to a technology for monitoring a monitored person. [Background Art]

[0002] Patent Document 1 describes a technology in which an index for determining the daily activities of a resident is obtained from output data of a plurality of sensors installed in a residence, whether the current condition is normal is determined by comparing the current index with a plurality of past indexes, and the information is notified if the condition is not normal. [Prior Art Literature] [Patent Literature]

[0003] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2000-90366 [Summary of the Invention] [Problem to be Solved by the Invention]

[0004] When sensor data indicating the living condition of a monitored person and notifications are provided to a watcher as in the technology described in Patent Document 1, the watcher can grasp the living condition of the monitored person by checking the sensor data and notifications. However, the sensor data and notifications do not necessarily include the information that the watcher desires. Therefore, when the watcher cannot sufficiently grasp the living condition of the monitored person only with the sensor data and notifications, the watcher has to make a direct check with the monitored person, which imposes a large burden on the watcher. Furthermore, for the monitored person, being repeatedly checked on their condition by the watcher may make them feel as if they are being constantly monitored, which can also become a burden.

[0005] An object of the present invention is to enable a watcher to easily obtain information that the watcher wants to know regarding the living condition of a monitored person. [Means for Solving the Problem]

[0006] One aspect of the present invention provides a monitoring system comprising: an acquisition unit that acquires monitoring data indicating monitoring elements for understanding the living situation of a person being monitored; a display unit that displays questions related to the monitoring elements; a reception unit that accepts the selection of questions from the person being monitored; and an answer generation unit that generates answers to the selected questions based on the monitoring data.

[0007] The monitoring system further comprises a determination unit that determines changes in the monitoring element over time, and the display unit may display a question of interest related to the monitoring element that has been determined to have changed in a way that makes it easier for the user to select it.

[0008] The display unit may, when displaying the featured question and other questions side by side, highlight the featured question or determine the arrangement of the featured question and other questions.

[0009] When the display unit determines that there is a change in any of the monitoring elements, it may display the multiple attention questions corresponding to each of the monitoring elements in a way that makes it easier to select the attention questions with higher priority.

[0010] The response generation unit generates the response using a large-scale language model, and the large-scale language model may be input to the monitoring data or the features of the monitoring data.

[0011] The monitoring system further includes a determination unit that determines changes over time for each monitoring element based on a comparison between monitoring data showing the monitoring elements of the person being monitored during the target period and monitoring data showing the monitoring elements of the person being monitored in the past prior to the target period, and the display unit may display the attention questions related to the monitoring elements for which changes have been determined to have occurred in a manner that makes it easier for the user to select them.

[0012] The large-scale language model may be input with information about the determined change as a feature of the monitoring data, and the large-scale language model may use the input information about the change to generate the response that explains the change in the monitoring element.

[0013] Another aspect of the present invention provides a program for a computer to perform the following steps: acquire monitoring data indicating monitoring elements for understanding the living situation of a person being monitored; receive information from a terminal device that displays questions relating to the monitoring elements, indicating the questions selected by the monitor who is monitoring the person being monitored; and generate answers to the questions indicated by the received information, based on the monitoring data. [Effects of the Invention]

[0014] According to the present invention, caregivers can easily obtain information they want to know about the living situation of the person being cared for. [Brief explanation of the drawing]

[0015] [Figure 1] This figure shows an example of the configuration of a monitoring system according to the embodiment. [Figure 2] This figure shows an example of the configuration of a monitoring device. [Figure 3] This figure shows an example of the configuration of a server device. [Figure 4] This figure shows an example of a question table. [Figure 5] This figure shows an example of a graph illustrating the changes in the activity levels of the person being monitored during the reference period and the target period. [Figure 6] This figure shows an example of a monitoring screen. [Figure 7] This figure shows an example of how to prioritize questions of interest. [Figure 8] This figure shows another example of prioritizing featured questions. [Figure 9] This is a diagram showing an example of the configuration of a terminal device. [Figure 10]It is a sequence chart showing an example of the operation of the monitoring system. Mode for Carrying Out the Invention

[0016] (Configuration) Figure 1 is a diagram showing an example configuration of the monitoring system 10 according to an embodiment. The monitoring system 10 provides a monitoring service for monitoring abnormalities such as intrusion into a monitored area such as a residence, and a watching service for monitoring the status of a watched person such as an elderly person residing in the monitored area. In the watching service, information indicating the activity status measured within a range that does not impose a burden on the watched person during the watched person's daily life is provided to a watcher such as a family member who lives separately from the watched person. Here, the watched person refers to a target person to be watched by the watcher. This allows the watcher to recognize the status of the watched person and watch over the watched person in an unobtrusive manner.

[0017] The monitoring system 10 includes a watching sensor 100, an intrusion monitoring sensor 150, a monitoring device 200, a server device 400, a terminal device 500, and a terminal device 600. Note that although FIG. 1 shows only one watching sensor 100 and one intrusion monitoring sensor 150 to prevent the diagram from becoming complicated, a plurality of each are actually provided. The watching sensor 100, the intrusion monitoring sensor 150, and the monitoring device 200 are installed in a monitored area such as the watched person's residence, and are connected via a communication network 11 such as a LAN (Local Area Network). The server device 400 is installed in a monitoring center operated by a business provider that provides the monitoring service and the watching service. The terminal device 500 is worn and used by the watched person. The terminal device 600 is used by the watcher. The monitoring device 200, the server device 400, the terminal device 500, and the terminal device 600 are connected via a communication network 12 including the Internet and a mobile communication network.

[0018] The monitoring sensor 100 detects the living conditions of the monitored person and outputs detection information. The detection information output from the monitoring sensor 100 is used for grasping the living conditions of the monitored person. The monitoring sensor 100 is installed at various locations in a monitored area which is the living space of the monitored person, such as a bedroom, a living room, and an entrance. The monitoring sensor 100 includes various types of sensors. For example, the monitoring sensor 100 includes human motion sensors, open / close sensors, radio wave sensors, ultrasonic sensors, TOF (Time Of Flight) sensors, cameras, bed sensors, and temperature sensors.

[0019] A human motion sensor is installed on a wall surface, a ceiling, or the like, and detects the presence and movement of a person mainly using infrared rays or ultrasonic waves. An open / close sensor is installed on an opening / closing body such as a door or a window, and detects opening / closing of the opening / closing body mainly using changes in magnetism or a switch. A radio wave sensor is installed on a wall surface, a ceiling, or the like, and detects the position and posture of the monitored person at predetermined intervals such as every second using radio waves such as millimeter waves and microwaves. Radio wave sensors include millimeter wave sensors. An ultrasonic sensor is installed on a wall surface, a ceiling, or the like, and detects the position and posture of the monitored person at predetermined intervals such as every second using ultrasonic waves. A TOF sensor is installed on a wall surface, a ceiling, or the like, and detects the position and posture of the monitored person at predetermined intervals such as every second using the time-of-flight of light. A camera is installed on a wall surface, a ceiling, or the like, and detects the position and posture of the monitored person at predetermined intervals such as every second by capturing and analyzing an image. The postures of the monitored person include, for example, standing, sitting, and lying down. For posture detection, for example, known methods are used, such as pattern matching between point cloud data obtained from a sensor and pre-registered point cloud data of each posture, pattern matching between a human figure extracted from an image by a background subtraction method or the like and pre-registered images of each posture, and a method using a trained machine learning model obtained by machine learning of images of various postures. A bed sensor is installed on a bed and detects whether a person is on the bed. A temperature sensor is installed indoors and measures the room temperature.

[0020] The intrusion monitoring sensor 150 detects intruders in the monitored area and outputs detection information. The intrusion monitoring sensor 150 includes both indoor and outdoor installations. Indoor intrusion monitoring sensors 150 are installed, for example, in openings such as front doors and windows, in passageways and other areas of movement within a residence, and in living rooms. Outdoor intrusion monitoring sensors 150 are installed, for example, in gardens and gates. The intrusion monitoring sensor 150, like the monitoring sensor 100, includes various types of sensors. For example, the intrusion monitoring sensor 150 includes the aforementioned motion sensors, open / close sensors, radio wave sensors, ultrasonic sensors, TOF sensors, and cameras.

[0021] The monitoring device 200 uses the intrusion monitoring sensor 150 to monitor for intrusion into the monitored area where the person being monitored is located. For example, if an intruder is detected by the intrusion monitoring sensor 150, the monitoring device 200 notifies the server device 400 that an intruder has entered the monitored area. The monitoring device 200 has three security modes: unmanned security mode, manned security mode, and deactivation mode.

[0022] The unmanned security mode is a security mode that monitors the entire monitored area. The unmanned security mode is used when the monitored person is out and the monitored area is unoccupied. In unmanned security mode, if detection information is output from any of the intrusion monitoring sensors 150, an abnormality signal is sent from the monitoring device 200 to the server device 400. When the unmanned security mode is set, it is preferable that the intrusion monitoring sensors 150 that monitor the monitored person's exit route begin detecting intruders after a predetermined time has elapsed since the unmanned security mode was set, in order to prevent false detection of a monitored person who is leaving the building as an intruder.

[0023] The manned security mode is a security mode that monitors only a portion of the monitored area. The manned security mode is used when there is a person being monitored in the monitored area, and the system needs to monitor for intruders in areas where the person being monitored is not present. For example, in manned security mode, to detect intrusion into the monitored area from the outside, predetermined outer perimeter areas such as the front door, windows, gates, and the exterior of the residence are monitored. Alternatively, from among multiple areas included in the monitored area, a designated area (such as the second floor) may be monitored according to the actions of the person being monitored. In manned security mode, when detection information is output from an intrusion monitoring sensor 150 installed in the outer perimeter area or the designated area, an abnormality signal is sent from the monitoring device 200 to the server device 400. On the other hand, even if detection information is output from an intrusion monitoring sensor 150 installed in a location other than the outer perimeter area or the designated area, this detection information is ignored, and no abnormality signal is sent from the monitoring device 200 to the server device 400.

[0024] The deactivation mode is a security mode in which the monitored area is not monitored. The deactivation mode is used when there is a person under surveillance in the monitored area and there is no need to monitor for intruders. In deactivation mode, intruders into the monitored area are not monitored, and even if detection information is output from the intrusion monitoring sensor 150, this detection information is ignored, and no abnormal signal is sent from the monitoring device 200 to the server device 400.

[0025] These security modes serve different purposes, and the location of the person being monitored can be determined from the security mode setting. For example, if the unattended security mode is set, it indicates that the person being monitored is out. On the other hand, if the manned security mode is set, it indicates that the person being monitored is in an area of ​​the monitored area that is not being monitored.

[0026] When the server device 400 receives an abnormal signal from the monitoring device 200, it outputs information about the monitored area and the person being monitored, prompting the controller of the monitoring sensor to confirm the person being monitored and the security guards who rush to the monitored area to take action.

[0027] The terminal device 500 is used to measure the activity level of the person being monitored. An example of the terminal device 500 is a wearable device. The terminal device 500 includes various sensors for measuring the activity level of the person being monitored. For example, the sensors of the terminal device 500 include an accelerometer, a GPS (Global Positioning System) receiver, and a heart rate sensor.

[0028] An acceleration sensor measures the acceleration of the terminal device 500. A GPS receiver receives satellite signals from multiple satellites and uses the received satellite signals to determine the position of the terminal device 500. A heart rate sensor measures the heart rate of the person being monitored. The terminal device 500 measures the activity level of the person being monitored based on the detection results of these sensors. For measuring the activity level, known methods such as scoring and adding up changes in steps taken, distance traveled, travel time, and heart rate per predetermined time obtained from the detection results of these sensors, or the method described in Japanese Patent Application Publication No. 2019-144730, are used. The terminal device 500 then transmits the activity level data indicating the measured activity level to the server device 400.

[0029] The terminal device 600 is used by the caregiver to view various information related to the monitoring service. Examples of terminal devices 600 include smartphones, mobile phones, personal computers, tablet devices, and wearable devices.

[0030] (monitoring device) Figure 2 shows an example of the configuration of the monitoring device 200. The monitoring device 200 comprises a control unit 201, a storage unit 202, a communication unit 203, an operation unit 204, a display unit 205, and a sounding unit 206. Each part of the monitoring device 200 is connected via a bus.

[0031] The control unit 201 is a processor that controls each part of the device and performs various processing tasks. For example, the control unit 201 includes a CPU (Central Processing Unit). The memory unit 202 is a memory that stores programs and various data necessary to realize the functions of the device. The data stored in the memory unit 202 includes information indicating the currently set security mode, an authentication code necessary for setting the security mode, and detection information from the monitoring sensor 100 and the intrusion monitoring sensor 150. For example, the memory unit 202 includes memory such as ROM (Read Only Memory) and RAM (Random Access Memory), and storage such as HDD (Hard Disk Drive) and SSD (Solid State Drive). The communication unit 203 is a communication interface for connecting the device to the communication networks 11 and 12. The communication unit 203 is used to communicate with other devices connected via the communication networks 11 or 12.

[0032] The operation unit 204 accepts various operations on the device. Operations accepted by the operation unit 204 include setting the security mode and reading an authentication code from an authentication medium. For example, the operation unit 204 includes a touch sensor that accepts operations on the screen of the display unit 205, an operation button that accepts setting the security mode, and a reader that reads an authentication code from an authentication medium such as an IC card held by the person being monitored. The display unit 205 displays various information. The information displayed by the display unit 205 includes the currently set security mode, the status of the monitoring sensor 100 and the intrusion monitoring sensor 150, and information indicating the installation locations of the monitoring sensor 100 and the intrusion monitoring sensor 150. For example, the display unit 205 includes a liquid crystal display. Alternatively, the display unit 205 may be configured as a touch panel in combination with the touch sensor. The sound unit 206 outputs various sounds, such as sounds to indicate the occurrence of an abnormality and message voices. For example, the sound unit 206 includes a speaker.

[0033] The control unit 201 functions as an acquisition unit 211, a setting unit 212, and a transmission unit 213. These functions are realized when the control unit 201 executes a program stored in the storage unit 202, performs calculations, or controls each part of the monitoring device 200.

[0034] The acquisition unit 211 acquires detection information output from the monitoring sensor 100 and the intrusion monitoring sensor 150. The detection information includes the type of monitoring sensor 100 or intrusion monitoring sensor 150, the date and time when the monitoring sensor 100 or intrusion monitoring sensor 150 detected the target, and the installation location of the monitoring sensor 100 or intrusion monitoring sensor 150.

[0035] The setting unit 212 sets one of several security modes in response to an operation by the person being monitored. Specifically, when an operation to set a security mode is performed using the operation unit 204, the setting unit 212 sets the security mode corresponding to this operation.

[0036] The transmitting unit 213 transmits the detection information from the monitoring sensor 100 and the intrusion monitoring sensor 150, acquired by the acquisition unit 211, to the server device 400. The transmitting unit 213 also transmits security mode information to the server device 400, indicating the security mode set by the setting unit 212. The security mode information includes information indicating the type of security mode set and the date and time that security mode was set.

[0037] (Server device) Figure 3 shows an example of the configuration of the server device 400. The server device 400 has the function of presenting example questions about the living situation of the person being monitored to the monitor and generating answers to the questions from the monitor. The server device 400 comprises a control unit 401, a storage unit 402, and a communication unit 403. Each part of the server device 400 is connected via a bus. The control unit 401, storage unit 402, and communication unit 403 are basically the same as the control unit 201, storage unit 202, and communication unit 203 of the monitoring device 200, respectively. However, the storage unit 402 stores a server program for realizing the functions of the server device 400 and a question table 421. The communication unit 403 is a communication interface for connecting the device to the communication network 12 and is used to communicate with other devices connected via the communication network 12.

[0038] Figure 4 shows an example of Question Table 421. Question Table 421 stores multiple monitoring elements that indicate the living situation of the person being monitored, and questions related to each monitoring element, in association with each element. Monitoring elements are elements that constitute the living situation and are focused on in order to understand the living situation of the person being monitored and to provide appropriate monitoring. Monitoring elements include elements related to activity status, elements related to health status, elements related to lifestyle habits, and elements related to the living environment. In the example shown in Figure 4, the multiple monitoring elements include activity level, posture ratio, time spent at home, bedtime and wake-up time, going out and returning home, visitors, sleep efficiency, whether or not meals were eaten, and room temperature.

[0039] The questions are sentences that express questions about monitoring elements, and more specifically, they are sentences that ask about the quantity, value, state, or presence or absence of monitoring elements. In the example shown in Figure 4, the question about activity level is "How active was the person?" The question about posture ratio is "What was the posture ratio?" The question about time spent is "How long were you in the room?" The question about bedtime and wake-up time is "Please tell me your bedtime and wake-up time." The question about going out and returning home is "Were you out?" The question about visitors is "Were there any visitors?" The question about sleep efficiency is "How was your sleep efficiency?" The question about whether or not you ate is "Were you eating?" The question about room temperature is "What was the room temperature?"

[0040] The control unit 401 functions as a communication control unit 411, an acquisition unit 412, a determination unit 413, a decision unit 414, a screen creation unit 415, and an answer generation unit 416. These functions are realized when the control unit 401 executes a program stored in the storage unit 402, performs calculations, or controls various parts of the server device 400.

[0041] (Communications Control Unit) The communication control unit 411 transmits and receives information for displaying and operating the monitoring screen 700 with the caregiver's terminal device 600. The monitoring screen 700 presents several example questions regarding the living situation of the person being monitored and accepts the caregiver's selection of questions from among these.

[0042] (Acquisition Department) The acquisition unit 412 acquires multiple monitoring data indicating the living conditions of the person being monitored. Specifically, the acquisition unit 412 receives detection information and security mode information transmitted from the monitoring device 200 via the communication unit 403. The acquisition unit 412 also receives activity level data transmitted from the terminal device 500 via the communication unit 403. Based on the received detection information, security mode information, and activity level data, the acquisition unit 412 generates multiple monitoring data. These multiple monitoring data represent multiple types of monitoring elements. These monitoring elements include the person being monitored's activity level, posture ratio, time spent at home, bedtime and wake-up time, going out and returning home, visitors, sleep efficiency, whether meals were eaten, and room temperature.

[0043] Activity level is an indicator of the degree of biological activity of the person being monitored, and includes biological information such as the intensity and frequency of physical movement, the amount of activity in daily life and work, and physiological activity within the body. Activity level is measured using detection information from a human presence sensor, an open / close sensor, or a millimeter-wave sensor. For example, when measuring activity level from detection information from a human presence sensor or an open / close sensor, the acquisition unit 412 accumulates the number of detections by the human presence sensor or the open / close sensor at predetermined intervals to measure the activity level of the person being monitored at predetermined intervals, and acquires the activity level data showing the measured activity level as monitoring data. In addition, by using detection information from radio wave sensors or cameras in combination, it is possible to recognize specific activities such as cleaning, sitting at the dining table, or relaxing on the sofa, and the acquisition unit 412 may take this activity content into consideration when measuring the activity level. Furthermore, the acquisition unit 412 may receive activity level data transmitted from the terminal device 500 via the communication unit 403 and use this as monitoring data.

[0044] The posture ratio indicates the proportion of time the person being monitored spent in each posture: standing, sitting, and lying down. The posture ratio is calculated, for example, based on detection information from a millimeter-wave sensor. The acquisition unit 412 acquires the calculated posture ratio data as monitoring data.

[0045] The dwell time indicates the length of time the person being monitored spent in each area, such as the bedroom or living room. The dwell time is calculated, for example, based on detection information from motion sensors installed in each area. The acquisition unit 412 acquires the calculated dwell time data as monitoring data.

[0046] The bedtime indicates the time the person being monitored went to sleep. The wake-up time indicates the time the person being monitored woke up. The bedtime and wake-up time are determined based on detection information from, for example, a millimeter-wave sensor or a bed sensor installed in the bedroom. The acquisition unit 412 acquires the data indicating the determined bedtime and wake-up time as monitoring data.

[0047] The "going out / returning home" function indicates that the person being monitored has gone out or returned home. This is determined based, for example, on security mode information or detection information from an open / close sensor installed on the front door. The acquisition unit 412 acquires data indicating this determination result as monitoring data.

[0048] The presence of a visitor indicates that there has been a visitor at the monitored person's residence. The presence of a visitor is determined, for example, based on detection information from a millimeter-wave sensor. For example, if the millimeter-wave sensor detects multiple people, it is determined that there has been a visitor at the monitored person's residence based on that detection information. The acquisition unit 412 acquires data indicating this determination result as monitoring data.

[0049] Sleep efficiency is an indicator of the quality of sleep of the person being monitored, for example, it shows the ratio of sleep time to time spent in bed. Sleep efficiency is calculated based on detection information from, for example, a millimeter-wave sensor or bed sensor installed in the bedroom. The acquisition unit 412 acquires the calculated sleep efficiency data as monitoring data.

[0050] Whether or not a meal has been eaten indicates whether the person being monitored has eaten. Whether or not a meal has been eaten is determined based on detection information from, for example, a millimeter-wave sensor installed in the dining room. For example, at predetermined times such as breakfast, lunch, and dinner, the millimeter-wave sensor detects whether or not a person is sitting at the dining table, and whether or not a meal has been eaten is determined based on that detection information. The acquisition unit 412 acquires data indicating the determination result as monitoring data.

[0051] The room temperature indicates the temperature of the room in the dwelling of the person being monitored. The room temperature is determined, for example, based on detection information from a temperature sensor installed in the room. The acquisition unit 412 acquires the data indicating the determined temperature as monitoring data.

[0052] (Judgment Department) The determination unit 413 determines changes in the living situation of the person being monitored according to predetermined conditions, based on a comparison of monitoring data during the target period and monitoring data during the reference period. This determination of changes in living situation is a concept that includes determining whether or not there is a change in living situation, the degree of change, the direction of change, and whether the change is good or bad. The predetermined conditions are set in advance for each monitoring element. Thus, in this embodiment, changes in living situation are determined on a rule basis.

[0053] The target period and reference period are determined based on the target date. The target date is the day on which the determination of changes in living conditions is made. Note that the target date may be changed according to the caregiver's actions. Alternatively, the target date may be set to the default target date based on the date last selected by the caregiver. The target period is, for example, the day before the target date. The reference period is a period longer than the target period prior to the target date, for example, the past month up to the day before the start date of the target period. In other words, the reference period is a period prior to the target period. This is to understand the usual daily life patterns of each person being cared for. For example, if the target date is March 20th, the target period will be March 19th, and the reference period will be from February 19th to March 18th.

[0054] The length of the target period is not limited to one day; for example, it could be several days, a week, several weeks, or a month. Similarly, the length of the reference period is not limited to one month; it could be a week, several months, six months, or a year. Furthermore, the target period and the reference period may partially overlap. For example, by setting the reference period to several months corresponding to one season, it is possible to reflect the patterns of daily life that change with each season. Also, if the target period is one day, by setting the reference period to multiple past days of the same day of the week, it is possible to smooth out the differences in patterns caused by different habits on each day of the week.

[0055] Specifically, the determination unit 413 calculates representative values ​​for each of the multiple monitoring elements at predetermined intervals for monitoring data during the target period and representative values ​​for each of the multiple monitoring elements at predetermined intervals for monitoring data during the reference period, and compares these to determine the changes over time for each monitoring element. Representative values ​​include, for example, the average value and the sum. Representative values ​​for each of the multiple monitoring elements at predetermined intervals during the target period indicate the most recent living situation of the person being monitored. Representative values ​​for each of the multiple monitoring elements at predetermined intervals during the reference period indicate the usual daily living situation of the person being monitored. The predetermined intervals are, for example, one hour, several hours, or one day. The multiple monitoring elements include the above-mentioned activity level, posture ratio, time spent, bedtime / wake-up time, going out / returning home, visitors, sleep efficiency, whether or not meals were eaten, and room temperature.

[0056] (activity amount) When determining changes in activity levels, the determination unit 413 first calculates, for example, the average value of activity level data for each hour during the target period and the average value of activity level data for each hour during the reference period, and then creates an activity level trend graph showing the changes over time.

[0057] Figure 5 shows an example of a graph illustrating the activity level trends of the person being monitored during the reference period and the target period. The vertical axis of the activity level trend graph represents activity level, and the horizontal axis represents time. The activity level trend graph includes line L1, which shows the average activity level of the person being monitored for each hour during the target period, and line L2, which shows the average activity level of the person being monitored for each hour during the reference period. Line L1 shows the person being monitored's most recent activity pattern. Line L2 shows the person being monitored's past activity pattern. Since the reference period is longer than the target period, the past activity pattern of the person being monitored shown by line L2 reflects the usual daily life pattern of each person being monitored.

[0058] Next, the determination unit 413 calculates a correlation value between the average value of activity level data during the target period and the average value of activity level data during the reference period. This correlation value indicates the degree of similarity between the person being monitored's recent living situation and their usual daily living situation. A correlation value closer to 1 indicates a high degree of similarity. Conversely, a correlation value closer to 0 indicates a low degree of similarity.

[0059] The determination unit 413 then compares the calculated correlation value with a threshold to determine a change in activity level. Here, the threshold is set to 0.7. For example, the determination unit 413 determines that there is no change in activity level if the correlation value between the average activity level of the person being monitored during the target period shown by line L1 in Figure 5 and the average activity level of the person being monitored during the reference period shown by line L2 is 0.7 or higher. On the other hand, the determination unit 413 determines that there is a change in activity level if this correlation value is less than the threshold of 0.7.

[0060] Furthermore, if there is a change in the activity level, the determination unit 413 may determine the degree of change in the activity level based on the correlation value described above. For example, if the correlation value is equal to or greater than the first threshold (e.g., 0.7), the determination unit 413 determines that the degree of change is small. If the correlation value is between or greater than the second threshold (e.g., 0.5) and less than the first threshold, the determination unit 413 determines that the degree of change is moderate. If the correlation value is less than the second threshold, the determination unit 413 determines that the degree of change is large.

[0061] Furthermore, if there is a change in the activity level, the determination unit 413 may determine the direction of the change in activity level. For example, the determination unit 413 determines that the activity level has increased if the average daily activity level of the person being monitored during the target period is greater than the average daily activity level of the person being monitored during the reference period. On the other hand, the determination unit 413 determines that the activity level has decreased if the average daily activity level of the person being monitored during the target period is less than the average daily activity level of the person being monitored during the reference period.

[0062] Furthermore, the determination unit 413 may determine whether the change in activity level is good or bad depending on the time of day when the change in activity level occurred. For example, if the average daily daytime activity level during the target period increases compared to the average daily daytime activity level during the reference period, the determination unit 413 will determine that change is good. On the other hand, if the average daily daytime activity level during the target period decreases compared to the average daily daytime activity level during the reference period, the determination unit 413 will determine that change is bad. Also, if the average daily nighttime activity level during the target period increases compared to the average daily nighttime activity level during the reference period, the determination unit 413 will determine that change is bad.

[0063] (Posture ratio) With respect to the posture ratio, the determination unit 413 determines whether or not there is a change in the posture ratio, for example, using data indicating the posture ratio, in the same manner as the method for determining changes in activity levels described above. Furthermore, if there is a change in the posture ratio, the determination unit 413 may determine the degree and direction of the change in the posture ratio, using data indicating the posture ratio, in the same manner as the method for determining the degree and direction of change in activity levels described above.

[0064] Furthermore, the determination unit 413 may use data indicating the proportion of postures to determine whether a change in the proportion of postures is good or bad, based on the time of day, location, and type of posture in which the proportion of postures changed. For example, if the proportion of standing in the living room during the day during the target period increases compared to the proportion of standing in the living room during the day during the reference period, the determination unit 413 will determine that change is good. On the other hand, if the proportion of lying down in the living room during the day during the target period increases compared to the proportion of lying down in the living room during the day during the reference period, the determination unit 413 will determine that change is bad. Also, if the proportion of either daytime or nighttime postures in the bedroom during the target period increases compared to the proportion of either daytime or nighttime postures in the bedroom during the reference period, the determination unit 413 will determine that change is bad.

[0065] (Duration of stay) With respect to the duration of stay, the determination unit 413 determines whether or not there has been a change in the duration of stay, for example, using data indicating the duration of stay, in a manner similar to the method for determining changes in activity levels described above. Furthermore, if there has been a change in the duration of stay, the determination unit 413 may determine the degree and direction of the change in the duration of stay, using data indicating the duration of stay, in a manner similar to the method for determining the degree and direction of change in activity levels described above.

[0066] (Bedtime / wake up time) Regarding bedtime and wake-up time, the determination unit 413 uses data indicating bedtime and wake-up time to determine whether there has been a change in bedtime and wake-up time, for example, in the same manner as the method for determining changes in activity levels described above. Furthermore, if there is a change in bedtime or wake-up time, the determination unit 413 may use data indicating bedtime and wake-up time to determine the degree and direction of the change in bedtime and wake-up time, in the same manner as the method for determining the degree and direction of change in activity levels described above. In addition, the determination unit 413 may use data indicating bedtime and wake-up time to determine whether the change is good or bad based on the direction of the change in bedtime and wake-up time. For example, if the bedtime in the target period is later and the wake-up time is earlier compared to the bedtime in the reference period, the determination unit 413 will determine that the change is a bad change.

[0067] (Going out / returning home) Regarding going out and returning home, the determination unit 413 uses data indicating going out and returning home, for example, to determine whether there is a change in whether or not the person goes out or the frequency of going out, in a manner similar to the method for determining changes in activity levels described above. Furthermore, if there is a change in whether or not the person goes out or the frequency of going out, the determination unit 413 may use data indicating going out and returning home to determine the degree and direction of the change in whether or not the person goes out or the direction of the change, in a manner similar to the method for determining the degree and direction of the change in activity levels described above. In addition, the determination unit 413 may use data indicating going out and returning home to determine whether the change is good or bad based on the direction of the change in whether or not the person goes out or the frequency of going out. For example, if the frequency of going out during the target period increases compared to the frequency of going out during the reference period, the determination unit 413 determines that the change is a good change. On the other hand, if the frequency of going out during the target period decreases compared to the frequency of going out during the reference period, the determination unit 413 determines that the change is a bad change. Furthermore, if the time of returning home during the target period is later than the time of returning home during the reference period, the determination unit 413 determines that this is an anomaly that warrants notification.

[0068] (Visitor) Regarding visitors, the determination unit 413 uses data indicating visitors to determine whether there is a change in the presence or absence of visitors or the frequency of visitors, in a manner similar to the method for determining changes in activity levels described above. Furthermore, if there is a change in the presence or absence of visitors or the frequency of visitors, the determination unit 413 may use data indicating visitors to determine the degree and direction of the change in the presence or absence of visitors or the frequency of visitors, in a manner similar to the method for determining the degree and direction of the change in activity levels described above. In addition, the determination unit 413 uses data indicating visitors to determine if a visitor occurs on a day when there are normally no visitors as an anomaly that warrants notification. Days when there are normally no visitors may be identified, for example, based on data indicating visitors during the reference period, or based on the schedule information of the person being monitored.

[0069] (Sleep efficiency) With respect to sleep efficiency, the determination unit 413 determines whether or not there is a change in sleep efficiency, for example, using data indicating sleep efficiency, in the same manner as the method for determining changes in activity levels described above. Furthermore, if there is a change in sleep efficiency, the determination unit 413 may determine the degree and direction of the change in sleep efficiency, for example, using data indicating sleep efficiency, in the same manner as the method for determining the degree and direction of change in activity levels described above. In addition, the determination unit 413 may determine whether the change is good or bad based on the direction of the change in sleep efficiency, using data indicating sleep efficiency. For example, if the sleep efficiency during the target period is higher than the sleep efficiency during the reference period, the determination unit 413 determines that the change is good. On the other hand, if the sleep efficiency during the target period is lower than the sleep efficiency during the reference period, the determination unit 413 determines that the change is bad.

[0070] (Meals included) Regarding the presence or absence of meals, the determination unit 413 uses data indicating the presence or absence of meals to determine whether or not there has been a change in the number of meals, in a manner similar to the method for determining changes in activity levels described above. Furthermore, if there is a change in the number of meals, the determination unit 413 may use data indicating the presence or absence of meals to determine the degree and direction of the change in the number of meals, in a manner similar to the method for determining the degree and direction of change in activity levels described above. In addition, the determination unit 413 may use data indicating the presence or absence of meals to determine whether the change is good or bad based on the direction of the change in the number of meals. For example, if the number of meals during the target period is less than three per day, the determination unit 413 may determine that the change is a bad change.

[0071] (Room temperature) Regarding the room temperature, the determination unit 413 determines whether or not there is a change in the room temperature, for example, using data indicating the room temperature, in the same manner as the method for determining changes in activity levels described above. Furthermore, if there is a change in the room temperature, the determination unit 413 may determine the degree and direction of the change in the room temperature, using data indicating the room temperature, in the same manner as the method for determining the degree and direction of change in activity levels described above. In addition, the determination unit 413 determines that if the room temperature deviates by a predetermined value or more from the standard temperature corresponding to the season, it is an anomaly that warrants notification.

[0072] (Decision section) The decision unit 414 determines questions regarding the living situation of the person being monitored based on the determination result of the judgment unit 413. If the judgment unit 413 determines that there has been no change in the monitoring elements, the decision unit 414 determines a predetermined number of general questions. General questions are general questions related to each monitoring element and are used to understand the overall living situation of the person being monitored. General questions are determined independently of any changes in the living situation of the person being monitored. For example, the decision unit 414 randomly selects general questions from among multiple question sentences stored in the question table 421 shown in Figure 4.

[0073] On the other hand, if the determination unit 413 determines that there has been a change in the monitoring element, the decision unit 414 determines the focus questions and general questions so that the total number is a predetermined number. The focus questions are questions related to the monitoring element that has changed and are used to grasp the changes in the living situation of the person being monitored. The focus questions are determined according to the determination result of the determination unit 413. For example, if it is determined that the activity level has changed, the decision unit 414 will select the question "How active is the person?" which is associated with the monitoring element "activity level," as the focus question from among the multiple question sentences stored in the question table 421 shown in Figure 4.

[0074] (Screen creation department) The screen creation unit 415 creates a monitoring screen 700 that includes the questions determined by the decision unit 414. The monitoring screen 700 presents the monitor with examples of specific questions to understand the situation of the person being monitored.

[0075] Figure 6 shows an example of the monitoring screen 700. The monitoring screen 700 is used by the caregiver to obtain information they want to know about the living situation of the person being monitored. The monitoring screen 700 includes a question presentation area 710 that presents several questions as examples, and an answer presentation area 720 that presents answers to questions selected by the caregiver. The monitoring screen 700 also accepts operations to change the target date.

[0076] The question presentation area 710 displays a list of questions. The question list contains multiple questions determined by the decision unit 414, arranged in a single row in order. Each question is displayed as a selectable button, for example, and accepts the operation to select a question. By performing the operation to select a question, the caregiver can choose a desired question from the multiple questions displayed in the question presentation area 710. The answer presentation area 720 displays the answer to the question selected by the caregiver.

[0077] Furthermore, the screen creation unit 415 prioritizes displaying the attention questions related to the monitoring elements that the judgment unit 413 has determined to have changed. Prioritizing display means displaying the attention questions in a format that makes it easy for the monitor to select them, and includes determining the layout and highlighting. Determining the layout means determining the arrangement of the attention questions and general questions so that the attention questions can be easily selected. Determining the layout includes, for example, changing the display order of the attention questions and displaying the attention questions at the top of the question list. Highlighting means displaying the attention questions in an emphasized manner. Highlighting includes, for example, changing the text color, changing the background color, adding bold text, adding an underline, changing the size, flashing display, adding a border, adding an icon, and two or more combinations of these.

[0078] Figure 7 shows an example of the priority display of a question of interest. For example, if a change in activity level is detected, the question "How active is the person?" becomes the question of interest. The screen creation unit 415 changes the display order of the questions included in the question list, placing the question of interest "How active is the person?" at the top of the question list. This makes it easier for the caregiver to select this question of interest.

[0079] Figure 8 shows another example of the priority display of attention questions. For example, if a change in activity level is detected, the question "How active are you?" becomes the attention question. The screen creation unit 415 changes the display color of the attention question "How active are you?" included in the question list to a different color from the other questions. This makes it easier for the caregiver to select this attention question.

[0080] (Answer generation part) The response generation unit 416 generates answers to questions selected by the caregiver based on the caregiver data. More specifically, the response generation unit 416 uses artificial intelligence (AI) and a large-scale language model to generate answers to the caregiver's questions. The large-scale language model is an AI model trained using a large amount of text data and generates natural sentences that humans can understand using advanced natural language processing capabilities. The large-scale language model may be stored on an external server device or in the storage unit 402.

[0081] The response generation unit 416 generates a prompt that includes the question selected by the caregiver and the judgment result of the judgment unit 413, i.e., information regarding changes in each monitoring element determined by the judgment unit 413, and inputs it to the large-scale language model. A prompt is an instruction sentence that causes the large-scale language model to perform a specific task. In addition to the question selected by the caregiver and the judgment result of the judgment unit 413, the prompt may also include basic information such as the installation location of the monitoring sensor 100 and the schedule information of the person being monitored. Furthermore, the prompt may include multiple monitoring data in place of or in addition to the judgment result of the judgment unit 413. The prompt may also include constraints such as not answering anything other than monitoring data for the target period or the question asked.

[0082] For example, a prompt instructs the system to generate an answer to a question selected by the caregiver, based on the determination result of the determination unit 413. In this prompt, the question selected by the caregiver constitutes a command, and the determination result of the determination unit 413 becomes the input data used to execute this command. The determination result of the determination unit 413 represents feature quantities such as activity level, posture ratio, and time spent, extracted from multiple monitoring data, and is therefore an example of "feature quantities of monitoring data" according to the present invention. The large-scale language model generates an answer that describes the changes in the monitoring element that is the subject of the question, according to the input prompt. Since the large-scale language model generates the answer according to the determination result of the determination unit 413 included in the prompt, it is possible to suppress variations in the meaning of the answer. Furthermore, by using the generation AI, even for the same question, natural answers with different nuances each time are generated, similar to those generated by a human.

[0083] Figure 9 shows an example of the configuration of the terminal device 600. Note that the terminal device 500 may have a similar configuration to the terminal device 600. The terminal device 600 comprises a control unit 601, a storage unit 602, a communication unit 603, an operation unit 604, and a display unit 605. Each part of the terminal device 600 is connected via a bus. The control unit 601, storage unit 602, communication unit 603, operation unit 604, and display unit 605 are basically the same as the control unit 201, storage unit 202, communication unit 203, operation unit 204, and display unit 205 of the monitoring device 200, respectively. However, the communication unit 603 is a communication interface for connecting the device to the communication network 12 and is used to communicate with other devices connected via the communication network 12. The storage unit 602 stores a monitoring application program for realizing the functions of the terminal device 600.

[0084] The control unit 601 functions as a communication control unit 611 and a display control unit 612. These functions are realized when the control unit 601 executes a program stored in the storage unit 602, performs calculations, or controls various parts of the terminal device 600.

[0085] The communication control unit 611 sends and receives information with the server device 400 for displaying and operating the monitoring screen 700.

[0086] The display control unit 612 displays the monitoring screen 700 on the display unit 605. The monitoring screen 700 is a screen displayed by, for example, a monitoring application program. As described above, the monitoring screen 700 presents several questions regarding the living situation of the person being monitored as an example, and accepts the selection of questions from the caregiver. When the caregiver selects a question on the monitoring screen 700, the display control unit 612 accepts the selection of questions from the caregiver via the monitoring screen 700. The display control unit 612 is an example of a "reception unit" according to the present invention.

[0087] (operation) Figure 10 is a sequence chart showing an example of the operation of the monitoring system 10. This operation is initiated when a predetermined time arrives, such as 6:00 every morning. Alternatively, this operation may be initiated each time new information used to acquire monitoring data is received.

[0088] In step S11, the acquisition unit 412 of the server device 400 acquires multiple monitoring data indicating the living conditions of the person being monitored during the target period and the reference period. Specifically, the acquisition unit 412 acquires multiple monitoring data for the target period and the reference period based on detection information, security mode information, and activity level data stored in the storage unit 402. The multiple monitoring data include activity level data, data indicating posture ratio, data indicating time spent at home, data indicating bedtime and wake-up time, data indicating going out and returning home, data indicating visitors, data indicating sleep efficiency, data indicating whether meals were eaten, and data indicating room temperature.

[0089] In step S12, the determination unit 413 of the server device 400 determines changes in monitoring elements indicating the living situation of the person being monitored, based on a comparison of the monitoring data for the target period acquired in step S11 with the monitoring data for the reference period. Here, the target date is March 20th. For example, as shown in Figure 5, the determination unit 413 creates an activity level trend graph that includes line L1, which shows the average value of activity level data for each hour during the target period on March 19th, and line L2, which shows the average value of activity level data for each hour during the reference period from February 19th to March 18th. Next, the determination unit 413 calculates a correlation value between the average value of activity level data for the target period on March 19th, shown by line L1, and the average value of activity level data for the reference period from February 19th to March 18th, shown by line L2. Then, if the calculated correlation value is 0.7 or higher, the determination unit 413 determines that there has been no change in the activity level of the person being monitored. On the other hand, the determination unit 413 determines that there has been a change in the activity level of the person being monitored if this correlation value is less than the threshold of 0.7.

[0090] Furthermore, if there is a change in activity level, the determination unit 413 determines the degree and direction of the change in activity level based on the correlation value described above, or a comparison between the daily average of activity level data during the target period and the daily average of activity level data during the reference period. In addition, the determination unit 413 determines whether the change in activity level is good or bad depending on the time of day when the change occurred. Similarly, the determination unit 413 determines whether there is a change in other monitoring elements, the degree of the change, the direction of the change, and whether the change is good or bad.

[0091] If it is determined in step S12 that there is no change in any of the monitoring elements (the determination in step S12 is NO), the process proceeds to step S13. In step S13, the decision unit 414 of the server device 400 determines a predetermined number of general questions to present to the monitor. Here, the predetermined number is set to five. For example, the decision unit 414 randomly selects five question sentences from the multiple question sentences stored in the question table 421 shown in Figure 4 and determines these to be general questions. After step S13, the process proceeds to step S15.

[0092] On the other hand, if it is determined in step S12 that there has been a change in at least one of the monitoring elements (the determination in step S12 is YES), the process proceeds to step S14. In step S14, the decision unit 414 of the server device 400 determines, based on the determination result in step S12, the number of focus questions and general questions to be presented to the monitor so that the total number is a predetermined number. For example, if it is determined in step S12 that there has been a change in the activity level of the person being monitored, the decision unit 414 determines the question sentence "How active is the person?" which is associated with the monitoring element "activity level" in the question table 421 shown in Figure 4, as the focus question. The decision unit 414 also randomly selects four question sentences from the multiple question sentences stored in the question table 421 shown in Figure 4 and determines these to be general questions.

[0093] In step S15, the screen creation unit 415 of the server device 400 creates a monitoring screen 700 that presents the questions determined in step S13 or S14 as examples. For example, if five general questions were determined in step S13, these general questions will be arranged vertically in order in the question presentation area 710 of the monitoring screen 700.

[0094] On the other hand, if the focus question "How active are you?" and the other four general questions are determined in step S14, these focus questions and general questions are arranged vertically in order. The screen creation unit 415 also prioritizes displaying the focus question determined in step S14. For example, as shown in Figure 7, the screen creation unit 415 may change the display order of the focus question "How active are you?" and place it at the top of the question list. Alternatively, as shown in Figure 8, the screen creation unit 415 may change the display color of the focus question "How active are you?" to red, which is different from the other questions.

[0095] When a caregiver wants to check the living situation of the person being cared for, they use the operation unit 604 of the terminal device 600 to instruct the display of the monitoring screen 700. In step S16, the communication control unit 611 of the terminal device 600 transmits a request to display the monitoring screen 700 to the server device 400 via the communication unit 603 in response to this operation. The communication control unit 411 of the server device 400 receives the display request transmitted from the terminal device 600 via the communication unit 403.

[0096] In step S17, the communication control unit 411 of the server device 400 transmits screen data for displaying the monitoring screen 700 created in step S15 to the monitor's terminal device 600 via the communication unit 403. The communication control unit 611 of the terminal device 600 receives the screen data transmitted from the server device 400 via the communication unit 603.

[0097] In step S18, the display control unit 612 of the terminal device 600 displays the monitoring screen 700 on the display unit 605 based on the screen data received in step S17. For example, the display control unit 612 displays the monitoring screen 700 shown in Figure 7 or Figure 8 on the display unit 605. In this monitoring screen 700, the attention question "How active is the person?" is displayed preferentially, making it easier for the caregiver to select this attention question. If the caregiver changes the target date on the monitoring screen 700, the processing from step S12 onwards described above will be repeated for the changed target date.

[0098] In step S19, the display control unit 612 of the terminal device 600 accepts the caregiver's selection of a question on the monitoring screen 700 displayed in step S18. For example, if the caregiver wants to know the activity level of the person being monitored, they use the operation unit 604 of the terminal device 600 to select the question "How active is the person?" displayed on the monitoring screen 700 shown in Figure 7 or Figure 8. The display control unit 612 accepts this selection of question in response to the caregiver's operation.

[0099] In step S20, the communication control unit 611 of the terminal device 600 transmits selection information indicating the question selected in step S19 to the server device 400 via the communication unit 603. The communication control unit 411 of the server device 400 receives the selection information transmitted from the terminal device 600 via the communication unit 403.

[0100] In step S21, the response generation unit 416 of the server device 400 generates an answer to the question indicated by the selection information received in step S20. For example, the response generation unit 416 generates a prompt that includes the question "How active was the person?" selected by the caregiver and the judgment result from step S12, and inputs the generated prompt into the large-scale language model. For example, if the target period is March 19th, and the judgment result from step S12 includes information such as the person being cared for having slightly decreased activity on March 19th, increased time spent in the living room, and no outings, the large-scale language model generates an answer based on the input prompt, such as "The person's activity level on March 19th appears to have been lower than usual. They did not appear to have gone out that day and spent a long time in the living room. This may have been influenced by their physical condition or the weather."

[0101] In step S22, the communication control unit 411 of the server device 400 transmits response information indicating the response generated in step S21 to the monitor's terminal device 600 via the communication unit 403. The communication control unit 611 of the terminal device 600 receives the response information transmitted from the server device 400 via the communication unit 603.

[0102] In step S23, the display control unit 612 of the terminal device 600 displays the response indicated by the response information received in step S22 on the monitoring screen 700. For example, as shown in Figure 6, the display control unit 612 displays the response, "The activity level on March 19th appears to have been lower than usual. It seems that the person did not go out that day and spent a lot of time in the living room. This may have been due to their physical condition or the weather," in the response display area 720 of the monitoring screen 700. By looking at this response, the caregiver can learn about the changes in the living situation of the person being cared for in detail without having to confirm with the person being cared for.

[0103] According to this embodiment, the caregiver can obtain answers to questions regarding the living situation of the person being cared for by selecting questions on the monitoring screen 700. This allows the caregiver to easily obtain information they want to know about the living situation of the person being cared for without having to confirm it with the person being cared for.

[0104] Furthermore, in the monitoring screen 700, attention questions related to monitoring elements where there has been a change in the living situation of the person being monitored are displayed preferentially, making it easier for the caregiver to select attention questions. This makes it easier for the caregiver to notice changes in the living situation of the person being monitored.

[0105] Furthermore, by using generative AI to generate responses, the nuances of the answers will differ each time, allowing for the generation of natural-sounding responses similar to those produced by humans. In addition, by using rule-based detection of changes in monitoring elements and inputting the results into a large-scale language model to generate responses, more appropriate responses can be produced.

[0106] (modified version) The present invention is not limited to the embodiments described above and can be implemented in various ways. The embodiments may be used individually, in combination with one or more modifications, or switched between as needed during implementation.

[0107] (Variation 1) In the above embodiment, an example was described in which the determination unit 413 of the server device 400 determines changes in monitoring elements based on rules, and the response generation unit 416 generates a response using a generation AI. However, the method for determining changes in monitoring elements and the method for generating responses are not limited to the example described in the embodiment.

[0108] For example, the determination unit 413 may use a generative AI to determine changes in the monitoring elements. In this case, the determination unit 413 generates a prompt that includes multiple monitoring data and basic information such as the installation location of the monitoring sensor 100, and inputs the generated prompt to a large-scale language model. This prompt instructs the model to determine changes in the monitoring elements based on the multiple monitoring data and basic information, and to output the determination result. The large-scale language model outputs the determination result of changes in the monitoring elements based on the input prompt. According to this modified example, by using a generative AI to determine changes in the monitoring elements, more accurate determination can be achieved.

[0109] Alternatively, the response generation unit 416 may generate responses based on rules. In this case, conditions for generating responses from changes in monitoring elements are predetermined and stored in the memory unit 402. The response generation unit 416 generates responses from the determination results of the determination unit 413 according to the conditions stored in the memory unit 402. With this modified version, since responses are generated based on rules, it is possible to generate responses that are highly predictable.

[0110] The determination unit 413 and the response generation unit 416 may either use the generation AI for processing, or both may use the generation AI for processing. Furthermore, the determination unit 413 and the response generation unit 416 may either use a rule-based approach for processing, or both may use a rule-based approach for processing.

[0111] Furthermore, in the above embodiment, the decision unit 414 of the server device 400 determined general questions and attention questions from the question sentences stored in the question table 421, but general questions and attention questions may be generated using a generation AI. In this case, the decision unit 414 generates a prompt that includes multiple monitoring data or the judgment result of the judgment unit 413 and basic information such as the installation location of the monitoring sensor 100, and inputs the generated prompt to a large-scale language model. This prompt instructs, for example, to generate general questions and attention questions based on multiple monitoring data or the judgment result of the judgment unit 413 and the basic information. The large-scale language model generates general questions and attention questions based on the input prompt. According to this modified example, by using a generation AI to generate questions, the questions will have different nuances each time, so natural questions similar to those generated by a human can be presented to the monitor.

[0112] Furthermore, in the above embodiment, the screen creation unit 415 may use a generation AI to prioritize the display of attention questions. In this case, the screen creation unit 415 generates a prompt that includes general questions and attention questions, and inputs the generated prompt to a large-scale language model. This prompt, for example, generates a list of questions based on general questions and attention questions, and instructs the model to prioritize the display of attention questions. The large-scale language model generates a list of questions that includes general questions and attention questions based on the input prompt. At this time, attention questions are processed so that they are prioritized for display by determining their placement or highlighting them. The screen creation unit 415 displays the list of questions generated by the large-scale language model in the question presentation area 710 of the monitoring screen 700. According to this modified example, by using a generation AI for priority display, more effective priority display can be achieved.

[0113] (Modification 2) In the above embodiment, an example was described in which the determination unit 413 of the server device 400 determines a change in a monitoring element based on the detection result of a single monitoring sensor 100. However, the determination unit 413 may determine a change in a monitoring element based on detection information from multiple monitoring sensors 100. For example, the determination unit 413 may determine a change in the activity level of the person being monitored by combining the activity level and posture ratio of the person being monitored based on the detection information from the human presence sensor and the detection information from the millimeter-wave sensor. In this case, the question table 421 may store a question sentence such as "What is the activity level and posture ratio?" associated with the monitoring element "activity level". According to this modification, by determining a change in a monitoring element based on detection information from multiple monitoring sensors 100, more accurate determination can be made.

[0114] (Variation 3) In the above embodiment, an example was described in which a single attention question is prioritized for display on the monitoring screen 700, but the embodiment is not limited to this. For example, if the determination unit 413 determines that there is a change in multiple monitoring elements, multiple attention questions related to each of these monitoring elements may be prioritized for display according to their respective priorities.

[0115] Specifically, when the determination unit 413 determines that there has been a change in multiple monitoring elements, the screen creation unit 415 sets priorities for multiple attention questions related to these monitoring elements. Priorities may be set in advance for each monitoring element. Here, it is assumed that the priority of activity level and sleep efficiency is set to "high," the priority of bedtime and wake-up time is set to "medium," and the priority of posture ratio is set to "low." In this case, the screen creation unit 415 sets the priority of attention questions related to activity level and sleep efficiency to "high," the priority of attention questions related to bedtime and wake-up time to "medium," and the priority of attention questions related to posture ratio to "low."

[0116] As another example, priority may be set according to the degree of change in the monitoring element. For example, the screen creation unit 415 sets the priority of focus questions related to monitoring elements with a large degree of change to "high," the priority of focus questions related to monitoring elements with a moderate degree of change to "medium," and the priority of focus questions related to monitoring elements with a small degree of change to "low."

[0117] As yet another example, priority may be set according to whether the change in the monitoring element is good or bad. For example, the screen creation unit 415 sets the priority of the attention question related to the monitoring element to "high" if the change in the monitoring element is bad, and sets the priority of the attention question related to the monitoring element to "low" if the change in the monitoring element is good.

[0118] The screen creation unit 415 then prioritizes displaying multiple attention questions so that those with higher priority are more likely to be selected. At this time, the screen creation unit 415 may prioritize displaying only the attention question with the highest priority, or it may prioritize displaying multiple attention questions in order of priority.

[0119] For example, the screen creation unit 415 displays high-priority questions in red, medium-priority questions in light red, and low-priority questions in gray. This highlights higher-priority questions, making them easier for caregivers to select.

[0120] According to this modified version, when there are changes in multiple monitoring elements, prioritizing the display of multiple attention questions related to these monitoring elements according to their priority makes it easier for the caregiver to select the most important attention questions.

[0121] (Modification 4) In the above embodiment, an example was described in which the monitoring screen 700 prioritizes the display of attention questions related to monitoring elements that have changed, while other general questions are displayed as is. However, the screen creation unit 415 may, in addition to prioritizing the display of attention questions related to monitoring elements that have changed, also restrict the display of other general questions. For example, the screen creation unit 415 may display only attention questions related to monitoring elements that have changed and hide other general questions. According to this modification, since only attention questions related to monitoring elements that have changed are displayed, it becomes easier for the caregiver to select attention questions.

[0122] (Variation 5) In the above embodiment, the caregiver selected a desired question from the options displayed on the monitoring screen 700, but the method by which the caregiver specifies a desired question is not limited to this. For example, the caregiver may input the question as text data into the monitoring screen 700 by operating the operation unit 604. In this case, the answer generation unit 416 of the server device 400 generates an answer to the input question in response to the caregiver's operation. This modification improves the freedom of the caregiver's questions. The caregiver may also input questions by voice. In this case, the caregiver's terminal device 600 converts the caregiver's voice into text data using, for example, voice recognition technology and transmits it to the server device 400. This makes it easier for the caregiver to input questions.

[0123] (Experimental variation 6) In the above embodiment, the multiple monitoring elements included activity level, posture ratio, time spent, bedtime / wake-up time, going out / returning home, visitors, sleep efficiency, whether or not meals were eaten, and room temperature. However, these monitoring elements are illustrative and not limited to them. For example, the multiple monitoring elements may include only some of the activity level, posture ratio, time spent, bedtime / wake-up time, going out / returning home, visitors, sleep efficiency, whether or not meals were eaten, and room temperature, or other monitoring elements may be included.

[0124] (Example 7) In the above embodiment, multiple activity levels may be measured based on different detection information output from multiple sensors, and the average value of these activity levels may be used as the activity level data of the person being monitored. For example, the person being monitored's terminal device 500 may measure the activity level based on the detection information from its sensor, and the server device 400 may measure the activity level based on the detection information from the human presence sensor, and the server device 400 may use the average value of these activity levels as the activity level data. This makes it possible to grasp the activity level more accurately.

[0125] (Variation 8) In the above embodiment, the activity level may be measured based on METs (Metabolic equivalents). METs are an index that represents the ratio of energy consumption during physical activity to energy consumption at rest, with resting energy consumption set to 1. METs are defined for each type of activity. Specifically, the server device 400 or terminal device 500 may calculate METs from the type of activity obtained from the sensor detection information and use that value as activity level data. According to this modification, a more accurate understanding of the activity level can be obtained by measuring the activity level based on METs.

[0126] (Extreme variation 9) In the above embodiment, both the monitoring sensor 100 and the intrusion monitoring sensor 150 were installed, but the monitoring sensor 100 may be used for intrusion monitoring purposes instead of the intrusion monitoring sensor 150. Specifically, the monitoring sensor 100 may detect an intruder into the monitored area and output detection information. According to this modification, since the monitoring sensor 100 also performs the function of the intrusion monitoring sensor 150, there is no need to install the intrusion monitoring sensor 150 separately, and costs can be reduced accordingly.

[0127] (Variation 10) In the above embodiment, an example was given in which the server device 400 measures the activity level. However, instead of the server device 400, the monitoring device 200 may measure the activity level and transmit the measured activity level data to the server device 400. With this modification, the processing load on the server device 400 can be reduced by having the monitoring device 200 measure the activity level.

[0128] (Variation 11) The configuration of the monitoring system 10 described in the above embodiment is illustrative and not limited thereto. Each of the monitoring system 10, monitoring sensor 100, intrusion monitoring sensor 150, monitoring device 200, server device 400, terminal device 500, and terminal device 600 may have a configuration different from the configuration described in the above embodiment, or may be configured without some of the components.

[0129] (Example 12) The functional configuration of the monitoring system 10 described in the above embodiment is illustrative and not limited thereto. Each of the monitoring system 10, monitoring sensor 100, intrusion monitoring sensor 150, monitoring device 200, server device 400, terminal device 500, and terminal device 600 may have functions different from those described in the above embodiment, or may be configured without some of the functions. For example, the monitoring device 200 may have the functions of the server device 400. Also, the functions of one device may be distributed among multiple devices, or the functions of multiple devices may be combined into one device. For example, in the above embodiment, the monitoring device 200 had both the function of monitoring intrusion into the monitored area and the function of monitoring the person being monitored, but instead of the monitoring device 200, a first monitoring device having the function of monitoring intrusion into the monitored area and a second monitoring device having the function of monitoring the person being monitored may be used.

[0130] (Example 13) The operation of the monitoring system 10 described in the above embodiment is illustrative and not limited thereto. The processing procedures of the monitoring system 10, monitoring sensor 100, intrusion monitoring sensor 150, monitoring device 200, server device 400, terminal device 500, and terminal device 600 may be rearranged or some processing procedures may be omitted, as long as they are consistent.

[0131] (Variation 14) Another aspect of the present invention may provide a method having steps of processing performed in at least one of the following: a monitoring system 10, a monitoring sensor 100, an intrusion monitoring sensor 150, a monitoring device 200, a server device 400, a terminal device 500, and a terminal device 600. Yet another aspect of the present invention may provide a program to be executed in the monitoring sensor 100, the intrusion monitoring sensor 150, the monitoring device 200, the server device 400, the terminal device 500, or the terminal device 600. This program may be provided stored on a computer-readable recording medium or provided by download via the Internet or the like.

[0132] A monitoring system according to one embodiment of the present invention can contribute to solving social issues such as extending healthy life expectancy, improving quality of life (QOL), and the increase in elderly people living alone due to the super-aging society. Furthermore, a monitoring system according to one embodiment of the present invention can contribute to achieving Goal 3 of the Sustainable Development Goals (SDGs) adopted by the United Nations, "Ensure healthy lives and promote well-being for all." [Explanation of symbols]

[0133] 10: Monitoring system, 100: Monitoring sensor, 150: Intrusion monitoring sensor, 200: Monitoring device, 201: Control unit, 202: Memory unit, 203: Communication unit, 204: Operation unit, 205: Display unit, 206: Sounding unit, 211: Acquisition unit, 212: Setting unit, 213: Transmission unit, 400: Server device, 401: Control unit, 402: Memory unit, 403: Communication unit, 411: Communication control unit, 412: Acquisition unit, 413: Judgment unit, 414: Decision unit, 415: Screen creation unit, 416: Answer generation unit, 500: Terminal device, 600: Terminal device, 601: Control unit, 602: Memory unit, 603: Communication unit, 604: Operation unit, 605: Display unit, 611: Communication control unit, 612: Display control unit

Claims

1. An acquisition unit that acquires monitoring data indicating monitoring elements to understand the living situation of the person being monitored, A display unit that displays questions related to the aforementioned monitoring element, A reception desk that receives selections of the aforementioned questions from the caregiver who is watching over the person being cared for, A response generation unit generates answers to the selected questions based on the monitoring data, A monitoring system equipped with these features.

2. The system further includes a determination unit that determines changes in the monitoring element over time, The display unit displays the relevant questions related to the monitoring element that has been determined to have changed in a way that makes it easier for the user to select them. The monitoring system according to claim 1.

3. When the display unit displays the featured question and other questions side by side, it highlights the featured question or determines the arrangement of the featured question and other questions. The monitoring system according to claim 2.

4. When the display unit determines that there is a change in any of the monitoring elements, it displays the multiple attention questions corresponding to each of the monitoring elements in a way that makes it more likely for the attention questions with higher priority to be selected. The monitoring system according to claim 2.

5. The response generation unit generates the response using a large-scale language model, The aforementioned large-scale language model is input to the aforementioned monitoring data or the features of the aforementioned monitoring data. A monitoring system according to any one of claims 1 to 4.

6. The system further includes a determination unit that determines the temporal changes for each of the monitoring elements based on a comparison between monitoring data showing the monitoring elements of the person being monitored during the target period and monitoring data showing the monitoring elements of the person being monitored in the past period prior to the target period. The display unit displays the relevant questions related to the monitoring element that has been determined to have changed in a way that makes it easier for the user to select them. The monitoring system according to claim 5.

7. The large-scale language model receives the information regarding the changes that have been determined as features of the monitoring data. The large-scale language model uses the input information regarding the changes to generate the response that explains the changes in the monitoring element. The monitoring system according to claim 6.

8. On the computer, The steps include acquiring monitoring data that shows monitoring elements to understand the living situation of the person being monitored, and The steps include: receiving information from a terminal device that displays questions related to the monitoring element, indicating the question selected by the monitor who is monitoring the person being monitored; The steps include generating an answer to the question indicated by the received information based on the monitoring data, A program to execute.

Images